Engineering Open-Source Medical Devices: A Reliable Approach for Safe, Sustainable and Accessible Healthcare 3030793621, 9783030793623

Citation preview

Arti Ahluwalia Carmelo De Maria Andrés Díaz Lantada   Editors

Engineering Open-Source Medical Devices A Reliable Approach for Safe, Sustainable and Accessible Healthcare

Engineering Open-Source Medical Devices

Arti Ahluwalia • Carmelo De Maria Andrés Díaz Lantada Editors

Engineering Open-Source Medical Devices A Reliable Approach for Safe, Sustainable and Accessible Healthcare

Editors Arti Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering University of Pisa Pisa, Italy

Carmelo De Maria Research Center “E. Piaggio” and Department of Information Engineering University of Pisa Pisa, Italy

Andrés Díaz Lantada Mechanical Engineering Department Universidad Politecnica de Madrid Madrid, Spain

ISBN 978-3-030-79362-3 ISBN 978-3-030-79363-0 https://doi.org/10.1007/978-3-030-79363-0

(eBook)

© Springer Nature Switzerland AG 2022 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Preface

This handbook, Engineering Open-Source Medical Devices: A Reliable Approach for Safe, Sustainable and Accessible Healthcare, is the result of an absorbing journey and has been prepared thanks to the contribution of a great team of inspiring engineering professionals. The journey started with editors’ and co-authors’ passion for biomedical engineering and medical devices, with the conception and implementation of several project-based learning courses focused on medical technology, principally at the University of Pisa and at Universidad Politécnica de Madrid. Many of the ideas and concepts presented in this handbook were nurtured by a unique set of innovative summer schools organized by the African Biomedical Engineering Consortium (ABEC) and supported by the United Nations Economic Commission for Africa (UNECA), in which the concept of “open-source medical devices” (OSMDs) was coined. Subsequently, the “UBORA: Euro-African Open Biomedical Engineering e-Platform for Innovation through Education” project, funded by the European Commission’s Horizon 2020 Programme (grant agreement n 731053, 2017–2019), enabled the creation of a unique collaborative e-infrastructure for open-source medical technologies (the UBORA platform: https://platform.uborabiomedical.org/). The platform has helped set the foundations for the systematic co-design engineering of OSMDs, underpinned by rigorous attention to the safety and efficacy of medical technologies. Beyond those who have directly contributed to the different chapters of the handbook, there is an impressive community of researchers, educators and students (the UBORA Community), who are now transforming the biomedical industry in different countries with a focus on healthcare equity. In the last 5 years, more than 1500 students and colleagues from around 40 countries have taken part in UBORAUNECA actions linked to the promotion of OSMDs as transformative technologies, and these experiences have been the fruitful soil for growing this text. For us, as editors of the handbook, it has been a privilege to distil the key good practices and challenges involved in the engineering of OSMDs in this book, which we hope becomes a comprehensive reference for colleagues in the field. v

vi

Preface

We would like to thank the editorial staff at Springer for their support with the handbook. Finally, we express our deepest gratitude to our families, friends and colleagues for their understanding, patience and endless support. Pisa, Italy Madrid, Spain

Arti Ahluwalia Carmelo De Maria Andrés Díaz Lantada

Contents

1

2

3

4

Open-Source Medical Devices: Concept, Trends, and Challenges Toward Equitable Healthcare Technology . . . . . . . . . . . . . . . . . . . Carmelo De Maria, Andrés Díaz Lantada, Licia Di Pietro, Alice Ravizza, and Arti Ahluwalia Towards a Harmonized Methodology for the Development of Safe and Regulation Compliant Open-Source Medical Devices . . . . . . . . Carmelo De Maria, Andrés Díaz Lantada, Licia Di Pietro, Alice Ravizza, and Arti Ahluwalia Getting Started with an Open-Source Medical Device Project: Systematic Needs Identification Techniques for Bottom-Up Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Philippa Ngaju Makobore, Andrés Díaz Lantada, Licia Di Pietro, Carmelo De Maria, and Arti Ahluwalia Design of Open-Source Medical Devices for Improved Usability and Risk Minimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Alice Ravizza, Noemi Stuppia, Federico Sternini, Luis Ignacio Ballesteros Sánchez, Rocío Rodríguez-Rivero, Enrique Chacón Tanarro, Juan Manuel Munoz-Guijosa, and Andrés Díaz Lantada

1

21

39

59

5

Human Centered Design Principles for Open-Source Medical Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Elizabeth Johansen, Mark Fisher, Andrés Díaz Lantada, Carmelo De Maria, and Arti Ahluwalia

6

Certification Pathways for Open-Source Medical Devices . . . . . . . . 127 Licia Di Pietro, Carmelo De Maria, Andrés Díaz Lantada, Alice Ravizza, and Arti Ahluwalia

vii

viii

Contents

7

Legislation for Open-Source Medical Devices: Current Scenario, Risks and Possibilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Maria Elena Lippi, Filippo Morello, Licia Di Pietro, Carmelo De Maria, and Valentina Calderai

8

Creativity Promotion in Open-Source Projects: Application to Open-Source Medical Devices and Healthcare Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Andrés Díaz Lantada and Juan Manuel Munoz-Guijosa

9

Methods and Technologies for the Personalized Design of Open-Source Medical Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Andrés Díaz Lantada, William Solórzano, Adrián Martínez Cendrero, Rodrigo Zapata Martínez, Carlos Ojeda, and Juan Manuel Munoz-Guijosa

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and Regulations of Medical Technologies . . . . . . . . . . . . 219 Licia Di Pietro, Gabriele Maria Fortunato, Ermes Botte, Arti Ahluwalia, and Carmelo De Maria

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through Innovative Open-Source Medical Devices . . . . . . . . . . . . . 243 Andrés Díaz Lantada, Rocío Rodríguez-Rivero, Ana Moreno Romero, Rafael Borge García, Luis Ignacio Ballesteros Sánchez, Licia Di Pietro, Carmelo De Maria, and Arti Ahluwalia

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Chapter 1

Open-Source Medical Devices: Concept, Trends, and Challenges Toward Equitable Healthcare Technology Carmelo De Maria, Andrés Díaz Lantada, Licia Di Pietro, Alice Ravizza, and Arti Ahluwalia

1.1

Introduction: The Social Product Development and the Modern Medical Technologies

Medical technology has transformed the practice of medicine and patient care, with a wide set of relevant breakthroughs achieved during the last decades high-precision medical imaging for improved diagnoses (European Society of Radiology, 2015), robotic-guided surgery and minimally invasive procedures for enhanced recovery after surgery (Ghezzi & Corleta, 2016), the progressive use of smartphones for diagnosing and monitoring patients (Freeman et al., 2020; Jamshidnezhad et al., 2019), and even 3D printing with biomaterials and biofabrication, as most innovative potential alternatives to conventional prostheses or organ transplants (Lanza et al., 2014; Atala & Joo, 2015; Moroni et al, 2018). However, in many cases, medical technology is developed in secrecy, and patients’ or medical professionals’ needs are considered as a minor part of the decision-making process, which is currently under the pressure of marketing and immediate payback, instead of being driven by needs and by the knowledge and long-sighted view of technology developers (Fasterholdt et al., 2018).

C. De Maria (*) · L. Di Pietro · A. Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy e-mail: [email protected] A. Díaz Lantada Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain A. Ravizza USE-ME-D srl, I3P Politecnico di Torino, Turin, Italy © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_1

1

2

C. De Maria et al.

While the economic growth of medical technology developers and manufacturers is fundamental for reaching more and more patients in a sustainable way, decisions taken on the basis of short-term incomes tend to limit the creativity of medical device designers, hindering also the personalization of medical technology, and to leave rare pathologies and low-resource settings unattended, to cite just some drawbacks of the current state-of-the-art in the biomedical industry (De Maria et al., 2018). In contrast with the biomedical industry, many product fields are now involving stakeholders and future users since the beginning of the product development process, embracing the new paradigm of “open innovation” (Ng & Jee, 2014; Gao and Bernard, 2017). In this new paradigm, the online sharing of information (concepts, blueprints, or other project documentation), with colleagues or even with developers outside the core team, is reinventing the product development process. In this sort of “social product development,” thematic communities co-develop and share innovative solutions working on online platforms, such as Thingiverse or GrabCAD. These collaborative and open-source design strategies have been widely explored in software development, bringing benefits in terms of accessibility, sustainability, lower costs, improved performance, and even safety (Lessig et al., 2005). Nowadays, the capillary diffusion of entry-level 3D printers, available in co-working spaces and FabLabs born with the “makers” movement (Gershenfeld, 2005), as well as the lower access cost to the “printing factories” have given the tools to physically build the projects, downloadable from online repositories. However, healthcare industry is still reluctant to taking advantage of the enormous potentials of open-source and collaborative approach toward a social development of medical devices, although it has the potential to increase the access to medical technologies (De Maria et al., 2020). In the medical industry, in fact, it is crucial to ensure the safety and efficacy requirements of medical technology, enforced by laws such as the European Medical Device Regulation 2017/775 and 2017/746. Indeed, despite several examples of healthcare technologies have appeared on the web (Niezen et al., 2016), only some of them have been designed to be compliant with medical device (MD) legislation (Arcarisi et al., 2019; Ferretti et al., 2017). In our perspective, to prove truly transformative, open-source medical devices and their boundaries should be adequately defined, the expected outcomes of opensource medical technologies should be analyzed and the characteristics of pioneering success cases should be understood, so as to follow their path. In addition, collaborative research and development online environments, capable of enabling collaboration, helping to match medical needs and technological offers and devoted to guiding medical technology developers in their endeavors, should be arranged. The UBORA e-infrastructure developed by our team and described also in this introductory chapter constitutes a relevant breakthrough in this direction. The following sections of this chapter deal with all aforementioned issues.

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

1.2

3

The Concept of Open-Source Medical Device (OSMD): Definition and Rationale

1.2.1

Reaching a Consensus Definition for OSMDs

Collaboration and information sharing are becoming fundamental in the biomedical field for developing technologies aimed at solving global health concerns. During the First International Conference on Collaborative Biomedical Engineering for Open-Source Medical Technologies (Pisa, September 2018), and in accordance with the principles of “The Kahawa Declaration” (Ahluwalia et al., 2018), an international focus group on open-source medical devices was established for working toward equitable access to healthcare technologies, by means of opensource approaches to the design of medical devices, and for helping to harmonize and articulate best practices in such emergent field. The first tasks assigned to the mentioned working group, involving all UBORA partners and key stakeholders from the medical industry with experience in open innovation, as well as policymakers, educators, and healthcare professionals, included (a) the gathering of successful examples of open-source medical devices and open-source initiative in BME to understand the state-of-the-art and its current limitations and (b) the elaboration of a consensus operative definition for the concept of open-source medical device. Relevant concepts were used for the elaboration of such definition, which tries to take into consideration several aspects present in the more common definitions of open-source software (Open-Source Initiative; Debian Project) and open-source hardware (Open-Source Hardware Association). However, in a way, the definition explained further on combines and expands both, in order to account for very specific and relevant issues present in medical technology development, for the fact that modern medical devices involve hardware and software, and for adequately incorporating recent trends in data management in collaborative projects (Wilkinson et al., 2016). To start with, let’s consider the concept of medical device:

1.2.1.1

Medical Device

According to the EU Regulation 2017/745 of the EU Parliament and of the Council on April 5, 2017, on medical devices, “medical device” means: Any instrument, apparatus, appliance, software, implant, reagent, material or other article intended by the manufacturer to be used, alone or in combination, for human beings for one or more of the following specific medical purposes: a) diagnosis, prevention, monitoring, prediction, prognosis, treatment or alleviation of disease, b) diagnosis, monitoring, treatment, alleviation of, or compensation for, an injury or disability, c) investigation, replacement or modification of the anatomy or of a physiological or pathological process or state, d) providing information by means of in vitro examination of specimens derived from the human body, including organ, blood and tissue donations, and which does not achieve its

4

C. De Maria et al. principal intended action by pharmacological, immunological or metabolic means, in or on the human body, but which may be assisted in its function by such means.

Consequently, medical devices range from contact lenses to Band-Aids, from pacemakers to implantable heart valves, and from surgical instruments to large medical imaging equipment, and the definition of open-source medical devices proposed in current document is direct consequence of adapting the open-source concept, and its implications on the software and hardware industries, to the previously cited definition of medical devices from the MDR 2017/745 of the EU. For operative purposes, working this EU definition covers also most medical devices worldwide, as defined by other regulatory bodies, such as the US Food and Drug Administration or the Chinese Food and Drug Administration, to cite just a couple of relevant examples.

1.2.1.2

Open-Source Software

The most commonly used definition of open-source software (currently the opensource definition v.1.9: https://opensource.org/docs/definition.php) derives from the Debian Free Software Guidelines created by Bruce Perens and the Debian developers. In short, open-source software is software with accessible source code, hence allowing peer-review and rapid evolution, complying also with criteria such as free distribution, distribution in source and compiled code, allowance of modifications and derived works, lack of discrimination against specific persons or groups, lack of discrimination against fields of use, lack of restriction to other software, lack of product specificity, and technological neutrality. Examples of open-source software licenses include “GPL,” “BSD,” and “Artistic,” among others.

1.2.1.3

Open-Source Hardware

The Open-Source Hardware Association (OSHWA) defines open-source hardware in its Statement of Principles 1.0 and Definition 1.0 as: “hardware whose design is made publicly available so that anyone can study, modify, distribute, make, and sell the design or hardware based on that design.” The hardware’s source, the design from which it is made, is available in the preferred format for making modifications to it. Ideally, open-source hardware uses readily available components and materials, standard processes, open infrastructure, unrestricted content, and open-source design tools to maximize the ability of individuals to make and use hardware. Open-source hardware gives people the freedom to control their technology while sharing knowledge and encouraging commerce through the open exchange of designs.” The definition is based on the previously mentioned open-source definition for open-source software.

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

5

However, as hardware differs from software by requiring the use of physical resources for the creation of physical goods, it is important to highlight that these Principles and Definition of OSHW also state that “persons or companies producing items (“products”) under an OSHW license have an obligation to make it clear that such products are not manufactured, sold, warranted, or otherwise sanctioned by the original designer and also not to make use of any trademarks owned by the original designer.”

1.2.1.4

Open-Source Medical Device

Working on the basis of previous concepts and definitions and considering the current state-of-the-art in the field of collaborative biomedical engineering, as well as ongoing international initiatives pursuing equitable access to healthcare technology, the UBORA consortium, in connection with the international focus group on OSMDs, proposed an operative definition for open-source medical devices as follows: An open-source medical device is a medical device whose design and product development information are made publicly available so that anyone can study, modify, distribute, make, and sell the medical devices, and their related software or hardware, based on the initial available design and information. The design of the open-source medical device should be shared in a format conceived for enabling validation, verification and modification. Opensource medical devices rely on widely available materials and components, benefit from being designed according to international safety standards and processes aimed at guaranteeing patients’ safety, take advantage of modularity, even being designed as inter-changeable and inter-operable kits, and rely on open e-infrastructures for information dissemination and promotion of collaboration. FAIR (findable, accessible, interoperable, reusable) data principles are proposed for open-source medical devices. Persons or companies producing and commercializing open-source medical devices are obliged to attribute to the original designers and to make clear that such medical devices are not manufactured, sold, warranted, or otherwise sanctioned by the original designer.

The definition has been officially submitted to the World Health Organization ICD-11 to be used as new concept for the biomedical industry and has been also presented in the new Handbook of Clinical Engineering (De Maria et al., 2020), in connection with the innovative paradigm of open-source biomedical engineering and with co-creation environments. In the opinion of the authors, the definition constitutes a relevant step in the harmonization of open initiatives, technologies, and resources that are emerging in connection with biomedical engineering and with the future of biomedical industry. The process for the elaboration of the definition has been supported by interesting debates that have also influenced decisions taken by the consortium responsible for arranging the UBORA community regarding the final guidelines for open licensing of designs through UBORA, the use of FAIR data principles, UBORA’s alignment with the “free as in freedom” concept, and the dissemination and communication strategies for UBORA devices, which have benefited from a clearer explanation of UBORA’s mission and of the meaning and potential impacts of OSMDs.

6

1.2.2

C. De Maria et al.

Rationale: The Reasons Behind Open-Source Medical Devices

Medical technologies are at the foundation of an efficient healthcare system. Despite “Good Health and Well-Being” as one of the Sustainable Development Goals (SDG), identified by the United Nations (United Nation 2015), the high costs of medical devices (MDs) can create a barrier for reaching this target. This cost derives from the long life-cycle of MDs (specification and planning, design, prototyping, manufacturing, certification, labeling and packaging, provision, installation, operation, maintenance, repair and disposal), in which each step is strictly regulated and controlled, to guarantee their efficacy and the safety of patients, healthcare providers, and bystanders. Even removing the charge on a single step could not make the difference. For example, the World Health Organization (WHO) estimates that in low-income countries, more than 80% of medical equipment is donated, but only 10–30% of these become operational, given the high operational costs, the lack of personnel and the frequent failures due to harsh environment, extreme climate conditions, humidity, dust, power instability, and lack of maintenance (WHO, 2010a, b; Malkin, 2007a, b). These conditions, not foreseen in the design phase, cause more frequent failures and determine a higher request for spare parts, which are expensive and difficult to find, making maintenance and repairing as problematic as the acquisition itself (Douglas, 2011). Compared to traditional medical device engineering methods, the social product development process based on the open-source and collaborative approach can be a possible alternative, both technically and economically viable. Open-source means making the design, documentation, source-code, blueprints, debated ideas, and results available for the general public. Having the software, electronic, and hardware design accessible under an open-source license and in the most suitable file format to study, modify, improve, and contribute to the design potentially leads to very rapid and more reliable innovation. In addition, the application of the opensource approach to medical device design has proven to offer a unique combination of advantages, such as increased safety, security, and reliability and reduced costs (De Maria et al., 2018). Until a few years ago, the development of medical devices was essentially linked to companies and large research institutions, but recently several examples of OSMDs have appeared on the web, in connection with the advent of the maker movement, as reviewed in the following section.

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

1.3

7

Brief Overview of Pioneering Success Cases in the OSMD Field

Recent pioneering cases of success in the OSMD field already reaching the market can be cited, including open-source electronic kits for medical signals, such as the solutions by Bitalino (Alves et al., 2006) and ProtoCentral Electronics (Whitchurch, 2019), open-source ECG systems (Gamma Cardio Soft, 2019), and varied opensource software to support diagnostic processes, in medical fields ranging from neurology and cardiology to dermatology and preventive medicine, in some cases benefiting from advances in smartphones as support to medical diagnoses (MIT’s Sana, 2020; Vlassi et al., 2017; Kassianos et al., 2015), to cite just a few. In connection with the maker movement, other inspiring pioneers have devoted themselves to explaining how to arrange DIY labs for prototyping (Pearce, 2014a) and detailed in an open-source way the development of varied solutions, such as varied laboratory equipment, adaptive aids for arthritis patients, or printable clubfoot bracer for children (Pearce, 2014b; Gallup et al., 2018; Savonen et al. 2019), while highlighting the potentials of distributed manufacturing to make medical devices reach those who need them most and analyzing the suitable business models for open hardware (Pearce 2017). Other inspiring initiatives, trying to promote OSMDs and sharing of information for improved medical technology and healthcare, can also be mentioned due to their remarkable growth and international projection, such as the “Enabling the Future” project and the “Autofabricantes” community, focused on personalized prostheses designs for children; the “Patient Innovation” and “Patients Like Me” networks, focused on shared information for solving complex pathologies; and the “Open Prosthetics” initiative and the “Open Bionics” environment, both concentrated on low-cost personalized prostheses, among others (Enable, 2019; Oliveria et al., 2019; Open Prosthetics, 2019; Open Bionics, 2019), some of them with more than a decade of dedication to the field of OSMDs. In a way, these initiatives evolve from analogous networks devoted to co-creation and to the promotion of collective intelligence in more conventional hardware and software development, which have already reshaped how common appliances for daily use are designed, manufactured, and continuously improved through cooperation. Among these well-established networks and environments, it is important to cite Thingiverse and GrabCAD for the sharing of computer-aided design files, GitHub, MyMiniFactory, and YouImaging for sharing complete design projects, the RepRap wiki for detailing how to build DIY 3D printers, or the FabLab and Shapeways networks for delocalized manufacturing of components. Describing in detail all the current initiatives in the field of OSMDs and the whole collections of OSMDs already available and shared online is beyond the purpose of present study. However, we have summarized a selected collection of initiatives, open-source hardware and software, ongoing communities working in the OSMD arena, and remarkable cases of success of OSMDs, in many cases also collaboratively developed. These initiatives, networks, communities, and solutions are listed

8

C. De Maria et al.

in Tables 1.1, 1.2, and 1.3 to inform researchers, developers, patients, and healthcare professionals interested in these novel approaches to medical technology development. First, Table 1.1 presents more than 30 selected examples of open-source medical devices of recent development with their purpose, area of application, and link or reference for details. Then, Table 1.2 lists down open-source hardware and software resources with potential application for the development of OSMDs. Finally, Table 1.3 presents online communities of developers and online research infrastructures for the co-creation of OSMDs. In accordance with all of the above, the OSMD field is already being explored worldwide, and several concepts and devices have been designed, manufactured, and tested. However, only some of them have been designed to be compliant with medical device legislation. Again we would like to highlight that it is crucial to ensure the safety and efficacy requirements of medical technology, and for this reason, the adoption of open resources must follow the standards and the current regulations (De Maria et al., 2015, 2018). To this end, a new e-infrastructure, UBORA (“excellence” in Swahili), which merges the open-source concepts with the safety and efficacy requirements enforced by the EU Regulation on medical devices (MDR) 2017/745, has recently been established (UBORA, 2020). The main features of UBORA and selected examples of OSMDs developed through it are presented in the following sections of this perspective.

1.4

The UBORA e-Infrastructure: Motivation and Purpose

Motivated by the fact that none of the aforementioned environments and collaborative design platforms provides designers with tools for a guided and systematic development process, in which open-source and collaborative design strategies play the central role, together with final safety promotion through harmonized standardization employment, our team decided to set up the UBORA platform. It is the first of a kind focused on the co-creation of medical devices compliant with EU Regulation on medical devices (MDR) 2017/745 and following internationally recognized standards. This platform or online e-infrastructure is, in consequence, developed for the promotion of collaboration through the whole development process of innovative open-source medical devices, whose complete development details, including specifications, design process, lists of components, computer-aided design files, and blueprints, among other relevant issues, are shared by means of an interactive and designer-oriented “wiki” structure, as explained in the following section. The collaborative design environment of UBORA provides quite unique features oriented to guiding developers through a systematic engineering design process, focused on patient safety and on achieving designs of medical devices compliant with international regulations, while fostering collaboration and joint decision-making for enhanced creativity, shared information, and peer-reviewed designs. A very singular aspect of this e-infrastructure is that it covers all types of medical devices (i.e., diagnostic tools, prosthetic devices, surgical tools, monitoring systems, therapeutic

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

9

Table 1.1 Selected examples of open-source medical devices (last access to reference website on August 2019) Open-source medical devices (selected examples) Software for management of medical records Software for management in dentistry Software for psycho experiments Eye tracking resources for communication Eye tracking resources for diagnosis Game for supporting diagnosis of malaria Framework for app development Dermatological database to support diagnosis Software for supporting dermatologic studies Software for personalized prosthesis design Ultrasound stethoscope

Otoscope with disposable specula 3D printed stethoscope DIY pulse oximeter linked to Arduino Multi-purpose platform for biosensing Multi-purpose platform for biosensing Multi-purpose monitoring platform Brain computer interface Biosensing electrical brain activity (EEG) Biosensing muscle activity (EMG) Biosensing heart rate (ECG)

Technology Software

Medical area Management

Link/Reference https://www.open-emr.org/

Software

Management

http://www.opendental.com/

Software

Psychiatry

http://www.psychopy.org/

Software

Neurology

http://www.pygaze.org/

Software

Neurology

http://www.pygaze.org/

Software

Preventive medicine

http://malariaspot.org/es/

Software Software

Medical research Preventive medicine Dermatology

Software

Dermatology

Software

Orthopedics

Mobilebased technology Screening device Screening device Monitoring device Monitoring device

Internal medicine emergency medicine

https://www.apple.com/ researchkit/ http://tkderm.sourceforge. net/index.html https://github.com/SageBionetworks/MoleMapper https://github.com/mtumost/most-3-d-customizer http://www.echopen.org/

Monitoring device Monitoring device Monitoring device Monitoring device Monitoring device Monitoring device

Internal medicine emergency medicine Internal medicine emergency medicine Neurology

Otorhinolaryngology

https://github.com/GliaX

Internal medicine emergency medicine Internal medicine emergency medicine Internal medicine emergency medicine

https://github.com/GliaX https://github.com/GliaX https://wearablesforgood. com/finalist-totem-openhealth/ http://bitalino.com/en/ http://www.libelium.com/ http://openbci.com/

Neurology

http://openbci.com/

Sports medicine rehabilitation Cardiology

http://openbci.com/ http://openbci.com/ (continued)

10

C. De Maria et al.

Table 1.1 (continued) Open-source medical devices (selected examples) Device for performing nano-immunoassay

Medical area Diagnostic medicine

Link/Reference https://metafluidics.org/ devices/ http://enablingthefuture.org/

Supporting equipment Supporting equipment

Pediatrics rehabilitation Pediatrics rehabilitation Pediatrics rehabilitation Rehabilitation traumatology Rehabilitation traumatology Pediatrics rehabilitation Pediatrics rehabilitation Internal medicine emergency medicine

IoT patient monitor

Supporting equipment

Internal medicine emergency medicine

Defibrillator

Supporting equipment Self-monitoring device

Internal medicine emergency medicine Oncology

Wrist-powered hand prostheses Elbow-powered hand prostheses Finger prostheses

Technology In vitro diagnostic device Prosthesis Prosthesis Prosthesis

Ankle prostheses

Prosthesis

Hand and forearm prostheses Hand and forearm prostheses MIDI percussion instrument IoT ECG-patch

Prosthesis

Wearable device for breast self-examination

Prosthesis

http://enablingthefuture.org/ http://enablingthefuture.org/ https://niatech.org/ technology/ https://openbionics.com/ https://github.com/ Autofabricantes/ https://github.com/ Autofabricantes/ https://github.com/ Protocentral/protocentral_ heartypatch https://github.com/ Protocentral/protocentralhealthypi-v3 Ferreti et al. Hardware X, 2017 Arcarisi et al. Applied Sciences, 2019

devices) and bioengineering systems supporting medical practice (i.e., supporting lab equipment, mobile apps, and software). The UBORA platform has been also developed following the principles of “The Kahawa Declaration” (Ahluwalia et al., 2018), a call of attention for pursuing the democratization of medical technology signed in the closure of the First International Design School of the EU-funded UBORA project (Nairobi, December 2017). UBORA was officially launched to the public, before representatives from ABEM, ABEM, UNECA, and WHO, during the First International Conference on Collaborative Biomedical Engineering for Open-Source Medical Technologies and the successive UBORA Design School 2018, held in Pisa from September 1–7. To date, the e-infrastructure, available at the address: https://platform.uborabiomedical.org, has around 500 users and 300 projects, at different stage of development. In turn, UBORA aims to arrange a diverse and truly global community of

Multidisciplinary and medical Medical

Multidisciplinary

Open-source hardware and software Open-source hardware

Open-source software

Open-source software

Open-source hardware

Open-source software

Open-source software

Open-source hardware

Libelium (& MySignals) Bitalino (PLUX) OpenSCAD

Slic3r

MEDIKits

3D Slicer

Ultimaker Cura

ProtoCentral

Multidisciplinary

Multidisciplinary

Medical

Medical

Multidisciplinary

Multidisciplinary

Open-source hardware

Arduino

Sector Multidisciplinary Medical

Type of resource Open-source software Open-source software

Name FreeCAD OsiriX

India

Netherlands

International team

US

Italy

2016

2013

2012

2011

2011

2010

2007

Portugal Austria

2006

2005

Year of release 2002 2003

Spain

Italy

Country of development Germany Switzerland

Open-source and low-cost prototyping kits with biomedical sensors

Open-source software for generating G-code for manufacturing Open-source low-cost biomedical prototyping kits Open-source software for managing medical images and personalizing biodevices Open-source software for generating G-code for manufacturing

Open-source and low-cost electronic prototyping kits e-Health and medical IoT development tools Open-source and low-cost biomedical prototyping kits Open-source CAD software

Additional comments Open-source CAD software Open-source software for DICOM files

(continued)

https://ultimaker.com/ en/products/ultimakercura-software https://www. protocentral.com/

https://d-lab.mit.edu/pro jects/150/medikit https://www.slicer.org/

http://www.openscad. org/ http://slic3r.org/

http://www.libelium. com/ http://bitalino.com/en/

Link/reference https://freecadweb.org/ http://www.osirixviewer.com https://www.arduino.cc/

Table 1.2 Open-source hardware and software resources with potential applications for the development of OSMDs (last access to reference website on August 2019)

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . . 11

Bioverse

BodyParts3D

GrabCAD

NIH 3D printing exchange Thingiverse

Redmine

Name GitHub

Multidisciplinary

Repository for sharing files and blueprints Repository for sharing files and blueprints Repository for sharing files and blueprints Repository for sharing files and blueprints

Medical

Medical

Multidisciplinary

Medical

Multidisciplinary

Sector Multidisciplinary

Repository for sharing files and blueprints

Type of resource Open-source collaborative version control software Open-source project management software

Table 1.2 (continued)

Sweden and online community

Global community Global community Japan

US

International team

Country of development US

2018

2009

2009

2008

2014

2016

Year of release 2005

Redmine is a free and open-source, web-based project management and issue tracking tool A Public Resource for Bioscientific and Biomedical 3D Prints. Created by the US National Institutes of Health Open CAD library of all sorts of industrial components and devices Open CAD library of all sorts of industrial components and devices Open CAD library of body structures generated from medical images Open CAD library of biodevices for the biofabrication field

Additional comments Collaborative version control. GitHub and Bitbucket offer git services for free

http://lifesciencedb.jp/ bp3d/ http://bioverse.co/

https://www.thingiverse. com/ https://grabcad.com/

https://3dprint.nih.gov/

https://www.redmine. org/

Link/reference https://git-scm.com/

12 C. De Maria et al.

Developers of open-source medical devices

Enabling the Future

Glia Free Medical Hardware

Gaza Strip

Global community

Slovenia

US-founded and global focus

Africa-based

Developers of open-source medical devices

Developers of open-source medical devices Developers of open-source medical devices

US-founded and global focus

Type of resource Community for sharing medical cases and information Community for sharing medical cases and information Community for sharing medical cases and information

Developers of open-source hardware and software Developers of open-source medical devices

SymbioLab

Open Prosthetics

Praekelt

Name Patients Like Me Patient Innovation Open-Source Imaging Initiative OpenMRS

Country of development US-UK and global focus Portugal-founded and global focus Germany-founded with global focus

2012

2011

2017

2006

2006

2004

2016

2014

Year of found. 2011

Open-source hardware and projects for bioprinting Global community; open-innovation based on needs and proposals; sharing of complete projects; files and blueprints; focused on hand prostheses Developers of free (or extremely affordable) medical hardware with complete projects on stethoscopes, otoscopes, pulse oximeters, among others

Open-source medical record system platform, open-source solutions for improved communication, open-source solutions for big-data health Open-source ICT solutions for helping with medical practice and patients’ management Search and posting of solutions linked to opensource prostheses

Additional comments Search and posting of solutions for medical issues with supporting forum Search and posting of solutions for medical issues with supporting forum Community collaborating in the development of open-source solutions for medical imaging

(continued)

https://glia.org/

https://www. praekelt.org/ https:// openprosthetics. org/ http://irnas.eu/ symbiolab http:// enablingthefuture. org/

Link https://www. patientslikeme.com/ https://patient-inno vation.com/ http://www. opensourceimaging. org/news/ https://openmrs.org/

Table 1.3 Online communities of developers and online research infrastructures for the co-creation of OSMDs (last access to reference website on August 2019)

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . . 13

Developers of open-source medical devices

Developers of open-source medical devices

Developers of open-source medical devices

Acceleration program for the affordable of medical devices Developers and e-infrastructures of open-source medical devices

Nia Technologies

Autofabricantes

Metafluidics

CamTech

UBORA

Soft Robotics Toolkit

Type of resource Developers of open-source medical of devices Developers of open-source medical devices

Name Open Bionics

Table 1.3 (continued)

Euro-African community and global focus

US

US

Spain

Canada

US

Country of development UK

2017

2014

2017

2015

2015

2014

Year of found. 2014 Additional comments Developers of low-cost personalized prostheses with open-source collaborative approach Open-source projects focused on soft robotics and its applications, many of them in the biomedical sector Developers of open-source solutions for personalized scan and imaging-based design of medical prostheses Focused on the collaborative design of personalized prostheses and biodevices; sharing of information Open-source community-driven microfluidic devices, many of them with diagnostic purpose, developed and shared in a collaborative way A Communication Platform for Medtech Innovators (not open-source) Global community; open-innovation based on needs and proposals; sharing of complete medical device projects; sharing of teaching materials

http://camtechmgh. org/ https://platform. ubora-biomedical. org/

http:// autofabricantes. medialab-prado.es/ https://metafluidics. org/

Link https://openbionics. com/ https:// softroboticstoolkit. com/ https://niatech.org/

14 C. De Maria et al.

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

15

engineers, healthcare professionals, patients and patient associations, members of the “maker” movement, amateur designers, families of patients, and citizens in general, so as to help them unite efforts and share their ideas, skills, and resources, toward patient-driven biomedical and regulation-supported engineering research, while pursuing equitable access to healthcare technology. Hence, the promotion of open innovation, placing patients and healthcare professionals in the center of medical technology development and matching technological offers and demands, constitutes a pivotal strength of the UBORA e-infrastructure (and related community). In the mid-term, UBORA will help to harmonize and systematize the way medical products are developed, as all projects developed within UBORA follow the same standardized structure, and promote the application of standards and the compliance with directives with worldwide recognition, even in low-resource settings and distant communities, where application of regulations and market overview campaigns should increase. Main UBORA’s features, some selected cases of study and forthcoming related projects and activities, covering from biomedical engineering education to detection of medical needs and design and deployment of effective, efficient, sustainable, and affordable medical technologies and devices, are described in the different chapters of the handbook. Such chapters discuss key challenges and expected trends linked to the innovative field of open-source medical devices, hoping that our views and efforts may inspire healthcare professionals and technology developers to join the UBORA community and support the equitable access to healthcare technology.

1.5

Perspective: Current Challenges and 5-Year View

Taking into consideration all mentioned issues and the understanding of OSMDs we have acquired during the last years, in parallel to the implementation of the UBORA e-infrastructure, to its validation through cases of study and to the arrangement of the international community, our personal 5-year outlook perspective regarding OSMDs can be summarized as follows: • Open-source medical devices are bound to reshape the biomedical industry in the next decade by letting patients, patient associations, healthcare professionals, and technology developers play more relevant roles in the planning, specification, and conception of innovative healthcare technologies. • Following the example of open-source approaches in other industries, opensource medical devices may well lead to safer and more affordable healthcare technologies, thanks to information sharing and peer-reviewed decisions along their development and through the regulatory compliance verification process. • Personalization of medicine will be promoted by means of patient-specific developments, including a special focus on rare pathologies, as open-source medical devices can prove technically and economically viable, and still be affordable, regardless the size of production series, thanks to technological design and

16

•

•

•

•

•

C. De Maria et al.

manufacturing advances applied in their development that allow a shift from mass production to mass personalization. The needs of remote and rural populations will be more adequately addressed and answered, thanks to innovative supply chains that may delocalize the production of open-source medical devices, placing the fabrication facilities in the point of care. This will generate also a synergic economic growth in poorer regions and bring medical technologies to where they are more needed. However, several challenges need to be faced and solved in a collaborative way, for supporting the growth of the OSMDs sector. Relevant questions linked to regulation, privacy, safety, traceability, intellectual property, sustainability, and policymaking, among others, still require to be understood, and their reliability demonstrated to a larger scale, in this new paradigm of medical device development. We expect that worldwide connected collaborative design environments or communities and related medical device project repositories, from which the UBORA e-infrastructure and community constitute a remarkable example, will enlighten the path toward affordable, safe, regulation compliant, and accessible healthcare technologies for all. The question about the possible conversion of selected online collaborative and open-source communities into global notified bodies for certifying medical devices using harmonized standards and directives remains still open but constitutes and interesting thread to follow. To achieve all this, reinventing biomedical engineering education, so as to prepare the biomedical engineers of the future for working in international contexts and for developing biomedical projects applying collaborative and open-source methodologies is essential.

Open-source medical devices constitute an emerging trend with the potential for completely transforming the way medical devices are developed and the whole biomedical industry, as some of the examples presented in this study have helped to illustrate. However, several challenges still need to be overcome, in order to deploy the power of open-source and collaborative bioengineering design strategies, toward affordable and equitably accessible healthcare technologies. In this direction, initiatives such as the UBORA e-infrastructure and related international and multidisciplinary communities, as described in detail in this perspective, may turn out to be truly transformative resources for supporting the endeavors toward a wellfounded future for biomedical engineering, which will be more open, collaborative, and equitable. The success of the field relies on the adequate gathering and fulfillment of real medical needs and on the safety of the medical devices developed and delivered to healthcare professionals and patients. To this end, the UBORA e-infrastructure provides a framework to develop ISO compliant medical devices starting from clinical needs by sharing ideas, blueprints, and data. If properly implemented, UBORA medical devices are aligned with the MDR 2017/745 from the design point of view and ready for screening and examination for certification. In a nutshell,

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

17

its final aim is to promote well-being for all, increasing access to medical devices and moving toward health equity in accordance with the United Nations 2030 Agenda and the Sustainable Development Goals. Considering all of the above, the answer to our driving question “Can we transform the medical industry toward healthcare equity through open-source medical devices?” is yes, we can: counting with the collaborative efforts of a new generation of medical device designers, understanding the benefits of the opensource paradigm, and adequately funneled to relevant medical needs and safe performance, with the support of online co-creation environments and communities, from which UBORA constitutes a one-of-a-kind example. The transformation is already happening. Acknowledgments This study has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No. 731053, UBORA: Euro-African Open Biomedical Engineering e-Platform for Innovation through Education (topic: INFRASUPP -032016-Support to policies and international cooperation for e-infrastructures).

References Ahluwalia, C., De Maria, A., & Lantada, D. (2018). The Kahawa Declaration: A manifesto for the democratization of medical technology. Global Health Innovation, 1(1), 1–4. Alves, A. P., Da Silva, H. P., Lourenco, A., & Fred, A. L. N. (2006). BITalino: A biosignal system acquisition based on Arduino. Proceeding of the 6th Conference on Biomedical Electronics and Devices (BIODEVICES), 2006.

18

C. De Maria et al.

Arcarisi, L., Di Pietro, L., Carbonaro, N., Tognetti, A., Ahluwalia, A., & De Maria, C. (2019). Palpreast: A new wearable device for breast self-examination. Applied Sciences, 9, 381. Atala, A., & Joo, J. J. (2015). Essentials of biofabrication and translation. Elsevier. Debian project. Debian Social Contract (version 1.1 ratified on April 26, 2004). Online, last access to web in May 2019. https://www.debian.org/social_contract.en.html. De Maria, C., Di Pietro, L., Lantada, A. D., Madete, J., Makobore, P. N., Mridha, M., . . . Ahluwalia, A. (2018). Safe innovation: On medical device legislation in Europe and Africa. Health Policy and Technology, 7(2), 156–165. De Maria, C., Di Pietro, L., Ravizza, A., Diaz Lantada, A., & Ahluwalia, A. (2020). Open source medical devices. In E. Iadanza (Ed.), Clinical engineering handbook. Elsevier., ISBN: 9780128134672. De Maria, C., Mazzei, D., & Ahluwalia, A. (2015). Improving African health care through open source Biomedical Engineering. International Journal on Advances in Life Sciences, 7(1), 10–19. Douglas, T. S. (2011). Biomedical engineering education in developing countries: Research synthesis. IEEE-EMBC. e-NABLE Community. Enabling the future: A passionate network of volunteers using 3D printing to give the World a helping hand. Online, last access to web in May 2019. http:// enablingthefuture.org. European Society of Radiology (ESR). (2015). Medical imaging in personalised medicine: A white paper of the research committee of the European Society of Radiology (ESR). Insights into Imaging, 6(2), 141–155. Fasterholdt, I., Lee, A., Kidholm, K., Yderstraede, K. B., & Pedersen, K. M. (2018). A qualitative exploration of early assessment of innovative medical technologies. BMC Health Services Research, 18, 837. Ferretti, J., Di Pietro, L., & De Maria, C. (2017). Open-source automated external defibrillator. HardwareX, 2, 61–70. Freeman, K., Dinnes, J., Chuchu, N., Takwoingi, Y., Bayliss, S. E., Matin, R. N., ... & Deeks, J. J. (2020). Algorithm based smartphone apps to assess risk of skin cancer in adults: Systematic review of diagnostic accuracy studies. BMJ, 368. Gallup, N., Bow, J. K., & Pearce, J. M. (2018). Economic potential for distributed manufacturing of adaptive aids for arthritis patients in the U.S. Geriatrics, 3(4), 89. Gao, J., & Bernard, A. (2017). An overview of knowledge sharing in new product development. International Journal of Advanced Manufacturing Technology, 94(5–8), 1545–1550. Gamma Cardio Soft S.r.l. Open source ECG. Online, last access to web in May 2019. http://www. gammacardiosoft.it/openecg/. Ghezzi, T. L., & Corleta, O. C. (2016). 30 years of robotic surgery. World Journal of Surgery, 40(10), 2550–2557. Gershenfeld. (2005). Fab: The coming revolution on your desktop-from personal computers to personal fabrication. Basic Books. Kassianos, A. P., Emery, J. D., Murchie, P., & Walter, F. M. (2015). Smartphone applications for melanoma detection by community, patient and generalist clinician users: A review. British Journal of Dermatology, 172, 1507–1518. Jamshidnezhad, A., Kabootarizadeh, L., & Hoseini, S. M. (2019). The effects of smartphone applications on patients self-care with hypertension: A systematic review study. Acta Informatica Medica, 27(4), 263. Lanza, R., Langer, R., & Vacanti, J. (2014). Principles of tissue engineering (4th ed.). Elsevier. Lessig, L., Cusumano, M., & Shirky, C. (2005). Perspectives on free and open source software. MIT Press. Malkin, R. A. (2007a). Design of health care technologies for the developing world. Annual Review of Biomedical Engineering, 9, 567–587. Malkin, R. A. (2007b). Barriers for medical devices for the developing world. Expert Review of Medical Devices, 4(6), 759–763.

1 Open-Source Medical Devices: Concept, Trends, and Challenges Toward. . .

19

MIT’s SANA project. (2020). Medical diagnostics over mobile phones. SANA site in GitHub. https://github.com/sanamobile. Moroni, L., Boland, T., Burdick, J. A., De Maria, C., Derby, B., Forgacs, G., . . . Mota, C. (2018). Biofabrication: A guide to technology and terminology. Trends in biotechnology, 36(4), 384–402. Ng, P. K., & Jee, K. S. (2014). Concurrent knowledge sharing and its importance in product development. Journal of Applied Sciences, 14, 2978–2985. Niezen, G., Eslambolchilar, P., & Thimbleby, H. (2016). Open-source hardware for medical devices. BMJ Innovations, 2, 78–83. Oliveira, P., et al. Patient innovation: Sharing solutions, improving lifes. Online, last access to web in May 2019. https://patient-innovation.com. Open Bionics. Online, last access to web in May 2019. https://openbionics.com/. Open Prosthetics. Online, last access to web in May 2019. https://openprosthetics.org/. Open source initiative. Open source definition. Online, last access to web in May 2019. https:// opensource.org/docs/definition.php. Open source hardware association. Open source hardware definition. Online, last access to web in May 2019. https://www.oshwa.org/definition/. Pearce, J. M. (2014a). Open-source lab: How to build your own hardware and reduce research costs. Elsevier. Pearce, J. M. (2014b). Laboratory equipment. Cut costs with open-source hardware. Nature (Correspondence), 505, 618. Pearce, J. M. (2017). Emerging business models for open source hardware. Journal of Open Hardware, 1(1), 2. Savonen, B. J., Gershenson, J., Pearce, J. M., & Bow, J. K. (2019). Open-source three dimensional printable infant clubfoot brace. Journal of Prosthetics and Orthotics. https://doi.org/10.1097/ JPO.0000000000000257 UBORA: Euro-African Open Biomedical Engineering Innovation e-platform for Innovation through Education. Online, as of January 2020. https://platform.ubora-biomedical.org/. United Nations General Assembly: Transforming our World: The 2030 Agenda for Sustainable Development, on 21 October 2015, A/RES/70/1. Online, last access to web in May 2019. https://www.un.org/sustainabledevelopment/sustainable-development-goals/. Vlassi, M., Mavraganis, V., & Asvestas, P. (2017). A software platform for the analysis of dermatology images. Journal of Physics, Conference Series, 937, 012011. Wilkinson, M. D., et al. (2016). Comment: The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3, 160018., 1–9. World Health Organization. (2010a). Medical devices: Managing the mismatch: An outcome of the priority medical devices project. World Health Organization. World Health Organization. (2010b). Barriers to innovation in the field of medical devices (Background paper 6). World Health Organization. Witchurch, A. Examples of open source medical devices. Protocentral site in GitHub. Online, last access to web in May 2019. https://github.com/Protocentral/.

Chapter 2

Towards a Harmonized Methodology for the Development of Safe and Regulation Compliant Open-Source Medical Devices Carmelo De Maria, Andrés Díaz Lantada, Licia Di Pietro, Alice Ravizza, and Arti Ahluwalia

2.1

Modern Product Development and Systematic Design Methodologies

One century ago, with the foundation of the Bauhaus and related design schools in Germany, the UK, Russia and the USA, among other countries, systematic engineering design principles were established (Droste, 2019). Until the beginning of the twentieth century, design had been considered an art and not a technical activity or science. Product development had been hence confined to arts and crafts workshops and kept separated from engineering sciences. Working on the principles established just before the World War II, more modern ideas on systematic product development were empowered by relevant figures (Kesselring, 1951, 1954; Tschochner, 1954; Matousek, 1957 or Niemann, 1975), whose proposals continue providing ways for the resolution and management of concrete tasks in engineering and product design projects in general (Kaiser & König, 2006). Kesselring in the 1940s and 1950s proposed engineering design methods based on successive approximations, through which technical and economic criteria were optimized by using varied principles (minimal costs, minimal weight and volume, minimal losses, optimal function). In the 1950s, Tschochner highlighted four essential design variables: function,

C. De Maria · L. Di Pietro · A. Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy A. Díaz Lantada (*) Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] A. Ravizza USE-ME-D srl, I3P Politecnico di Torino, Turin, Italy © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_2

21

22

C. De Maria et al.

material, form and size; as afterwards, Matousek would also do by focusing on working principle, materials, manufacturing methods and geometries. The approach of Niemann, developed along the 1960s and 1970s, proposes starting the development process by defining a general product scheme, upon which the details are progressively developed working from subsystems towards individual components. The global design is divided in subsystems that can be developed in parallel, until their integration into a final system, by systematic variation and combination of possible solutions. Nowadays, the principles accepted for engineering innovative products in an efficient and successful path are based on the ideas of the aforementioned pioneers, as well as on the development stages and pathways proposed by other authors (i.e. Hansen, 1956; Wächtler, 1967 or Kuhlenkamp, 1971). In general terms, these researchers proposed four basic product development stages referred to as “prior studies”, “definition of fundamental principle”, “basic design” and “detailed design”. This step-by-step process is also detailed in design guidelines, as those developed by organizations such as the “Verein Deutscher Ingenieure” (VDI, i.e. VDI 2221 and 2206) or the “International Organization for Standardization” (ISO), as relates to testing, validation and overall quality management. The results of these studies, validated through countless products developed in a satisfactory way, after some modifications (Pahl & Beitz, 1988; Roozenburg & Eeckels, 1995; Ulrich & Eppinger, 2007), led to a working structure that includes “product specification and planning”, “conceptual design”, “basic engineering” and “detailed engineering”, although the boundaries between these stages are not always clearly detectable. Such stages correspond almost biunivocally with those from the educational CDIO “conceive-design-implement-operate” model (Crawley et al., 2007), although opting for dividing the conceptual stage in two stages for its special relevance. Among more recent models, for the development of specially complex engineering products and systems, it is interesting to highlight the “V-Model” of VDI 2206 for the design of mechatronic systems (Gausemeier & Moehringer, 2002; VDI, 2004) and related extended models (Graessler, 2017) that subdivide the global problem in subproblems that are solved and integrated towards the global solution, as in the previous methods, following a “divide and conquer” strategy. However, with respect to 20 years ago, when products were typically designed in secrecy and with the silo mentality of traditional corporate research labs, many industrial fields have experienced a paradigm shift from a “closed” to an “open innovation”, by involving stakeholders and future users since the beginning of product development process (Ng & Jee, 2014; Gao & Bernard, 2017). In this context, the internet and social computing technologies have allowed a better understanding of customers’ needs and preferences (Wang et al., 2007), and thanks to the support of cloud-based design and prototyping (Wu et al., 2015), the desires of the crowds can be rapidly transformed into new products.

2 Towards a Harmonized Methodology for the Development of Safe and. . .

23

This new concept of social product development has taken place in fashion, jewellery and furniture design (Perilli, 2017; Sarmah & Rahman, 2017) with the creation of virtual communities that actively develop innovative solutions, freely shared through online repositories, such as Thingiverse1 or GrabCAD2, born in the wake of the “makers” movement (Gershenfeld, 2005; Rosenfeld & Sheridan, 2014). The benefit of this collaborative and open-source design, in terms of accessibility, sustainability, lower costs, improved performance and safety, is widely exploited in software development (Lessing et al., 2005). Recently it has also been under consideration in several academic research fields from biology to nanotechnology (Oberloier & Pearce, 2018; Mushtaq & Pearce, 2018).

2.2

Special Considerations for Medical Devices and Challenges Linked to Open-Source Medical Devices: A New Paradigm and the Need for Harmonization

Medical technology has transformed the practice of medicine and patient care, with a wide set of relevant breakthroughs achieved during the last decades: from highprecision medical imaging for improved diagnoses and robotic-guided surgery to minimally invasive procedures for enhanced recovery after surgery, and the progressive use of smartphones for diagnosing and monitoring patients. The engineering design of successful medical devices depends not only on technological aspects but relies on human factors, including orientation to patients’ needs, collaboration with healthcare professionals throughout the whole development process and the compromise of multidisciplinary research and development (R&D) teams formed by well-trained professionals, especially biomedical engineers, capable of understanding the connections between science, technology and health and guiding such developments. Along decades, these requirements have inspired a number of systematic design methodologies, specifically thought for medical devices. The “Biodesign” method (Yock et al., 2015), for example, is based on three main steps, namely, “Identify”, “Invent” and “Implement”, of which the first requires a close interaction with clinicians and patients during a first-hand observation period. Interestingly, each step foresees a refinement or screening phase of the generated ideas, designs and business models. The spiral innovation approach (Yazdi & Acharya, 2013) has been initially described as “a new model of graduate education in bioengineering innovation and design”. This approach ensures early, staged and repeated examination of 1

Thingiverse: Digital design for physical objects, https://www.thingiverse.com [last access: January 2020]. 2 GrabCAD: Design Community, CAD Library, 3D Printing Software, https://grabcad.com [last access: January 2020].

24

C. De Maria et al.

all key elements of a successful medical device: clinical (perspectives of healthcare workers and patients); commercial (regulatory or compliance issues, intellectual property); technical (prototyping and testing); and organizational (fund raising and team management). In addition, the specific needs and constraints present in low-resource settings have led to the definition of teaching/design methodologies specific to low- and middle-income countries, such as the Frugal Biodesign approach (Sivarasu, 2019). However, it should be noted that in the medical field, the situation is more complex than in other industrial areas. Important requirements, due to the special quality and safety criteria, must be considered when engineering systems to interact with the human body. Two particular ISO standards are proposed to meet these requirements: 13485:2016 “Medical devices – Quality management systems – Requirements for regulatory purposes” (ISO, 2016) and 14971:2017 “Medical devices – Application of risk management to medical devices” (ISO, 2017). So important are these aspects that these two standards have been included in the list of Harmonized Standards that manufactures can use to demonstrate that their products, services, or processes comply with the EU legislation MDR 2017/745 on medical devices (The European Parliament, 2017a) and the MDR 2017/746 on medical devices for in vitro diagnosis (The European Parliament, 2017b). With respect to the trend of open innovation, in many cases medical technology is developed behind closed doors, and patients’ or medical professionals’ needs have been considered as a minor part of the decision-making process, which is currently under the pressure of marketing and immediate payback (Fasterholdt et al., 2018). The economic growth of medical technology developers and manufacturers is of course fundamental for reaching patients in a sustainable way. However, decisions taken on the basis of short-term incomes tend to leave rare pathologies and clinical needs of low-resource settings unattended and to hinder the creativity potential of designers (Rautenstraucht et al., 2002; Fogliatto & Da Silveria, 2011). These are just some drawbacks of the current state-of-the-art in the biomedical industry. Several examples of open-source healthcare-related technologies have appeared on the web (Niezen & Eslambolchilar, 2016), but although it is mandatory to meet the safety and efficacy requirement for their use in the clinical routine, only some of them have been designed to be compliant with medical device legislation (Ferretti et al., 2017; Arcarisi et al., 2019). To prove truly transformative, together with a solid framework to guarantee safety and efficacy, open-source medical devices (OSMDs) and the expected outcomes of their design approach need to be adequately defined. Open-source means making the design, documentation, source-code, blueprints, ideas and results available for the public in the most suitable file format to use, study, modify, improve and contribute to the design (see Chap. 1). The definition of OSMDs points out that online collaborative research and development environments, which help match medical needs and technological offers, could be the key for their mainstreaming. Indeed, relevant challenges appear when trying to incorporate this new design approach to the medical field. Not only

2 Towards a Harmonized Methodology for the Development of Safe and. . .

25

do the engineering design methodologies have to be reformulated accordingly, but engineers need to be trained to adopt new methods of approaching and solving problems. Clearly, the coordination of multiple actors is also crucial for the transformation of an invention into an economically advantageous and marketable innovation which can be adapted and adopted in different contexts and territories (Bonaccorsi & Rossi, 2003; Malkin, 2007). The medical device landscape is changing due to social and demographic pressures as well as to developments and advances in technology and our access to it. In this context, the UBORA e-platform was inspired and developed, facilitating safe innovation, education and equitable access to quality medical technology (De Maria et al., 2018; Ahluwalia et al., 2018a). In the next sections, we illustrate the UBORA learning/teaching and project development methodology, which underlines this unique e-infrastructure and can support the sustainable growth of the nascent field of open-source medical devices.

2.3

Training a New Generation of Biomedical Engineers in the Systematic Development of OSMDs Through Project-Based Learning Methods

Preparing engineers in general and biomedical engineers in particular to work in the medical industry, in connection with the design of medical devices, is a challenging process. In fact, the trainee should acquire a broad overview of the biomedical field and industry, a well-balanced combination of general and specific knowledge, according to the chosen specialization, several technical abilities linked to modern engineering tools and professional skills. Furthermore, starting from the consideration that biomedical engineering (BME), if adequately applied to the development of equitable healthcare technologies, may constitute a fundamental resource to achieve global health coverage (Ahluwalia et al., 2018a), BME trainees should be aware of their social responsibility, and ethical issues should be always considered in the BME field and its education. Ideally, fulfilling the 2030 Agenda for Sustainable Development (UN, 2015), especially as regards the Sustainable Development Goals (SDGs) numbers 3 and 4 on “Good Health and Well Being” and “Quality Education”, should become the driving context for the biomedical engineers and the BME educators of the future. Problem-based learning, project-based learning (PBL), experiential learning, game-based learning, learning in collaborative project and environments, among others, are just different versions of highly formative and integrative learning experiences that place students in the centre of the teaching-learning process, in accordance with a desire for a more holistic training for the twenty-first century, especially in engineering education (Larmer, 2014). In all these project-related teaching-learning methodologies, student teams face a real life (typically

26

C. De Maria et al.

engineering) problem, more or less simplified, and perform the specification, design, prototyping, and testing of a product, a process, an event or, generally speaking, a system. In some cases, prototyping and testing is achieved just virtually, but there is a critical analysis of results and a public exposition and subsequent debate for increased learning throughout the groups of students taking part in the course(s). In addition, creativity, decision-making and critical thinking are fostered, and professional tools for engineering practice (i.e. design and simulation software, prototyping tools, etc.) are applied, so as to prepare students, as globally as possible, for their professional and personal lives. Knowledge acquisition is necessary, but the development of specific professional skills and transversal abilities, for more adequately applying the acquired knowledge to solve real challenges, is fundamental in modern education (UNESCO, 1998). All this is in connection to what accreditation agencies, i.e. ABET and ENAEE, have been proposing in the last decades. This education based on learning objectives and professional outcomes is also essential for the recently implemented European Area of Higher Education and aligned with the UNESCO’s World Declaration on Higher Education for the twenty-first century (De Graaf & Kolmos, 2003). The varied types of PBL experiences mainly differ in the level of depth, to which the project, product, process, or system is specified, designed, implemented, and managed or operated, and in the proposed context and desired level of realism, which depends also on the time and resources available for students living through the formative experience. The “conceive-design-implement-operate” or “CDIO” approach to project-based learning encompasses all the aforementioned types of active learning experiences (Crawley et al., 2007). In fact, the complete CDIO cycle involves the whole life cycle of any engineering project or system, from specification and planning, through the design, engineering, and construction, towards full operation, maintenance, and end of life. Among the more recent project-based and active learning educational models in engineering education (Graham, 2018), it is also important to note the MIT’s “NEET” or “New Engineering Education Transformation model”, whose main elements are its project-centric curriculum and its organization in “threads” around a series of projects focused on new machines and systems. In fact, the NEET model is similar to CDIO but focuses on highly innovative fields: digital cities, autonomous machines, living machines, advanced materials, or renewable energies. The permanent search for educational contexts with an increased level of realism (when compared to more classic project- and problem-based learning experiences) and, therefore, with a higher social impact is one of the characteristics of these PBL educational experiences. In a way, this aspect links PBL with service learning, defined by Jacobi as “a form of experiential education, in which students engage in activities that address human and community needs together with structured opportunities for reflection designed to achieve desired learning outcomes” (Jacoby, 1996). Ideally, the developed solutions reach society and transform it. This projectbased service learning model can be added to the previously listed types of active and integrative learning experiences. This hybridization between service learning and PBL can have additional impact if open-source and collaborative approaches to

2 Towards a Harmonized Methodology for the Development of Safe and. . .

27

engineering and its education are also involved and promoted, as recent international “express CDIO” learning experiences have put forward (Ahluwalia et al., 2018b). PBL, connected to the development of real medical devices, is arguably the best strategy for training engineers, from a wide set of environments, towards successful professional practice in the medical industry (Yock et al., 2015; Yazdi & Acharya, 2013; Sivarasu, 2019). PBL and challenge-based instruction prove very appropriate in BME, not just for organizing single courses and providing an introduction to the medical industry and to medical technology but also for adequately implementing and being the backbone of whole programmes of studies and for career development, as presented and discussed in an excellent review (Abu-Faraj, 2008). Hybridizing PBL with service learning and connecting it with the needs of the developing world may enhance the learning outcomes and better connect students with a more guided transition to professional practice, especially if the results are shared (Sienko et al., 2013; Ahluwalia et al., 2018a, 2018b). Accordingly, the UBORA methodology, which interweaves systematic engineering design methods with the CDIO approach and with standardized strategies for medical device development, while focusing in depth in the more recent open-source and collaborative paradigms, is described in the following section.

2.4

The UBORA e-Platform: Tools and Methodology for Harmonizing the Design of Open-Source Medical Devices

UBORA is the first web-based infrastructure focused on the co-creation of OSMDs compliant with EU Medical Devices Regulation MDR 2017/745, achieved by providing a unique methodical approach to their design and development. The UBORA methodology is inspired by the “Conceive-Design-Implement-Operate” educational model, empowered by the EN ISO 13485:2016, for the identification of needs, risk class and relevant standards. The UBORA e-platform has been developed with the aim of both training and mentoring biomedical engineers and medical device developers in the engineering design of OSMDs and for industrial innovation purposes. This e-infrastructure and its projects, in fact, are articulated through a common metastructure (a set of guided work packages and tasks), which should help to harmonize the OSMDs development process, to promote safety and the application of relevant standards. UBORA is developed for the promotion of collaboration through the whole development process of innovative OSMDs, whose complete development details, including specifications, design process, lists of components, computer-aided design files and blueprints, among other relevant issues, are shared by means of an interactive and designer-oriented structure.

28

C. De Maria et al.

The UBORA platform covers all types of medical devices (i.e. diagnostic tools, prosthetic devices, surgical tools, monitoring systems, therapeutic devices) and bioengineering systems supporting medical practice (i.e. supporting lab equipment, mobile apps and software). In turn, UBORA has helped to arrange a diverse and truly global community of engineers, healthcare professionals, patients and patient associations, members of the “maker” movement, amateur designers, families of patients and citizens in general, so as to help them unite efforts and share their ideas, skills and resources, towards patient-driven biomedical and regulation-supported engineering research, while pursuing equitable access to healthcare technology. Hence, the promotion of open innovation, placing patients and healthcare professionals in the centre of medical technology development and matching technological offers and demands constitutes a pivotal strength of the UBORA e-infrastructure and its community. In fact, UBORA is arranged in a way that all the information (concepts, design files, documentation, source code, blueprints and prototypes, testing results and all collected data) of the medical devices, developed within its boundaries by the growing UBORA community, may be shared and straightforwardly found, so as to allow their reproduction, utilization and potential production by manufacturers. In these terms, UBORA can be considered as a tool for connecting in a more direct way with companies and the BME community, which may support their R&D strategies in the biomedical field. Projects are presented as a collection of biomedical engineering resources and medical devices, listed in form of clickable boxes with a representative title and highlight images (see Fig. 2.1) for rapid identification. Searching tools by keywords, defined when the projects are initially registered as ideas, provide a smooth interaction with the e-infrastructure and promotes findability of projects and solutions for different medical disciplines and varied pathologies.

2.4.1

The UBORA Methodology

With respect to the standard CDIO model, which lays at its foundation, the UBORA methodology starts from the identification of clinical needs, as an explicit step in the design process, and foresees intermediate check points for assessing the compliance to ISO standards. This project development metastructure, schematized in Fig. 2.2, guides the design workflow and enables harmonizing and systematizing the method whereby open-source medical products are developed. The entire process is supported by mentors who are assigned to projects. Mentors come from an academic and/or professional background and have well-proven experience in the field of designing and testing medical devices. Their expert review ensures that the designers correctly identify all the users’ and clinical needs, while also helping the developers to integrate all the applicable safety standards in the design requirements. Mentors assist the developers by commenting the state-of-theart and well-known technical burdens to the project, thus guiding the developers towards effective innovation.

2 Towards a Harmonized Methodology for the Development of Safe and. . .

29

Fig. 2.1 UBORA e-infrastructure. (a) Landing page. (b) Some examples of medical device projects, selected through specific keywords

2.4.2

UBORA Project Management Metastructure

The UBORA methodology has been implemented into the platform, which has two main sections devoted to the development of OSMDs named “Clinical needs” and “Project management”, respectively, and a section for the creation and strengthening

30

C. De Maria et al.

Fig. 2.2 The UBORA project development methodology

of the “Community” of developers, mentors and stakeholders. A “resource section” contains selected teaching-learning materials on biomedical engineering, as well as the documentation and video tutorials to use and further develop the e-infrastructure, whose source code will be released as open-source by the end of the UBORA project. The clinical needs section is aimed at identifying bioengineering design specifications linked to a particular clinical need. The section provides a unique environment for healthy discussion between patients, healthcare providers and engineers, to ensure that clinical needs are turned into safe and affordable projects. The project management section is a structured framework composed of six work packages (WPs, Table 2.2), which implements the whole UBORA methodology, offering tools for the management of computer-aided modelling files and preparation of the pre-production device dossier. WP1 or “Medical Need and Product Specification” work package is designed to specify the clinical need and the scope of the medical device and to identify the main product requirements, as well as clinical area related to the technology. A list of suggested keywords for indexing the projects is also provided, on the basis of identified clinical need (e.g. rehabilitation, prevention, etc.), clinical area (e.g. cardiology, neurosurgery, etc.) and technology (e.g. mobile-based, ergonomic support, active medical device etc.): this scheme is one of the strategies implemented to have a FAIR data management ensuring findability, accessibility, interoperability and reusability of data (Wilkinson et al., 2016). Moreover, in WP1, the risk class and applicable standards for the specific design are identified. The device classification, structured through a checklist, follows the rules described in Annex VIII of MDR 2017/745. In order to improve usability and minimize human error, the rules are arranged as a decision tree, which leads the designer(s) through a flow of simple (mainly yes/no) questions. UBORA provides also a list of questions concerning the medical technology, in order to select the basic safety standards (also known as horizontal standards), which are usually difficult to identify through keywords in a search engine. Table 2.1 presents a selection of the internationally recognized standards that are used as reference and proposed to developers.

2 Towards a Harmonized Methodology for the Development of Safe and. . .

31

Table 2.1 Selection of internationally recognized standards proposed for UBORA projects Standard EN ISO 13485:2016

EN ISO 14971:2012

MEDDEV 2.7.1 rev 4. Clinical evaluation: A guide for manufacturers and notified bodies under Directives 93/42/EEC and 90/385/EEC IEC 62366-1

EN ISO 15223-1:2016

Description This standard specifies requirements for all entities involved in medical devices, in all stages of the product life cycle: from design to manufacture to installation to disposal. UBORA e-infrastructure is structured to be a guideline for design activities in compliance to this standard This standard specifies requirements for designers and manufacturers of medical devices, in order to minimize the risk of the device itself. There is no “risk zero” device, but many activities can be implemented to reduce and manage risk. This standard provides useful checklists and also guidance on risk management techniques such as FMEA This guideline provides information on methods used to assess the clinical performance and the clinical benefit of a medical device This standard provides guidance on how to manage the human factors while designing a medical device (usability engineering) This standard lists a series of symbols that may be applicable in labels of medical devices

After specification and planning, the design starts: WP2 or “conceptual design” work package guides the designers through the development of different embodiments/ideas of the same medical device, which are analysed and ranked in a polling section on the basis of feasibility, performance, usability and safety aspects. After the identification of the most promising conceptual design, the infrastructure helps in structuring the idea, by requesting information, for example, on facility requirements, estimated life, shelf life, request of consumables. Through WP3 or “design and prototyping” work package, the designer prepares documents related to the medical device concept proposed in WP2: the project management system suggests its organization in subsections related to mechanical components, electronics/firmware, software and their integration. Once WP3 is concluded, the WP4 or “implementation” work package guides the developers in the testing phase, so as to demonstrate compliance with relevant standards. At the end of this stage, developers can download the preproduction device dossier, which summarizes the main achievements reached during the project development. WP5 or “operation” work package is aimed at completing the technical documentation of the device, in order to publish the project for downloading, and provides a lean business model canvas (Trimi & Berbegal-Mirabent, 2012), customized for OSMDs, to help the potential transfer of the technology to industry. The e-infrastructure provides guidance in the

32

C. De Maria et al.

Table 2.2 The UBORA medical device engineering methodology implemented through the project management metastructure

Needs

Conceive

Design

Implement

Work packages (WP.x.) and tasks (Tx.y.) WP.1.- Medical need and product specification T.1.1.- Clinical needs T.1.2.- Existing solutions T.1.3.- Intended users T.1.4.- Product requirements T.1.5.- Device classification T.1.6.- Regulation checklist T.1.7.- Formal review WP.2.- Conceptual design T.2.1.- Physical principles T.2.2.- Voting T.2.3.- Concept description T.2.4.- Structured information on the device WP.3.- Design and prototyping T.3.1.- General product description T.3.2.- Design for ISO testing compliance T.3.3.- Instruction for fabrication of prototypes WP.4.- Implementation

T.4.1.- Prototypes and safety assessment T.4.2.- Quality criteria T.4.3.- ISO compliance T.4.4.- Results from vitro/ vivo tests T.4.5.- Structured information on the device T.4.6.- Preproduction document

Descriptions and main purposes Specifying the clinical need and the scope of the medical device and identifying the main product requirements to be fulfilled Description of the medical need to be solved Analysis of related medical devices and market Explanation regarding users and attributes List of requirements to be fulfilled. Classification achieved through e-questionnaire Proposed standards for the development process Check by mentors of classification and specs Guiding designers through the development of different embodiments/ideas of the same medical device Rationale for the medical device Proposing and analysing different product ideas Describing the best product idea or concept Basic information on device, users, and environment Helping developers to materialize a first design and preliminary functional prototype of the medical device concept Hardware and software designed and used Fine-tuning the design for safety and quality tests Describing how to prototype the proposed device Guiding developers in the testing phase, so as to demonstrate compliance with relevant standards. Creation of a preproduction document Improved prototypes and tests information Quality criteria towards production Evaluation of standards fulfilment Summary of tests performed with prototypes Detailed info on device, users, and environment Dossier for interaction with notified bodies (continued)

2 Towards a Harmonized Methodology for the Development of Safe and. . .

33

Table 2.2 (continued)

Operate

Work packages (WP.x.) and tasks (Tx.y.) WP.5.- Operation

T.5.1.- Technical documentation T.5.2.- Lean canvas and business model WP.6.- Project closure

Descriptions and main purposes Completing the technical documentation of the medical device, so as to publish a first version of the project for downloading. Approaching a preliminary business plan Improved product documentation and blueprints Business plan for technology transfer Guiding developers in creating promotional videos and brochures, towards crowdfunding campaigns and commercialization after adequate certification. Final product technical and economical dossier

selection of the most appropriate license scheme, as well as advising in legal issue related to the use of the developed technology. Finally, WP6 or “project closure” work package guides the developers in creating promotional videos and brochures, towards crowdfunding campaigns and commercialization after adequate certification, for which the completely documented process provided by UBORA in form of dossier is essential.

2.5

UBORA as a Model for Education in Biomedical Engineering

The UBORA educational model has been constructed as an evolution of modern PBL approaches to engineering education, including the CDIO model, but empowered with the support of the UBORA e-platform following the exposed methodology. It is important to mention some main keys of this model for transforming medical industry through the collaborative development of OSMDs: 1. Focus on open-source devices and collaborative methods. Project-based learning is intrinsically collaborative, as it normally deals with teams of students working together for a common objective. However, the concept of “collaborative PBL” makes reference to experiences, of which UBORA is a pioneering example, especially concentrating on collaborative design and project management methods and on co-creation strategies employing online resources involving international teams (Diaz Lantada & De Maria, 2019). The project development pipeline incorporates regulated control points and includes relevant issues within the open-source paradigm (use of FAIR data principles, traceability of the design changes, connection to fundraising environments), which are clearly presented to students using UBORA. Moreover, UBORA provides students with a clear

34

2.

3.

4.

5.

C. De Maria et al.

pathway for straightforward and sustainable medical device innovation. These methods, also related to recent trend changes in product engineering, from mass productions to mass personalization, are of special relevance for medical devices and the future of the biomedical industry. Focus on regulations and standards. The relevance of regulations and of designing according to standards is very often ignored in PBL experiences, which tend to concentrate on the more technical or technological issues. Medical devices, for their particular human and health focus, should be developed within highly regulated environments and following the more relevant harmonized or internationally recognized standards. In the devices developed through UBORA as a teaching-learning instrument, a focus on regulations and standards is promoted. Thus, the platform can be employed to improve the learning experience in relation to issues in which it is typically difficult to engage students (Di Pietro et al., 2019). Degree of multidisciplinary and international component. UBORA’s educational experiences are open to all types of higher education students. Although the standard profile is that of engineering students, any student focusing on biomedical studies is welcome. In fact, the competitions and schools delivered by the UBORA consortium have counted with students from a variety of engineering disciplines (e.g. industrial, mechanical, chemical, biomedical, electrical, energy, automation and robotics) as well as with medical students (Ahluwalia et al., 2018b). All of them were able to use the platform and interact with other students and mentors while developing their projects. Hybridization with service learning. Medical devices can be proposed on the basis of collaboration with patient associations, hospital or healthcare professionals and devoted to delivering real solutions to selected relevant medical needs. These solutions, shared through UBORA, may reach society and help to evolve medical practice and biomedical industry, while helping also to create richness in low- and middle-income settings. Long-term impact through UBORA. Results from UBORA can be shared both as open teaching materials, which can be used for capacity building in universities starting to implement their BME programmes and as cases of study of OSMDs to be used in any medical device design course.

2.6

Conclusions

Open-source medical devices constitute an emerging trend with the potential for completely transforming the way medical devices are developed and the whole biomedical industry. However, several challenges still need to be overcome, in order to deploy the power of open-source and collaborative bioengineering design strategies, towards affordable and equitably accessible healthcare technologies. In this direction, initiatives such as the UBORA e-infrastructure and related international and multidisciplinary communities, as described in detail in this perspective,

2 Towards a Harmonized Methodology for the Development of Safe and. . .

35

may turn out to be truly transformative resources for supporting the endeavours towards a well-founded future for biomedical engineering, which will be more open, collaborative and equitable. The success of the field relies on the adequate gathering and fulfilment of real medical needs and on the safety of the medical devices developed and delivered to healthcare professionals and patients. To this end, the UBORA e-infrastructure provides a framework to develop ISO-compliant medical devices project starting from clinical needs by sharing ideas, blueprints and data. If properly implemented, UBORA medical devices are aligned with the MDR 2017/745 from the design point of view and ready for screening and examination for certification. In a nutshell, its final aim is to promote well-being for all, increasing access to medical devices and moving towards health equity in accordance with the United Nations 2030 Agenda and the Sustainable Development Goals.


Medical devices presented as cases of study along the handbook follow the explained methodology and are developed and shared as open-source medical devices through the UBORA e-infrastructure.

References Abu-Faraj, Z. O. (2008). Bioengineering/biomedical engineering education and career development: Literature review, definitions and constructive recommendations. International Journal of Engineering Education, 24(5), 990–1011. Ahluwalia, A., De Maria, C., & Díaz Lantada, A. (2018a). The Kahawa Declaration: A manifesto for the democratization of medical technology. Global Health Innovation, 1(1), 1–4. Ahluwalia, A., De Maria, C., Madete, J., Díaz Lantada, A., Makobore, P. N., Ravizza, A., Di Pietro, L., Mridha, M., Muñoz-Guijosa, J. M., ChacónTanarro, E., & Torop, J. (2018b). Biomedical engineering project based learning: Euro-African design school focused on medical devices. International Journal of Engineering Education, 34(5), 1709–1722.

36

C. De Maria et al.

Arcarisi, L., Pietro, L. D., Carbonaro, N., Tognetti, A., Ahluwalia, A., & De Maria, C. (2019). Palpreast – A new wearable device for breast self-examination. Applied Sciences, 9(3), 381. Bonaccorsi, A., & Rossi, C. (2003). Why open source software can succeed. Research policy, 32(7), 1243–1258. Crawley, E. F., Malmqvist, J., Östlund, S., & Brodeur, D. R. (2007). Rethinking engineering education: The CDIO approach (pp. 1–286). Springer. De Graaf, E., & Kolmos, A. (2003). Characteristics of problem-based learning. International Journal of Engineering Education, 19(5), 657–662. De Maria, C., Di Pietro, L., Díaz Lantada, A., Madete, J., Makobore, P. N., Mridha, M., . . . Ahluwalia, A. (2018). Safe innovation: On medical device legislation in Europe and Africa. Health Policy and Technology, 7(2), 156–165. Díaz Lantada, A., & De Maria, C. (2019, April). Towards open-source and collaborative project based learning in engineering education: Situation, resources and challenges. International Journal of Engineering Education, accepted for publication. Di Pietro, L., Botte, E., Granati, R., Moroni, S., Tomasi, M., Vozzi, G., & De Maria, C. (2019, July). Teaching design standards and regulations on medical devices through a collaborative projectbased learning approach. International Journal of Engineering Education, accepted for publication. Droste, M. (2019). Bauhaus. Aktualisierte Ausgabe. Taschen. Fasterholdt, I., Lee, A., Kidholm, K., Yderstraede, K. B., & Pedersen, K. M. (2018). A qualitative exploration of early assessment of innovative medical technologies. BMC Health Services Research, 18, 837. Ferretti, J., Di Pietro, L., & De Maria, C. (2017). Open-source automated external defibrillator. HardwareX, 2, 61–70. Fogliatto, F., & Da Silveira, G. (Eds.). (2011). Mass customization (Springer Series in Advanced Manufacturing). Springer. Gao, J., & Bernard, A. (2017). An overview of knowledge sharing in new product development. International Journal of Advanced Manufacturing Technology, 94(5-8), 1545–1550. Gausemeier, J., & Moehringer, S. (2002). VDI 2206: A new guideline for the design of mechatronic systems. IFAC Proceedings, 35(2), 785–790. Gershenfeld, N. (2005). Fab: the coming revolution on your desktop-from personal computers to personal fabrication. Basic Books. Graessler, E. (2017). A new V-model for interdisciplinary product engineering. 59th Ilmenau Scientific Colloquium, TU Ilmenau. Graham, R. (2018). The global state of the art in engineering education. MIT Press. Hansen, F. (1956). Konstruktionssystematik. VEB-Verlag Technik. International Organization for Standardization. (2016). ISO 13485:2016: Medical devices – Quality management systems – Requirements for regulatory purposes. ISO. International Organization for Standardization. (2017). ISO 14971:2017: Medical devices – Application of risk management to medical devices. ISO. Jacoby, B. (1996). Service-learning in higher education: Concepts and practices. Jossey-Bass. Kaiser, W., & König, W. (2006). Geschichte des Ingenieurs. Ein Beruf in sechs Jahrtausenden. Carl Hanser Verlag. Kesselring, F. (1951). Bewertung von Konstruktionen. VDI Verlag. Kesselring, F. (1954). Technische Kompositionslehre. Springer. Kuhlenkamp, A. (1971). Konstruktionslehre der Feinwerktechnik. Hanser. Larmer, J. (2014). Project-based learning vs. Problem-based learning vs. X-BL. Edutopia. Lessig, L., Cusumano, M., & Shirky, C. (2005). Perspectives on free and open source software. MIT Press. Malkin, R. A. (2007). Design of health care technologies for the developing world. Annual Review of Biomedical Engineering, 9, 567–587. Matousek, R. (1957). Konstruktionslehren des allgemeinen Maschinenbaus. Springer.

2 Towards a Harmonized Methodology for the Development of Safe and. . .

37

Mushtaq, U., & Pearce, J. M. (2018). Open source appropriate nanotechnology. In Nanotechnology and global sustainability (pp. 220–245). CRC Press. Ng, P. K., & Jee, K. S. (2014). Concurrent knowledge sharing and its importance in product development. Journal of Applied Sciences, 14, 2978–2985. Niemann, G. (1975). Maschinenelemente. Springer. Niezen, G.; Eslambolchilar, P.; Thimbleby, H..- “Open-source hardware for medical devices”. BMJ innovations, 2(2), 78-83, 2016. Oberloier, S., & Pearce, J. (2018). General design procedure for free and open-source hardware for scientific equipment. Designs, 2(1), 2. Pahl, G., & Beitz, W. (transl. & ed. Wallace, K.). (1988). Engineering design: A systematic approach. The Design Council/Springer. Perilli, A.. From creative AI to open source sculpture: how tech is changing art. It’s Nice That, 23rd May, 2017. Online, as of May 2019: https://www.itsnicethat.com/ Rautenstrauch, C., Seelmann-Eggebert, R., & Turowski, K. (Eds.). (2002). Moving into mass customization. Springer. Roozenburg, N., & Eeckels, J. (1995). Product design: Fundamentals and methods. Wiley. Rosenfeld Halverson, E., & Sheridan, K. M. (2014). The maker movement in education. Harvard Educational Review, 84(4). Sarmah, B., & Rahman, Z. (2017). Transforming jewellery designing: Empowering customers through crowdsourcing in India. Global Business Review, 18(5), 1325–1344. Shuman, L. J., Besterfield-Sacre, M., & McGourty, J. (2005). The ABET professional skills, can they be taught? Can they be assessed? Journal of Engineering Education, 94, 41–55. Sienko, K. H., Sarvestani, A. S., & Grafman, L. (2013). Medical device compendium for the developing world: A new approach in project and service-based learning for engineering graduate students. Global Journal of Engineering Education, 15(1), 13–20. Sivarasu, S. (2019). Frugal Biodesign: An approach for developing appropriate medical devices in low-resource settings. Biomedical Engineering for Africa, 47. The European Parliament and the Council of the European Union. (2017a). Regulation (EU) 2017/ 745 on medical devices. https://eur-lex.europa.eu/legal-content/en/TXT/?uri¼CELEX% 3A32017R0745. Last access Jan 2020. The European Parliament and the Council of the European Union. (2017b). Regulation (EU) 2017/ 746 on in vitro diagnostic medical devices. https://eur-lex.europa.eu/legal-content/EN/TXT/? qid¼1565171051278&uri¼CELEX:32017R0746. Last access Jan 2020. Trimi, S., & Berbegal-Mirabent, J. (2012). Business model innovation in entrepreneurship. International Entrepreneurship and Management Journal, 8(4), 449–465. Tschochner, H. (1954). Konstruieren und Gestalten. Abrißeiner Konstruktions- und Gestaltungslehre unter besonderer Berücksichtigung von Maschinenbau und Feinmechanik. Girardet, Essen. Ulrich, K., & Eppinger, S. (2007). Product design and development (4th ed.). Mc-Graw Hill/Irwin. UNESCO. World declaration on higher education for the twenty-first century: Vision and action, adopted by UNESCO’s World Conference on Higher Education on 9 October 1998. United Nations General Assembly. Transforming our World: The 2030 Agenda for Sustainable Development, on 21 October 2015, A/RES/70/1. Verein DeutscherIngenieure. (1993). VDI 2221: Systematic approach to the development and design of technical systems and products. VDI (rev. 2017). Verein DeutscherIngenieure. (2004). VDI 2206: Design methodology for mechatronic systems. VDI. Wang, F. Y., Carley, K. M., Zeng, D., & Mao, W. (2007). Social computing: From social informatics to social intelligence. IEEE Intelligent Systems, 22(2), 79–83. Wächtler, R. (1967). Beitrag zur Theorie des Entwickelns (Konstruierens). Feinwerktechnik, 71, 353–358.

38

C. De Maria et al.

Wilkinson, M. D., et al. (2016). Comment: The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3, 160018., 1–9. Wu, D., Rosen, D. W., Wang, L., & Schaefer, D. (2015). Cloud-based design and manufacturing: A new paradigm in digital manufacturing and design innovation. Computer-Aided Design, 59, 1–14. Yazdi, Y., & Acharya, S. (2013). A new model for graduate education and innovation in medical technology. Annals of Biomedical Engineering, 41(9), 1822–1833. Yock, P. G., et al. (2015). Biodesign: The process of innovating medical technology (2nd ed., pp. 1–952). Cambridge University Press.

Chapter 3

Getting Started with an Open-Source Medical Device Project: Systematic Needs Identification Techniques for Bottom-Up Strategies Philippa Ngaju Makobore, Andrés Díaz Lantada, Licia Di Pietro, Carmelo De Maria, and Arti Ahluwalia

3.1

Needs Identification in Open-Source and Collaborative Biomedical Engineering

Traditional product development starts with a product planning stage where the decision-making process in the preliminary phases usually involves questions including the following: • • • • •

Which are the market trends? Which are the most profitable sectors and locations to invest in? Which technologies are currently more stylish? Are the investments in previous products from the portfolio already recovered? Are there fiscal benefits for investing in specific sectors and products?

In the medical industry this leads in many cases to delayed applications of technological breakthroughs. Furthermore, the needs of patients and healthcare professionals have only a reduced impact on the actual decisions to undertake new developments. There are more serious implications especially in resource constrained settings where product development must adhere to design constraints and cater for an end user in a unique healthcare setting.

P. N. Makobore Uganda Industrial Research Institute, Kampala, Uganda A. Díaz Lantada (*) Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] L. Di Pietro · C. De Maria · A. Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_3

39

40

P. N. Makobore et al.

However, in the open-source and collaborative biomedical engineering (BME) paradigm, as materialized by the “UBORA bioengineering design model” for innovating open-source medical devices (OSMDs), the product planning starts by considering the following issues: • What are the needs of patients and patient associations with which we have interacted? • What do healthcare professionals from our surroundings need? • What are the priorities the WHO proposes to focus on? • Are there rare pathologies linked to our field of expertise not adequately addressed? • Are there relevant technologies not yet translated to medical practice due to nonmedical motivations? • Are there technologies not yet reaching patients and healthcare providers in low-/ middle-income settings or remote locations? • Could we partner with collaborators in the field for more efficient and straightforward tackling of global health concerns? Medical technology developers, especially those focusing on OSMDs, should be aware that interaction with end users (patients and their families, medical doctors, nurses, surgeons, etc.) and their active involvement in the decisions taken throughout the whole innovation and development process is a key to success. Multidisciplinary teams are more relevant than ever before in collaborative and open-source approaches to medical technology development. International online communities of engineering and medical experts, but involving also policymakers, social scientists, lawyers and even educators, can support the configuration of specific teams for developing concrete medical device projects, hence working on a quite efficient project-based basis. Their interaction and co-creation through dedicated e-infrastructures for innovation and research is also advisable, as they can help to match technological demands and offers and help with delocalized teamwork.

3.2

Strategies and Techniques for Systematically Identifying and Screening Medical Needs Solvable by the Co-creation of Medical Technologies

Probably the most relevant step of the whole innovation process, especially in the OSMD arena, is adequately selecting a suitable medical need and collecting the desires and expectations of users. To this end, it is important to systematically search for priority medical needs and users’ expectations about medical technology and count with such information together with the experience and views of technology developers, who better understand the scientific/technical state-of-the-art and its potentials.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

41

Different approaches and techniques support such systematic finding of needs and expectations. Some of them are coincident with tools used by marketing departments in most engineering sectors to fine-tune product planning and specifications (street surveys, online questionnaires, monitoring of social networks, brainstorming with groups of interest, focus group meetings, among others). However, in collaborative product development for the biomedical field, face-to-face discussions and meetings with final users may be preferred, due to the relevance of bottom-up development pathways. Again the relevance of e-infrastructures for education, innovation and research, through which communities of researchers, future bioengineers, medical device developers and key stakeholders interact along the whole development process, should be highlighted as powerful resources for finding needs and associating them with ad hoc created development teams of experts for their resolution. Some techniques and strategies have already proven successful in needs identification for the co-creation and collaborative design of innovative OSMDs or for its fine-tuning to the specific needs of concrete patients, specific populations or settings with special boundary conditions. Indeed, OSMDs are very well suited for the promotion of personalized approaches to medical practice, as previously reported (Ahluwalia et al., 2018a) and as some of the following techniques and examples help to highlight: (a) Motivational Questionnaires Combined with Interviews Online surveys do not always prove useful for the engineering design of new products and processes, if not adequately analysed in wrap up face-to-face meetings with the relevant stakeholders. Therefore, for the expansion of the OSMD field and for finding relevant health issues and clinical needs to work on, following a bottomup and collaborative processes, we propose the combined use of surveys with recapitulation meetings involving healthcare professionals, mainly doctors and nurses, and with representatives from patient associations. Policymakers and members of the management team at hospitals may also join but, in our experience, the interaction with ground professionals dealing day-by-day with the actual health challenges is most appropriate. During the recapitulation meetings, it is interesting to follow a guided process with a meeting leader playing the role of “creativity promoter”. The meeting leader should briefly present the survey results in a time slot of around 10–15 min. Then a guided debate should be conducted. Creativity promotion tools, such as brainstorming (better in a written mode as brainwriting), Philips 6-6 or lotus flowers-brainstorming upon the results of an initial brainstorm, among others, can be applied for listing down medical needs in different sectors, departments or regions, during another stage or time slot of around 20–30 min. Association of ideas through a guided debate of 20–30 min and a final prioritization and selection of needs, for which OSMDs are to be implemented, again in 20–30 min, would lead to a relevant work plan.

42

P. N. Makobore et al.

This successful approach is illustrated by the case study presented in Sect. 3.3, dealing with a needs assessment process in Ugandan hospitals for the development and delivery of open-source solutions. (b) Innovation Through Education “Innovation through education” is another key strategy for the nascent field of OSMDs and for collaborative BME (as well as the basic motto of the UBORA community). Indeed, the most interesting transformations that take place in higher education are in most cases motivated by students’ desire to learn and to transform the world by applying the outcomes from their formative experiences. A cohort of science and technology students, future engineers, involved in project-based learning (PBL) activities and adequately guided and motivated can provide transformative solutions and breakthroughs to any engineering discipline, including BME, as the UBORA international competitions and design schools have proven, among many other relevant examples. Innovation through education is very appropriate for the rapid generation of highly innovative concepts, for the analysis of “crazy ideas” and for the validation of out-of-the-box solutions in an extremely resource-efficient way. Besides, this strategy is especially focused on the long-term, as related to the gathering of talent and to the implementation of communities with shared views about the future of technology development, which in our opinion will be based on open-source and collaborative paradigms. This alternative approach is also illustrated in Sect. 4 by presenting a teachinglearning experience, in which PBL and service-learning are hybridized, in connection with medical device innovation and open-source and collaborative approaches to global health concerns. (c) Online Infrastructures for Matching Demands and Offers in the OSMD Field Another option for the detection of technological demands associated to medical needs, which may provide interesting ideas for development at any moment, relies on the use of online infrastructures for research and innovation. An advantage of this procedure is that solutions for specific patients and their families may be developed, hence promoting personalized approaches to medical device development and to medical practice. However, it is true that this approach may not be the most efficient or the one providing optimal societal impacts with limited resources. In any case, sharing personalized solutions, as open-source projects with public blueprints obtained following parametric design resources, may help to promote their further usability, after personalized design adaptation for the requirements of patients or users with similar needs. To cite an example, the Enabling the Future community (http://enablingthefuture.org/), focusing on open-source developments of hand prostheses, offers some technological solutions (prosthetic hand kits) at very reasonable prices and with options for personalization of design, size, colour for enhancing the user experience and making the final devices more aesthetic or ergonomic, according to users’ preferences.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

43

This alternative option is additionally illustrated in Sect. 3.5, in which results from needs assessment and related OSMDs proposed and developed through the UBORA e-infrastructure are presented.

3.3

3.3.1

Application Case: Assessing Clinical and Medical Equipment Challenges in Ugandan Hospitals to Start Up Needs-Based Open-Source Medical Device Projects Needs Assessment

During the UBORA project, to support the first implementation of the UBORA e-infrastructure and to deliver examples of OSMDs developed according to systematic needs assessment processes, an assessment exercise was carried out in five regional referral hospitals and three health centre IV’s in northern, western and eastern Uganda. Uganda’s healthcare system works on a referral basis starting with the lowest level as village health teams responsible for community-based preventive and promotive services and next is health centre II providing preventive, promotive and out-patient curative health services and outreach care. Health centre III includes maternity, in-patient and laboratory services, and health centre IV further includes emergency surgery and blood transfusion services (see WHO document, (World Health Organization, Primary healthcare systems, case study from Uganda, 2017)). From a health centre IV, most cases are referred to general hospitals, regional referral hospitals and the highest and last the national referral hospital. The study aimed at investigating and prioritizing relevant clinical needs and urgent challenges with existing medical equipment. The data collected from the study are serving as a baseline resource for the UBORA e-infrastructure for the design of OSMDs, especially for Sub-Saharan Africa and other low- and middle-income resource settings. However, it is important to highlight that the same strategy applied and validated through this case study can be successfully put into practice for any medical device development project, in which the needs of patients and healthcare professionals are considered the starting point. Summarizing the methodology, the study was conducted during March 2017 with a focus on paediatrics, obstetrics and gynaecology, intensive care units (ICU), emergency, surgery and laboratory departments through primarily face-to-face interviews aided by questionnaire. The questionnaire was designed to adequately capture the current clinical needs and communicate challenges that clinicians face when using medical devices and equipment. In addition to the clinical status of health facilities, information was collected from biomedical technicians to understand the more common reasons why medical equipment breaks down and the hurdles faced with maintaining and repairing them. Ethical review was not applicable to this study which accessed and collected information already in the public domain. No ‘study subjects’ participated in the

44

P. N. Makobore et al.

Fig. 3.1 Needs assessment study at the tuberculosis ward at Fort Portal RRH. (Fort Portal, Uganda)

study, no human data or materials were collected, and no interventions were performed. All persons interviewed agreed to do so by prior appointment as part of their professional day-to-day work obligations to the hospitals. A total of 27 clinicians were interviewed that included 5 doctors, 3 medical officers and 19 nurses. At least four biomedical technicians were also interviewed (Fig. 3.1). Findings across the eight health facilities surveyed indicate that paediatric and maternal wards report a higher burden of disease or clinical condition in comparison to other departments such as surgery and general out-patient services. Figure 3.2 represents a tally of the various clinical conditions by department/ward showing paediatric and maternal health reporting a higher disease burden/clinical conditions. Common clinical conditions for paediatrics ranged from complications during childbirth such as birth asphyxia, malaria, pneumonia, anaemia (most often induced by malaria), neonatal sepsis and diarrhoea. Sickle cell anaemia was a condition that was predominately in the eastern part of Uganda based on data collected from Mbale RRH and Amuria HC IV. Figure 3.3 shows statistics by health facility for the prevalence of the top five clinical conditions for the paediatric population. Providing warmth for preterm, low-birth weight and malnourished infants is a challenge as they usually have to rely on their mothers for warmth, which is often insufficient. Water availability is scarce which often leads to increased hospital infections. For neonates and children suffering from malnutrition, intravenous (IV) medication is prescribed based on their weight. Manual regulation of IV medicines is prone to error, and this often leads to morbidity and mortality of infants. Medication for neonates and children suffering from malnutrition is closely correlated with their weight.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

45

Fig. 3.2 Greatest disease burden/clinical condition by wards surveyed in eight health facilities

Fig. 3.3 Top five paediatric clinical conditions by health facility. Results from surveys at different Ugandan hospitals are presented

In the neonatal section, it was noted that children could possibly require oxygen for several weeks, but the oxygen supply was usually inadequate and oxygen concentrators unavailable. The most common cause of death was the lack of oxygen and blood to transfuse children with anaemia. In Fort Portal RRH (western Uganda), for instance, has an anaemia prevalence of over 85% in children, most requiring blood transfusion. Unfortunately, blood is usually in short supply, and as a result

46

P. N. Makobore et al.

there are four deaths on average per day in their paediatric ward. For maternal health, the most common clinical challenges included post-partum haemorrhage (PPH), post-abortion-related complications, pre-eclampsia, severe anaemia, ruptured uterus and malaria during pregnancy that results into preterm labour or miscarriages. In addition to the clinical challenges, clinicians were severely overworked with an average clinician to patient ratio of 1:45 to 1:60 on a daily basis across all health facilities. Overall essential medical equipment (https://www.who.int/medical_devices/ priority/en/, https://www.who.int/medical_devices/publications/interagency_med_ dev_list/en/) was lacking in all the eight health facilities analysed in the study. Hospital workshops were grossly understaffed with a ratio of 2 to 3 biomedical technicians serving up to 17 healthcare departments for the health facilities surveyed in northern, eastern and western Uganda. Donated equipment was particular problematic as they lacked requisite manufacturer support, not calibrated, with spare parts and consumables extremely difficult to source. In addition, training of end users on how to use the devices was not provided. The hospital workshop at Fort Portal RRH had received support from the Japanese International Cooperation Agency (JICA) that trained biomedical technicians and introduced the 5S-CQI-TQM quality management system which significantly improved the efficiency of the workshop despite being severely understaffed. The “Medical Equipment” workshop at Mbarara RRH seemed to have the most structured corrective and preventive maintenance system in place. At this site, preventive maintenance was carried out every 3 months, corrective maintenance as often as the breakdowns occur and user training carried out for every cycle of new interns usually every 3 months. Figure 3.4 depicts the biggest challenges with equipment as the overall lack of appropriate devices, staff constraints, procurement constraints and malfunctioning equipment. The most critical medical equipment in the paediatric departments were oxygen concentrators, pulse oximeters for pneumonia diagnosis, infant warmers, phototherapy units, continuous positive airway pressure (CPAP) devices, thermometers and syringe pumps. Oxygen concentrators were critical for both paediatric and maternal departments. Most hospitals utilize refillable oxygen cylinders and then distribute oxygen to the patient. There appeared to be a critical need for a carefully regulated oxygen delivery system that can supply multiple patients at a time. In maternal departments, delivery instruments were few, with a limited number of autoclaves and suction machines to remedy pulmonary aspiration of expectant mothers caused by a number of factors including delayed gastric emptying. There was also a need for additional vacuum extractors for assisted labour. Medical imaging was inaccessible to most of the regional referral hospitals. Fort Portal RRH lacked an X-ray machine for children, Gulu RRH (Northern Uganda) only had two imaging centres that offered X-ray and ultrasound services, which were prohibitively expensive for any average patient. There is no computed tomography (CT) scan in the entire Gulu district, meaning patients would have to travel over 300 km to Kampala to access the service.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

47

Fig. 3.4 Challenges with medical equipment for all health facilities surveyed

Furthermore, the lack of an appropriate and effective infection control method has contributed to equipment malfunctions and corrosion causing a risk to patient safety. Currently bleach is used to clean wards and some medical instruments, but the concentrations are not carefully monitored, and as a result severe corrosion is a frequent occurrence. The results of the methodical needs identification process presented painted a clear picture of the clinical and medical device gaps faced in Ugandan government health facilities. These statistics provided useful information to conceptualize appropriate needs-driven OSMDs and to start up such projects with the support of the UBORA e-infrastructure harmonized development process. Regarding personnel, the work burden for clinicians was grave as depicted by the extremely high patient to doctor ratios across all eight health facilities. In addition to the clinician work burden, biomedical technicians suffered similar workload challenges as they were few in number and had to repair and maintain medical equipment across numerous districts. Figs. 3.5 and 3.6 illustrate some common detected problems.

3.3.2

Application to an Open-Source Medical Device: Portable Neonate Warmer

Using the information gathered from the needs finding, a portable neonate warmer (PNW) was designed responding to mortality due to hypothermia in rural low-income communities and the delay in seeking care at a health facility. The PNW (Figs. 3.7 and 3.8) was designed to provide the appropriate thermal therapy

48

Fig. 3.5 Unused medical equipment at Gulu RRH Workshop

Fig. 3.6 Poor disposal of decommissioned autoclaves at Gulu RRH

P. N. Makobore et al.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

49

Fig. 3.7 Portable neonate warmer prototype: (a) Padded transport carrier for the PNW. (b) Wire mesh frame of the transport carrier for the PNW. (c) Testing temperature sensors while simulating the weight of the baby. (d) Sodium acetate heating pad in solid state (pre-activation)

Fig. 3.8 Fabrication of the portable neonate warmer: (a) Sleeping bag adjustable hood design for the PNW. (b) Measurement of sodium acetate for the heating pad. (c) Preparation of a super cooled solution of anhydrous sodium acetate for the heating pad

and temperature monitoring for neonates with hypothermia during transportation to a health facility and while waiting for medical care on arrival. The lack of thermal protection is still a major hindrance for newborn survival in developing countries. Although hypothermia is rarely a direct cause of death, it contributes to a substantial

50

P. N. Makobore et al.

Fig. 3.9 Fabrication of the portable neonate warmer: (a) partitioning of the heating pad. (b) 3D printing of the cylindrical casing

proportion of neonatal mortality globally, mostly as a comorbidity of severe neonatal infections, preterm birth, and asphyxia. The portable neonate warmer has four distinct modules: (1) temperature monitoring unit, (2) sleeping bag, (3) transport carrier and (4) heating pad. The sleeping bag utilizes biocompatible heat absorbent and resistant materials that include polyester, cotton and polyethylene terephthalate. The sleeping bag was designed and sewn to have adjustable hoods to cover the neonate or infant’s head, a slot underneath for the insertion of the heating pad as shown in Fig. 3.7 (images a and b). The sodium acetate crystals are weighed and used to form a supercooled solution of anhydrous sodium acetate (C2H3NaO2) in the lab as shown in Fig. 3.7 (images c and d). The sodium acetate is enclosed in a nylon casing that is partitioned to ensure an even distribution of the sodium acetate in the phase change material (PCM) pad as shown in Fig. 3.3 part (a). A stainless steel spring activator with a 3D printed casing is enclosed in the pad embedded in the sodium acetate. The cylindrical casing was 3D printed with FMD technology using a PLA filament as shown in Fig. 3.3 part (b). Each heating pad was fitted with activators on opposite sides for an efficient activation process (Fig. 3.9). The described PNW project can be consulted in the UBORA e-infrastructure (https://platform.ubora-biomedical.org/) and is open to interactions from colleagues, medical technology developers, medical professionals, associations, patients, families and even sponsors. Please visit the following website for additional information: https://platform.ubora-biomedical.org/projects/753b19aa-266f-4e63-bc8b9488d220a36f

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

3.4

51

Application Case: Project-Based Service Learning Experience Connected to the Engineering of Open-Source Medical Devices for Solving Real Medical Needs

Bottom-up strategies for open innovation may benefit also from the establishment of connections between academia, industry and healthcare stakeholders. To illustrate this, another application cases is presented, in connection with project-based service learning activities: Several courses at Universidad Politécnica de Madrid (i.e. “Development of medical devices” at the BSc in Biomedical Engineering, “Bioengineering” at the MSc in Mechanical Engineering, “Bioengineering Design” at the MSc in Industrial Engineering and “MedTech” at the MSc in Organizational Engineering) deal with the development of medical technologies and guide students through the complete life cycle of medical device innovation, following the CDIO “conceive-designimplement-operate” model, as recently published (Díaz Lantada et al., 2018). Along the last decade, these courses have led to around 1000 students involved in the development of more than 250 medical device concepts, from which more than 100 have reached a validation stage in the form of functional prototypes manufactured with the support of the UPM’s Product Development Laboratory. Since 2017, this model, combining industrial innovation with education, has been adapted to focus on OSMDs and on collaborative design methodologies in BME, and students’ interactions have been articulated through the UBORA e-infrastructure. This singular online platform has also helped to promote a harmonized approach to medical technology innovation, to guide students through the classification of medical devices and the application of standards and to share the solutions achieved as open-source projects with downloadable files and blueprints. Besides, a hybridized PBL and service-learning model has been promoted, motivating students to innovate by focusing on medical needs selected through interactions with relevant stakeholders. In spite of letting students decide upon the needs to address and the related medical devices to develop in these courses, the relevance of collaborating with patients and healthcare professionals in any medical technology project has been highlighted, and additional efforts have been put into providing students with a more realistic context. To this end, during the needs identification phase, so as to select the topics for the medical technology projects to be developed by student teams, contact with different patient associations and clinical areas has been fostered. Furthermore, connection to open-innovation approaches to medical technology has been supported by proposing students to join the UBORA community, to use the UBORA e-infrastructure as open-source tool for guided medical technology development and to participate in the UBORA design competitions (Ahluwalia et al., 2018b). Moreover, the involvement of a team of clinicians focused on organ transplants, of general surgeons and obstetricians and of a couple of associations focused on disabilities has been achieved thanks to the proactivity and motivation of our

52

P. N. Makobore et al.

students. In addition, a seminar with participants from hospital innovation units, with surgeons as users of advanced medical tools, with medical technology entrepreneurs and with experts from notified bodies has made students more aware of the complex context of the medical industry. All this has led to a more careful selection of medical needs and to the consequent proposal of very relevant and innovative medical technologies, to be developed during these courses. The potential impact of these technologies is increased, not only because they address more realistic needs but also because the patients and professionals associations involved can constitute fundamental links of the chain towards eventual technology transfer, which may take place beyond the temporal framework of these courses, hence providing students also a path for professional development. In order to promote impacts and sustainability beyond the courses, students are sharing their developments through the UBORA e-infrastructure, which also supports designers in their decision-making process towards safer and EU regulation compliant medical devices. Finally, some teams are considering spin-off creation and all students participate annually in the Actúa-UPM ideas challenges for technological enterprises and in the UBORA Design Competitions. Continuously evolving PBL experiences keep them alive and are directly connected to quality improvement cycles. As a result of this continued update of PBL experiences in BME, student motivation and learning outcomes have importantly increased in the last years, and the developed medical devices are now more complete, professional, regulation-compliant and focused on real needs. As example of innovation through education in the aforementioned courses, UBORA device “Lazarus Biotech Standing Frame” is presented in the images below: The device is a standing frame for children with mobility problems developed according to a need assessment process, performed in collaboration with ASPADIR foundation, by a team of students at Universidad Politécnica de Madrid led by Eng. Adrián Martínez (UBORA mentor). The foundation works with children with physical and sensory disabilities, and the developed device (see Fig. 3.10) was finalist of the 2019 International ABEC-UBORA Design Competition. Normally a whole academic year, with a semester for focusing on the needs identification, product planning and conceptual design and a second semester for designing, prototyping and testing proves very adequate and highly formative for these types of experiences. However, examples from single-semester courses and even 1-week express courses or hackathons have been also devoted to OSMDs with remarkable success (Ahluwalia et al., 2018b). The described approach is extremely powerful for transforming education and, through it, the biomedical industry, as it is connected with capacity building, and several students end up wishing to further industrially develop their medical devices. Involving healthcare professionals and patient associations in these formative experiences is part of the success keys towards solving global health concerns through open innovation and through modern engineering education.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

53

Fig. 3.10 UBORA device “Lazarus Biotech Standing Frame”, design and prototype. Acknowledgements: Eng. Adrián Martínez and team

3.5

Application Case: Online Needs Detection Through the UBORA e-Infrastructure

The UBORA e-infrastructure is a remarkable example of open-innovation platform in the biomedical field, which also supports the detection of relevant needs for OSMD projects. Indeed, the platform is aimed at identifying bioengineering solution linked to specific clinical needs, which may support an adequate approach to providing affordable technologies to low-resource and remote settings, to developing innovative solutions for rare pathologies and to helping with a more personalized medical practice from the technological point of view. The UBORA community understands that good bioengineering solutions should match needs from the real world: they should be beneficial to the patient, technically feasible and sustainable and should empower the local community. To identify the needs expressed by both patients and healthcare professionals, UBORA counts with a needs identification section, open to all, which provides an ample space for description and also prepares the subsequent development process and organization within the e-infrastructure, through a set of keywords and identification resources (clinical need types, areas of interest and type of technology are selected among versatile lists). These keywords and areas of interest are very useful to match needs, projects, developers and mentors and to find related solutions through the e-infrastructure. It is important to note that the keywords have been selected according to FAIR (findable, accessible, interoperable, reusable) data principles (Wilkinson et al., 2016) that guided the implementation of the UBORA e-infrastructure, so as to promote the findability of solutions. As example, Table 3.1 presents the complete description of a medical technology, an open-source multi-purpose kit for cell culture, which could be used for solving several needs in the fields of cellular pathology and molecular biology.

54

P. N. Makobore et al.

Table 3.1 Example of information provided for describing a technological demand or a medical need, in this case associated to cell culture systems, within the UBORA e-infrastructure Title Description

Clinical need type Area Technology Keywords

Open-source multi-purpose kit for cell culture Advances in tissue engineering and biofabrication rely on innovative microfluidic systems (labs- and organs-on-chips) capable of interacting with cells and enabling different types of studies (cell-material interactions, cell-tocell communication, response to drugs and external stimuli, etc.) for fine-tuning regenerative strategies. Normally, cell culture kits, labs-on-chips and organs-onchips for studying cells and imitating physiological interactions in vitro, tend to be highly specialized. Counting with an open-source multi-purpose kit for cell culture would constitute an interesting alternative for enabling more studies at a fraction of final cost and as support for educational and training purposes Support to laboratory practice Cellular pathology / molecular biology In vitro diagnostic device Lab-on-a-chip, organ-on-a-chip, tissue engineering, biofabrication

In addition, Table 3.2 presents some selected examples of medical needs recently incorporated to the UBORA e-infrastructure, waiting for teams of developers to be arranged for providing solutions. These may also serve as input topics for projectbased learning activities in BME in connection with the OSMD paradigm or as seeds for research and innovation projects, ideally performed in collaboration among multidisciplinary and international teams. The developed solutions, shared through UBORA, may also constitute cases of study for educational and training purposes. Some of the needs proposed in Table 3.2 are already being solved, such as the affordable Braille display, thanks to the collaborative efforts of the UBORA community, which is already shared as public project in UBORA. Additional needs can be found in the UBORA e-infrastructure’s needs identification section: https:// platform.ubora-biomedical.org/clinical-needs

3.6

Conclusions and Future Trends

Successful medical technologies rely on adequately detecting unsolved or partially solved socially relevant medical challenges or needs, as different well-known methods for the engineering of medical devices have highlighted and explained in detail in the last decades (Kucklick, 2007; Yock et al., 2015). Identifying needs in a precise way and clearly specifying product requirements constitutes the planning stage, which is fundamental for the success of any engineering products, processes or systems under development. In the area of medical technology, the planning stage is even more critical, due to the very special considerations that apply to the development of medical devices, due to their interactions with the human body and their working environment (Davis, 2003).

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

55

Table 3.2 Selected examples of medical needs recently incorporated to the UBORA e-infrastructure, some of them waiting for teams of developers to be arranged for providing solutions, others already shared as open-source medical devices through UBORA Detected need Wound care dressing indicating infection Device for the application of drugs to children

Clinical need type Support to medical practice Non-surgical therapy/ administration of drugs

Area Clinical microbiology Paediatrics

Device for preventing stroke Prostheses in composite materials

Monitoring purpose

Cardiology

Rehabilitation

Rehabilitation

Eyelid cleaner

Rehabilitation

Dermatology

Braille converter

Rehabilitation

Rehabilitation/ ophthalmology

Multi-purpose kit for cell culture Affordable automatic dialyzer machine

Support to laboratory practice Support to medical practice

Cellular pathology

Navigator for supporting blind people

Rehabilitation

Rehabilitation/ ophthalmology

Device for fast diagnosis of sepsis App for pregnancy monitoring App for detecting and preventing anaemia Glucose monitoring patch

Prevention of pathology or disease Monitoring purpose Remote or self-diagnosis

Clinical immunology/paediatrics Obstetrics and gynaecology Haematology

Monitoring purpose

Public health

Nephrology

Technology Monitoring device Other supporting equipment Monitoring device Other supporting equipment Preventive device Other supporting equipment In vitro diagnostic device Other supporting equipment Other supporting equipment In vitro diagnostic device Mobile-based technology Mobile-based technology Monitoring device

Even if the more traditional engineering design approaches are still valid in our continuously evolving societies and rapidly changing world, a shift of trend can be appreciated in several design areas, which now benefit from the collaborative efforts of co-creators and from bottom-up innovation strategies. Indeed, traditional in-house innovation lacks dynamism, creative power and attention to end users’ needs, as compared with more recent open-innovation environments. In the medical industry, the emergent area of OSMDs, developed in collaboration and shared with colleagues, medical professionals, patients and patient associations and citizens, can benefit from the employment of systematic needs identification techniques oriented to bottom-up strategies.

56

P. N. Makobore et al.

In this chapter, we have tried to present some options for a methodical identification of needs, as a support to getting started with open-source medical device projects. We have focused on illustrating such techniques by means of selected cases of studies, in which the potentials of an intimate collaboration with medical professionals, patients and patients associations have been highlighted. Besides, the relevance of bridging the gap between academia, industry, healthcare professionals and patients has been discussed, as a key for transforming the medical device industry towards a more equitable paradigm. In our opinion, the future of medical technology development is collaborative, and the needs should be identified following bottom-up approaches and interacting with the end users. More than market-driven, which is important, medical technology should be always user-centred and respond to socially relevant questions and to emerging healthcare concerns.

References Ahluwalia, A., De Maria, C., & Díaz Lantada, A. (2018a). The Kahawa Declaration: A manifesto for the democratization of medical technology. Global Health Innovation, 1(1), 1–4. Ahluwalia, A., De Maria, C., Madete, J., Díaz Lantada, A., Makobore, P. N., Ravizza, A., Di Pietro, L., Mridha, M., Munoz-Guijosa, J. M., Chacón Tanarro, E., & Torop, J. (2018b). Biomedical engineering project based learning: Euro-African design school focused on medical devices. International Journal of Engineering Education, 34(5), 1709–1722. Davis, J. R. (2003). Handbook of materials for medical devices. ASM International. Díaz Lantada, A., Ballesteros Sánchez, L. I., Chacón Tanarro, E., Moreno Romero, A., Borge García, R., Peláez, M. A., Ramos, R., & Juan Ruíz, J. (2018). Coordinated design and implementation of Bioengineering Design and MedTECH courses by means of CDIO projects linked to medical devices. 14th International CDIO Conference, Kanazawa, Japan. Enabling the future community: http://enablingthefuture.org/ Kucklick, T. R. (2007). The medical device R&D handbook. CRC Press. UBORA platform for open source medical devices: https://platform.ubora-biomedical.org/ Wilkinson, M. D., et al. (2016). Comment: The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3, 160018., 1–9.

3 Getting Started with an Open-Source Medical Device Project: Systematic. . .

57

World Health Organization. Primary healthcare systems, case study from Uganda, 2017. World Health Organization. List of priority and medical devices: https://www.who.int/medical_ devices/priority/en/ World Health Organization. Interagency list of medical devices for essential interventions for reproductive, maternal, newborn and child health: https://www.who.int/medical_devices/ publications/interagency_med_dev_list/en/ Yock, P. G., et al. (2015). Biodesign: The process of innovating medical technology (2nd ed., pp. 1–952). Cambridge University Press.

Chapter 4

Design of Open-Source Medical Devices for Improved Usability and Risk Minimization Alice Ravizza, Noemi Stuppia, Federico Sternini, Luis Ignacio Ballesteros Sánchez, Rocío Rodríguez-Rivero, Enrique Chacón Tanarro, Juan Manuel Munoz-Guijosa, and Andrés Díaz Lantada

4.1

Introduction

Safe performance is essential for any medical technology that biomedical engineers and technology developers in general would like to introduce into the market. Without safe use and operation, the technology is, in plain words, useless or, even worse, dangerous, with both social and legal implications for the developers. No societal transformation can be achieved based on a technology that is not designed for safe use and operation. Therefore, several methodologies, techniques and international standards focus on the analysis and assessment of potential use errors and malfunctions during the operation of healthcare technologies and deal with strategies for detecting and minimizing related risks. Open-source medical devices (OSMDs) are typically designed in a collaborative way, involving truly international and multicultural development teams and following development life cycles, through which the information is shared, and also manipulated and updated, within large communities of “makers”. Besides,

A. Ravizza (*) · N. Stuppia USE-ME-D srl, I3P Politecnico di Torino, Torino, Italy e-mail: [email protected] F. Sternini USE-ME-D srl, I3P Politecnico di Torino, Torino, Italy DET, Politecnico di Torino, Torino, Italy L. I. Ballesteros Sánchez · R. Rodríguez-Rivero · E. Chacón Tanarro · J. M. Munoz-Guijosa · A. Díaz Lantada (*) Mechanical Engineering Department, ETSI Industriales, Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_4

59

60

A. Ravizza et al.

OSMDs are sometimes manufactured and delivered through still unconventional supply chains. For example, a design that is shared online can be downloaded from the cloud and manufactured in a set of collaborative delocalized laboratories – like fablabs – close to the point-of-care for more easily reaching users, mainly healthcare professionals, and patients, in remote regions. All this, as compared with traditional in-house developments, brings new uncertainties to the process and potential hazards, if systematic methods for promoting usability and minimizing risks are not adequately shared in the open-source community, applied and documented. At the same time, the openness of information throughout the whole development process should, ideally, promote design peer-review and failure detection at earlier stages, if the design principles for OSMDs and their special features are correctly considered. In consequence, this chapter is prepared with the intention of summarizing the more relevant techniques and standards for usability promotion and risk minimization in biomedical engineering (BME) tasks linked to healthcare technology development. In addition, the final purpose is presenting an innovative and straightforward – user friendly – methodology for developing intrinsically safe medical technologies, based on internationally accepted techniques and standards, and directly applicable to open-source medical devices. Thanks to illustrating the different methods, techniques, standards and the global strategy with different examples and case studies, the chapter may also constitute an educational resource for colleagues wishing to incorporate to their BME courses and programme aspects linked to usability, risk management and safety promotion, both in connection with OSMDs and with more traditional approaches to medical technology development. This chapter evolves from a position paper prepared by BME Alice Ravizza and cols. for the 12th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC – Biodevices, 2019: (Ravizza, A.; Díaz Lantada, A.; Ballesteros Sánchez, L.; Sternini, F. and Bignardi, C. Techniques for Usability Risk Assessment during Medical Device Design.). The position paper, according to the conference’s indications, “portrayed a short report of work in progress or an arguable opinion about an issue discussing ideas, facts, situations, methods, procedures or results of scientific research”. In a way, it discussed the relevance of combining designed methods for enhanced usability in connection with risk minimization techniques and presented, in a preliminary way, the main steps of a possible methodology. In the present chapter, the methodology is presented in its complete form, linked to the field of open-source medical devices, adequately adjusted to the singular characteristics of OSMDs, and illustrated through different examples and case studies. Basic Definitions: Users’ Needs and Usability Biomedical engineers routinely include users’ needs in the design requirements of medical devices. But what is a user need? Not only the patient clinical condition, of course. It is also the need to be provided with a device that is adequate to his skills, education and capabilities and can be used safely by all those in connection with the device or interacting with it along its whole life cycle.

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

61

Usability is defined, by the international standard IEC 62366, as “the characteristics or features of the user interface that facilitate use and thereby effectiveness, efficiency and user satisfaction in the intended environment of use” (International Electrotechnical Commission [IEC], 2015). It is consequently an essential concept to understand and systematically consider in the design process of any medical device, for the benefit of healthcare professionals, patients and all relevant stakeholders. Fundamental design decisions may be driven not only by performance, cost or environmental impacts of the device but also by its ergonomic and use experience aspects, in connection with usability, safety and user experience, as happens in several industries and as we explore and discuss in the following sections. Why Is Usability Important in BME? Usability tests are part of an engineering process that sees the collaboration of a team of experts in the specific medical sector (physicians, nurses, medical personnel), clinical and biomedical engineers and product designers and include analysis of past adverse events, related to the use of the device, design thinking of devices in accordance with the repetitive and repeatable mental patterns of the human user, considerations on experience and technical knowledge of different types of user (laymen or professional) and the application of ergonomic principles on the design of the devices. Poor manipulation and low intuitiveness of the interface not only can lead to discomfort and pain in the user but also can be the source of different types of errors. The most common fall into the categories of the procedure violation, incorrect interpretation of the information provided by the machine and incorrect setting of the machine operating parameters, among others. The need therefore arises to place, in the design of medical devices, a particular attention to the needs, expectations and level of technical skills of the end user and the need to personalize the physical interface with respect to specific characteristics of the body or the specific user expectations on the software user interface. The classical techniques of human factors engineering count with important advantages to help systematize the approach to medical device design with a view on usability, because they allow to describe the different types of users and to build around them a personalized interface. In fact, the entire process of usability assessment and promotion makes it possible to put the patient and his/her needs at the centre of the medical device design towards enhanced healthcare.

4.2

Relevant Regulations and Standards

In the new Medical Device Regulation EU 2017/745 (European Parliament and Council of the European Union, 2017), the usability of medical devices acquires extreme importance in the process of certification. If fact, the essential requirement 5 states that in eliminating or reducing risks related to use error (EU 2017/745), the manufacturer shall:

62

A. Ravizza et al.

(a) reduce as far as possible the risks related to the ergonomic features of the device and the environment in which the device is intended to be used (design for patient safety), and (b) consider the technical knowledge, experience, education, training and use environment, where applicable, and the medical and physical conditions of intended users (design for lay, professional, disabled or other users). Usability is also considered a crucial risk factor in other leading regulations, for example, the term “usability” appears also in FDA guidelines (US Food and Drug Administration, 2016). Besides, usability assessment is interwoven with the entire design and development process of any medical device (from the prototype to the post market evaluations) including SaMD (i.e. software as medical device). Usability-centred design is aimed at preventing, eliminating and/or reducing human use errors. Regarding relevant standards, ISO 13485 describes the management of design activities thanks to the definition of inputs, the subsequent definition of outputs and their verification. The outcome of this process is a finalized solution that can be subject to validation, in order to give proof that it satisfies all the requirements. In this design and development process, particular attention is given to the user interface, defined as “means by which the user and the medical device interact” (IEC 62366, 2015), which includes all elements of the device with which the user interacts and all source of information transmitted by the device. Aspects of the interface that ensure usability, lower the probability of human error and provide an efficient experience of use are therefore important inputs for the first phase of the design control process. Usability input requirements include means of ensuring the humanmachine communication, a list of actions that should be performed by the human and others that should be performed by the system, means to ensure that the system is matching with real world experience and expectations. For example, inputs include the definition of how to operate and to transport the system: a knob or an audio command? A handle or wheels? Additionally, the inputs may define what actions should be automated: should the system require the user to input a password for silencing an alarm or should the system autonomously mute the alarm after a certain period? Lastly, how should the designers try to comply with user expectations in terms of colour coding, icon appearance and meaning and so on? Do we really need to custom design the “start/stop” button or should we use the internationally recognized symbol? In the world of collaborative, open-source design, the possibility to refer to existing projects is therefore a great enhancement to the designers’ capability to identify correct and efficient solutions to the user needs. Once the user requirements have been identified and design variants are proposed, these should be verified. Usability becomes therefore a test requirement for safety, together with the common industry standard tests, chosen during each design verification according to the specific medical technology. Usability is tested at the same level of design verification as biocompatibility, haemocompatibility, electrical

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

63

safety and sterility. It is also included in the requirements for safety, effectiveness and quality assurance required for regulatory compliance. In addition, ISO 14971 for risk management also provides important links to usability design steps. The most important application of ISO 14971 to usability is the provision of tools to identify and assess potentially hazardous situations. For this purpose, the standard provides some questions in annex A of ISO/TR 24971:2020 that are specifically related to usability. Namely, these questions aim to investigate the modalities through which the intended user may utilize the medical device and acquire information from it. The importance of usability is mentioned both in a direct manner, mentioning the term usability and user error and in an indirect manner, questioning whereas the medical device, for example, has a menu or whereas it displays information indicating all the elements that concur to inform the user on the status and functionality of the devices. An example answer of question A.2.31 for an MR equipment device follows. Question “A.2.31 Is successful application of the medical device dependent on the usability of the user interface?”

4.3 4.3.1

Possible answer The successful use of the medical device is highly dependent on human factors as the quality of the produced image relies on the correct positioning of the patient in the magnetic gantry by the intended user

Error Definition, Identification and Assessment What Is a User Error?

User error is any error made by the user in interfacing or interacting with a device, i.e. any situation caused by the user that leads to device use outcomes unintended by the manufacturer. It includes two distinct types of error: use error and abnormal use, as explained further on. Abnormal Use Abnormal use is a “conscious, intentional act or intentional omission of an act that is counter to or violates normal use and is also beyond any further reasonable means of user interface-related risk control by the manufacturer” (IEC 62366, 2015). Two examples are nurses muting device’s alarm by padding the siren with cloth so that it does not disturb the patient and intentionally not completing the logout in a software intended to manage the accesses and the patient’s health records in an emergency department, in order to allow a quicker triage phase. From the experience of USE-ME-D, an example is provided by the physicians who does not remove a central venous catheter that should be changed and cut it, in order to ease the insertion of the new CVC and minimize the impact on the patient tissues and organs.

64

A. Ravizza et al.

Use Error Use error is “user action or lack of user action while using the medical device that leads to a different result than that intended by the manufacturer or expected by the user” (IEC, 2015, p. 11). It is different from abnormal use because is not intentional and is part of normal use. Two examples are users not responding to a device prompt to create their password because they do not know how to input letters instead of numbers in the keypad and users inserting a patient weight in pounds instead of kilograms. From the experience of the research conducted by USE-ME-D, various types of use errors are: Medical device Software as medical devices (SaMDs)

Serological IVD test that requires capillary sampling

Use error An intended user that cannot access the patient’s search functionality in a software for ER management because he does not recognize the icon The capillary sampling method requires the user to apply the capillary’ point on the blood, and for capillary principle, the blood is attracted into the sample chamber. A foreseeable use error is a user who is not familiar with the capillary’s working principle therefore disassembles the capillary’s component making it unusable

Example

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

4.3.2

65

The Two Steps of the Usability Assessment

Usability is complex and its assessment cannot and should not be performed with a single straightforward test. Assessments regarding usability start early during the design and are iteratively performed in order to increase knowledge about user needs and expectations, interface solutions that better match those needs, risks and their mitigation measures (Fig. 4.1). The international standard defines two main steps of usability assessment: a formative (typically iterative) phase that is integrated in the development and guides further iterations of the interface and then a summative phase that is intended to validate and provide objective evidence of the safety of the latest (approved) iteration of the interface design. Formative Evaluation Step Formative evaluation is a “user interface evaluation conducted with the intent to explore user interface design strengths, weaknesses, and unanticipated use errors” (IEC 62366, 2015) (Fig. 4.2). It is generally iterative and should be performed until the manufacturer is confident that the acceptance criteria will be met during summative evaluation. Formative evaluation improves user interface design recognizing and focusing on issues in preliminary analysis. During formative iterations, it may be useful to identify early phase studies and late phase studies. Early phase studies are characterized by a higher uncertainty in the possible device variants, with many specifications not yet completely defined. At this stage,

Fig. 4.1 Usability engineering workflow

66

A. Ravizza et al.

Fig. 4.2 Steps for formative evaluation

many prototypes are still available, and they can be radically different one to another, so the employment of rapid and low-cost prototyping techniques (i.e. 3D printing, cardboard modelling) proves quite beneficial for first conceptual assessments, as in the example of the protecting splint shown further on. During early phase, the focus should be on the identification of the most probable human errors and in the definition of safe-by-design solutions in the user interface so that such errors are prevented (Fig. 4.3). The outcome of this phase is a rough prototype that is focused on the mitigation of critical risks, and that includes core features and functionalities that should be therefore tested for their effectiveness. Late phase studies are characterized by a better-defined list of requirements and of specifications, which leads to a smaller list of device variants, with potentially small but very significant differences. The outcome of this phase is the finalized device, typically manufactured in a small o laboratory scale. The finalized device features all the safe-by-design measures that have been identified in the earlier stages of development, is provided by all the required alarms and protections and is accompanied by a finalized draft of the instructions for safe use. For example, in the following pictures (Fig. 4.4), the differences between the early and late stage of a software as a medical device visualization is provided.

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

67

Fig. 4.3 Two design iterations of the same medical device intended for difficult intubations. Prototypes are obtained by additive manufacturing

Fig. 4.4 Outcome of the late phase study of a software for the intraoperative visualization of patient specific organ models. It can be noted in the image on the right the improvement of the visualization of the menu and of the information provided on the screen as result of the feedback collected during a late phase focus group (Sternini et al., 2021)

Summative Evaluation Step The summative evaluation, according to the definition given in IEC 62366-1 (IEC, 2015), is “an evaluation of the user interface that is conducted at the end of the development, on the finalized user interface, with the intent to obtain objective evidence that the user interface can be used safely”. This definition means that manufacturers shall conduct a summative evaluation to make their final evaluation of a medical device in order to determine whether or not the user interface has acceptable risk-benefit profile under a usability point of view and also to determine and confirm the expected effectiveness and where relevant, the

68

A. Ravizza et al.

clinical benefit whose endpoints can be measured by usability testing. This is particularly the case of endpoints related to the perceived quality of life and to Health Technology Assessment endpoints, when expressed in terms of efficiency of the device use. As the summative evaluation is intended to confirm all the aforementioned aspects of the medical device, the study should be designed to cover all the relevant user categories, environments and use scenarios. Example(s) In a summative evaluation planning for a medical device like a serological test kit intended for non-professional users, the usability testing should comprise multiple user groups classified by age and education level as it is foreseeable that the usability performances and risk profile may vary according to manual dexterity and the ability of comprehending the information for use. During the usability assessment of a serological test, it was apparent that young adults (19–24 years old) had significantly less difficulties utilizing the device both in terms of test execution time and number of components necessary to complete the test. Therefore, following the primary endpoint of the assessment of the usability risk profile, another endpoint that can be measured by usability scales and methods are those related to the time efficiency of use of the device and commercial estimate use-experienced based on the number of components required to successfully execute the test (Fig. 4.5). Another method of assessing the user’s learning curve is to track and evaluate their performance through the task analysis results. In the usability evaluation of the

Fig. 4.5 Comparison between first and second test execution time for a serological test. The recorded time for the first use attempt averaged to 6 min, and on the second attempt it dropped to 3 min and 16 s

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

69

Fig. 4.6 The bar diagram presents the summary of performance of users for each macrotask, after a data cleaning process. During the user test, each task completed by the user was labelled as “ok” if properly completed, “te” when impacted by a technical error, “ue” in case of use errors and “c” in case of critical errors. The summary here presented shows the improvement of the average performance of users during the device use, in terms of reduction of the prevalence of the use errors and in terms of increase of the prevalence of tasks completed correctly. (Sternini et al., 2021)

previously mentioned intraoperative software device, the simulated use was designed to provide users with three opportunities to interact with the device: firstly through a tutorial section, secondly with a hip model and finally with a liver model. The task associated with the use remained identical, and by comparing the number of correctly executed tasks and use error, it was possible to investigate the device’s learning curve (Fig. 4.6).

4.3.3

Methods and Techniques for Usability Assessment in IEC 62366

For medical devices design and evaluation, with the aim of guaranteeing safety and usability, we propose the following “Human Error Assessment” process. Use Specification It is an analysis of the main users and scenarios where the device will be used. Main functions and subfunctions of the device should be well defined and understood. In addition, all the elements that compose the user interface should be identified, including informative material, data that is presented by the device and components that should be manipulated or assembled

70

A. Ravizza et al.

The medical device use specification is a document prescribed by IEC 62366 – 1 clause 5.1 with a precise structure that, if followed, improves regulatory compliance. It is the foundation for defining and designing the user interface of the device; use specification allows identifying the known and foreseeable hazards and hazardous situations related to the user interface. At this early initial stage, it contains some necessary information: The user groups The use environment The intended medical indications of the device The use specification is refined over time, even in a post-market phase, while more knowledge is gained through user research and its level of detail and accuracy increases. New user groups might be discovered during the user research. If so, these would be added to the use specification and can trigger new user research activities. IEC 62366-1:2015 requires the use specification of a medical device to contain at least the following elements: Intended medical indication, which are conditions or disease(s) to be screened, monitored, treated, diagnosed, or prevented Intended patient population Intended part of the body or type of tissue applied to or interacted with Intended user profile Intended use environment Operating principle of the device Additional information can be present as these can be helpful to support subsequent usability engineering activities. They are as follows: Anticipated tasks of users in the operation of the medical device The set of user needs derived from the anticipated task Much or all the use specification is likely to be a key input to design and development of the device, in particular of the user interface. Moreover, as the user interface development process goes on, the use specification should be reviewed and updated as needed. User Profile and Environment Profile Human and machine are the real actors in human factors engineering (HR engineering), so it is essential to understand their characteristics and focus the attention of the entire engineering process on these two main characters. The machine, although it may be complex, has a technical description provided and supplied by the manufacturer, and therefore available. The humans, instead, for their peculiarity, can have extremely different characteristics, from the physical point of view (height, physical size, strength, stamina, dexterity, coordination), sensorial point of view (visive, auditory, tactile sensitivity, which are abilities often attributable to age) and cognitive point of view (language skills, health literacy and level of

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

71

education, memory, emotional state) in addition to their state of health, willingness, motivation and ability to learn and adapt to a new medical device. The human interface under assessment can be either used by the person who receives the clinical benefit from the use of the medical device, i.e. the patient, or the person who interacts, handling and operating, with the medical device, in different steps of the device life cycle. Users are then classified as the main user and other users, such as persons involved in transport, installation, maintenance, cleaning and disposal. In the usability study of medical devices, a core step of the HF engineering process is to define, understand, analyse and describe the characteristics of the primary user of the medical device, that is, the user for whom the intended use of the device is though defining in that way the user profile. A user profile (as defined in IEC 62366-1:2015, 3.29) typically describes and summarizes the characteristics of a single distinct user group, definable if users share characteristics (mental, cultural, physical and demographic traits in addition to expertise and type of interaction with the particular medical device) likely to influence usability. User profiles help in evaluating the general knowledge of the user on the medical device under assessment and on similar types of medical device, plus they evaluate learning abilities and learning styles. User profiles provide designers with information that they can use to shape a medical device’s user interface so that it facilitates effective medical device use for the people who are intended to use it. Task Analysis Task analysis broadens the system description, identifying all the relevant human interventions in the use of the device and where errors can occur. It also defines the automated system functions in terms of their inputs and outputs. It has the objective to understand and represent in an organized manner the set of tasks that the human element carries out in the use of the device. Task analysis may include analysis of cognitive processes and performance shaping factors (individual, social and ergonomic) influencing on the device use. Some methods (Kirwan & Ainsworth, 1992) for this purpose are divided in “task data collection” techniques and “task description” techniques. Task data collection techniques are techniques which are primarily used for collecting data on human-system interactions and which are then fed into other techniques. Some of these techniques are walk-through, talk-through, critical incident technique, observation, questionnaires and structured interviews. Task description techniques represent and structure the information collected into a systematic format, serving as a reference material for other techniques that are used in risk identification. Some of these techniques are hierarchical task analysis, tabular task analysis, timeline analysis and decision-action diagram. Human Error Analysis Including the identification of possible human errors based on previous task analysis. Human error modes can be analysed at two layers: external errors (actions) or internal errors (cognitive). Some techniques to allow human error identification are human hazard and operability study [HAZOP].

72

A. Ravizza et al.

Incident analysis or use of taxonomies and checklists of possible generic human errors might occur during the use of any device. Human error analysis includes assessment of probability and severity of each error identified so that the risk associated can be assessed and prioritized by importance. In the medical device industry, it is not always possible to associate a quantitative data to the justification of probability of occurrence, conversely to what happens, e.g. in the aeronautical industry, where there are statistics put in place to assess the occurrence of failure in a component. In situations where the probability of occurrence is too complex to define, it is best to simply use the severity criteria. Human Error Reduction and Mitigation Based on previous stages, a set of recommendations and requirements for the device design are proposed to reduce and mitigate human errors, thus guaranteeing safety and enhance usability. Combining the previous aspects analysed, it is possible to establish orders of priorities and propose strategies for the human error reduction process. The European Medical Device Regulation in the fourth general safety and performance requirement defines the adequate order of the implementation of the risk control measures, prioritizing safe-by-design risk control measures, namely, requesting manufacturers to eliminate or reduce risks as far as possible. After this first step, the regulation requires manufacturers to implement the appropriate protection measures, including alarms, to mitigate the risks that cannot be eliminated. The last resource that the regulation foresees is the provision of information for safe use and the training of the users. The order of risk control measures is of particular importance in usability studies as information for safe use is typically provided as user manuals and leaflets and as labels. It is common knowledge in the state of the art that professional users do not dedicate time to the reading of labels and manuals (Mayberry, 2007). Interviewed medical professionals using complex medical devices as laparoscopic columns referred that the user manual is accessed only in case of problems, while clinical professionals that participated to the usability testing of the aforementioned intraoperative software did not open the user manual even if it was made available to them in the simulated setting, preferring asking questions to the moderators. During usability testing of devices that require a high level of informativeness of the user manual, moderators are trained to ask participants to read IFU, as we expect that otherwise they would not refer to informative material. For this reason, the provision of information that is crucial for patient and user safety should be carefully designed and, if needed, included in the graphical user interface of the medical device, and the usability testing should always include a verification that the information is provided in a complete and timely manner to the user. In addition, an evaluation regarding clarity and meaningfulness of the information for the user should be completed, as the consistency of the information provided to users with the strengths and issues of the device experienced in the clinical practice strengthens the perception of appropriateness of the safety information, thus increasing the trustworthiness of the medical device and reducing deviations from the intended use.

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

73

A paramount example scenario where the design of the information material shall be an element of analysis during the usability assessment is if a medical device undergoes a change in the intended user: for example, going for professional use only to including lay users. The importance of assessing the clarity, completeness and consistency of the informative material for users who do not possess a background in health sciences acquires a new meaning in the risk assessment evaluation. Therefore, the design of the informative material shall be drafted according to the user educational background, physical abilities and more simply by considering the user’s feedback. For example, if in the case of a serological test kit, a one-page brochure will be deemed adequate as informative material, but if the intended users also comprise non-health operators, a necessary adaptation to better deliver the information for safe use content may require the revision of the linguistic registry, tone of the text and overall method of delivering information to the users, for example, in the design of informative icons that describe the use procedure. Moreover, the informative material may be integrated with the development of a quick guide or an explanatory video. Risk mitigation techniques are further analysed in Sect. 4.4, and a case study on risk mitigation, focused on an open-source medical device, is presented in Sect. 4.5.

4.3.3.1

Which Technique for Which Phase?

The international standard IEC 62366 (IEC, 2016) presents a series of techniques for the evaluation of usability features. Each technique has been assessed in terms of applicability to the following steps of usability: – Early formative – Late formative – Summative We divided the iterations that are commonly expected for design in these three steps, according to our experience, as they are characterized by very different approaches. In early formative, a quick-and-dirty approach that can be described as a “fail early, fail often, fail forward” approach is preferred to identify the best interface. Subsequently, in late feasibility, a more structured approach may help to refine the interface. Lastly, during the summative phase, a frozen version of the interface is validated to confirm its risk-benefit profile. Criteria to choose the most appropriate technique(s) for each phase include: – – – – –

Need to involve experts in the technology Need to involve real user(s) Time required to assess and time required to report Qualitative results (opinions) vs. quantitative results (usability scores) Depth of analysis

74

A. Ravizza et al.

As an outcome of this analysis, we have identified some techniques that we consider particularly appropriate for each step of the usability assessment. A detailed list of evaluation techniques is shown in Tables 4.1 and 4.2. Usability techniques typically require the presence of some sort of prototype for assessment. Sketches and drawings, while very useful at the very early stages of development, cannot substitute a physical model or a complete software interface. It is important to note that the use of rapid prototyping and rapid tooling techniques, including common 3D printers, laser stereolithography systems, selective laser sintering machines, mockups for the software design and low-cost replication tools such as soft moulds, for short test series proves interesting for the straightforward creation of physical models, which can be used to support most of the aforementioned techniques for usability assessment. These prototypes or models can support decision-making processes for selecting among different product ideas, on the basis of ergonomics, aesthetics, basic performance, overall usability and safety, to reach the device concept in the first stages of the development process. They can also support in the creation of a first minimally viable product for interacting with healthcare professionals, patients, families and citizens, among different types of users, so as to analyse and improve design features. Finally, their application to generating short series of the conceptual or even the final designs is key for performing benchmarking studies before investing in marketability and production. Prototypes according to different design iterations, consequence of the different decisions taken to mitigate risks and to improve usability, can, consequently, support the whole methodology and approach we propose here. This evaluation has shown that a risk-based approach is easily adapted to a resource-wise approach, as further described. During early formative, low-resource review techniques such as expert reviews, standard analysis, cognitive walkthrough are easily performed on documentation and by design experts. They do not require the participation of a large number of real users nor the availability of a finalized prototype, while low-cost replication tools may provide effective samples to boost discussion. Later stages of formative assessments may benefit from more structured techniques, such as a detailed task analysis that is linked to the FMEA technique for the identification of risk. User tests with five to ten users may be planned at later formative steps in order to allow refinement.

4.3.3.2

Which Technique for Which Device?

Medical devices belong to varied categories in terms of technology, intended use, intended users (layperson or professional), invasiveness in the human body or expected useful life, which affect design decisions in connection with usability and safety. For this reason, we have also assessed each technique presented by the norm IEC 62366 (IEC, 2016) in terms of adequateness to different kinds of devices.

Need to involve real user(s) No No No No Yes No No No Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes

Need to involve experts in the technology Yes Yes Yes Yes Yes

No Yes Yes No No Yes Yes Yes Yes No Yes No Yes No

*FMEA failure modes and effects analysis, FTA failure tree analysis

Method as per Table E.1 of IEC 62366-2:2016 Advisory panel reviews Brainstorms and use scenarios Cognitive walkthroughs Expert reviews FMEA and FTA, coupled with task analyses/functional analyses* Focus groups Functional analyses Heuristic analysis Observations One-on-one interviews Participatory designs PCA analysis Simulations Standards reviews Surveys Task analyses Time-and-motion studies Usability tests Workload assessment Low Medium Medium Medium Medium Medium High High Low Low High Medium High High

Time required to assess Medium Low Low Low High

Table 4.1 Assessment of each evaluation technique according to predefined criteria

Low Low Medium Medium Medium Medium High High Low Low High Medium High High

Time required to report Low Low Low Low High Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Qualitative results (opinions) Yes Yes Yes Yes No No Yes Yes Yes No no Yes Yes Yes Yes Yes Yes Yes No

Quantitative results (usability scores) No No No No Yes

Low High High Medium Medium Medium High High Medium Low High Medium High Medium

Depth of analysis Low Low Medium Low High

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . . 75

76

A. Ravizza et al.

Table 4.2 Assessment of each evaluation technique in proper usability engineering phase Method as per Table E.1 of IEC 62366-2:2016 Advisory panel reviews Brainstorms and use scenarios Cognitive walkthroughs Expert reviews FMEA and FTA, coupled with task analyses/functional analyses Focus groups Functional analyses Heuristic analyses Observations One-on-one interviews Participatory design PCA analysis Simulations Standards reviews Surveys Task analyses Time-and-motion studies Usability tests Workload assessment

Adequate for step Early formative Early formative Early formative Early formative and summative Late formative Early formative and late formative Early formative Late formative Early formative Early formative and late formative Late formative Late formative Late formative and summative Early formative Late formative and summative Late formative and summative Late formative and summative Late formative and summative Late formative and summative

A detailed evaluation of techniques and devices is shown in Table 4.3, in which the same technique is considered as adequate or inadequate for devices that may be apparently very similar from the usability point of view. However, when taking a better look, this is explained by the technological differences in the device. As an example, let’s consider the technique “standard review”, which proves “adequate” for very different devices such as heart valves and nasogastric tubes but is considered “adequate with reserve” for Software as a Medical Device (SaMD). This derives from the poor standardization that is still present in the SaMD sector, while traditional devices can be assessed by very consistent and complete international standards and guidelines. Also consider, for example, the technique “participatory design” that is considered “not appropriate” for traditional electromedical devices for the layperson, such as pulse oximeters, but on the other hand is “adequate” for SaMD and apps for the layperson. This again is justified by the fact that software is day by day more pervasive in our life and lay users have more chances to gain confidence with such instruments. Therefore, participatory design may allow the designers to align the medical app to users’ expectations, by allowing users to design an intuitive and userfriendly app, with a user interface as similar as possible to a consumer app.

Method as per Table E.1 of IEC 62366-2: 2016 Advisory panel reviews Brainstorm, use scenarios Cognitive walkthrough Expert reviews FMEA Focus groups Function analysis Heuristic analysis Observation One-on-one interviews Participatory design PCA analysis Simulation

Implantable, not electromedical, e.g. heart valve Yes

Yes

Yes

Yes

Yes Yes Maybe

Yes

No Yes

Maybe

Maybe Yes

Implantable, electromedical, e.g. implantable defibrillator Yes

Yes

Yes

Yes

Yes Yes Yes

Yes

No Yes

Maybe

Yes Yes

Yes Yes

Yes

Yes Yes

Yes

Yes Yes Yes

Yes

Yes

Yes

Electromedical for professional use, e.g. ECG Yes

Yes Yes

Maybe

Yes No

Yes

Yes Maybe Yes

Yes

Yes

Yes

Electromedical for use by the layperson, e.g. home thermometer Yes

Table 4.3 Assessment of each evaluation technique related to device kind

Yes Yes

Yes

Yes Yes

Yes

Yes Yes Yes

Yes

Yes

Yes

SAMD for professional use, e.g. surgical planning Yes

Yes Yes

Maybe

No Yes

Yes

Yes Maybe No

Yes

Yes

Yes

SAMD for use by the layperson, e.g. app for control of treatment adherence Yes

No Yes

Yes

Yes Yes

Yes

Yes Yes No

Yes

Yes

Yes

Not active device for professional use, e.g. nasogastric tube Yes

(continued)

No Yes

Maybe

No No

Yes

Yes Maybe No

Yes

Yes

Yes

Not active device for use by the layperson, e.g. contact lenses Yes

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . . 77

Method as per Table E.1 of IEC 62366-2: 2016 Standards reviews Surveys Task analysis Time-andmotion studies Usability tests Workload assessment

Implantable, not electromedical, e.g. heart valve Yes

No Yes Maybe

Maybe

No

Implantable, electromedical, e.g. implantable defibrillator Yes

No Yes No

No

No

Table 4.3 (continued)

Maybe

Yes

No Yes Yes

Electromedical for professional use, e.g. ECG Yes

No

Yes

Maybe Yes Maybe

Electromedical for use by the layperson, e.g. home thermometer Yes

Maybe

Yes

No Yes Maybe

SAMD for professional use, e.g. surgical planning Maybe

No

Yes

Maybe Yes Maybe

SAMD for use by the layperson, e.g. app for control of treatment adherence Maybe

Maybe

Yes

No Yes Maybe

Not active device for professional use, e.g. nasogastric tube Yes

No

Yes

Maybe Yes Maybe

Not active device for use by the layperson, e.g. contact lenses Yes

78 A. Ravizza et al.

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

4.3.3.3

79

How Many Users?

A user test shall be an appropriate probability of observing a use error caused by a design defect, and this probability is related to the number of test participants. A methodology proposed by AAMI HE75:2009, Human factors engineering – Design of medical devices, and recalled also in ISO 62366-2, represents recommendations for sample size selection in usability tests as it explains a correlation between the number of participants used in the test, called sample size, and the probability of observation. According to this methodology, the probability is cumulative, that is, it takes into account the probabilities of the individual users to commit use error, according to an exponential relationship. Consequently, confidence in the test findings of the adequacy of a user interface increases when the sample size is increased as the cumulative probability of a usability defect increases too. The methodology also suggests that large sample sizes of 100 are not always needed for usability tests, but that there are trade-offs that need to be considered when using various sample sizes. Moreover, to determine the appropriate sample size, it is important that the manufacturer considers the potential consequences of use error, the complexity of the design and degree of similarity to existing medical devices as well as the expected heterogeneity of each user group, meaning that the users group used in user test should be representative of all users and therefore should include users who differentiate in occupational background, expected knowledge and skill levels and medical device use patterns. This methodology is depicted in ISO 62366-2 in both graphical and table-based forms. Figure K.1 of ISO 62366-2 shows the exponential decrease of the cumulative probability of detected problems as the number of participants increases when sample size is greater than 10. It is generated from the following equation (Equation K.1 of ISO 62366-2): R ¼ 1  ð1  P Þ N where R is the cumulative probability of detecting a usability problem P is the probability of a single test participant having a usability problem (or the underlying usability defect probability) N is the number of test participants in the evaluation Figure K.1 of ISO 62366-2 illustrates the probability of observing at least one instance of a use error as a function of sample size and underlying population use error rates. It is important to underline that the underlying population use error rate can never be known and has to be estimated. Table K.1 illustrates how small sample sizes can be used satisfactorily to identify usability defects. It shows the cumulative probability of a usability defect being

80

A. Ravizza et al.

detected in a usability test, given the underlying probability that a single test participant would show a particular problem for a given task. This table applies to all kinds of populations and types of usability tests. According to this table, for example, if the underlying (assumed true population value) probability of a single test participant having a usability problem is 0,25, then the cumulative probability of detection is 0,82 with six test participants. To sum up, as depicted by Figure K.1 and Table K.1, many usability defects can be discovered with sample sizes in the range of five to eight participants. Moreover, it can be recommended including at least 15 participants per distinct user group in a usability test, which is in accordance with the methodology proposed in ISO 62366-2.

4.3.4

Linking Usability to Risk Identification

Each usability evaluation technique allows the designers to identify risks and potentially hazardous situations. We describe here some of the techniques identified above, in terms of capability of the assessment to be easily linked to a formal risk analysis as per ISO 14971. The preferred methods for early feasibility help the designers to identify the main risks in general terms. A quick-and-dirty approach usually identifies main risk areas and is potentially adequate to determine risk severity in terms of worst-case consequences of the risk scenario. For example, at very early stages of ideation of a electromedical device to be used in emergencies (e.g. a defibrillator), designers may already be aware of the importance of high visibility and audibility of the device, since it is expected to be used in loud, dark, confusing environments. The consequences of not being able to power on a defibrillator, simply because the start button is not visible in the dark, are easy to be estimated in terms of severity, even at early stages. While estimation of probability is not easy during early feasibility, this should not worry the designers, as they should focus only on main risks. During late feasibility, we propose designers to use a more structured method, by application of the failure modes and effects analysis (FMEA) technique. The application of the FMEA technique yields the best results if the question “what happens if. . .” is posed at each application step or phase. For this reason, we propose to assess risks by an integrated technique: designers may firstly describe the use of the device in very fine detail and then apply the FMEA technique to each step. We propose that the description of the device use is defined by a very detailed task analysis and, where applicable, also by a function analysis or use flowchart. Description of the intended use interface by a flowchart is particularly adequate for medical device software, both standalone and integrated in an electromedical device. The use of this integrated method allows for a very precise assessment of risk severity, thanks also to the possibility of obtaining a description of the chain of events that arise from a hazardous situation, for example, thanks to brainstorming or focused expert reviews. If the designers do not have enough past data or experiencebased estimations to determine risk-related probability, a user test can be very useful

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

81

to estimate probability of each hazard. If the user test is planned in this phase, the task list and use flowchart already available to designers from the FMEA activity will be used to plan and record the user tests. For each use error or use uncertainty observed during user tests, designers can determine severity and estimate probability. In late feasibility, user tests can also be integrated to other techniques to help designers to reach a new and refined iteration of the device interface. For this reason, we encourage designers to plan, after user testing session, additional sessions with the users to gather information through interviews, SUS questionnaires (Brooke, 1996) and open-ended questions intended to encourage participatory design. These interactive sessions with end users are also very useful to gather information about expected probability of each encountered error or uncertainty. For example, late feasibility studies of a surgically invasive device for professional use, e.g. catheter for angioplasty, may include the definition of a task list based on standard reviews, guidelines, state of the art and interviews. For each task, designers may identify potential hazardous situations and their consequences. Subsequently, user tests on a simulator or dummy may confirm or improve the estimated risk list; the same users may be involved after the test to discuss their errors, determine root causes and suggest improvements in the catheter shape, pliability or accessory list. During user tests in the formative phase, assessments and integrations to reports, perception-cognition-action (PCA) technique is also very common and useful; users can also be invited to express their thoughts and impressions verbally while they perform the tasks, up to comments as part of participatory design. A detailed description of the use of different techniques is given in Table 4.4.

4.3.5

Special Considerations for OSMDs

An integrated approach to usability and risk management, while complex in general terms, is typically a teamwork experience that includes the participation in the identification of the core usability risks by different stakeholders, for example, usability experts, patients and professional users. A heterogeneous team provides valuable inputs especially during the formative stage, where, for example, the clinical point of view provided by a clinical expert can add insightful consideration during the usability risk assessment process. The team then collaborates in the definition of various prototypes, that are designed by the engineer and tested by the stakeholders. Any usability assessment technique can be easily chosen as based on the available resources and adapted to the phase of design, kind of device under assessment and capabilities of the team. Designers should be provided with a complete usability toolbox and be able to choose adequate tools for each of their designs. More rightly so, open-source devices may benefit from a collaborative approach in the phases of identification of the most important risks, choice of the most

82

A. Ravizza et al.

Table 4.4 Risk identification related to each evaluation technique Method as per Table E.1 of IEC 62366-2:2016 Advisory panel reviews

Brainstorms and use scenarios

Cognitive walkthroughs

Expert reviews

FMEA and FTA

Focus groups

Functional analyses

Heuristic analyses

Observations

One-to-one interviews

Risk identification By brainstorming and review of past experiences, panels may identify potentially hazardous situations, assess probability and severity and describe the risk minimization measures present in the state of the art Designers involved in the brainstorming may identify user errors and misuse/abnormal use; designers may also identify risk control measures Designers involved in the brainstorming may identify usability pitfalls in the design and describe the hazardous situations that may arise; designers may also identify risk control measures Experts may point out usability strengths and pitfalls during their review. Usability pitfalls may then be linked to the hazardous situations that may arise; designers may also identify risk control measures. We recommend that experts answer to questions listed in the ISO 14971 Annex C FMEA technique is a very thorough method for the identification of risks. We recommend that this method is used in conjunction with a very detailed task analysis During a focus group, designers and usability experts may guide the discussions with users leading to the identification usability pitfalls in the design and describe the hazardous situations that may arise; designers may also identify risk control measures and ask participants to the focus group to comment the proposed measures A functional flow diagram, commented with the identification of machine functions and user functions, may be used in conjunction with a FMEA technique for a thorough identification of risks During a heuristic analysis, usability experts may use heuristic principles to identify and give usability scores to usability pitfalls. They may describe the hazardous situations that may arise; designers may also identify risk control measures and ask experts participating to the heuristic review to comment the proposed measures and score their capability to lower the risk During observation, designers may identify user uncertainties or errors; root cause should be discussed with the users to ensure that the hazardous situation is well understood by the designers; we believe that observation alone cannot provide sufficient information regarding risk and that it should be backed up with interviews or surveys as a de-brief activity Interviews are a very powerful tool when used in conjunction with techniques involving users that perform actual tasks on the device, from observation to cognitive walkthrough to usability tests. Interviews are best used as de-briefing activities as they allow to identify not only the hazardous situations but also their root causes (continued)

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

83

Table 4.4 (continued) Method as per Table E.1 of IEC 62366-2:2016 Participatory designs

PCA analyses

Simulations

Standards reviews

Surveys

Task analyses

Time-and-motion studies

Usability tests

Risk identification Participatory design is a very powerful tool when used in conjunction with techniques involving users that perform actual tasks on the device, from observation to cognitive walkthrough to usability tests. Participatory design should be focused on defining risk mitigation measures in terms of their perceived effectiveness PCA analyses can be integrated in the task analysis and therefore in the FMEA analysis to provide a complete evaluation of risk; it is most applicable to complex tasks and/or complex interfaces We believe that simulations among core techniques, as they can easily be adapted to all devices thanks to the use of mockups, dummies, animal models and other simulated settings. This allows the planning of all usability assessment activities in a cost-effective and ethical fashion We believe that standards review should be applied whenever an internationally recognized document is available, be it an ISO norm, a guideline from a scientific society or a local procedure. Non-fulfilment of standard requirements is a potential source of significant risk Surveys are useful tools in some situations, where the use of the medical device is difficult to observe, typically if it is used by the layperson as part of private life (contact lenses, in vitro testing for pregnancy, and so on). Surveys are not adequate to investigate root causes of hazardous situations. If properly designed, it can help identifying additional characteristics of the usability of the medical device even in terms of being consistent with heuristic principles, without requiring prior knowledge of the heuristic principles to the users Task analyses are the most powerful tools for linking usability assessment to risk management. They are best used as an input to the FMEA technique but can also be used during preliminary steps of the device design to determine the user needs and consequent testable technical requirements. Non-fulfilment of one of those requirements shall be treated as significant risk We believe that time and motion studies are most adequate to assess risk of those devices in which the time of execution is a risk control measure, e.g. if a fast execution improves patient safety (e.g. lowering chances of bacterial contamination or improving chances of patient recovery) Usability tests are a very powerful tool to determine those risks that are not identified by the designers, using techniques that do not directly involve users (such as brainstorming, standard reviews, etc.). Usability tests allow to estimate the probability of a hypothesized hazardous situation; they also allow designers to consolidate the task list (continued)

84

A. Ravizza et al.

Table 4.4 (continued) Method as per Table E.1 of IEC 62366-2:2016 Workload assessment

Risk identification Workload assessment reviews may allow designers to identify some kind of use errors related to overload or environmental distractions; we believe that this technique is most appropriate when professional users are involved, as they are more prone to burn out and also more aware of the impact of overload on their performance at work.

effective risk control measures, drafting or readable and understandable instructions for use, drafting of effective information for safe use, identification of the correct profile of user for the user testing. Additionally, an open approach to formative testing (especially for a device intended for the lay user) may involve the participation of a significant number of different users, with a wide variety of backgrounds, thus enhancing the probability of detecting a core usability risk in the early stages of the iterative design. Integration of usability assessments in the wider risk management leads to safer and more intuitive medical devices, for the benefit of patient and professional users alike. Hence, open-source design allows the designers to test their interface iteration with an ample pool of potential users, that represent the population in terms of age, scholarity, medical literacy, dexterity and other physical capabilities. The inclusion of the medical professionals and of the patient representatives in the open-source community allows for an easy sharing of different needs and difficulties related to the medical device use. The inclusion of such actors in the design process of opensource medical device allows to include in a transparent and reproducible process the requirements and the needs of the stakeholders that will eventually use the medical device, thus enabling design of medical devices that will be capable of facing these requests. In a similar way, the stakeholders could share the common issues that they experience during the device use, providing the designers with the chance to take into consideration a greater number of hazardous situations and a greater number of examples of foreseeable misuse, thus resulting in an increased depth and extension of the risk analysis, that will eventually result in safer devices and in the definition of good manufacturing practices. Open-source medical devices projects benefit from methods for easy information exchange and for gathering or summarizing relevant design decisions and inputs, considering that these may come from a wide set of collaborators with different backgrounds, countries, cultures and regulatory environments, working together for developing innovative medical technologies. To this end, co-creation platforms, such as the UBORA e-infrastructure, provide designers with ad hoc implemented tools or guided design procedures, which help to organize information regarding applicable regulations and standards, technical and economical specification and usability-related considerations, among others. To complement these guided-design processes for the OSMDs field and for healthcare technologies in general, the use of summary tables (like Table 4.5) may prove interesting. These may help to

External errors Human error 1 Human error 2 ... Human error N

User/ scenario

Tasks

Internal error (cognitive)

Performance shaping factor Severity

Probability

Error reduction and usability requirements

Table 4.5 Example of a typical template for integrating information of the process, as discussed in Sect. 4.3.4, for improved information exchange in OSMDs projects

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . . 85

86

A. Ravizza et al.

summarize potential human errors, while interacting with the medical devices, and to detail aspects including usability conditions, tasks involved, failure mode, severity and probability of different errors and related failures, and mitigation strategies and measures.

4.4 4.4.1

Risk Mitigation Techniques Risk Control Measures in Usability

Regulatory requirements (e.g. Medical Device Regulation EU 2017/745 for medical devices in Europe, Annex I) on risk minimization are clearly indicating a preferred order in the identification and selection of risk minimization measures. Safe-bydesign solutions are preferred and, if not available or not sufficient, other measures shall be added in terms of protections and alarms. Moreover, information for safe use shall be provided. Designers shall plan in eliminating the most severe risks by safe-by-design solutions from very early stages of design. To continue with the defibrillator example given for early feasibility, designers may decide to place all the interface commands on the same (front) side of the device and review standards for colours and icons at a preliminary stage of the ideation. Alarms and protections can be included during all iterations of the formative stage even evaluation adjunct to safe-by-design measures. For example, while designing a software interface of an electromedical device, the designers may allow only an “admin” user (e.g. a qualified medical professional) to set performance parameters in a predefined interval, as based on state-of-the-art clinical guidelines. Then designers may place adequate screens for password input as protection measures for the “admin” access. Moreover, for all interface screens, designers may provide information for safe use in terms of reference to the allowed interval for clinical parameters, tips to proceed to the next clinically relevant step of the therapy and so forth. It should also be noted that user manuals and labels, being an integral part of the interface and the main source of information for safe use, shall be an integral part of the risk analysis during late formative iterations. Examples During a usability test of a SaMD used in the management of the ER department, one nurse detected a critical error in the therapy administration session tab. The user refused to perform a task associated with the drug administration due to the lack of time reference associated with it. The drug in question was an antibiotic. Even though there was a work plan window where the time references for some medicines are reported, the one concerning the antibiotic was missing as it was wrongly classified as an extemporaneous therapy. Considering that the effectiveness of antibiotic therapy relies on the correctness of the administration time, this

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

87

Fig. 4.7 Example of safe-by-design risk control measure. On the left, a screenshot where the drug reference was missing, on the right the modified version where the drug administration time is correctly reported

Fig. 4.8 On the left picture is displayed the icon associated with a patient that has an allergy, the special precaution is indicated with a red exclamation point. In the centre scenario, the allergy screening of the user was not yet conducted and therefore the allergy anamnesis is unknown; the red exclamation point is no longer present next to the icon. In the picture on the right, the allergy icon is paired with a green tick sign indicating that the patient does not have allergies. By providing users with informative text box reference, we have minimized two risks: not recognizing the allergy icon, being misled on the allergy anamnesis status

characteristic of the user interface was considered critical and required a risk control measure. Another typical source of use error in a SaMD is a misleading design choice for a digital icon. During the usability test for the same software, a substantial number of users were misled by both the allergy’s icon shape (a green leaf, visible in Fig. 4.7) and the text reference associated with it. Namely, where the green leaf icon was paired with a red exclamation mark it indicated that no allergies were reported. This design choice created some confusion for the users as in the Italian translation this could either mean that the user did not have allergies or that allergies were not reported yet. This confusion could potentially have a critical impact on the patient’s level of health and therefore was modified by implementing a different text reference that clearly distinguished among all the possible scenarios (Fig. 4.8). Another safe-by-design example solution can be illustrated for a medical sterilizer device. During the normal use flow, one of the more critical tasks is the safe and effective closing of the sterilizer’s door. To prevent hazardous scenarios of entrapped limbs, the digital user interface was provided with a safe-by-design solution that requires both hands to be on the screen during closing (see Fig. 4.9)

88

A. Ravizza et al.

Fig. 4.9 Example of positioning for the door closing, with two buttons to be activated at the same time, was deemed an adequate risk control measure for safety. Moreover, the door has no sufficient strength to crush hands. No additional risks or uncertainties detected

4.4.2

Summative as Part of Device Validation

The goal of device validation is to determine if the device is adequate for its intended purpose and to confirm its estimated risk-benefit profile. No major modifications are expected at this phase. While not all parts of the interface may be subject to summative evaluation, designers should plan to validate most, including all the critical ones. For example, summative assessment of the interface for the assistance and maintenance personnel of an electromedical device, when personnel are directly trained by the device legal manufacturer, may not be needed. The choice of which parts of the device interface shall be tested and of which users should be involved is based on risk assessment and should be clearly justified. In an example from our team, the evaluation of the usability of kits for haemodialysis, different aspects of the kits are evaluated for the identification of the most critical components. In particular: • Kit complexity in terms of number of pumps and lines • Kit configuration depending on the therapy in terms of the number of clamps, bags, pouches, critical tasks during normal use, tasks for the kit assembly and filters that have to be assembled on the kit • Criticality of the use scenario, defined by the environment, user and patient condition The decision process was structured as follows:

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

89

1. The most complex kits are identified, and these are selected as candidates for the usability evaluation, being sure of covering all possible therapies. 2. For each therapy, the worst-case scenario is selected, using the most complex kit available for the specific therapy. Finally, adequacy of the selected worst-case scenario is verified, ensuring that the use scenarios properly cover all environments and users. By including the two patients’ conditions, the usability assessment will also investigate any differences in use-experience for patients in acute and chronic conditions. We propose to plan the summative evaluation by mirroring activities of the late formative step, on the final and frozen iteration of the interface. For this reason, a complete task analysis should be available and checked for coherence to the user manual or instruction leaflet. Moreover, if applicable to the kind of device, also a complete use flowchart should be available. Summative evaluation should be performed with real users and in a very well simulated or real use environment, depending on device kind and ethics considerations. During user tests, additional techniques may be integrated to determine the length of time needed for each task (by time-and-motion studies) and the workload of the user. It should be noted that, while very adequate for summative activities, time of use and workload assessment are not easily evaluated during formative tests. The interface is still under modification, and, more often than not, the tasks may be interrupted for clarifications and comments from the users. This is most common if the users that participate to the user test are aware that the device is under development and not under validation: most users are very keen to provide their feedback and opinions as part of participatory design activities. Interrupted and commented tasks disrupt the workload assessment and the time estimation. The outcome of the summative step is the confirmation of all parts of the device interface, including the information for safe use. No additional risks should be encountered and all the foreseen risks should be confirmed in terms of severity and probability. Risk control measures should be formally reviewed for final implementation and effectiveness and the risk-benefit profile confirmed. The most common technique for summative evaluation is the user testing in either real or simulated conditions. The simulation is completed in the most representative environmental condition available. Nevertheless, real patients are typically not included, as it would be unethical to involve patients and therefore exposing them to risks with a device which is still under technical validation. The moderators should be trained to recognize possible errors and possible hazardous situations while they observe the users completing tasks with the medical device. Usually, moderators ask test participants to think aloud to better understand the cognitive process underlying each user action, by application of the PCA technique. Nevertheless, the moderators should be completely impartial during the execution of the test and should not offer advice, corrections or any additional information regarding the device and its use, as it would impact the performance

90

A. Ravizza et al.

of the test. The user will be tempted to ask questions to moderators, as users will see the device for the first time during the usability test. The moderator should be able to evaluate the difficulties faced by the user and understand if they are blocking problems or not. In this latter case, the moderator can help the user in order to complete the usability testing, while in case of a simple user uncertainty, no advice or help should be provided. All the questions posed by the user should be answered only after the complete execution of the test and should be accompanied by additional questions posed by the moderator, aimed at a deeper knowledge of the root causes of the user uncertainties and errors. The questions shall never include a moderator opinion or perception but should only invite the user to express their thoughts and feelings. These unstructured interviews could also be intended to collect comments and additional suggestions for future improvements, while more structured questionnaires could be provided to users in order to assess and evaluate the perception of the users regarding a specific aspect of the user interface.  To illustrate how combined risk identification techniques may be applied to generate design iterations for risk minimization, aiming at safe-by-design solutions, the following section presents a case study that explains how FMEA and finite element method (FEM) simulations are synergically applied to detect and prevent risks.

4.5

Case Study: Risk Mitigation in an Open-Source Face Protecting Splint

Chapter 9 of this handbook deals with the personalized design of OSMDs and presents as case study, in Sect. 9.3, the personalized design methods and resources for creating a conceptual face protecting splint, a medical device for safe sport practice during the recovery of a broken nose. These devices are commonly developed on purpose for athletes (i.e. football/basketball players) but can benefit from open-source methods for reaching a larger amount of users. Now, such concept is analysed from a usability and risk minimization perspective and redesigned accordingly, for illustrating some of the techniques presented in this chapter: FMEA and finite element method (FEM) simulations are synergically applied. To promote safety, in medical device design, the use of the Failure Modes and Effects Analysis technique and related methodologies proves interesting for re-engineering purposes, as already advanced among techniques presented in previous sections. The technique works following a “what happens if?” approach focusing on the functions that the medical devices should perform and related possible failures due to incorrect design, processing, performance and usability. For each

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

91

Fig. 4.10 Results from FMEA analysis and redesign or reprocessing proposals. Probability (P), detectability (D) and severity (S) are assessed from 1 to 10, and the priority risk index (PRI) is calculated PRI ¼ PDI. Impact of improvements is quantified

possible failure and for each related effects, a priority risk index (PRI) is calculated by multiplying the probability (P), detectability (D) and severity (S) of the failure. More probable, more severe and more difficult to detect wrong operating conditions or failures get higher values, typically in a 1–10 scale. Once the PRIs are calculated, re-design or re-engineering decisions are proposed, and the PRIs are calculated in the new situation. The methodology is applied to present face protecting splint, and results from the FMEA analysis are summarized in Fig. 4.10. We have focused on the more relevant failures according to their PRIs and tried to reach situations, in which the overall PRI is below 30, considered by us a minor risk situation. Among design and processing (manufacturing and post-processing) changes proposed to the original concept and preliminary prototypes, it is necessary to highlight the following: (1) the free region surrounding the eyes has been enlarged to avoid potential harms; (2) the incorporation of flexible pads in the zone in contact with the forefront and cheeks has been proposed, so as to avoid direct skin contact, eventual abrasion and better performance in terms of impact; (3) post-processing after autoclaving is demanded, for leading to softer outer surfaces of the composite laminate, and quality check with randomized trials is proposed; and (4) in special cases, use a textile or PDMS cover applied to the whole splint to completely encapsulate it. The overall structure of the redesigned splint can be shown in Fig. 4.11. FEM simulations are employed to check the adequate number of plies to be laminated and autoclaved to achieve the final splint: NX 8.5 (Siemens PLM

92

A. Ravizza et al.

Fig. 4.11 Proposed design after FMEA analysis: Improved comfort and visibility together with minimized risks for eyes

Solutions) is used both as CAD modelling software and for the FEM simulations aimed at such geometry optimization (mainly thickness optimization, as related to the number of plies used, for enhanced production, while withstanding a desired level of mechanical performance). The NX Nastran structural solver of the advanced simulation module of NX-8.5 has been employed to perform the mechanical static analyses presented in the following images. The final design of the splint, based on laminating 4 plies, 6 plies and 8 plies of pre-impregnated carbon fibre mats, has been evaluated. In all cases, a two-dimensional mesh has been used, due to the laminar geometry, which is also the adequate mesh for defining the properties of the different plies and for applying and studying the impact of ply number on final performance. Material properties are applied to the plies according to manufacturer’s data and to common CFRP (carbon fibre reinforced polymer) datasheets, as estimation for non-available data. Each ply is simulated as orthotropic material with Ex ¼ Ey ¼ 130 GPa, Gxy ¼ 4.2 GPa, Poisson ratios of 0.25 and specific mass 1500 kg/m3. Boundary conditions to represent contact with forefront and cheeks are applied and a load of 200N perpendicular to the nose region (Fig. 4.12). This static loading is a quite demanding condition, which would lead to a nose breakage upon an unprotected case. Results are checked with the support of post-processing tools to highlight zones deforming beyond 1.25 mm, considered as a reasonable limit for avoiding additional damages to the injured nose of the patient. Regions beyond 140 MPa, which is the compressive strength of the ply in the weakest direction and just above the interlaminar strength of the selected plies, are also detected and highlighted in red. Results are discussed further on. Figure 4.13 presents the post-processing of the performed FEM analysis. The displacement fields for the different face protecting splint configurations are shown. Optimization of the number of plies is studied, and displacement results for different

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

93

Fig. 4.12 Pre-processing step for finite-element modelling of the face protecting splint. First 2D mesh is applied and the composite material laminate is defined, by introducing the properties of the different plies sequentially laminated and finally autoclaved to obtain the final laminate or composite splint. Then contact with forehead and cheeks is applied, as well as a 200N force upon the nose region, as boundary and loading conditions

splints, laminated using 4 plies (upper), 6 plies (middle) and 8 plies (lower image), are presented. Displacements higher than 1.25mm are marked in red. The improved results by increased number of plies (and manufacturing time due to lamination by hand upon the moulds) can be appreciated. Finally, 8 plies are considered adequate for the solicitation, and from the FEM simulation, the improved design of Fig. 4.12 is correct. Figures 4.14, 4.15, and 4.16 present the maximum principal stress results, as selected from alternative stress fields calculated, for the face protecting splints

Fig. 4.13 Post-processing of FEM analysis: Displacement fields for different face protecting splint configurations. Optimization of the number of plies and displacement results for splints laminated using 4 plies (upper), 6 plies (middle) and 8 plies (lower image). Displacements higher than 1.25 mm are marked in red. Improved results by increased number of plies can be appreciated, and 8 plies are considered adequate for the solicitation

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

95

Fig. 4.14 Post-processing of FEM analysis: Maximum principal stress result for face protecting splint obtained by laminating 4 plies of carbon fibre pre-impregnated fabric. Selection of results for different representative plies of the laminate: Outer ply (upper image), central ply (middle image) and inner ply (lower image)

96

A. Ravizza et al.

Fig. 4.15 Post-processing of FEM analysis: Maximum principal stress result for face protecting splint obtained by laminating 6 plies of carbon fibre pre-impregnated fabric. Selection of results for different representative plies of the laminate: Outer ply (upper image), central ply (middle image) and inner ply (lower image)

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

97

Fig. 4.16 Post-processing of FEM analysis: Maximum principal stress result for face protecting splint obtained by laminating 4 plies of carbon fibre pre-impregnated fabric. Selection of results for different representative plies of the laminate: Outer ply (upper image), central ply (middle image) and inner ply (lower image)

98

A. Ravizza et al.

obtained by laminating 4, 6 and 8 plies of carbon fibre pre-impregnated fabric. Selection of results for different representative plies of the laminate are included for additional information, again showing that the 8-ply laminate performs the best.

4.6

Conclusions

An integrated approach to usability and risk management, while complex in general terms, can be easily adapted to the design step, kind of medical device under assessment and available resources. Designers should be provided with a complete usability toolbox and be able to choose adequate tools for each of their designs. Integration of usability assessments in the wider risk management leads to safer and more intuitive medical devices, for the benefit of patient and professional users alike. While our group has tested this method in multiple instances, we wish that it would be used widely. With more experience, this method can be refined, adapted to different cultural settings and various technical skills and updated with devicespecific tools. Moreover, these techniques may be integrated with the risk mitigation measures required for the adequate management and protection of patient data. To illustrate the described methods and techniques, several case studies for electromedical devices, invasive devices, standalone software and protecting splints, among others, have been presented. These examples may serve to give proof of the flexibility and scalability of this method. Besides, some key aspects linked to applying these methods to projects emerging from open-innovation schemes and delivered as open-source medical devices and technologies have been put forward. Interestingly OSMDs provide the opportunity to include a wider range of potential users in the steps of participatory design and iteration testing. In the authors’ opinion, the methods of OSMDs can be easily integrated in the usability design process, thanks to their iterative approach. Acknowledgements Authors acknowledge the UBORA “Euro-African Open Biomedical Engineering e-Platform for Innovation through Education” project, funded by the European Union’s “Horizon 2020” research and innovation programme under grant agreement No. 731053. Adrián Martínez Cendrero and Rodrigo Zapata Martínez are acknowledged for their support with personalized design procedures linked to the face splint case study.

References Brooke, J. (1996, September). SUS – A quick and dirty usability scale. Usability Evaluation in Industry, 189, 4–7. Council of the European Communities. (1993, July 12). COUNCIL DIRECTIVE 93/42/EEC of 14 June 1993 concerning medical devices – ANNEX I. Official Journal of the European Communities, 36, 13–18.

4 Design of Open-Source Medical Devices for Improved Usability and Risk. . .

99

European Parliament and Council of the European Union. (2017, May 5). Regulation 2017/745 on medical devices, amending Directive 2001/83/EC, Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and repealing Council Directives 90/385/EEC and 93/42/EEC. Official Journal of the European Union, 1–175. International Electrotechnical Commission. (2015). IEC 62366-1:2015 Medical devices – Part 1: Application of usability engineering to medical devices (1st ed.). International Electrotechnical Commission. (2016). IEC TR 62366-2:2016 Medical devices – Part 2: Guidance on the application of usability engineering to medical devices (1st ed.). International Organization for Standardization. (2007). ISO 14971:2007 Medical devices – Application of risk management to medical devices (2nd ed.). International Organization for Standardization. (2016). ISO 13485:2016 Medical devices – Quality management systems – Requirements for regulatory purposes (3rd ed.). Kirwan, B., & Ainsworth, L. (1992). A guide to task analysis: The task analysis working group. Taylor & Francis. Mayberry, J. F. (2007). The design and application of effective written instructional material: A review of published work. Postgraduate medical journal, 83(983), 596–598. https://doi.org/10. 1136/pgmj.2006.053538 Ravizza, A., Díaz Lantada, A., Ballesteros Sánchez, L., Sternini, F., & Bignardi, C. (2019). Techniques for usability risk assessment during medical device design. In Proceedings of the 12th International Joint Conference on Biomedical Engineering Systems and Technologies – Volume 1: Biodevices. ISBN 978-989-758-353-7, pp. 207–214. https://doi.org/10.5220/ 0007483102070214. Sternini, F., et al. (2021). Usability assessment of an intraoperative planning software. In Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies 483–492 (SCITEPRESS – Science and Technology Publications. https://doi.org/ 10.5220/0010252904830492. U.S. Food & Drug Administration. (2016, February 3). Applying human factors and usability engineering to medical devices. Guidance for Industry and Food and Drug Administration Staff. Retrieved from internet website (last access in December 2021). https://www.fda.gov/ downloads/medicaldevices/.../ucm259760.pdf

Chapter 5

Human Centered Design Principles for Open-Source Medical Devices Elizabeth Johansen, Mark Fisher, Andrés Díaz Lantada, Carmelo De Maria, and Arti Ahluwalia

5.1

Introduction

According to ISO 9241-210:2019, “human-centered design (HCD) is an approach to interactive systems development that aims to make systems usable and useful by focusing on the users, their needs and requirements, and by applying human factors/ ergonomics, and usability knowledge and techniques. This approach enhances effectiveness and efficiency, improves human well-being, user satisfaction, accessibility and sustainability; and counteracts possible adverse effects of use on human health, safety and performance.” However, in a way, human-centered design integrates and transcends the concepts of usability, safety, and ergonomics, entering the realm of hedonomics, searching for more satisfactory user experiences, and considering human desires and sociocultural aspects, as intrinsic to successful product engineering. Medical devices, for their intimate interaction with the human body, for their combined physical and psychological effects, and for their extreme relevance for health and

E. Johansen (*) Spark Health Design, Hanover, MA, USA Design that Matters, Redmond, WA, USA e-mail: [email protected] M. Fisher Department of Biomedical Engineering, Northwestern University, Evanston, IL, USA Mission Product Development, Highland Park, IL, USA A. Díaz Lantada Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain C. De Maria · A. Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_5

101

102

E. Johansen et al.

well-being, are very singular products, to which the application of human-centered design principles is of special relevance. Furthermore, HCD is essentially linked to reliability: apart from being usable, safe, and user-friendly, medical devices must dependably work within a given environment, workflow, and skill set. The ability to adapt to users’ needs in very different settings is crucial for making medical technologies succeed in the majority of the world and make truly global impact. To cite some examples, the need for privacy may affect the external shape of a device, to make it less visible and minimize potential stigmas (Krista Donaldson’s D-Rev TED talk); cultural diversity may influence use and even lead to unexpected risks, so different configurations may be required for international markets; healthcare technologies designed for children can importantly benefit from improved playability, more user-friendly interfaces, and supporting devices for explaining procedures through role plays (Philips MRI for children example); or the wish for personalization, even with both ergonomic and aesthetic purposes, can radically affect the selected manufacturing technologies and thus reconfigure the supply chain. The present chapter complements aspects put forward in Chap. 3, which focused on involving patients, healthcare professionals, and users, in general, for identifying medical needs, as starting point for designing relevant medical devices. The topics from this chapter are also related to those from Chap. 4, which presented innovative approaches toward safer medical devices, by linking usability enhancement techniques with risk minimization methods. The human-centered strategies explained here connect with those topics but detail in more depth how human factors apply to the engineering of novel medical devices and how they are deeply interwoven within the whole life cycle of healthcare technologies. Along the chapter, the most relevant internationally recognized standards, principles, and methods linked to HCD are detailed, together with special considerations and methods/techniques for their application to medical devices. Such principles, methods, and techniques are described following a medical device development workflow, which is common to reliable engineering design methodologies, including steps closely related to those described in ISO 13485 and in the UBORA-CDIO model for open-source medical devices (see Chap. 2). In essence, team formation, detection of user needs, synthesis, ideation, prototyping, assessment, and detailed design stages are covered and enhanced through the application of HCD strategies. The benefits of HCD for developing medical devices are illustrated through a case study, which describes the complete engineering of a phototherapy device for treating jaundice in under-resourced settings in the majority of world1, 2. The Firefly device was developed with East Meets West Foundation, MTTS and Design that Matters (DtM) under an agreement that waived all confidentiality and left the parties free to use the resulting intellectual property without restriction. Traditional

1

The principles of human-centered design and the case study have been previously published as an open-source paper under a Creative Commons CC-BY license: Fisher, M. and Johansen, E. “Human centered design of medical devices and diagnostics in global health”, Global Health Innovation 3(1), 1–15, 2020. The paper is reprinted here for illustrating Sects. 2, 3, and 5, with permission from the journal, considering that human-centered design strategies apply to all types of healthcare technologies, including OSMDs. 2 The term “majority world” refers to the 80% of humanity who live on $10 or less a day.

5 Human Centered Design Principles for Open-Source Medical Devices

103

frameworks for confidentiality, product licensing and royalties were incompatible with an open collaborative design process that involved hundreds of students and professional volunteers, dozens of domain experts in the US and in emerging markets and an overall project focus on global health impact. In 2020, MTTS announced an agreement with Africa Health Supplies to expand manufacturing of the full MTTS jaundice kit in Ghana, including Firefly phototherapy. There are no patents protecting the design, so it can be replicated without fearing a penalty. It is important to highlight that HCD principles are universal and apply to all areas of product engineering and to all kinds of medical innovations. The beneficial synergies between human-centered and open-source approaches, for developing accessible and socially relevant healthcare technologies, are also discussed.

5.2

Principles and Impacts of Human-Centered Design and Related Methods

The word design often evokes images of fashion and art. In HCD, the word design denotes the act of intentionally creating something meaningful. Common industry practices of human factors, usability engineering, and user experience are encompassed within HCD (Harte et al., 2014). However, HCD expands the focus beyond the direct user to inquire into the motivations, preferences, and viewpoints of all stakeholders who will affect the distribution, uptake, and impact of a product. In the context of medical technology (medtech), the array of stakeholders can be wide-ranging. In its most impactful form, HCD asks practitioners to reach beyond the question of whether a device will be safe, effective, and sellable, to ask whether a product will deliver intended impacts in communities. Some call the application of HCD to positive social impact “Design for Social Impact.” Authors propose that building upon the knowledge and perceptions of communities who will adopt devices is paramount to achieving significant positive impact. With medtech designed for particular contexts and specific users, there is potential to improve long-standing global health challenges, ranging from reducing childhood mortality from pneumonia to managing chronic diabetes. In the authors’ experience, HCD contains a powerful set of methods and mindsets to achieve the goal of high-impact technology. Using HCD, teams can: • Uncover the essential needs of diverse stakeholders involved in the success of medical technology for global health • Generate creative solutions • Evaluate product alternatives quickly and effectively without months of testing • Arrive at an optimal solution quickly that satisfies the critical needs of direct users and other stakeholders The HCD methods detailed in this chapter have roots in multiple design traditions described well by Giacomin (2014), in particular, the traditions of ethnography, participatory design, co-design, contextual design, and empathic design which are presented and described by Steen (2011).

104

E. Johansen et al.

The practice of HCD became adapted to business by companies such as IDEO, Frog, Continuum, and others in the 1980s and 1990s (Szczepanska, 2017). Since then, the methods have been widely adopted among organizations such as healthcare system Kaiser Permanente (Kachirskaia & Mate, 2018); medical device companies Johnson & Johnson and Roche Diabetes Care (Nilsson & Sheppard, 2018); and global health funders such as USAID and the Bill and Melinda Gates Foundation (Cheney, 2018). In addition, the HCD approach and some of its methods are recognized by the International Standards Organization (ISO) in 13407: “Human centered design processes for interactive systems,” as well as in 241-210: “Ergonomics of human-system interaction-Human-centered design for interactive systems,” to cite a couple of examples (see: ISO, 1999, 2019). Evidence of human factors and usability engineering are increasingly required for medical device regulatory approvals including US FDA approval and international CE marking (ANSI/ AAMI, 2018; US FDA; 2016, ISO/IEC, 2015).

5.2.1

The Impact of Human-Centered Design

Many global health practitioners seek proof that HCD will lead to greater positive impact for under-resourced communities. Bazzano et al. (2017) analyzed 21 publications from peer-reviewed and grey literature sources selected because they discussed aspects of HCD as connected to health outcomes. One of their conclusions was that existing publications do not quantitatively study the connection between HCD and health outcomes. Bazzano et al. (2017) also noted other gaps in existing literature, including a lack of publications describing a full project cycle, little description of HCD methodologies used, and absence of gathering stakeholder feedback in context. The 21 selected publications focused on software applications, clinical decision tools, and websites used in global health, but no medical devices or diagnostics. In this chapter, we present sources illustrating the impact of HCD in several domains which may also be considered relevant for health impacts. There are analyses indicating that user-experience design and human factors engineering have impact on the financial success of products. The primary argument is that a small investment of time and money in early application of HCD creates outsized savings later in development (Mantei & Teorey, 1988; Bias & Mayhew 2005; Rajanen, 2003; Conklin & Yakemovic, 1991; Karat, 1993; Mauro, 1994). For example, Karat (1993) references a study of a software product for which $20,700 spent on usability resulted in a $47,700 return on the first day the improvements were implemented, while for another system, $68,000 spent on usability resulted in a $6,800,000 return in the first year. In addition, the US FDA medical device regulatory agency studied the impact of human factors engineering on medical device design, issuing this statement (US FDA, 2011):

5 Human Centered Design Principles for Open-Source Medical Devices

105

With increased safety, the likelihood of your incurring expenses associated with product recalls or liability is reduced; when Human Factors Engineering/Usability Engineering approaches are used in the design of devices, particularly if the perspective of users is taken into account, the overall ease of use and appeal of a device can simultaneously be enhanced.

Notably, these analyses concern themselves with the financial success of a product and not positive impact for communities. Product financial success is not always correlated (and can be inversely correlated) to positive impacts for communities (Jewett, 2019). However, in the authors’ experience, designing products with communities saves time and money in costly redesigns, an extremely vital benefit to underfunded global health initiatives that often do not make it from the prototype to a final product (Johansen, 2018). Indeed, HCD may be used to generate multiple medical device and diagnostic concepts and arrive more quickly at a solution that is desirable to stakeholders. Both correct use and positive product perception are necessary for safe and effective healthcare technologies. We believe that HCD increases chances of positive impact through better honing the design concepts to meet stakeholder needs early in the design process.

5.2.2

The Human-Centered Design Process

When properly implemented, HCD permeates all phases of product development and involves a wide range of skills. From forming the right team to product launch, HCD increases the chances of successful product uptake in healthcare. As shown in Fig. 5.1, HCD starts by considering what is desirable to stakeholders. Then the team must find opportunities at the intersection of viable businesses and feasible technologies. While technical feasibility and business viability are not discussed in detail in this paper, engineers and scientists will typically understand how to address technical feasibility, but not business viability. Teams founded by engineers and scientists should also include business experts as advisors or members of their team early in the process. Finding a solution that is desirable, feasible, and viable is critical to the success of any product for medical or consumer use. The HCD process helps to ensure all user needs are identified and addressed before the lengthy and expensive process of detailed development. Figure 5.2 shows a simplified overall product development process. The items in gray describe steps that include HCD processes which are described in detail below. As practitioners with experience in applying HCD to over 20 global health medtech projects targeting under-resourced communities, we provide documentary evidence of the value of HCD. The methods and benefits of HCD using the case study of Design that Matters’ Firefly phototherapy to treat newborn jaundice, as well as short examples from other projects, are illustrated. Elizabeth Johansen was

106

E. Johansen et al.

Fig. 5.1 The human-centered design approach. (Ideo.org, 2015)

Fig. 5.2 Human-centered design process and how it integrates with product development

Director of Product Development at Design that Matters during Firefly development. Mark Fisher was Director of Engineering for Northwestern Global Health Foundation focusing on development of point-of-care (POC) infant HIV antigen tests and a POC molecular platform product targeted at HIV viral load, mTB, and hepatitis C. Beyond medical devices, HCD can also be used to design diagnostics (Johansen, 2018, Linnes et al., 2017), and it can be used to design intangible innovations such as systems, services, and processes within global health (USAID, 2019). The methods recommended in this paper are particularly directed at the design of medical devices and diagnostics for global health.

5 Human Centered Design Principles for Open-Source Medical Devices

5.3

107

Human-Centered Design of Healthcare Technologies and the Case Study of Design That Matters’ Firefly Phototherapy for Newborn Jaundice

Globally, over 100,000 newborns die, and more than 50,000 are left permanently disabled due to jaundice each year. The burden of death and disability due to jaundice lies primarily in Sub-Saharan Africa and South Asia (Olusanya et al., 2018). An estimated 10% of all newborns worldwide need phototherapy treatment to avoid the ill effects of jaundice (Bhutani et al., 2013). Phototherapy is a minimally invasive treatment that requires shining as much blue light as possible on the newborn’s skin. This enables newborns to pass excess bilirubin out of their body safely through urine and stool so it will not cross the blood-brain barrier (Maisels & McDonagh, 2008). Nonprofit organization Design that Matters (DtM) developed Firefly phototherapy in collaboration with Medical Technology Transfer and Services (MTTS) and East Meets West Foundation to address the need for better tools to treat newborn jaundice in under-resourced hospitals (Johansen, 2018). At present, Firefly is estimated to have treated over 200,000 newborns with jaundice in over 20 majority world countries throughout Africa, Asia, and the Caribbean.

5.3.1

Team Formation and Background Research

As can be seen from Fig. 5.2 above, identifying the right expertise is one of the most critical elements of designing a successful product; this is where the HCD process starts. Required expertise typically includes, but is not limited to, design, clinical, scientific, engineering, process development, logistics, manufacturing, regulatory, quality, and business expertise. All the team members need to approach the problem with an open mind regarding the underlying user needs and challenges. The best product requires collaboration; no one individual has all the experience and expertise needed to design the best possible product. In order to instill HCD into the development process, all team members must have an awareness of HCD, and at least one team member must become an HCD champion who is proficient in HCD practices. Human factors engineers, usability engineers, industrial designers, user-experience designers, and product designers are professionals who are likely to have past experience in HCD techniques as they apply to medical devices. However, the HCD champion could be from any discipline: scientist, engineer, clinician, business, or others. The HCD champion must lead the process for when and how stakeholders are engaged in iterative feedback and how this feedback loops feed into the product development. HCD closely resembles the scientific method including steps of developing a base understanding (secondary research, user research, observations); forming a hypothesis (product concept); testing the hypothesis (user testing); analyzing the results (synthesis, frameworks for design); revising the hypothesis (iterate the design); and

108

E. Johansen et al.

retesting as required (Newton, 1686). One key difference between HCD and the scientific method is that the latter uses more qualitative than quantitative techniques to test hypotheses. Applying HCD in medical device development therefore requires engineers and scientists to develop skills in collecting and analyzing qualitative data. While the team is forming, background research grounds the project planning process and assists the team in identifying the appropriate team members, users, stakeholders, available technologies, and existing approaches.

5.3.1.1

Team Formation and Background Research: Project Firefly

To bring Firefly newborn phototherapy from an idea to impact, DtM convened a diverse team. The company included engineers and designers. The lead engineer with deep HCD experience led the HCD process. Experts from manufacturing and commercialization partner MTTS Asia provided insights around existing newborn equipment successes and challenges and prepared to finalize the product design and introduce the technology to the market. Newborn health program implementation experts at NGO East Meets West Foundation provided access to doctors, nurses, and maintenance technicians at multiple hospitals for interviews and prepared to incorporate the device into their newborn health programs. Over 100 volunteers in academia and industry spanning engineering, design, business, and healthcare contributed to stakeholder interviews, brainstorming sessions, design reviews, and concept exploration. Reduced rate engineering consultants with design for manufacture experience were engaged in detailed design. During initial background research, the team identified a range of existing approaches to jaundice treatment, including traditional overhead phototherapy lights and flexible phototherapy blankets.

5.3.2

Stakeholder Observations and Interviews

When considering the range of user needs, the team first needs to identify the range of stakeholders. For example, let us consider a simple rapid diagnostic test like an HIV rapid test. Stakeholders certainly include patients and their families, lab technicians, and nurses. Stakeholders also include Ministries of Health, central laboratories, purchasing organizations, logistics (tests and related items like lancets, gloves), people reporting results, and others. A good design will account for the needs of all stakeholders. If there is not an existing product or process, consider something with similar use steps. For instance, in the development of a new point-ofcare molecular diagnostic for tuberculosis, the development team considered a modern and complex existing system like the GeneXpert used for centralized testing, as well as smear microscopy, the current point-of-care, gold-standard culture solution. The team considered all phases of the diagnosis from providing a sample through result (Doerfler et al., 2018; Fisher et al., 2017).

5 Human Centered Design Principles for Open-Source Medical Devices

109

User observation is a qualitative approach. The goal is not to have a statistically significant result but rather to uncover the range of important, unstated needs in more depth with a limited number of observations. HCD research is hence structured to engage a range of stakeholders and contexts that might influence the purchase, use, and maintenance of medical technologies. With HCD, designers focus on more than identifying usability issues. However, a helpful benchmark for number of stakeholders to engage comes from Virzi (1992) and Nielsen & Landauer (1993). They found that evaluations with only five users would identify 80% of usability issues in a simple piece of software. According to authors’ experience, it has been found that five to ten sets of observations in different locations is normally sufficient, if the locations are selected to represent extremes (e.g., including sites that range from rural to urban, busy to slow, and large to small). Similarly, speaking with five to ten stakeholders at each site, who represent a variety of roles and experience levels, usually gives a sufficient picture of the primary needs to satisfy with the product. The intent is to efficiently identify most needs, by covering a range of users and stakeholders with a manageable number of observations. Often, the most important information comes from questions one would not think to ask. For this reason, visits typically combine observations with open-ended interviews. The visits should be done at the location of intended product use wherever possible. Observations include understanding workflows, hand-off points, and detailed processes. The most powerful method to document observations is through pictures and video with permission of patients and healthcare workers. If photo and video are not possible, observations can be documented in notes. This allows the information to be effectively shared with other team members and partners during the multi-year device development process. For interviews, it is important to prepare with potential questions to ask, but the interview should not follow a rigid script. The most informative interviews are conversational, normally starting with very open-ended questions (e.g., “Describe your workflow,” “Describe a typical day,” “What is the biggest problem in the process?”). When conducting user visits, the purpose is always to uncover the core underlying needs rather than simply recording an answer to a question. Asking follow up questions like “Why do you do that?” can provide unexpected answers. The interviewer should also feel open to asking spontaneous questions based on the conversation. A 30 min interview will often provide all the desired information and more, just by beginning with only three or four open-ended questions.

5.3.2.1

Stakeholder Observations and Interviews: Project Firefly

To comprehensively gather information relevant to the design of Firefly phototherapy, DtM identified key stakeholders through background research and interviews with partners MTTS and East Meets West Foundation. Device end users included newborns with jaundice; their parents and grandparents who may stay near the device and turn it on and off during breaks for breastfeeding; nurses who will

110

E. Johansen et al.

operate and clean the device; doctors who will recommend and develop treatment protocols using the device; and biomedical technicians who may inspect, repair, and clean the device. Additional stakeholders included purchasers, such as ministries of health, nongovernmental organizations, and public and private hospitals; manufacturers; local regulatory bodies within target countries; and international regulatory bodies to obtain CE Mark. This case study focuses on end-user observations. To learn more about the needs and use patterns of end users, DtM conducted observations in a range of primary-, secondary-, and tertiary-level hospitals in Southeast Asia treating newborns. The following are several examples of observations DtM conducted that uncovered important insights that guided product design. At some hospitals, DtM observed newborns covered with a blanket because nurses or parents did not realize it is important for phototherapy light to reach as much skin as possible (Fig. 5.3). At one hospital, DtM found that newborns were being treated with a white observational light built into the phototherapy device because hospital staff did not realize the white light did not provide phototherapy. In many hospitals, phototherapy intensity was set to standard level instead of intensive brightness, even though most patients were severely jaundiced. The intention was to save electricity and lengthen the useful life of the fluorescent bulbs, which tended to last only 1 month.

Fig. 5.3 An overhead phototherapy device treats newborns at a hospital in Vietnam. (Photo CC BY-SA 4.0 Design that Matters)

5 Human Centered Design Principles for Open-Source Medical Devices

111

Fig. 5.4 Overhead phototherapy devices treat newborns at a tertiary hospital in the Philippines. (Photo CC BY-SA 4.0 Design that Matters)

During a visit to a tertiary hospital in the Philippines, the DtM team observed newborns with jaundice sharing limited overhead phototherapy in crowded beds (Fig. 5.4). These overhead phototherapy lamps are designed to be bright enough to provide treatment to only one newborn in the center of the lighting field. Newborns at the outskirts of the light may not receive effective phototherapy. Interviews with healthcare professionals did not reveal this issue because the staff did not realize the phototherapy light is too dim at the edges of the light field.

5.3.3

Synthesis

Once a round of observations and interviews has been completed, the next step is to consolidate the observational data, the literature search information, and competitive offering information into a set of summary frameworks and product requirements that will best guide the development process to achieve positive impact. Design principles, patient and healthcare provider journey maps, stakeholder maps, and user or stakeholder personas are all examples of useful frameworks. An example of design principles is provided in the case study; the other common frameworks are covered in detail in the Design Kit from Ideo.org (2015). The product requirements document may be a more formal document for a regulated medical device or simply a set of bullet points describing what needs the

112

E. Johansen et al.

product should to satisfy. A simple and common approach involves generating a series of one-line or one-phrase needs sourced from the observations and interviews. These needs along with any requirements from international standards come together to form a product requirements document and a market requirements document (ISO, 2016). In the early stages of development, the goal of the product requirements document is to describe only the essentials of what the end product should do, while not detailing a specific solution. For instance, if the team is designing a better way to communicate test results from a central laboratory to patients, a generative requirement would be to “communicate the result within 7 days from sample draw.” An overly specific requirement would be “the patient receives a text message within 7 days of whole blood draw.” If the team designs to the second requirement, it would only design for patients with a mobile phone who live where mobile service exists, and whole blood would be the only sample type considered. At the earliest stages, the product requirements should give the development team as much flexibility in their solution as possible, while reducing the scope only where options have been intentionally considered and ruled out. As development proceeds, the requirements may become more and more specific as a leading design solution is selected and pursued.

5.3.3.1

Synthesis: Project Firefly

After completing initial user and stakeholder research in seven majority world countries across Southeast Asia, DtM created a set of high-level design principles and a set of detailed product requirements to guide design. The following are the four design principles that inspired Firefly Phototherapy development. • Effective: not only should the device provide intensive phototherapy, but it should also look like it can provide intensive phototherapy. Perceptions of efficacy are as important as actual efficacy in driving stakeholder adoption. • Comforting: the head of neonatology at a district hospital in Vietnam inspired the team to make the device like a “high-tech bird’s nest”; comforting and fitting snugly to the newborn, like a nest for baby birds, while demonstrating that it is high-tech and effective. • Maintainable: the need for maintenance beyond cleaning should be eliminated, by designing for durability. In Firefly’s target settings, easy-to-clean means wipeable with isopropyl alcohol and other disinfectants to reduce infection. Durability was increased by eliminating moving parts that would easily wear out. • User-friendly (hard-to-use-wrong): the design should ensure that overburdened healthcare professionals at district-level hospitals feel confident using Firefly. The device should not only be easy to use but also hard to use wrong for busy professionals. The DtM team also created a product requirements document that included elements from the international standards for newborn phototherapy equipment

5 Human Centered Design Principles for Open-Source Medical Devices

113

expressed in a way that reflected the design principles (IEC, 2009). To make Firefly hard to use wrong, one requirement was to provide only intensive-level phototherapy. With the advent of high-power, long-lasting LEDs that use little electricity and the knowledge that it is not possible to overdose on phototherapy, DtM saw an opportunity to simplify the controls and prevent busy healthcare professionals from choosing a setting that was too low. In addition, DtM included product requirements beyond international standards that reflected the hard-to-use-wrong design principle. One example was to size the device so only one newborn would fit directly centered under the brightest part of the lighting field.

5.3.4

Ideation and Prototyping

Once the team has synthesized learnings from fieldwork and updated the requirements, the next step is to develop a range of concepts or approaches to accomplish the goal. At this stage, the goal is not to try to find the single best solution but rather to consider the range of potential solutions. The final design may be one of these ideas but, more likely, will be a combination of ideas with other features added as the team learns more about the designs. Brainstorming is one approach to developing a wide range of ideas to consider. Many approaches exist for an effective brainstorm (see Chap. 8 on creativity promotion for OSMDs), but this chapter describes complementarily the approach used by IDEO.org (2015). The IDEO.org Design Kit provides the following list of brainstorm rules and a description of what they mean: • • • •

Defer judgment – anything works for the duration of the brainstorm. Encourage wild ideas – crazy ideas often have a nugget of a great idea. Build on the ideas of others – be positive and encouraging. Stay focused on the topic – it is very easy to stray from the topic in a wild brainstorm. • Conduct one conversation at a time – it is important to hear everyone for joint inspiration. • Be visual – sketches tell a better story than words. • Go for quantity – do not worry about perfection. A good brainstorm lasts 15 min to 1 h and may yield 35–100 ideas, from which perhaps the half may be feasible and five to ten may lead to solid approaches. In this stage, as many ideas as possible should be produced, regardless of quality. Any idea is welcome in a brainstorm; judgment is applied later. Generally, five to eight people are a good number for an effective brainstorm. The group should include the extended team, and perhaps an outside person or two who has expertise in an area of relevance to the project. The ideas should be captured on sheets of paper and posted in the room. Sketches are highly recommended. After the brainstorm, the ideas should be sorted into themes. Themes are solutions that have something in common as defined by the team.

114

E. Johansen et al.

After the brainstorm, the goal is to, quickly and affordably, test critical assumptions about stakeholder desirability, technical feasibility, and business viability. The team should develop three to five rough concepts based on the brainstorms. These concepts are almost certainly combinations of brainstorm ideas and new ideas. These concepts should be embodied in a fast and rough way to test critical assumptions. In some cases, a two-dimensional sketch or a photo-realistic product image can help to explore aspects of the concept. For example, a mobile app might be embodied by a few sketched screens known as a wireframe. In other cases, a three-dimensional prototype may be warranted. Even simple representations of the idea made from cardboard and tape can be useful. The goal of these early prototypes is not to make the real product or meet every requirement in one prototype but rather to test targeted assumptions across the domains of stakeholder desirability, technical feasibility, and business viability. Concept sketches, images, and prototypes are extremely helpful assets to solicit stakeholder feedback. The assets must be at a sufficient level of fidelity to convey the idea to stakeholders who may be unfamiliar with the product design process but need not be technically sound. These assets are used by the team to verify that the innovation addresses the needs identified in stakeholder interviews and observations. The team can use images and prototypes to preview the product architecture, gather feedback about the user interface options, and explore reactions to varied aesthetics even before solving the technical challenges of the product. The value of the rapid, iterative prototyping process is to discover and address possible failure points early in the development process before it is too expensive to correct them. A good design results much more quickly from a series of less refined prototypes than a single prototype developed in more detail.

5.3.4.1

Ideation and Prototyping: Project Firefly

In one of the earliest brainstorm sessions for Firefly, DtM considered the best product architecture for hard-to-use wrong phototherapy. To seed the brainstorm, the team researched various types of lighting technology including LEDs, transparent materials including plastic and glass, and plastic light pipes that transmit LED light from one end to another. Brainstorm ideas included a light pipe device that could enable mothers to hold their newborns during therapy, configurations with transparent materials enabling phototherapy from below, and various ways to incorporate a single-sized infant bed. After the brainstorm, the team combined the ideas that best represented the design principles and product requirements into photo-quality rendering concepts shown in Fig. 5.5. The renderings provided at a sufficient level of detail to clearly express the idea for early stakeholder feedback, before solving all the technical details of how each concept would work. DtM’s partner, East Meets West Foundation, showed these concepts to 88 healthcare professionals at a jaundice training conference in Vietnam. The result: ideas like “cocoon” at top left in Fig. 5.5, which would require caregivers to hold the device in their arms, were immediately discarded. Moms are

5 Human Centered Design Principles for Open-Source Medical Devices

115

Fig. 5.5 Early phototherapy device concepts. (Image CC BY-SA 4.0 Design that Matters)

not able to hold babies all day, and nurses do not have time given the low staff-topatient ratios. The preferred architecture was the shade as it appeared to fit in small spaces in crowded hospitals and looked reliable and familiar.

5.3.5

User and Stakeholder Feedback

In returning to the field for additional rounds of feedback, the team can visit a range of users, similar to those who were visited in initial observations and interviews. Visiting the same stakeholders multiple times throughout product development can be beneficial when they are a proven source of articulate feedback. In addition, revisiting may help develop partners to help launch the technology when it is ready. It is also beneficial to include new stakeholders who have never been exposed to the idea before. To gather user input, at least two members of the team ideally visit the users where the product might be employed. During the visit, the idea of a prototype is explained briefly, and the user is asked to step through the process they expect to use with the prototype, ideally in the location where they would use it. If the product or process would normally include instructions, those should also be prototyped and included. In a real use environment, the user is likely to have little more than the

116

E. Johansen et al.

product and instructions. This approach allows the team to better understand what parts of the process are difficult to understand or may be easily misunderstood. In addition, the user should be told that the team is in the middle of the design process and their feedback is needed to incorporate into the final design. People are often reluctant to criticize a design if they think it is complete. Prototyping and user feedback can most efficiently occur in parallel with technical design and development. Many teams, with which authors have worked, want to wait until they have a fully functional design before gathering stakeholder feedback. There is often a reticence to spend budget to show stakeholders ideas that may not ultimately be technically feasible or viable for the business. Authors’ experience is that the product design process is more efficient if multiple early ideas are shown to stakeholders in parallel with technical exploration. Teams can engage technical and business experts to help choose early ideas that seem most likely to be technically feasible and viable for the business. We also found that conducting at least one round of feedback with multiple concepts, representing a variety of possible product directions, enables the most honest and comprehensive stakeholder feedback, because it prompts stakeholders to share information they may not have otherwise thought to share before seeing the range of what is possible. This enables the team to focus technical efforts on one or two leading product directions that are highly likely to satisfy stakeholders. Once a fully functional design has been created, it is much more expensive to change based on stakeholder feedback. In addition, the team is less willing to change the design once a single fully functional prototype has been created because they are more psychologically invested in the design. Stakeholder desirability, technical feasibility, and business viability must all be considered for a successful product. Time and money may be saved in the long run by adding these cycles of stakeholder feedback to the development process.

5.3.5.1

User and Stakeholder Feedback: Project Firefly

After gathering feedback from neonatologists and neonatal nurses in Vietnam about multiple early concepts, the DtM team decided to build a “looks-like” and “interactslike” prototype to convey the “shade” concept to hospitals in Vietnam. Based on feedback and prior field observations, DtM added lighting below the infant bed in hopes it would make the device more effective and “hard to use wrong.” The prototype used low-power blue LEDs made for hobby electronics and a painted medium-density fibreboard housing to simulate the experience of what it might be like to interact with the product. DtM had not yet solved the technical challenge of how to provide uniform, intensive, blue lighting from above and below, but engineering experts on the team felt it may be possible. By bringing this more detailed “interacts-like” model before substantial technical development had taken place, DtM learned invaluable lessons about how the device would be used, which became inputs to technical development.

5 Human Centered Design Principles for Open-Source Medical Devices

117

Fig. 5.6 Unexpected use discovery in Vietnam. (Photo CC BY-SA 4.0 Design that Matters)

The DtM team asked nurses to use the prototype and a baby doll to show how they would set up the device for use in the neonatal intensive care unit. One nurse suggested she would place a sheet over the device to shield bystanders from the bothersome blue light. Because the interview took place in the hospital, the nurse was able to grab the material she would ideally use and show the team how she would drape it on the device as shown in Fig. 5.6. Allowing for draping a sheet over the top light became a product requirement. This requirement compelled the team to overcome technical challenges of ensuring the top light would not overheat in warm hospitals, even when covered with a sheet. DtM’s “interacts-like” phototherapy prototype was also helpful in determining the appropriate size of the Firefly bassinet. The team’s goal was to design a bassinet small enough to fit only one newborn. The purpose was to avoid cross-infection or reduction in therapy effectiveness due to multiple infants sharing one phototherapy device. The team based the size of the prototype bassinet on ergonomic data about the range of typical newborn sizes. Though the sizing may be suitable for a newborn, bringing a physical prototype early in development helped explore aspects of device use that were not available in ergonomics tables. These activities include inserting the baby into the bed and drawing blood from the baby while in the bed. During the user input session, a particularly large infant was at the hospital (Fig. 5.7). With the permission of the mother, the clinicians placed the infant in the bassinet and pantomimed different procedures such as a blood draw to determine

118

E. Johansen et al.

Fig. 5.7 Observations in Vietnam to determine appropriate device size. (Photo CC BY-SA 4.0 Design that Matters)

the appropriate bassinet size. The prototype was too small to be convenient, leading to product requirements related to the height and angle of the bassinet side walls that also drove the overall size of the device.

5.3.6

Detailed Design

An extensive description of the detailed design process of a medical or diagnostic device is not provided here, as this process is already described in many books, periodicals, and international standards. As regards HCD, iterative stakeholder feedback should continue in the detailed design phase. These evaluations take the form of formative and summative human factors tests that focus on the correct and effective use of a detailed design. Formative and summative human factors testing is described in IEC 62366 (IEC, 2015). The process of detailed design and pacing of iterative feedback varies depending on the type of innovation. Major changes are not simple at this point. Continued stakeholder engagement provides an opportunity to prototype training and communication materials. It also provides insight into how the product may be rolled out in the field.

5 Human Centered Design Principles for Open-Source Medical Devices

5.3.6.1

119

Final Design: Project Firefly

The final Firefly design is shown in Fig. 5.8. From the earliest stages of the project through licensing to MTTS Vietnam, DtM engaged over 100 healthcare professionals and parents in 7 majority world countries through iterative stakeholder feedback. By determining the ideal product requirements early, the DtM team was able to complete detailed design and build devices for clinical evaluation in Vietnam in only 5 months. Having very early stakeholder feedback meant the device was designed to be the appropriate size the first time. The team avoided creating expensive, plastic injection molding tooling for the bassinet and base housing that would have had to been scrapped, redesigned, and retooled. Identifying early that professionals would drape the top light with a sheet meant that MTTS has not received reports of any issues with the Firefly top light overheating over 7 years of use. The final product is “hard-to-use-wrong,” enabling it to be used safely and effectively by the busiest professionals at the most rural hospitals. The design includes a small bed size to reduce cross-infection and promote effective treatment by restricting use to a single infant (Fig. 5.8). The bassinet locks into a base to hold the newborn centered on intensive lighting from above and below. The intensive lighting from top and bottom covers a larger skin surface area, increasing treatment speed by 45% compared to traditional overhead phototherapy lights (Arnolda et al., 2018). Reduced treatment time may shorten costly hospitals stays and enable

Fig. 5.8 The final Firefly phototherapy device, manufactured and sold by MTTS Asia. (Photo CC BY-SA 4.0 Design that Matters)

120

E. Johansen et al.

discharging the newborn earlier to the comfortable home environment to bond with mother and family. The iterative HCD process identified opportunities early in the design process that paved a direct path to a reimagined phototherapy experience that better fits an under-resourced environment. The team was able to bridge the divide between prototype and launched product by identifying necessary product requirements early, cutting cycles of technical development.

5.4

Synergies Between Human-Centered Design and Open-Source Medical Devices

HCD is intrinsic to open-source approaches to healthcare technologies. From team formation, through specification and conceptual design, toward prototyping, testing, and product launch, the sharing of information along the life cycle of OSMDs and the promotion of an active involvement of patients, patient associations, healthcare professionals, potential users, and citizens in general, play fundamental roles, if the development is performed according to systematic methodologies (see Chap. 2). In fact, ideally, one of the most relevant benefits of OSMDs for transforming healthcare is a better comprehension of user’s needs, which may lead to more userfriendly, safer through increased inspection, and more equitable medical devices, as all potential users are considered relevant, regardless of their social status, country of origin, race, religion, political opinions, sex, sexual orientation, or type of pathology. HCD prevails when, for instance, the decision to develop a medical device for solving a rare pathology or to design a personalized solution is not hindered by an apparently reduced market niche. HCD also stands out for a desire to interact with users well beyond the commercialization stage, who can further co-create with designers and become emotionally involved in the future success of the humancentered initiative. It is important to note that communities of “makers” devoted to OSMDs precisely focus on these types of challenges, on giving voice to underrepresented populations and making the access to healthcare technologies more equitable. To this end, several communities, platforms, and websites that contribute to opening the development process of medical devices, connect people, and apply HCD strategies have emerged in the last years, as previously reviewed (see Chap. 1). In a way, closed intellectual property strategies lose sense when applying HCD methods, as truly successful innovation is a non-stop process empowered by learning through continuous interactions with users. With this perspective, open-source and human-centered may become unbeatable, not only from the perspective of social impact and equity but also as regards long-term economic sustainability. Supporting HCD strategies with open-source approaches may have positive impacts by further reducing costs (e.g., minimizing the travel costs for on-site visits) and by providing additional and unexpected feedback, which may result in new ways of using technologies or lead to innovative applications for certain medical devices’ subsystems.

5 Human Centered Design Principles for Open-Source Medical Devices

121

Among remarkable and pioneering examples in the open-source arena, in which user and HCD methods have already proven transformative, the medical field stands out as one of the most promising. For example, Enabling the Future has opened new horizons in the prosthetic sector, by involving users and their families in the personalization of their open-source hand prostheses, which now reach thousands of children and are continuously reinvented, by incorporating additional functionalities proposed by the community of users. The achieved solutions stand out for being more user-friendly, affordable, and personalized. Patient Innovation, another revolutionary initiative focused on the sharing of solutions and patients’ experiences in relation to healthcare issues, showcases also some examples of medical devices clearly designed following HCD methods, like their “Ostom-i Alert Sensor,” a connected medical device for ostomy care, or their “Shower ShirtTM,” a shirt that helps breast cancer patients have a shower. Within the UBORA community, some solutions are not only interesting from an aesthetic perspective but also connect with hedonomics, focusing on a more pleasurable user experience, including UBORA’s Med-Pass, an open-source medical passport, designed to be fitted in a jewel-like structure, like a collar, bracelet, or brooch, and UBORA’s baby pacifiers for medicine administration and minimizing sudden death syndrome. Other examples from the UBORA community follow frugal engineering principles, also related to HCD, for the reconfiguration of medical devices developed for the richer economies, with the purpose of making them succeed also in the majority of the world, in which working and environmental conditions may be more challenging (due to power fluctuations, high temperatures, high humidity, dust, insect infiltration, poor availability of spare parts, high-cost consumables, and low staff-topatient ratio, as introduced in the abstract). Among interesting solutions designed for the majority of the world, it is interesting to highlight UBORA’s “Ventconnect” ventilator splitter or UBORA’s aid system for prone positioning of patients with acute respiratory distress syndrome, to cite a couple. More recent examples, in which OSMDs have clearly followed HCD methods, include most of the open-source solutions generated by makers, along the spring of year 2020, in the quest to provide preventive, protective, diagnostic, monitoring, and therapeutic medical devices for fighting the COVID-19 pandemic. Several of these solutions are shared as open-source medical devices, including around 25 UBORA concepts and devices (i.e., face shields, protective masks, ventilators, sanitizing systems, diagnostic sensors, apps for supporting patients, families, and citizens, etc.), which have been co-created with users and following relevant regulations and standards. With the purpose of recapitulation, Table 5.1 below presents a final summary of mutual synergies between HCD principles and innovative approaches from the emergent open-source medical devices field. They are presented according to the key stages of any medical device development life cycle.

122

E. Johansen et al.

Table 5.1 Summary of mutual synergies between HCD principles and innovative approaches from the emergent open-source medical devices field. OS open-source Medical device development stage Team formation

Needs identification and specification

HCD principles positively affected by open-source development schemes OSMDs are emerging thanks to international communities of designers, “makers,” patients and associations, healthcare professionals and nonprofits with truly multidisciplinary backgrounds. Through collaboration within these communities, teams with the necessary expertise and local connections for longterm success promotion can be rapidly arranged The information gathered by means of interactions with relevant stakeholders and through visits to rural and urban hospitals, potential manufacturers and suppliers, associations, notified bodies, and Ministries of Health, among others, can be complemented by launching online surveys within the e-infrastructures and communities focused on OSMDs

Conceptual design Basic engineering and prototyping

Through the application of HCD principles, ideation and prototyping are empowered, thanks to co-creation and preliminary testing of ideas, designs, and prototypes with end users and key stakeholders. Teams implemented according to HCD are multidisciplinary and experienced. However, they can still benefit from existing OS solutions shared online, especially for the straightforward implementation of different subsystems in the first design and prototyping stages

Detailed engineering Product’s life

Medical technologies developed following HCD principles can be continuously updated and improved, even with the incorporation of new functionalities, not only through maintained interactions with the key stakeholders but also through complementary subsystems developed by users and shared through OSMDs platforms and e-infrastructures. Being shared as OSMDs, technologies developed by application of HCD may also reach more end users

OSMDs enhanced through the application of HCD principles Selecting a HCD champion from the beginning of an OSMD project will result in a better application of HCD principles along the development. HCD principles should become central to teaching-learning activities within OS communities, and all team members should receive basic training in HCD methods The systematic incorporation of varied expertise and user types through HCD methods leads to well-identified medical needs and to adequately specified products. This may be a relevant complement for other ways of needs identification, common within OS communities, including the posting of personal needs online or the interaction in online forums OSMDs should start from existing medical needs and should be specified together with users for improved success changes. However, more often than not, OS technologies shared online do not follow systematic engineering design principles, which leads to technologies searching for users, instead of technologies co-created with users. Promotion of HCD within OS communities is necessary In many cases, online repositories of open-source technologies – GitHub, UBORA – share solutions reaching an interesting state of development, but not adequately being transferred to end users (e.g., projects resulting from educational experiences, with which developers lose connection after the end of the course). Systematic application of HCD may support the involvement of industrial partners and promote business viability and social impacts in the long run

5 Human Centered Design Principles for Open-Source Medical Devices

5.5

123

Conclusions

Usable and economically sustainable medical devices and diagnostics are critical for advancing healthcare worldwide, but most existing technology is not designed for the majority of the world, which prevents healthcare technology equity. HCD can provide scientists, clinicians, engineers, Ministries of Health, and policymakers with the tools to bridge the gap, from promising prototypes to launched products, by creating better alignment with stakeholders’ needs early in the process. In our experience, gathering stakeholder insights when the cost of changes is still low can increase the likelihood that much-needed global health innovations reach the market. Although HCD activities may add a modest amount to cost and time early in the project, studies outside of global health domains have shown the up-front investment in HCD can lead to a tenfold return on investment in the long run (Bias & Mayhew, 2005). The advantages include yielding medical devices that are better accepted by stakeholders and avoiding expensive redesigns and recalls. For these reasons, many healthcare, medical device, and global health funders are integrating HCD into their organizations. Incorporating HCD into a product development process requires more than reading this resource, although authors expect that it may be a good starting point in some cases. Furthermore, in the short term, we recommend accessing the references and finding opportunities to work with experienced HCD champions to learn by doing. In the long term, we strongly encourage the integration of HCD principles in the core teaching programs for biomedical engineers so that it becomes second nature to any medical device designer. While more work can be done to study the connection between HCD and positive health impact in communities, we believe there is ample evidence to include an iterative approach to stakeholder engagement in any biomedical engineering program, for training the medical technology developers of the future, especially where positive health impact is the end goal and in agreement with the United Nations’ Sustainable Development Goals. Further exploring and developing the positive synergies between HCD strategies and open-source approaches to medical devices is also proposed.

References ANSI/AAMI. (2018). HE75:2009 R2018: Human factors engineering – Design of medical devices. Available: https://webstore.ansi.org/Standards/AAMI/ANSIAAMIHE752009R2018 Arnolda, G., Chien, T. D., Hayen, A., Hoi, N. T. X., Maningas, K., Joe, P., Cavallin, F., Trevisanuto, D., & Moccia, L. (2018). A comparison of the effectiveness of three LED phototherapy machines, single-and double-sided, for treating neonatal jaundice in a low resource setting. PLOS ONE, 13(10), e0205432. https://doi.org/10.1371/journal.pone.0205432 Bazzano, A., Martin, J., Hicks, E., Faughnan, M., & Murphy, L. (2017). Human-centered design in global health: A scoping review of applications and contexts. PLOS ONE. https://doi.org/10. 1371/journal.pone.0186744

124

E. Johansen et al.

Bhutani, V., Zipursky, A., Blencowe, H., Khanna, R., Sgro, M., Ebbesen, F., Bell, J., Mori, R., Slusher, T., Fahmy, N., Paul, V., Du, L., Okolo, A., Almeida, M., Olusanya, B., Kumar, P., Cousens, S., & Lawn, J. (2013). Neonatal hyperbilirubinemia and rhesus disease of the newborn: Incidence and impairment estimates for 2010 at regional and global levels. Pediatric Research, 74(1), 86–100. https://doi.org/10.1038/pr.2013.208 Bias, R., & Mayhew, D. (Eds.). (2005). Cost-justifying usability: An update for the Internet age. Elsevier. Cheney, C. (2018). Gates Foundation and USAID team up to bring design to health. Devex. Available: https://www.devex.com/news/sponsored/gates-foundation-and-usaid-team-up-tobring-design-to-health-92084 Conklin, E., & Yakemovic, K. (1991). A process-oriented approach to design rationale. HumanComputer Interaction, 6(3), 357–391. https://doi.org/10.1207/s15327051hci0603&4_6 Doerfler, M. D., Truitt, K., Fisher, M., Theron, G., & McFall, S. (2018). A novel collection cup for tuberculosis sputum in South Africa: engineering design and user testing considerations. Global Health Innovation, 1(2). https://doi.org/10.15641/ghi.v1i2.642 Donaldson, K. (2013). The 80-dollar knee prosthesis that’s changing lives. TED Talk. Fisher, M., & Johansen, E. (2020). Human centered design of medical devices and diagnostics in global health. Global Health Innovation, 3(1), 1–15. Fisher, M., Dolby, T., Surtie, S., Omar, G., Hapeela, N., Basu, D., DeWalt, A., Kelso, D., Nicol, M., & McFall, S. (2017). Improved method for collection of sputum for tuberculosis testing to ensure adequate sample volumes for molecular diagnostic testing. Journal of Microbiological Methods, 135, 35–40. Giacomin, J. (2014). What is human centred design? The Design Journal, 17(4), 606–623. https:// doi.org/10.2752/175630614X14056185480186 Harte, R., Glynn, L., Broderick, B., Rodriguez-Molinero, A., Baker, P., McGuiness, B., O’Sullivan, L., Diaz, M., Quinlan, L., & ÓLaighin, G. (2014). Human centred design considerations for connected health devices for the older adult. Journal of Personal Medicine, 4, 245–281. https:// doi.org/10.3390/jpm4020245 IDEO.org. (2015). The field guide to human-centered design: Design kit. Available: http:\crwww. designkit.org International Electrotechnical Commission (IEC). (2009). IEC 60101-2-50:2009 Medical electrical equipment - Part 2-50: Particular requirements for the basic safety and essential performance of infant phototherapy equipment. Available: https://webstore.iec.ch/publication/2669 International Organization for Standardization (ISO). (1999). ISO 13407:1999 Human-centred design processes for interactive systems. Available: https://www.iso.org/standard/21197.html International Organization for Standardization (ISO). (2015). ISO/IEC 62366-1:2015: Medical devices – Part 1: Application of usability engineering to medical devices. Available: https:// www.iso.org/standard/63179.html International Organization for Standardization (ISO). (2016). ISO 13485:2016 Medical devices – Quality management systems – Requirements for regulatory purposes. Available: https://www. iso.org/standard/59752.html International Organization for Standardization (ISO). (2019). ISO 9241-210 Ergonomics of humansystem interaction – Part 210: Human-centred design for interactive systems. Available: https:// www.iso.org/standard/77520.html Jewett, C. (2019). Hidden FDA reports detail harm caused by scores of medical devices. Kaiser Health News. Available: https://khn.org/news/hidden-fda-database-medical-device-injuriesmalfunctions/ Johansen, E. (2018). Making human factors affordable for medical device and global health startups. International Symposium on Human Factors and Ergonomics in Health Care, Boston, MA, USA, March 26–28, 2018. 7(1): 140–147. Kachirskaia, I., & Mate, K. (2018). Human-centered design and performance improvement: Better together. New England Journal of Medicine Catalyst. Available: https://catalyst.nejm.org/hcdhuman-centered-design-performance-improvement/

5 Human Centered Design Principles for Open-Source Medical Devices

125

Karat, C. (1993). Cost benefit and business case analysis of usability engineering. ACM SIGCHI Conference on Human Factors in Computing Systems, New Orleans, LA, April 1993. Linnes, J., Johansen, E., & Kumar, A. (2017). Incorporating the needs of users into the development of diagnostics for global health: A framework and two case studies. In F. Piraino & S. Seliović (Eds.), Diagnostic devices with microfluidics (pp. 219–250). CRC Press. Maisels, M., & McDonagh, A. (2008). Phototherapy for neonatal jaundice. New England Journal of Medicine, 358(9), 920–928. https://doi.org/10.1056/NEJMct0708376 Mantei, M., & Teorey, T. (1988). Cost/benefit analysis for incorporating human factors in the software lifecycle. Communications of the ACM, 31(4), 428–439. https://doi.org/10.1145/ 42404.42408 Mauro, C. (1994). Cost-justifying usability in a contractor company. In Cost-justifying usability (pp. 123–142). Academic. Newton, I. (1686). Philosophiae Naturalis Principia Mathematica. Nielsen, J., & Landauer, T. K. (1993). A mathematical model of the finding of usability problems. In Proceedings of ACM INTERCHI Conference, Amsterdam, The Netherlands, 24–29 April 1993, pp 206–213. https://doi.org/10.1145/169059.169166 Nilsson, T., & Sheppard, B. (2018). The changing face of medical device design. McKinsey Insights. Available: https://www.mckinsey.com/industries/pharmaceuticals-and-medical-prod ucts/our-insights/the-changing-face-of-medical-device-design# Olusanya, B., Teeple, S., & Kassebaum, N. (2018). The contribution of neonatal jaundice to global newborn mortality: Findings from the GBD 2016 study. Pediatrics, 141(2), e20171471. https:// doi.org/10.1542/peds.2017-1471 Patient Innovation: sharing solutions, improving life. https://patient-innovation.com/ Philips Healthcare. (2013). Taking better care of children: ambient experience makes a visit to the pediatric department a more positive experience. Rajanen, M. (2003). Usability cost-benefit models–different approaches to usability benefit analysis. In Proceedings of the 26th information systems research seminar in Scandinavia (IRIS26), Haikko, Finland. Steen, M. (2011). Tensions in human-centred design. CoDesign, 7(1), 45–60. https://doi.org/10. 1080/15710882.2011.563314 Szczepanska, J. (2017). Design thinking origin story plus some of the people who made it all happen. Medium. Available: http://www.medium.com/@szczpanks/design-thinking-where-itcame-from-and-the-type-of-people-who-made-it-all-happen-dc3a05411e53 US Food and Drug Administration (FDA). (2011). Applying human factors and usability engineering to optimize medical device design. FDA Center for Devices and Radiological Health. US Food and Drug Administration (FDA). (2016). Applying human factors and usability engineering to medical devices: Guidance for industry and food and drug administration staff. Available: https://www.fda.gov/media/80481/download USAID, Bill and Melinda Gates Foundation. (2019). Design for health. Available: https://www. designforhealth.org/ Virzi, R. (1992). Refining the test phase of usability evaluation: How many subjects is enough? Human Factors: The Journal of the Human Factors and Ergonomics Society, 34(4), 457–468. https://doi.org/10.1177/001872089203400407 World Health Organization. (2010). Medical Devices: managing the mismatch: An Outcome of the priority medical devices project. World Health Organization. Available: https://www.who.int/ medical_devices/publications/med_dev_man-mismatch/en/

Chapter 6

Certification Pathways for Open-Source Medical Devices Licia Di Pietro, Carmelo De Maria, Andrés Díaz Lantada, Alice Ravizza, and Arti Ahluwalia

6.1

Introduction

Medical devices (MDs) contribute to the attainment of the highest standards of health for individuals. In May 2007, during the World Health Assembly (WHA 60.29), the World Health Organization (WHO) adopted the resolution on health technologies which set out the framework on MDs. In particular, WHO had a mandate of encouraging “Member States to draw up national or regional guidelines for good manufacturing and regulatory practices, to establish surveillance systems and other measures to ensure the quality, safety and efficacy of medical devices and, where appropriate, to participate in international harmonization” (World Health Organization, 2007). Additionally, during the 67th WHA in 2014, they stressed the importance of the regulation of MDs to ensure access to safe, effective and quality medical products. Analysing the current situation, the regulation of MDs varies greatly across the world, ranging from comprehensive to poor. According to a recent study published by WHO in 2017 (World Health Organization, 2017), the global picture of the current status of medical device regulation, visualized in Fig. 6.1, clearly shows that developing countries lack of an appropriate legislation on MDs. Furthermore, these

L. Di Pietro · C. De Maria (*) · A. Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy e-mail: [email protected] A. Díaz Lantada Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain A. Ravizza USE-ME-D srl, I3P Politecnico di Torino, Turin, Italy © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_6

127

128

L. Di Pietro et al.

Fig. 6.1 Global current status of medical device regulations; existence of a national legal framework for medical devices, World Health Organization (2017). Global atlas of medical devices. *Yes includes countries no matter how limited the coverage of the medical device legal framework was. https://apps.who.int/medicinedocs/en/m/abstract/Js23215en/

countries have neither the financial resources nor the technical expertise to transition successfully from an unregulated market to a comprehensive medical device legislation in a single programme or single strategic action. Because of the lack of appropriate regulatory landscape, several countries all over the world, in particular in low-resource settings, are orienting their regulatory procedures on the European system, which is being updated with the recent Medical Device Regulation 745/2017. Regulation of medical products is a necessary means to protect patients and healthcare users from those devices that are unsafe or ineffective in order to guarantee access to high-quality healthcare. When appropriately implemented, the regulation contributes to a better public health pushing towards a Universal Health Coverage recognized as a unifying platform for making progress on Sustainable Development Goal (SDG) 3 for health (Nations, 2015). The necessity to guarantee safety and efficacy has brought to strict norms for controlling each step of the long-life cycle of a MD (design, prototyping, installation, operation, maintenance, repair and disposal), resulting in an increased cost. Additionally, the lack of harmonization between countries obliges manufacturers to prepare different dossier for each country. This is a long and expensive process that limits the innovation and discourages investments and trades and ultimately limits the access to high-quality devices. Open-source medical devices (OSMDs) have the potential to reshape medical industry and to rethink our approach to healthcare, advancing towards universal health coverage by means of a more equitable access to technologies for good health

6 Certification Pathways for Open-Source Medical Devices

129

and well-being. Pioneering examples have demonstrated that open-source medical devices can reach patients in remote, low- and middle-income settings, through innovative approaches to supply chain management. In some cases, production in the point-of-care is enabled, working directly from cloud-shared blueprints. However, more often than not, first experiences with open-source medical technologies have progressed without an adequate consideration of the regulatory landscape and without analysing certification pathways, which can place patients at risk and hinder the emergence of open-source medical devices as reliable and sustainable alternatives. The bright future of open-source technologies, especially in connection with healthcare, relies on their being designed for compliance with relevant legislations and in accordance with internationally accepted standards, as further analysed. It is important to highlight that having a medical device available under an opensource license, together with the sharing of safety criteria and performance data, allows anyone to contribute to its design, making the device easier to repair and maintain, allows the production of compatible spare parts and consumables, and increases its safety, security and robustness (Lessig et al., 2005; De Maria et al., 2018). To be considered as safe and effective MDs, these products must be compliant to the current legislation. Considering the landscape depicted at the beginning of this chapter, the European (EU) Regulation has be considered as a point of reference in the following analysis. Once demonstrated the compliance with EC legislation, it is possible to obtain the CE mark and place the device on the EU market. Several examples of open-source medical devices (OSMDs) appear on the web (Niezen et al., 2016; Alves et al., 2006; Witchurch, 2020), but despite the legal obligation to manufacture safe and effective devices, only some of them have been designed to be compliant with the current regulation (Ferretti et al., 2017; Arcarisi et al., 2019). Considering the current state-of-the-art for OSMDs, which lack of an appropriate regulatory framework, they should be adequately defined to have a real impact on the medical industry and healthcare systems. For this reason, the UBORA consortium has proposed the definition of OSMD that has been officially submitted to the World Health Organization ICD-11 (Iadanza, 2020). Providing environments, such as the UBORA platform (UBORA Platform, 2020), that offer guided and systematic development process to design medical devices compliant with the EU Regulation and with the harmonized standards, when properly implemented, as guaranteed by authorized notified bodies, these products can be used in hospitals and on patients.

130

6.2

L. Di Pietro et al.

OSMDs in the Context of European Regulation on Medical Devices

Almost everyone will be exposed to a MD in her/his lifetime, and for this reason, the EU has established a modernized and more robust legislative framework to ensure better protection of public health and patient safety. In this context, the Medical Devices Regulation (MDR) 2017/745 entered in force in Europe in May 2017 in order to ensure safety and effectiveness of MDs. According to the guidance proposed by WHO, compliance with the regulation should be implemented in all phases of the device’s life span (Fig. 6.2), in which it is possible to identify three stages: • Pre-market control: performed on the device to ensure that the product, to be places on-market, complies with regulatory requirements. It includes: – Definition of a medical device – Risk classification – Essential principles of safety requirements • Placing on-market control: performed to ensure establishment registration, device listing and after-sale obligations • Post-market control: performed to ensure that the device in use continues to be safe and effective Harmonization in medical devices regulation is considered a fundamental instrument for ensuring the safety of patients and healthcare providers other than a fair competition within the healthcare industry. In order to promote harmonization strategies in the medical device field , the UBORA consortium has put in place not only networking and capacity building in the Biomedical Engineering field but also collaborative and open-source approaches, such as design tool and a methodology, to design medical devices compliant with the EU MDR through the application of harmonized standards.

Fig. 6.2 Phases in the life span of medical devices (blue arrow) with related regulation control (orange arrows) and persons who directly manage the different phases (green arrows)

6 Certification Pathways for Open-Source Medical Devices

131

The activity phases in Fig. 6.2 are simplified. For example, the development phase includes development and planning, design verification and validation, and prototype testing. The manufacturer usually manages the first three phases of the medical device’s life span in which is possible to operate on the design in order to obtain a device design compliant with the regulation. Compared to the traditional methods of design, the open-source and collaborative approach can be a possible alternative. In the “pre-market” phase, for example, the open-source approach can lead a better design, increasing safety, security and reliability and reducing costs. In the “placing on-market” phase, the OS approach guides towards an easier demonstration of compliance to the requirements established by the MDR. Moreover, in a context where MD suppliers and manufacturers do not always have a satisfactory solution in many maintenance cases involving spare parts (e.g. no more after-sales service with still operating old MDs, sales of complete kits instead of single spare parts, overpricing of spare parts) and considering the constraints of maintenance workshop in hospital (e.g. limited storage capacity and long delivery times), the availability of blueprints under OS license can be a valuable solution for biomedical maintenance team in the “post-market” phase.

6.2.1

Pre-market Control

The pre-market control includes all steps that are necessary to design and manufacture a MD compliant with the regulation. The open-source approach in the pre-market phase allows anyone to inspect and improve the device, leading to very rapid innovation compared to traditional methods. Making the device under an open-source license enables the design to be modified for specific uses other than increasing safety, security and robustness. As aforementioned, the design process is fundamental to obtain a safe and effective medical product. Using the open-source approach proposed by UBORA (UBORA bioengineering design model), focused on the co-creation of OSMDs compliant with MDR 2017/745, it is possible to apply the safety standards since the beginning of the design considering the main product requirements. Guiding the designer(s) through work flows and providing free tools to correctly identify risk class and applicable standards for a specific device, it is possible to minimize the human errors related to the design process. In the pre-market control, three key aspects are fundamental and they are described below.

132

6.2.1.1

L. Di Pietro et al.

Definition of Medical Devices

According to MDR 2017/745, defining the intended purpose of the device is the starting point for all decision, including whether the product is a medical device or not. If properly described, the intended purpose will provide: • confirmation, or not, of whether the product being considered fits the definition of a “medical device” and therefore whether or not the regulation applies; • the basis for the classification of the future planned device into one of the four classes of device, as required by Article 51; • core text which is needed for the future labelling, instructions, promotional or sales materials, the clinical evaluation and the technical documentation. Intended purpose is defined in Article 2 (12) and means: “the use for which a device is intended according to the data supplied by the manufacturer on the label, in the instructions for use or in promotional or sales materials or statements and as specified by the manufacturer in the clinical evaluation”. The intended purpose is usually a short statement of two or three sentences that focuses on what the device is intended to be used for. It is a required item in the Technical Documentation (Annex II, 1.1). It is necessary that intended purpose fulfils the definition of a medical device in Article 2 (1): ‘medical device’ means any instrument, apparatus, appliance, software, implant, reagent, material or other article intended by the manufacturer to be used, alone or in combination, for human beings for one or more of the following specific medical purposes: • diagnosis, prevention, monitoring, prediction, prognosis, treatment or alleviation of disease, • diagnosis, monitoring, treatment, alleviation of, or compensation for, an injury or disability, • investigation, replacement or modification of the anatomy or of a physiological or pathological process or state, • providing information by means of in vitro examination of specimens derived from the human body, including organ, blood and tissue donations, and which does not achieve its principal intended action by pharmacological, immunological or metabolic means, in or on the human body, but which may be assisted in its function by such means.

The definition of OSMD (Chap. 1) includes the description of the Article 2 and extends by focusing on sharing and accountability, proper of the open-source approach. To give an impact to the OSMD regulatory framework, the UBORA consortium has articulated a definition described in Chap. 1.

6.2.1.2

Risk Classification

If the device fulfils the definition above, Article 51 of MDR 2017/745 requires all medical devices to be classified into one of four classes. Annex VIII contains the

6 Certification Pathways for Open-Source Medical Devices

133

22 rules to determine the classification of the device. The classification is a “riskbased” system based on the vulnerability of the human body taking into account of the potential risks associated with the device. The devices are divided in classes, I, IIA, IIB and III, according to their intended use. To classify a MD according to the current MDR is necessary to follow the rules listed in Annex VIII. Before starting with the classification, it is important to define the duration of use and if the devices is considered to be invasive or an active device. The definitions, described in Chap. 1 of Annex VIII, are as follow: • Duration of use: – “Transient” means normally intended for continuous use for less than 60 min. – “Short term” means normally intended for continuous use for between 60 min and 30 days. – “Long term” means normally intended for continuous use for more than 30 days. • Invasive and active devices: – “Surgically invasive device” means: an invasive device which penetrates inside the body through the surface of the body, including through mucous membranes of body orifices with the aid or in the context of a surgical operation; and a device which produces penetration other than through a body orifice. – “Reusable surgical instrument” means: an instrument intended for surgical use in cutting, drilling, sawing, scratching, scraping, clamping, retracting, clipping or similar procedures, without a connection to an active device and which is intended by the manufacturer to be reused after appropriate procedures such as cleaning, disinfection and sterilization have been carried out. – “Active therapeutic device” means: any active device used, whether alone or in combination with other devices, to support, modify, replace or restore biological functions or structures with a view to treatment or alleviation of an illness, injury or disability. – “Active device intended for diagnosis and monitoring” means: any active device used, whether alone or in combination with other devices, to supply information for detecting, diagnosing, monitoring or treating physiological conditions, states of health, illnesses or congenital deformities. The classification rules, listed in Chap. III of Annex VIII are as follow: • Rules 1–4: Non-invasive devices These devices either do not touch the patient or only contact skin which is intact. • Rules 5–8: Invasive devices An invasive device is any MD that is introduced into the body, either through a break in the skin or an opening in the body. • Rules 9–13: Active devices

134

L. Di Pietro et al.

CLASS I Stethoscope

CLASS IIA

CLASS IIB

CLASS III

SELF-ASSESSMENT CLASS IM Thermometer

Dental fillings Surgical clamps Tracheotomy tubes

Condoms Lung ventilators Bone fixation plate

Pacemaker Heart valves

CLASS IS Sterile gauze

NOTIFIED BODY APPROVAL REQUIRED

NOTIFIED BODY APPROVAL REQUIRED

LOW RISK

INCREASING RISK

HIGH RISK

Fig. 6.3 Medical devices classification according to MDR 2017/745

An active MD is any device relying on a source of electrical energy or any source of power other than that directly generated by the human body or by gravity. • Rules 13–22: Special rules These are rules which cannot be categorized into the other sets previously mentioned. It is important to document the decision on the classification and the supporting information. It is another required item in the Technical Documentation (Annex II, 1.1). The classification determines the conformity assessment route for the device. As shown in Fig. 6.3, if the device falls into Classes IIA, IIB or III it has implications for the notified body. The UBORA platform, based on the UBORA bioengineering design model and empowered by the ISO 13485, provides a guided questionnaire which leads the developers towards the identification of the medical device class (I, IIa, IIb, III) based on Annex VIII of the MDR 2017/745.

6.2.1.3

Compliance with the General Safety and Performance Requirements

As established by the MDR, the compliance with Annex I – General Safety and Performance Requirements (GSPRs) is the most fundamental preconditions to placing any medical device on the EU market. The requirements are set of product characteristics and ensure that any new device will be safe and perform as intended throughout its life. Manufacturers must be able to demonstrate to the regulatory authority that their products comply with the GSPRs and have been designed and manufactured to ensure high level of safety and performance. GSPRs apply to all medical devices and include the following chapters:

6 Certification Pathways for Open-Source Medical Devices

135

• Chapter 1: General requirements • Chapter 2: Requirements regarding design and manufacture • Chapter 3: Requirements regarding the information supplied with the device Demonstrating that a medical device conforms to all GSPR is the responsibility of the manufacturer. The evidence of conformity, recorded in the technical documentation, may be subjected to review by the regulatory authority which has a different involvement depending on the class of the device. In many cases, the most straightforward way of fulfilling these requirements is to ensure the device conforms to an applicable harmonized European standard (norm). Recently, several examples of OSMDs have appeared on the web; however, only some of them have been designed to be compliant with the current regulation. It is crucial to ensure the GSPRs of medical technology, and for this reason, the use of the open-source approach must take into consideration the applicable harmonized standards to design MDs for the market. To this end, another interesting tool is implemented within UBORA platform for supporting the design of regulation-compliant and safe OSMDs, the decision tree for the identification of applicable standards to a specific device. Among them, the standard for biocompatibility assessment (ISO 10993), standard for minimizing the risk factors (EN ISO 14971), standard for packaging and labelling (EN ISO 15223) and standard for electromedical equipment in case of active OSMDs (EN ISO 60601). To ensure the compliance with the standards, which have been identified during the first step of the development, the entire design process must be supported by experts with a well-proven experience in the field of designing and testing medical devices. These “skilled persons” assist the design, commenting the projects and guiding (mentoring) the developers towards an effective and proper device development. Having OSMDs with a pre-production documentation, in a format ready for streamlined scaling up to manufacture as certified devices following the fabrication guidelines, can reduce the entire cost of the product other than fostering innovation and the development of new products.

6.2.2

Placing On-Market Regulation

In order to place a MD on the European market, the manufacturers have to apply a CE mark on their products. The CE mark can be seen as a declaration of the manufacturer that the product is compliant to the relevant legislations including those related to safety. The CE marking consists of several processes that start from the manufacturer’s choice of the conformity assessment route, which itself depends on the classification of the medical device. The following charts will describe the conformity routes to consider based on the class type.

136

L. Di Pietro et al.

Fig. 6.4 (a) Conformity assessment routes for Class I medical devices. (b) Conformity assessment routes for Class Is/Im/Ir medical devices

6.2.2.1

Class I Medical Devices

The only route for Class I devices (Fig. 6.4a) is the self-declaration. Compliance is demonstrated through a self-certification route in which the manufacturer self-affixes the CE mark as a legally binding attestation. Class I “measuring”, Class I “sterile” and Class I “re-usable surgical instruments” require a notified body to review the technical documentation and issue the CE mark.

6.2.2.2

Class IIa Devices

Manufacturers of Class IIa devices may choose two different routes as visualized in Fig. 6.5. Compliance is demonstrated not only evaluating the product but also considering the production quality. Class IIa MDs require a notified body to review the technical documentation, assess per device category and issue a CE mark.

6.2.2.3

Class IIb Devices

Manufacturers of Class IIb devices may choose two different routes as visualized in Fig. 6.6. Compliance is demonstrated not only evaluating the product but also considering the production quality. Class IIb MDs require a NB to review the technical documentation, assess per generic device group and issue a CE marking.

6 Certification Pathways for Open-Source Medical Devices

137

Fig. 6.5 Conformity assessment routes for Class IIa medical devices

6.2.2.4

Class III Medical Devices

Manufacturers of Class III devices may choose two different routes as visualized in Fig. 6.7. Compliance is demonstrated not only evaluating the product but also considering the production quality and the clinical evaluation. Class III MDs require a NB to review the technical documentation, assess per each device and issue a CE mark.

6.2.2.5

Custom-Made Medical Devices

“Custom-made device’ means any device specifically made in accordance with a written prescription of any person authorised by national law by virtue of this person's professional qualifications which gives, under his responsibility, specific design characteristics, and is intended for the sole use of a particular patient exclusively to meet their individual conditions and needs” MDR 2017/745 Article

138

L. Di Pietro et al.

Fig. 6.6 (a) Conformity assessment routes for Class IIb medical devices (Annex VII, Rule 12); (b) conformity assessment routes for Class IIb implantable well-established technologies (WET) and non-implantable no Rule 12 non-WET; (c) conformity assessment routes for Class IIb implantable non-WET

2 (3). However, mass-produced devices which need to be adapted to meet the specific requirements of any user shall not be considered to be custom-made devices. In the open-source word, custom-made MDs are very common. The reason is because now it is possible to produce MDs that are individualized, for example, using additive manufacturing methods based, for example, on patient CT scans. Custom-made devices have been introduced with the intention to cover special cases where commercially available mass-produced products are inadequate for the needs and requirements of a particular patient. They are not a special class but, as well as the common MDs, they must be classified into the aforementioned classes (I, IIa, IIb, III). The main difference is that the custom-made MDs do not need the CE mark (Fig. 6.8a), except for “custom-made class III MDs” where a CE certificate is required (Fig. 6.8b).

6 Certification Pathways for Open-Source Medical Devices

139

Fig. 6.7 (a) Conformity assessment routes for Class III (non-implantable) medical devices. (b) Conformity assessment routes for Class III (implantable ) medical devices

Fig. 6.8 (a) Conformity assessment route for custom-made medical devices. (b) Conformity assessment routes for custom-made Class III implantable medical devices.

6.2.3

Post-Market Control

All conformity assessment procedures require the manufacturer to implement a postmarketing surveillance and vigilance system. These systems are regulatory requirements. The manufacturer is expected to establish an active set of procedures and processes for collecting data for devices on the market, called post-marketing surveillance. An aspect of post-marketing surveillance (PMS) is vigilance. The PMS in the MDR (Article 2, Sec 6) is defined as: “all activities carried out by

140

L. Di Pietro et al.

manufacturers in cooperation with other economic operators to institute and keep up to date a systematic procedure to proactively collect and review experience gained from devices they place on the market, make available on the market or put into service for the purpose of identifying any need to immediately apply any necessary corrective or preventive actions”. In a worldwide market, assistance service and supply of consumables and spare parts profoundly affect the final costs of medical devices and technology, discouraging both manufactures, vendors and hospital managers from exploring the potential market. This is particularly true for developing countries where the lack of MD spare parts is one of the most important design barriers Malkin (2007). Most medical equipment all over the world is manufactured in developed countries, and it has been designed to operate in clean, sterile, climate-controlled environments, with reliable electricity. However, in low-resource settings (LRS), MDs are subjected to harsh environmental conditions including extreme climates, humidity, dust and power instability. These conditions cause more frequent failures and determine a higher request for spare parts, which are expensive and difficult to find. As a result, maintenance and repairing are as problematic as acquisition Malkin (2007). In a context where MD suppliers and manufacturers do not always have a satisfactory solution in many maintenance cases involving spare parts (e.g. no more after-sales service with still operating old MDs, sales of complete kits instead of single spare parts, overpricing of spare parts) and considering the constraints of maintenance workshops in hospitals (e.g. limited storage capacity and long delivery times), the open-source approach to design MD spare parts can be a valuable solution for biomedical maintenance teams.

6.3

Software as Medical Device

As technology continues to advance all facets of healthcare, software has become an important part of all products, integrated widely into digital platforms that serve both medical and non-medical purposes. Software in medical device field can be classified as: • Software as a part of a medical product, e.g. as embedded software of a medical device • Software as medical product itself (standalone software) • Software as accessories of a medical product • Discrete software, that is not a medical product Standalone software is a software that is intended to be used, alone or in combination, for a purpose as specified in the definition of a “medical device” in the MDR 2017/745. In other words, think of software as a medical device (SaMD) as a software which, on its own, is a medical device. In the same way of all the MDs, SaMD must be classified according to Annex VIII, and the compliance to the GSPR is required. In order to demonstrate the

6 Certification Pathways for Open-Source Medical Devices

141

compliance of the software, a particular standard related to its management must be applied, and it is described below. IEC 62304 “Medical device software — Software life cycle processes” applies to the development and maintenance of medical device software when: • The software is itself a medical device; • Or the software is an embedded or integral part of the final medical device. This standard covers safe design and maintenance of software. It provides processes, activities and tasks to ensure safety. There are nine parts of the safety standard: • • • • • • • • •

Part 1: Scope Part 2: Normative references Part 3: Terms and definitions Part 4: General requirements Part 5: Software development process Part 6: Software maintenance process Part 7: Software risk management process Part 8: Software configuration management process Part 9: Software problem resolution process

Software classification is based on potential for hazards that could cause injury to the user or patient. According to IEC 62304, software can be divided into three classes: • Class A: no injury or damage to health is possible. The software system cannot contribute to a hazardous situation, or the software system can contribute to a hazardous situation which does not result in unacceptable risk after consideration of risk control measures external to the software system. • Class B: non-serious injury is possible. The software system cannot contribute to a hazardous situation, or the software system can contribute to a hazardous situation which does not result in unacceptable risk after consideration of risk control measures external to the software system. • Class C: death or serious injury is possible. The software system can contribute to a hazardous situation which results in unacceptable risk after consideration of risk control measures external to the software system, and the resulting possible harm is death or serious injury. For the purpose of this classification, serious injury is defined as injury or illness that directly or indirectly is life-threatening; results in permanent impairment of a body function or permanent damage to a body structure; or necessitates medical or surgical intervention to prevent permanent impairment of a body function or permanent damage to a body structure.

142

6.4

L. Di Pietro et al.

Brief Description of Other Regulations

Promoting the economic and social well-being are interests of any government, and the regulations are important tools to achieve these objectives. Promoting regulatory harmonization plans and streamlining the regulatory process could reduce the legislative burden, lower costs and reduce delays to new products reaching patients in low-income countries. A brief description about the most relevant medical device legislation is described in the current section. The key procedures in Japan, China and the United States and how they differ from current EU Regulation are described. • Japan Japan’s Ministry of Health, Labour and Welfare (MHLW) is the regulatory body that oversees food and drugs in Japan, which includes creating and implementing safety standards for MDs and drugs. In conjunction with the MHLW, the Pharmaceutical and Medical Devices Agency (PMDA) is an independent agency that is responsible for reviewing drug and MD applications. The PMDA works with the MHLW to assess new product safety, develop comprehensive regulations and monitor post-market safety. Under Japan PMDA regulations, a medical device can be classified as a general medical device (Class I), controlled medical device (Class II) or a specially controlled medical device (Class III and Class IV), depending on the risk level. For general medical devices, only a notification/self-declaration is required, and the product does not need to undergo the approval process by the MHLW and PMDA. Controlled medical devices can be designated to be certified by an authorized third-party certification party or reviewed by the PMDA. Specially controlled medical devices must be reviewed and approved by the PMDA and MHLW. • China China National Medical Products Administration (NMPA) regulates MDs and pharmaceutical products across China. MDs in China are categorized into three different classes – Class I, II and III, with risk levels ranging from low to high. The criteria for classification are based on the purpose of use, the structural features of the device, whether the device has direct contact with the human body and the methods and status of use. This classification is used predominantly for risk management where different rules are adopted: class I (low level of risk, safety and effectiveness can be ensured through routine administration); class II (medium level of risk, further control is required to ensure their safety and effectiveness); and class III (high levels of risk, special measures with strict control and administration must be enforced to achieve safety and effectiveness). • The United States The institution responsible for regulating MDs in the United States is the Food and Drug Administration (FDA). The FDA uses a risk-based classification system,

6 Certification Pathways for Open-Source Medical Devices

143

which classifies medical devices into the following three categories: Class I, Class II and Class III. Class I devices are associated with the lowest risk, while Class III devices are associated with the highest risk. Each device is assigned to a panel (cardiovascular, anaesthesiology, etc.). The panel determines the class and special controls and exemptions applicable to the device. Class I devices are defined as non-life sustaining and present minimal harm potential to user. These devices are typically simple in design, manufacture and have a history of safe use. Class II medical devices are devices where general controls are not sufficient to assure safety and effectiveness and existing methods/ standards/guidance documents are available to provide assurances of safety and effectiveness. Class III devices usually support or sustain human life. For Class III medical devices, sufficient information is not available to assure safety and effectiveness through the application of general controls and special controls. Typically, a pre-market approval submission to the FDA is required to allow marketing of a Class III medical device.

6.5

Open-Source Medical Devices: A Key Towards Harmonization

In 1992, the Global Harmonization Task Force (GHTF) was formed to promote worldwide harmonization of medical devices regulatory practices. In 2012, the GHTS was disbanded and replaced by the International Medical Device Regulators Forum (IMDRF) that, with the support of World Health Organization (WHO), aims at generating debates and agreements towards the future directions in medical device regulatory harmonization. Despite these activities, the harmonization plan is still far from completion, and several countries all over the world, in particular developing countries, are orienting their regulatory procedures following the examples of the most relevant regulatory frameworks. The EU MDR 745/2017 is emerging as inspiring model for many low- and middle-income settings. Developing sustainable technologies to make healthcare affordable to a larger population, thus reducing global inequalities, can only be performed taking into consideration the cultural conditions and environmental-climate constraints in which these will be applied. These conditions are usually not considered during the design phase, in traditional product development approaches, which causes frequent failures and determines a higher request for spare parts, hence making maintenance and repairing very problematic. Using open-source approaches, the devices are not only shared and follow a conceptual scheme for presenting their documentation, which is oriented towards demonstrating their compliance with the relevant standards, but also have the potential to increase the access to medical technology, thanks to a feasible reduction in design, management, maintenance and repairing costs, due to the open access to device blueprints. These open-source medical devices should be designed for compliance with the most widespread regulations and comply also with internationally

144

L. Di Pietro et al.

accepted medical devices standards, so as to guarantee safety and usability and, consequently, promote impacts. If designed following commonly accepted methodologies and for compliance with relevant regulations and standards, as proposed by the “UBORA bioengineering design model”, open-source medical technologies, through information and good practices sharing, may constitute an additional key towards harmonization.

References Alves, A. P., Da Silva, H. P., Lourenco, A., & Fred, A. L. N. (2006). BITalino: A biosignal system acquisition based on Arduino. In Proceeding of the 6th Conference on Biomedical Electronics and Devices (BIODEVICES), 2006. Arcarisi, L., Di Pietro, L., Carbonaro, N., Tognetti, A., Ahluwalia, A., & De Maria, C. (2019). Palpreast: A new wearable device for breast self-examination. Applied Sciences, 9, 381. De Maria, C., Di Pietro, L., Lantada, A. D., Madete, J., Makobore, P. N., Mridha, M., . . . Ahluwalia, A. (2018). Safe innovation: On medical device legislation in Europe and Africa. Health Policy and Technology, 7(2), 156–165. Ferretti, J., Di Pietro, L., & De Maria, C. (2017). Open source automated external defibrillator. HardwareX, 2, 61–67. De Maria, C., Di Pietro, L., Ravizza, A., Lantada, A. D., & Ahluwalia, A. D. (2020). Open-source medical devices: Healthcare solutions for low-, middle-, and high-resource settings. In Clinical Engineering Handbook (pp. 7-14). Academic Press. Lessig, L., Cusumano, M., & Shirky, C. (2005). Perspectives on free and open source software. MIT Press. Malkin R. A. (2007) Barriers for medical devices for the developing world. In: Expert review of medical devices 4.6, pp. 759–763. Nations, U. (2015). Sustainable development goals. Available online: Sustainabledevelopment.un. org. Last access March 2020. Niezen, G., Eslambolchilar, P., & Thimbleby, H. (2016). Open source hardware for medical devices. BMJ Innovations, 2(2), 78–83. UBORA Platform. Available online https://platform.ubora-biomedical.org/. Last access: March 2020. Witchurch, A. Examples of open source medical devices. Protocentral site in GitHub. https://github. com/Protocentral/. Last access: March 2020. World Health Organization. (2007). Resolution WHA60. 29: Health technologies. 60th World Health Assembly, Geneva, 14–23. Last access: March 2020. World Health Organization. (2017). Global atlas of medical devices.

Chapter 7

Legislation for Open-Source Medical Devices: Current Scenario, Risks and Possibilities Maria Elena Lippi, Filippo Morello, Licia Di Pietro, Carmelo De Maria, and Valentina Calderai

7.1

A Philosophy (and a Legislation) for the Future

Since the appearance of its first definition at the end of the 1990s, the open-source (OS) approach to software technology embodied a different way of conceiving our life on the Internet and a new route for researchers. The sharing of individuals’ ideas as a way of improving knowledge, the encounter of minds made possible by the openness and the impulse of maximizing individual potentialities would constitute the background for the construction of a digital environment in which innovation is truly at once the motor and the aim. The availability of flexible licensing method – preferred to rigid proprietary schemes – is the juridical translation of this “philosophy of the openness”, with considerable advantages whenever an OS software is applied to a related hardware. The interaction of many people who work on the same source code eases the process of innovation: since the costs in terms of money and time are cut, the costs of end products can be sensibly reduced too. The perspective disclosed by the OS movement seems particularly desirable in those sectors where innovation at contained costs allows for the realization of public goals. This is the reason why the use of the OS methodology in the designing of medical devices is an enrichment and an opportunity to improve patients’ wellness. In particular, the abandon of the proprietary framework would lead to the creation of a sustainable environment for medical devices’ circulation and diffusion. Intellectual property is expensive due to the protection costs it produces. As briefly recalled, the expensiveness of a patented product is reflected by its final price without the certainty of a return in terms of innovation. The price reflects the right to exclude others from the monetization of the product: it is not a guarantee of the technological

M. E. Lippi · F. Morello · L. Di Pietro · C. De Maria · V. Calderai (*) University of Pisa, Pisa, Italy e-mail: [email protected] © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_7

145

146

M. E. Lippi et al.

extraordinariness of the product. This right should constitute a way of incentivizing the development of new ideas, in a strictly utilitarian way, but, de facto, it can discourage the entrance of new minds in the market, especially in sensitive fields such as the one under scrutiny. Born as a sort of immaterial reflection of the ordinary material proprietary framework, intellectual property law has already been seen as an unwieldy presence (Boldrin & Levine, 2002). As pointed out by Boldrin and Levine in 2002, intellectual property is composed by two elements: “the right to own and sell ideas” (“the right of first sale”) and “the right to control the use of those ideas after sale” (the “downstream licensing”). The second form of control has curbed competition and has mostly contributed to transform the market of ideas into a monopoly. The consequences are clearly negative in terms of technological innovation, because, as briefly anticipated, the absence of competition and intervention on ideas prevent the development of ideas themselves. Monopolization cannot stop creativity, but it certainly delays the achievements of new “intellectual” outputs. This is a completely undesirable outcome for the improvement of healthcare through the creation of new medical devices. Differently, through an OS license, both a software and a hardware can be reshaped and improved over time without any proprietary restrictions and can be declined in order to accomplish individual needs as well. In the end, a collective, creative process would lead to a remarkable, still collective result – a goal for a wider community, not just for companies and scientists. Legal reality does not always perfectly match the philosophy of openness. Medical devices must conform to a precise regulatory framework, set by legislation, at national and international level, and common standards. This is due to the importance of the right to health, globally recognized by the Universal Declaration of Human Rights (Article 25) and specifically protected by the Constitution of the World Health Organization. Common standards, in particular, are deemed the easiest way to settle a proper and universal environment in which medical devices can circulate and operate. The drafting of procedures that manufacturers must follow and the individuation of competent authorities that must approve and supervise manufacturers’ creation constitute a key bulwark in order to prevent physical harms to users. Moreover, security should not be meant only in a “material” sense. The information collected in the healthcare sector conveys an individual’s personal story. Stockpiled and processed on a massive scale, these data (name, age, sex, address, work, clinical situation, etc.) provide the means to cluster people into groups, so that risks of discrimination inevitably rise, both at individual and collective level. Data protection and privacy legislations are indispensable to prevent these further harms. Even if in periods of emergency – as the Covid-19 pandemic crisis – these instruments can be relatively put on hold, it is important to remember that they cannot be disregarded. In other words, if balance is possible, oblivion is unacceptable. New innovations enter the scene every day, from the improvement of the so-called wearable medical devices to the development of apps that can be downloaded on a smartphone to create an updating database for every patient or

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

147

even a space for diagnosis and therapy advices. Legal scholars and practitioners are becoming more aware of this evolving reality, as they try to determine the new concepts that law has already the duty to face.

7.2

A Possible Cooperation Between OS and Medical Devices Regulation

The major aim in the field of open-source medical devices (OSMDs) is the achievement of a balance between safety and innovation. A medical device must be safe and effective. This is the reason for the profusion of rules and standards that have been discussed and adopted in the world over the last years. It must be remembered that there is no specific trace of OS in the regulation of medical devices. In particular, there is no specific reference to OS technology even in the last and most applied regulation – the European MDR 2017/745. The current solution consists in inscribing every medical device into the most suitable class of risk in accordance with the rules provided by regulators. Identifying the features of a device and its belonging to a particular class of risk reintroduces the problem of costs that OS aims to cut. The preliminary finding of the perfect nomenclature that accompanies the pre-market phase is followed by the post-market surveillance. As shown in Chap. 6, this form of surveillance signifies a constant monitoring of the device also once placed on market. Manufacturers must constantly be aware of the possible risks and malfunctions of their products in order to prevent any possible harms to users (patients, in this case). Regulation is expensive, but necessary: safety represents the basis for the circulation of medical devices and their survival in the market. OS can find its place even in this apparently hostile landscape. In an OS environment, there is a variety of possible auditors with the competence required to individuate and minimize the risks that current regulations are aimed to face, both at the pre-market and at the post-market surveillance stages. From this pragmatic point of view, OS can be seen as a structure, more than an autonomous tool or technology, and it can enlarge the vision married by legislators. In fact, incorporating the OS approach in a medical device project would signify the achievement of relevant benefits. Hazards and bugs can be rapidly identified and fixed by the OS community, that would become the final operator for risk management. A constant cheap surveillance that would accompany the entire life of the device, actively facing the challenges posed by regulation, is needed. Since quality and reliability are the essential core for the existence of an OS service, high levels of performances should be assured as well. In this sense, the OS community would represent an efficient hub to sustain the development of a reliable medical device. From a legal point of view, OS can be functional for a deeper and more effective fulfilment of regulation thanks to the dynamism that characterizes it. Again, since it

148

M. E. Lippi et al.

conveys a commercial meaning as well, adopting an OS approach signifies a different perception of the market of medical devices. In any case, apart from any revolutionary intents, what is undoubtable at the moment is that OS developers must be compliant with the standards provided both at national and global level. After having mentioned some of the most critical issues raised by the security exigence in the field of medical devices, it is important to examine them a little deeper.

7.3

Software “in” Medical Devices and Software “as” Medical Devices

Although there is no specific definition of OSMDs in Europe or elsewhere, software is not forgotten in the harmonizing process. Harmonization has often a wide perspective. Since the common general bases in medical devices regulation are linked to categories of risk, the specific technologies are inscribed into comprehensive provisions and examined only when necessary or when useful to facilitate the whole process. Software adds a new layer of risk in a device management. A software controls a hardware. In this case, the potential physical harms are not just linked to possible defects and design faults but also to software malfunction and miscalculation. As a hardware design is made widely accessible, so the source code is shared, and many data circulate in a medical device that relies its functioning on the presence of a software. The material and the immaterial components are strictly connected. The collection and processing of patients’ health data extend the risk scenario with types of not physical harms. The essential role of software in the OSMDs environment makes necessary to refresh the general lines of software regulation.

7.3.1

IMDRF Definition

At a global level, IMDRF (IMDRF/SaMD WG/N10FINAL:2013) stresses the difference between software “in” medical devices and software “as” medical devices (SaMDs). The former is a software just “embedded” (or “part of”) in a device. The latter is a software that responds to the following definition “software intended to be used for one or more medical purposes that perform these purposes without being part of a hardware medical device”. At the same time, this particular kind of software can be “interfaced with other medical devices, including hardware medical devices and other SaMD software, as well as general purpose software”. Even mobile apps responding to the previous definition can be considered as SaMDs. SaMDs can be extremely multifunctional, since they can:

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

149

– provide means and suggestions for mitigation of a disease; – provide information for determining compatibility, detecting, diagnosing, monitoring or treating physiological conditions, states of health, illnesses or congenital deformities; – be an aid to diagnosis, screening, monitoring, determination of predisposition; prognosis, prediction, determination of physiological status.

It is a quite well-resembling portrait of what a software (also OS) can be and work in the world of medical devices.

7.3.2

The United States

The distinction among different types of software relatable to medical devices has been adopted by FDA (not surprisingly since it has been one of the promoters of the 2013 IMDFR drafting). The North American authority recalls three categories: 1. Software in medical devices 2. Software as medical devices (indispensable for the functioning of the device itself) 3. Software that are used just in the manufacturing phase and in the maintenance of the device A fundamental rule is Sect. 520(o) of FD&C Act. It illustrates the cases in which a software cannot be considered a device according to Sect. 210(h).

7.3.3

The European Union

The European legislator is well aware of the role reserved to software in healthcare. In several disposition of Regulation 2017/745, software is considered as either a medical device or a component of medical devices (“software in medical devices”). In fact, according to Article 2, a software can be part of the plethora of medical devices. The general criterion adopted by the European legislator consists in the inclusion of the software in the same class of the device, if the software “drives or influences the use of” the device (Annex VIII, Chap. II). Accordingly, software whose operations are not limited to merely supporting the hardware qualify as devices. This is the case of software used for diagnostic (see para 3.7) or therapeutic purposes, since they could even substitute the intellectual process performed by a physician. The latter feature leads to a specific classification. Software “intended to provide information which is used to take decisions with diagnosis or therapeutic purposes” normally owes to Class IIa, whereas “active therapeutic devices with an integrated or incorporated diagnostic function which significantly determines the patient management by the device, such as closed loop systems or automated external defibrillators, are classified as class III” (Classification rule n. 22). These rules perfectly suit the

150

M. E. Lippi et al.

reality of OSMDs, especially if we think that the application of new tools (such as artificial intelligence (AI) functions) to these devices can improve and multiply their performances, by implementing – for instance – the diagnostic phase. Just as in the US federal law, European regulation portraits the scenario of software as medical devices with an also legal distinction between a software working as mere support for hardware and software that could even replace the activity of a doctor. When the operation of the software directly affects the health of a patient (as in the case of a defibrillator), the class must be the riskiest one.

7.4

Health Data and the Right to Data Protection: A Focus on GDPR

A software is an intangible tool that can collect and analyse huge amount of data on a daily basis. The management of health data and health records is essential: an effective medical device must be supported by constantly updated information to reach the hoped results. Also, it is fundamental to understand how the collected personal information can be processed and protected. It is a matter of ethics that could not have been ignored by regulation since it is also a tragically concrete problem. As reported by 2019 Healthcare Data Breach Report (https://www. hipaajournal.com/2019-healthcare-data-breach-report/), in 2019, there was a 196% increase from 2018 in health data breaches. The situation could be even worse, because it is known that in the United States not every data breach is reported, as it happens when the leak regards a business associate. Data breaches (i.e. the illegitimate and illegal disclosure and use of personal data) are a reality and a risk. The North American Department of Health and Human Services’ Office for Civil Rights has individuated and enumerated the main causes of data breaches: hacking activity or informatic incidents; unauthorized access and disclosures; theft; mere data loss; and improper disposal. It is evident that also in the medical devices market, a conspicuous part of the risks connected to software is due to privacy issues and the collection of patients’ personal data. The securitization of the collection procedure and data processing in general must be considered as part of the risk management manufacturers must be aware of. Again, the most advanced discipline in this field is provided by the EU through the General Data Protection Regulation (EU) 2016/679 (GDPR).

7.4.1

Personal Data

GDPR is perhaps the most coherent and complete regulation on data protection until now. It is the result of a great work of harmonization among Member States which has become a model for US legislation as well. Its Article 4 provides the lexis for the

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

151

matter. Starting from the beginning, “‘[p]ersonal data’ means any information relating to an identified or identifiable natural person (‘data subject’); an identifiable natural person is one who can be identified, directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data, an online identifier or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural or social identity of that natural person”. In particular, health data constitute a relevant category of personal data. Health data are considered “sensitive”, as they were defined in the former European Directive 95/46/CE. GDPR dedicates to them a “special category” (Article 9). More generally, Article 4 states that “‘[d]ata concerning health’ means personal data related to the physical or mental health of a natural person, including the provision of health care services, which reveal information about his or her health status”. The identifiability of the data subject (i.e. the subject to which personal information pertains) is the key concept that dominates the entire data protection discipline. To be personal, information must be connectable to the individual it belongs to. The connection can be made both directly and indirectly and determined by the presence of an “identifier”. The broad definition provided by the aforementioned disposition gives the possibility of regulating the processing of large amounts of data, including those that can be just indirectly linked to specific persons, so as to make them “identifiable”.

7.4.2

Processing of Data

The same Article 4 provides a definition of “processing of data”, the main activity taken in consideration by GDPR. The operations recalled can be found also in the healthcare sector, both at a basilar stage and with the presence of more advanced technologies: “[p]rocessing of data means any operation or set of operations which is performed on personal data or on sets of personal data, whether or not by automated means, such as collection, recording, organisation, structuring, storage, adaptation or alteration, retrieval, consultation, use, disclosure by transmission, dissemination or otherwise making available, alignment or combination, restriction, erasure or destruction”.

7.4.3

Subjects Involved in the Processing

In addition to the data subject, there are also other figures that must be recalled: • Controller: the natural or legal person, public authority, agency or other body which, alone or jointly with others, determines the purposes and means of the processing of personal data. If the purposes and means of the processing are

152

M. E. Lippi et al.

determined by EU or Member State law, the controller or the specific criteria for its nomination may be provided by EU or Member State law itself. • Processor: the natural or legal person, public authority, agency or other body which processes personal data on behalf of the controller. • Recipient: the natural or legal person, public authority, agency or another body, to which the personal data are disclosed, whether a third party or not. • Third party: a natural or legal person, public authority, agency or body other than the data subject, controller, processor and persons who, under the direct authority of the controller or processor, are authorized to process personal data. For what concerns OSMDs, data subjects are certainly represented by patients, while manufacturers can be considered as the controller of the processing. A doctor or another professional figure supposed to be informed of patients’ health could be seen as the necessary recipient in this case. The role of the OS community members is more difficult to individuate. They could be seen as personal data processor, due to the changes they can produce in the code structure, or they could even be seen as an additional and innovative figure. Probably, a clear legislative stance would be necessary, in order not to leave to the interpreter what should be clarified by law in the first place.

7.4.4

Principles for Data Protection

GDPR is a remarkable and articulated regulation. There are some general principles that permeate the entire data protection discipline and that represent its essential structure. They can be summarized as following, having regard to the content of Article 5: • Lawfulness, fairness and transparency: personal data must be processed in a lawful, fair and transparent way in relation to the data subject. • Purpose limitation: data must be collected for specified, explicit and legitimate purposes and processed in a manner that is compatible with those purposes. Further processing for achieving purposes in the public interest, scientific or historical research purposes or statistical purposes would not be considered incompatible with the initial purposes. • Data minimisation: personal data must be adequate, relevant and limited to what is necessary in relation to the purposes for which they are processed. • Accuracy: personal data must be accurate and updated when necessary. At the same time, personal data which are inaccurate, having regard to the purposes for which they are processed, must be erased or rectified without delay. • Storage limitation: personal data must be kept in a form which permits identification of data subjects for a period which must be no longer than is necessary for the purposes for which the personal data are processed. Longer periods can be tolerated in case of personal data processed solely for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes –

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

153

still having regards to the appropriate technical and organizational measures required by GDPR to safeguard the rights and freedoms of the data subject. • Integrity and confidentiality: personal data must be processed in a manner capable of ensuring appropriate security, including protection against unauthorized or unlawful processing and against accidental loss, destruction or damage. In order to accomplish these aims, appropriate technical or organizational measures must be used. • Accountability: the controller is responsible for, and must be able to demonstrate compliance with, the aforementioned principles. In addition to this fundamental core, Article 25 of GDPR illustrates two of the most important principles in the new regulatory data protection framework: privacy by design and privacy by default. They are intended to convey the idea of structural privacy, privacy as an essential component for every activity that implies the use or collection of personal data. Outside the EU, they have been originally well described and promoted by Ann Cavoukian, former Information and Privacy Commissioner for the Canadian province of Ontario (http://dataprotection.industries/wp-content/ uploads/2017/10/privacy-by-design.pdf). GDPR represents the first time they have been embodied in a legally binding document in Europe: • Privacy by design: “Taking into account the state of the art, the cost of implementation and the nature, scope, context and purposes of processing as well as the risks of varying likelihood and severity for rights and freedoms of natural persons posed by the processing, the controller shall, both at the time of the determination of the means for processing and at the time of the processing itself, implement appropriate technical and organisational measures, such as pseudonymisation, which are designed to implement data-protection principles, such as data minimisation, in an effective manner and to integrate the necessary safeguards into the processing in order to meet the requirements of this Regulation and protect the rights of data subjects”. • Privacy by default: “The controller shall implement appropriate technical and organisational measures for ensuring that, by default, only personal data which are necessary for each specific purpose of the processing are processed. That obligation applies to the amount of personal data collected, the extent of their processing, the period of their storage and their accessibility. In particular, such measures shall ensure that by default personal data are not made accessible without the individual’s intervention to an indefinite number of natural persons”. These principles provide some points of reference for data controllers who want to be compliant with regulation in order to establish a safe environment for data subjects: the state of the art; the cost of implementation; the nature, scope, context and purposes of processing; the risks of varying likelihood; and severity for rights and freedoms of natural persons posed by the processing. Some of them (such as “state of the art”) are broad notions which can be used as a general container. Data controllers have to add meaning to these expressions, having regard to the context in which they operate and to the processing of data realized. As specified in the

154

M. E. Lippi et al.

Guidelines adopted by the European Data Protection Board in 2019 (par. 19), data controllers must “take account of the current progress in technology that is available in the market” and “must have knowledge of and stay up to date on technological advances, how technology can present data protection risks to the processing operation, and how to implement the measures and safeguards that secure effective implementation of the principles and rights of data subjects in face of the technological landscape” (https://edpb.europa.eu/our-work-tools/public-consultations-art704/2019/guidelines-42019-article-25-data-protection-design_it). In other words, the legislator has delegated to technics the implementation of the same paradigms that they have to follow. Regulation (EU) 2017/745 has already incorporated the general data protection legislation. In Annex I, Chap. II (Requirements regarding design and manufacture), Article 14.2.d states: “Devices shall be designed and manufactured in such a way as to remove or reduce as far as possible [. . .] the risks associated with the possible negative interaction between software and the IT environment within which it operates and interacts”. The rule is clearly a transposition in the world of medical devices of the principles of privacy by design and privacy by default. Manufacturers should display all the tools they possess to secure both the body of the users and their data. Privacy by default means that data controllers should take care of collecting only the data necessary for every specific purpose of the data processing and for the pursued operation (https://edps.europa.eu/data-protection/our-work/subjects/pri vacy-default_en). Privacy by design describes those situations in which the attention for privacy issues starts from the beginning, from the planning of the product – the design – so that privacy can be “incorporated” in a software or in a device. The European legislator has provided some technical “tricks” in order to implement privacy by design and privacy by default. Data minimization has already been recalled; pseudonymization is still defined by article 4(5), GDPR: • Pseudonymization: “the processing of personal data in such a manner that the personal data can no longer be attributed to a specific data subject without the use of additional information, provided that such additional information is kept separately and is subject to technical and organisational measures to ensure that the personal data are not attributed to an identified or identifiable natural person”. It is important to underline that pseudonymization is a reversible process since it is still possible to recover the connection between data and the data subject – so the latter is still “identifiable”. Different from pseudonymization is “anonymization”, another technique whose objective is excluding the possibility of relinking information to a specific individual, in order to assure an irreversible process. GDPR has taken in consideration the previous difference, since it states that “[p]ersonal data which have undergone pseudonymisation, which could be attributed to a natural person by the use of additional information should be considered to be information on an identifiable natural person”. Data processed through pseudonymization are still personal data. Differently, “[t] he principles of data protection should therefore not apply to anonymous information, namely information which does not relate to an identified or identifiable natural

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

155

person or to personal data rendered anonymous in such a manner that the data subject is not or no longer identifiable. This Regulation does not therefore concern the processing of such anonymous information, including for statistical or research purposes”. Due to their non-personal nature, anonymized data is specifically treated in Regulation (EU) 2018/1807 on a framework for the free flow of non-personal data in the European Union. This Regulation takes in consideration the AI/ML scenario as a fundamental source of non-personal data, represented particularly by aggregate and anonymized dataset for Big Data analytics purposes. It is also explicitly stated that, if it is technologically possible to turn anonymized data into personal data, the resulting data must be treated as personal data, and their processing must be disciplined by GDPR. Then, in case of datasets with an inextricable combination of personal and non-personal data, Regulation (EU) 2018/1807 would not “prejudice the application of Regulation (EU) 2016/679” (Article 2). Two other and last principles must be presented. They are intended to take in consideration the problem of control for data subjects, so they should be part of the risk management system. The first one is represented by the consent the data subject must be always free to give in order to allow for the processing of data. • Consent of the data subject: “any freely given, specific, informed and unambiguous indication of the data subject's wishes by which he or she, by a statement or by a clear affirmative action, signifies agreement to the processing of personal data relating to him or her” Informed consent means that data subjects have the possibility of taking their decisions with all the tools required to ponder them. It is already at the roots of the relationship between patients and physicians in healthcare, since patients have the right of knowing the details of the medical treatments they have to undergo. Perhaps, it is the most fragile principle in GDPR. In fact, although they can be adequately provided with all the possible information, it is extremely difficult to verify that ordinary data subjects have perfectly understood the technical aspects implied in a data processing. That informed consent could be an ineffective data protection tool is not a recent idea: in 1994, Schuck defined “informed consent idealists” those who “emphasize the qualitative dimension of physician-patient interactions concerning treatment” (Schuck, 1994). Consent can be rather considered as a portal through which an individual enters in a legal relationship with another entity (individual or collective). It is an element through which we can distinguish between a legitimate and an illegitimate behaviour. At this first stage of “protection”, the problem of control is still open. Another more concrete solution to the control dilemma raised by personal data protection is represented by one of the “minor” rights that orbit around the personal data protection: the right to data portability, disciplined in the EU by GDPR (Article 20, par. 1 and par. 2): • Right to data portability: “1.The data subject shall have the right to receive the personal data concerning him or her, which he or she has provided to a controller, in a structured, commonly used and machine-readable format and

156

M. E. Lippi et al.

have the right to transmit those data to another controller without hindrance from the controller to which the personal data have been provided, where: (a) the processing is based on consent pursuant to point (a) of Article 6(1) or point (a) of Article 9(2) or on a contract pursuant to point (b) of Article 6(1); and (b) the processing is carried out by automated means. • 2.In exercising his or her right to data portability pursuant to paragraph 1, the data subject shall have the right to have the personal data transmitted directly from one controller to another, where technically feasible.” Data subjects have the right to transfer their data from a context to another. In the digital environment, this means the transfer could interest a platform like a social network or a social medium in general. Broadly speaking, every “space” where we have a personal data processing and where data is processed automatically should be considered. Through data portability, the user of a software or a device can regain at least some control, even if it is true that this principle cannot be considered as a definitive remedy: once the information is disclosed, it is impossible to recover a total control on it. Besides, portability finds its roots in market necessities (Wong & Henderson, 2019): it is intended to improve competitiveness in an environment like the Internet, where the risk of monopolies is always under scrutiny. The usefulness of portability is contested by many scholars even through this mostly commercial perspective, because of the proprietary aura that it seems to bring with it. Paradoxically, data portability could reveal as an obstacle to the same freedom of commerce it is supposed to promote. Still, the difference remains quite clear between property and portability, since the second was not created with the intent of introducing a true ius excludendi alios – the right to exclude other people from the fruition of a good, a typical proprietary right shared by many civil law systems. Portability constitutes a light form of control for data subjects, and it can be used in context such as the utilization both of a social network and of an app (e.g. an app linked to a medical device).

7.4.5

GDPR Principles in OSMDs

Having regards to the various previously treated principles, the OS approach can become a useful strategy in order to implement GDPR. As we have already recalled, the OS scientific community is able to offer the constant surveillance required by GDPR and MDR. Every programmer can individuate and eliminate bugs in the data security system of a device or an app linked to a device. Out of the centralized structure of a company, the “state of the art” can be monitored and updated at zero costs for patients. Risks and eventual harms could rapidly be prevented to solve critical situations, since OS hosts an almost unlimited variety of “minds”. The exigence of control could be satisfied as well. OS can be seen as a form of shared control among “peers”. In fact, there are two level of peers: the ones who shared the technical knowledge to act on the source code and to correct the device

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

157

functionalities, and the plethora of subjects (patients, physicians, health operators) who could have a constant access to the data concerning a device structure. The two categories could even intersect. This is the core of openness. Of course, those subjects who do not share a technical knowledge will still find difficult to interpret and really understand the information they have to face, a matter of understanding that deals with the basic knowledge people are provided with (and often not provided when it comes to informatic and technology). But this is the “true transparency dilemma” that is somehow structural in data protection. In any case, the aim of data reachability in terms of code and structure would be perfectly represented through an OS approach. Principles such as lawfulness, fairness and transparency can constitute the OS community’s pillars and guide its members. As for hardware security implementation, the widespread and capillary control of the community represent a solid safeguard for what concerns software’s issues as well.

7.4.6

A FAIR Path: A Non-legislative Approach

Outside the regulatory framework, even before the entry into force of the GDPR, a special mention must be reserved to the FAIR Guiding Principles for scientific data management and stewardship, proposed by a group of scholars in 2016 (Wilkinson et al. 2016). The acronym “FAIR” stands for Findability, Accessibility, Interoperability and Reusability. Four foundational principles whose aim is implementing “transparency, reproducibility and reusability” in data management, also with regard to “the algorithms, tools and workflows that led to that data”. The Guiding Principles has a practical function that aims to combine the use of machines with a reliable system of data collection and use. In 2018 the European Commission have recognized the importance of FAIR principles through a non-mandatory report (https://ec.europa.eu/info/sites/info/files/ turning_fair_into_reality_0.pdf). By recalling European Commission’s presentation, the FAIR principles can be summarized as following: • Findability: findable data must be described by sufficiently rich metadata and registered or indexed in a searchable resource that is known and accessible to potential users. A unique and persistent identifier should be assigned such that the data can be unequivocally referenced and cited in research communications. • Accessibility: data should be obtained by humans and machines upon appropriate authorization and through a well-defined and universally implemented protocol. In other words, anyone with a computer and an Internet connection should be able to access at least the metadata. • Interoperability: the data are described using normative and community recognized specifications, vocabularies and standards that determine the precise meaning of concepts and qualities that the data represent. It is this that allows the data

158

M. E. Lippi et al.

to be “machine actionable” so that the values for a set of attributes can be scrutinised across a vast array of data sets in the sound knowledge that the attributes being measured or represented are indeed the same. Interoperability is an essential feature in the value and usability of data. • Reusability: there is necessity of rich metadata and documentation that meet relevant community standards and provide information about provenance. This covers reporting how data was created (e.g. survey protocols, experimental processes, information about sensor calibration and location) and information about data reduction or transformation processes to make data more usable, understandable or “science-ready”. In this way, data can be reproduced and combined. The European Commission has also taken in consideration the possibility of inscribing FAIR principles into an Open Science framework. In particular, it has underlined that “[t]he greatest benefits come when data are both FAIR and Open, as the lack of restrictions supports the widest possible reuse, and reuse at scale”. Moreover, “[t]o maximise the benefits of making FAIR data a reality, and in the context of Open Science initiatives, the FAIR principles should be implemented in combination with a policy requirement that research data should be Open by default that is, Open unless there is a good reason for restricting access or reuse”. In fact, “[d]ata should be made as open and as FAIR as possible, relative to legal and ethical requirements, and informed by the judgements and culture of the research communities about what is appropriate and practical when providing access”. Following the European Commission’s advice, the UBORA project has adopted the FAIR principles to draft its data management plan.

7.5

US Suggestions for AI Regulation and the Opportunity Offered by OS

Software and in particular OS software create a scenario in which it is almost impossible to foresee all the possible technological innovations. It would be difficult to introduce a quality management system respectful of current regulation without knowing in advance how the programmers would implement it. It would be logical thinking that every implementation would require a new certification through the accomplishment of a new evaluation procedure. There is also another “dilemma” which is linked to the use of algorithms that make a device “intelligent”. AI implements a machine to perform on a human brain model. Thanks to the use of “machine learning algorithms”, AI devices “learn” from the input they receive and actively respond to external stimulation. The machine transforms a given input into a new output, so that the use, the circulation and the management of data become particularly relevant. In addition, AI could completely substitute a human professional – for example, for what concerns the analysis of the collected data and the formulation of a diagnosis or a therapy.

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

159

In the last few years, the attention of institutions for AI has sensibly grown. Because of the ethical issues raised by AI, in 2018, the European Commission drafted the Ethics Guidelines for Trustworthy AI, which are a suggestion for legislators and a stimulus to open a new global debate on the ethical aspects of AI. The purpose is defining the general lines of an AI which should be lawful, ethical and robust (in a technical meaning as well, to prevent eventual harms). For what concerns privacy, the European Commission underlines the necessity of a data governance where the first form of privacy and data protection is preventive. The avoidance of discrimination through data collection and the importance of maintaining data integrity and accessibility are equally stressed. The European institution also wishes for the adoption of specific, universally accepted technical standards for a trustworthy AI. In the United States, we find a more specific approach to AI in the sector of medical devices. Relying on the IMDRF model for SaMDs we have already recalled, FDA has drafted a precertification model in order to define a regulatory framework for SaMDs. In its 2019 Developing a Software Precertification Program: A Working Model (https://www.fda.gov/media/119722/download), FDA has shown a peculiar interest for AI-based SaMDs in the strict sense of the term, since it has clearly stated that Version 1.0 of the program is focused on the establishment of “processes for SaMD technologies, which may include software functions that use artificial intelligence and machine learning algorithms”. The pre-cert program follows a Total Product Life Cycle (TPLC) approach to evaluate and monitor a software “from the premarket development to the postmarket performance” (https://www.fda.gov/ medical-devices/digital-health/digital-health-software-precertification-pre-cert-pro gram) through four steps. 1. Excellence Appraisal: every organization determined to develop and produce a SaMD should present some basic features, i.e. product quality, patient safety, clinical responsibility, cybersecurity responsibility and proactive culture (the last linked to an organization’s capacities in terms of surveillance and respect of users’ needs). 2. Review Pathway Determination: still related to the IMDFR categorization and linked to the development of a risk-based framework. The main elements that would have to be recalled in this case are the intended medical purpose of the SaMD; the individuation of the healthcare situation or condition the SaMD is intended for; and the description of the SaMD’s core functionality in terms of features and functions. 3. Streamlined Review: this is for those organizations that have already obtained the excellence appraisal. The purpose of the streamlined process is to use the information provided by organizations in the second phase. The aim is “understanding” the product in order to analyse a software’s performance and the safety measures that have been adopted. With all the collected information, FDA can finally provide a premarket decision. 4. Real-World Performance: a feedback in terms of usage after the launch of the SaMD must be provided. In particular, the attention is focused on users’

160

M. E. Lippi et al.

experience and the effectiveness of the SaMD: its performance is analysed in order to evaluate its accuracy, reliability and security. Through this first attempt, FDA has proved its attention for the relevance of software in healthcare. As reported in the discussion paper Proposed Regulatory Framework for Modifications to Artificial Intelligence/ Machine Learning (AI/ ML)Based Software as a Medical Device (SaMD), FDA has recognized its pacific approval of SaMD whose algorithms are “locked” (i.e. unchangeable even through use), while the specific topic of AI applied to SaMD is still underregulated. This is why in this last unbinding document FDA suggests a new approach in order to decidedly distinct locked algorithms from the “learning” algorithms. In other words, FDA has stressed the importance of taking in consideration the capacity of modifications proper of AI- and ML-based SaMDs. Modifications that have been included into three categories, depending on the element they concern: – Performance – clinical and analytical performance; – Inputs used by the algorithm and their clinical association to the SaMD output; and/or – Intended use – The intended use of the SaMD, as outlined above and in the IMDRF risk categorization framework, described through the significance of information provided by the SaMD for the state of the healthcare situation or condition.

Some modifications may not need a review, while, for example, “if the modification is beyond the intended use for which the SaMD was previously authorized, manufacturers are expected to submit a new premarket submission”. The main idea advanced by FDA is the planning of modifications that a product developer should provide – the “predetermined change control plan”. The manufacturer should anticipate the modifications to performance, input or the indented use of the software in the “SaMD pre-specifications” (SPS). An Algorithm Change Protocol (ACP) should also show the methods adopted by the manufacturer in order to manage the risks even in presence of a modification. The manufacturer should provide “the data and procedures to be followed so that the modification achieves its goals and the device remains safe and effective after the modification”. As FDA underlines, this model is a suggestion. It must also be said that probably it would work properly in a proprietary framework, where modifications are managed in a centralized way, instead of a free environment as open-source is. It still can be considered as a first attempt to take the first steps in the AI/ML medical devices environment from a regulatory point of view. Again, it is still true that OS can provide a stronger framework in order to prevent discrimination without having to wait until reports concerning users’ experience. The basically unlimited and undiscriminated access to the source code is already a guarantee of a variegated approach, thanks to programmers’ different expertise and different personal stories and identities. As previously affirmed, a wide and variegated community can individuate and intervene in short times to eliminate possible indices of discrimination among data subjects.

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

7.6

161

A Problem of Liability: Distributing Costs in the OSMDs Community

The aforementioned regulations and regulatory proposals provide rules, principles and suggestions at a general level. Reconnecting to what we have affirmed at the beginning, it is important to understand how the OS approach can be part of the current normative scenario and how it can deal with the development of medical devices and the need of regulation. OS is the perfect support for a constant flux of ideas and improvements and the most suitable framework in order to give a constant implementation to the principles expressed in MDR and connected regulations (like the GDPR). As an innovative framework, OS requires also a different legal approach and brings new questions for the future. The struggle between innovation and safety finds one of the most critical aspects in liability. Generally, the most relevant features of liability law are the nexus between the behaviour of a subject and the damage that occurs, the damaging part’s fault. At the current moment, medical devices manufacturers are held responsible for the malfunctions of the device as stated in MDR Aarticle 10) and according to the European Directive 85/374/EEC on liability for defective products (https://eurlex.europa.eu/legal-content/EN/TXT/?uri¼celex%3A31985L0374). This Directive shows the basis of what is usually named “strict liability”: a liability assessed without regard to criteria like fault or negligence from the side of the producer. Producers can be held as liable even when they have applied the standards of behaviour for the productive process. It is the “injured person” that must “prove the damage, the defect and the causal relationship between defect and damage” (article 4). The idea is protecting consumers and their properties against the risks connected to the use of products which are not “safe”. In fact, as in Article 6: 1. A product is defective when it does not provide the safety which a person is entitled to expect, taking all circumstances into account, including: (a) the presentation of the product; (b) the use to which it could reasonably be expected that the product would be put; (c) the time when the product was put into circulation. 2. A product shall not be considered defective for the sole reason that a better product is subsequently put into circulation.

While, according to Article 3: 'Producer' means the manufacturer of a finished product, the producer of any raw material or the manufacturer of a component part and any person who, by putting his name, trade mark or other distinguishing feature on the product presents himself as its producer.

The responsibility for the possible defect shifts the burden of the damage from the side of the consumer to the side of the producer. The latter is considered to be the most suitable to bear the costs derived from the damage, an approach intended to assure the highest protection of consumers, who have been seen as the “weakest” party of the relationship due in primis to the technical information asymmetries that cannot be easily solved.

162

M. E. Lippi et al.

The European legislator does not ignore the presence of a variety of subjects involved in the production and the commercialization of products. As well as producers, the Directive indicates as, other possible liable parts, the importer and, “when the producer of the product cannot be identified”, every supplier of the product (Article 3). One of the preliminary recitals also states that “in situations where several persons are liable for the same damage, the protection of the consumer requires that the injured person should be able to claim full compensation for the damage from any one of them”. In this way, the need of protecting consumers is intended to be satisfied by the various actors involved in every phase of the circulation of the defective product. When it comes to OS, additional problems raise with regard to programmers. They undoubtedly represent a type of the “several persons” recalled by the Directive. It would be important to understand and define their role in the field of OSMDs, especially when the use of AI tools is implied. The OS community is always in development, with always potentially new and unpredictable subjects who enter the scene. The role of its members must be clarified in terms of legal status and in order to establish the relevance of individual contributes. According to the recalled European Directive, the members of the OS community could be seen as other possible liable addressees. This could provide another solution to the problem of distributing costs on the part that can better sustain them, still in order to protect consumers as “weak parts”. If single actors know they could be held as responsible for possible malfunctions, the incentives to operate on the source code and the design of a medical device would probably be sensibly reduced, if not vanished. A system of individual liability would multiply OS expensiveness, aggravating the risk (and the costs) of legal proceedings. From this perspective, OS could prompt legislators and legal scholars to rethink liability or to apply current liability schemes to innovative tools and devices. It is a matter that has already been discussed more broadly in other innovative contexts, such as robotics. Here a rich discourse concerning liability and how to deal with it has already been raised, pertaining to different aspects such as the level of autonomy of a device and the way it is designed. It could be helpful to recall some of the ideas that has already started circulating among legal scholars. Reaching the roots of liability, for what concerns the “relationship” between devices and programmers, it can be underlined the loss of control that severs the nexus between the output produced by an autonomous device (developed through the use – for example – of neural nets or genetic algorithms) and programmer’s liability (Matthias, 2009). On the contrary, it could be said that this loss of control is just apparent, because of the role played by programmers in the design phase, when the learning skills of the device must be trained as well (Bertolini, 2013). From a strictly pragmatic point of view, solutions can be found in terms of balancing the incentives that move a programmer to intervene “on the code” and the need of safety. Through this perspective with regard to robots and their manufacturers, Bertolini (2015) has summarized three possible paths that can be declined for what concerns programmers and the OS community:

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

163

1. Liability cap: by introducing a limit to the amount of the eventual award in compensation, programmers could know the risk in advance and adopting an adequate insurance to cover it. 2. Compulsory insurance for devices users: this kind of insurance would shift the bearing of the costs from programmers to users. Here the negative aspect would rely in the shifting itself, which ends to substitute the reaching of a true solution to the problem of assessing liability. 3. No-fault plans: in the form of a fund (public or private), in order to indemnify the victims reducing litigation complexity. In this way, liability schemes in the traditional sense would be excluded, since the no-fault plans would respond to the demand of compensation, while the exigence of safety would be assured by the adoption of high standards of certification (ibid). These suggestions and in particular no-fault plans do not solve the risk of “moral hazard and safety investments on the side of the responsible party” (ibid). However, in the OS approach, the risk of moral hazard can be reduced for the same reasons previously shown. Then, the respect of standardization constitutes a strong litmus test for safety thanks to current regulation. Finally, it must be held in mind that a normative stance is desirable in this case as well. Not only to clear the relationships among programmers and consumers (patients) but also those between programmers and manufacturers. At an EU level, this operation would be necessary in order to obtain a completely harmonized legal framework.

7.7

Conclusive Observations

The rules and provisions previously exposed constitute the current point of reference for OSMDs manufacturers, designers and programmers. As shown, data protection and privacy issues represent one of the major causes of concern even in the medical devices “cyber market”. OSMDs are still in progress just as the normative infrastructure that accompanies technological discoveries and innovations. It must also be noted that at present the number of juridical cases reported concerning OSDMs is low, so that, from a legal point of view, most of the prospected hypotheses are just hypotheses. This is not helpful for legislators, who have to individuate and balance the values and rights at stake, and this is also why constant reviews and interaction between experts and rulers are essential. In addition to the sensible lack in terms of specific regulations, OSMDs would require a more detailed analysis also in a very pragmatic sense. OS has already been taken in consideration in the public discourse as a possible support to implement public information structures and public databases. An excellent and recent example of a new OS-based approach is represented by the guidelines (https://docs.italia.it/ italia/developers-italia/lg-acquisizione-e-riuso-software-per-pa-docs/it/bozza/index. html) adopted by the Italian AGID (Agenzia per l’Italia Digitale) in 2019. The Italian public administration (PA) has perceived the importance of OS in terms of low costs

164

M. E. Lippi et al.

and saving of time. The idea prospected by the guidelines is that PA would be still the monitoring authority for the source code. At the same time, due to the transparency of the code itself, everyone with the necessary expertise can check it and report (or even correct) eventual errors and malfunctions. The same could be done in healthcare sector with constant and collective surveillance and implementation through programmers’ activity. As said at the beginning of this chapter, OS can be seen like a methodology, even a philosophy. It would be important to focus the attention on the opportunities that it can offer to solve the issues we have discussed before. The interaction of different minds can be seen not as a cause of concern but as a way of improving both innovation and the surveillance process that accompanies it. As seen, legislators and public actors like the European Union Agency for Cybersecurity (ENISA, https://www.enisa.europa.eu/) had already not despised the collaboration between private subjects, national, and international authorities: on the contrary, they are currently seen as desirable and necessary. Moreover, the instances of control that rise in the area of data protection could be satisfied, thanks to a collaborative model where control is more and more “decentralized”. It is not a trivial result, since digital and cyber questions need answers as early as possible. In the era of “digital platforms economy”, where control over personal information relies in the hands of a few companies, a non-proprietary approach can adjust the balance of power and regain the same control that ordinary people usually loose in digital spaces. The final goal is finding the right balance between safety and innovation more than between safety and legislation. In this sense, the creation of an OS community referring to medical devices could be an enrichment and a support for legislative operations that are just a part of a developing world. The same idea of standardization has introduced a new path in the legal discourse. The search for identical standards and equal basis among countries – standards that can and must be adapted over time – could be considered as new and innovative through a normative perspective as well. What we have and we are going to increasingly have in the future is a law in evolution: a law that must be drafted with the indispensable help of technics and that cannot be thought to be set for an undefined period of time. It is impossible to think of code in the old-fashioned style for particular sectors like the medical devices one. Future legislation for technology will probably be constituted mostly by mandatory standards: ductile, flexible laws dialoguing with innovation, instead of hindering it.

References “On the patent front, more time and energy seem to be spent on nuisance and defensive patenting of the obvious than on innovation”, Boldrin, M., & Levine, D. (2002). The case against intellectual property. The American Economic Review, 92(2), 210. IMDRF/SaMD WG/N10FINAL:2013. According to section 210(h):

7 Legislation for Open-Source Medical Devices: Current Scenario, Risks and. . .

165

“The term “device” (except when used in paragraph (n) of this section and in sections 301(i), 403(f), 502(c), and 602(c)) means an instrument, apparatus, implement, machine, contrivance, implant, in vitro reagent, or other similar or related article, including any component, part, or accessory, which is— (1) recognized in the official National Formulary, or the United States Pharmacopeia, or any supplement to them, (2) intended for use in the diagnosis of disease or other conditions, or in the cure, mitigation, treatment, or prevention of disease, in man or other animals, or (3) intended to affect the structure or any function of the body of man or other animals, and which does not achieve its primary intended purposes through chemical action within or on the body of man or other animals and which is not dependent upon being metabolized for the achievement of its primary intended purposes. The term “device” does not include software functions excluded pursuant to section 520(o).” According to sect. 520(o), a software cannot be considered a device when it is used: “(A) for administrative support of a health care facility [. . .] (B) for maintaining or encouraging a healthy lifestyle and is unrelated to the diagnosis, cure, mitigation, prevention, or treatment of a disease or condition; (C) to serve as electronic patient records, including patient- provided information, to the extent that such records are intended to transfer, store, convert formats, or display the equivalent of a paper medical chart [. . .] (D) for transferring, storing, converting formats, or displaying clinical laboratory test or other device data and results, findings by a health care professional with respect to such data and results, general information about such findings, and general background information about such laboratory test or other device, unless such function is intended to interpret or analyze clinical laboratory test or other device data, results, and findings [. . .] (2) In the case of a product with multiple functions that contains— (A) at least one software function that meets the criteria under paragraph (1) or that otherwise does not meet the definition of device under section 201(h); and (B) at least one function that does not meet the criteria under paragraph (1) and that otherwise meets the definition of a device under section 201(h), the Secretary shall not regulate the software function of such product described in subparagraph (A) as a device.” From sect. 520(o) it appears clear that the main criterium to identify a software as an autonomous device lies in its capability of both storing and analyzing the data collected during the activity of the device. See para. 3.7: “A device is considered to allow direct diagnosis when it provides the diagnosis of the disease or condition in question by itself or when it provides decisive information for the diagnosis”. Rule n. 11 (par. 6.3). This typology owes to Class IIa, apart for software capable of causing “death or an irreversible deterioration of a person's state of health” (the riskiest devices, assigned to Class III) and those which can cause “a serious deterioration of a person's state of health or a surgical intervention” (Class IIb). Also, a software “intended to monitor physiological processes” owes to Class IIa, unless it is thought “for monitoring of vital physiological parameters, where the nature of variations of those parameters is such that it could result in immediate danger to the patient” (in this last case, it would owe to Class IIb). A final statement reconnects all the other residual typologies to Class I. https://www.hipaajournal.com/2019-healthcare-data-breach-report/ http://dataprotection.industries/wp-content/uploads/2017/10/privacy-by-design.pdf https://edpb.europa.eu/our-work-tools/public-consultations-art-704/2019/guidelines-42019-article25-data-protection-design_it https://edps.europa.eu/data-protection/our-work/subjects/privacy-default_en Schuck, P. H. (1994). Rethinking Informed Consent in Yale Law School Legal Scholarship Repository, p. 915.

166

M. E. Lippi et al.

“Before the GDPR, data portability was grounded in competition law under Article 102 of the Treaty on the Functioning of the European Union (TFEU) for abuse of dominance and exclusionary conduct45 as well as the Sherman Act46 and Clayton Act47 in the US. With the potential for service providers to ‘lock-in’ consumers and make it more difficult for them to leave the platform, data portability is seen as a solution allowing users to move from one service to another”, Wong, J., & Henderson, T. (2019). The right to data portability in practice: Exploring the implications of the technologically neutral GDPR in International Data Privacy Law. Oxford University Press, p. 11. “Specifically to data portability as a right, Graef et al. consider how the RtDP clashes with competition law and consumer protection law where data portability is seen as a duty and a form of property-like control respectively”, ibidem, p. 12. Wilkinson, M. D. et al. (2016). The FAIR Guiding Principles for scientific data management and stewardship. Available at https://www.nature.com/articles/sdata201618. https://ec.europa.eu/info/sites/info/files/turning_fair_into_reality_0.pdf https://www.fda.gov/media/119722/download “Non-device software functions are not subject to regulation and are not within the scope of the Software Pre-Cert Pilot Program. In particular, software functions intended (1) for administrative support of a health care facility, (2) for maintaining or encouraging a healthy lifestyle, (3) to formats, or displaying data without interpreting or analyzing clinical laboratory test or other device data, results, and findings or (5) to provide certain limited clinical decision support are not medical devices and are not subject to FDA regulation”, p. 10. A statement which is perfectly coherent with section 520(o) of FD&C Act. https://www.fda.gov/medical-devices/digital-health/digital-health-software-precertification-precert-program https://eur-lex.europa.eu/legal-content/EN/TXT/?uri¼celex%3A31985L0374 Matthias, A. (2009). From coder to creator: Responsibility issues in intelligent artifact design, p. 17. Bertolini, A. (2013). Robots as products: The case for a realistic analysis of robotic applications and liability rules in law. Innovation and Technology, 233. Bertolini, A. (2015). Robotic prostheses as products enhancing the rights of people with disabilities. Reconsidering the structure of liability rules. International Review of Law in Computers & Technology, 127–128. As recalled by Bertolini, “[n]o fault schemes are applied in different legal systems, either radically replacing the tort law system (New Zeland) or substituting it in one or more specific fields, often motor vehicle circulation and work-related injuries”, ibid., p. 128. Ibid. https://docs.italia.it/italia/developers-italia/lg-acquisizione-e-riuso-software-per-pa-docs/it/bozza/ index.html

Chapter 8

Creativity Promotion in Open-Source Projects: Application to Open-Source Medical Devices and Healthcare Technologies Andrés Díaz Lantada and Juan Manuel Munoz-Guijosa

8.1

Introduction

Engineers have, from the dawn of technological development, focused on the invention, design and construction of ingenia for supporting societies in their progressive path to well-being and improved life quality. Moreover, in Spain, France and Germany, engineers are known as ingenieros (as), ingenieurs (es) and ingenieure (innen), respectively, having these words a common root from Latin ingenium, also derived from gene, and a meaning related to ingenuity, that is, ability to use intelligence and wisdom to provide with creative, novel solutions, approaches or means (Vérin, 1984). Mentoring the creative process is deeply rooted within engineering practice, and modern engineers are usually involved in creativity promotion tasks in research, development and innovation projects in all kinds of industries and sectors, very often embedded in a framework of substantial temporal, economic and organizational restraints. The fact that engineers are creative and resilient persons is well supported by their shared ability of finding innovative and high-value solutions to societal and technical challenges. These aptitudes are not innate but based on a comprehensive scientific-technological wisdom supported by some professional skills and attitudes, which can indeed be trained through practice and importantly promoted by the application of straightforward principles and techniques. In the last decades, our world has lived through unprecedented technological changes, which have importantly helped to improve living standards, to increase life expectancy and to reduce poverty worldwide. Whereas many of them are a result of the “market pull”, others are directly relatable to the “technology push”, which in A. Díaz Lantada (*) · J. M. Munoz-Guijosa Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_8

167

168

A. Díaz Lantada and J. M. Munoz-Guijosa

turn is associated to universal human curiosity and creativity. However, inequalities are still present, our environment has been importantly compromised, and the global challenges ahead, summarized as the Sustainable Development Goals of the Agenda 2030 (https://sustainabledevelopment.un.org/post2015/transformingourworld), require from the most creative and best possible trained engineers, which will be those capable of providing sustainable answers to those non-desirable states, in collaboration with many other professionals and interacting with all relevant stakeholders. Considering the Sustainable Development Goal 3 on “Good health and well-being”, among possible ways of progressing towards universal healthcare coverage and medical technology equity, it is important to highlight the transformative potentials of open-source medical devices (OSMDs), collaboratively developed and shared for enhanced replicability and more direct and cost-effective application, as widely discussed along present handbook. The engineering design of open-source medical devices, as happens with medical technologies in general, benefits from creativity promotion mentors and from systematic application of techniques and tools for fostering innovation, along the whole life cycle, but very especially in the specification and concept design stages, as further analysed in this chapter. It is necessary to highlight that one of the most remarkable characteristics of OSMDs is that the innovation process is radically reformulated, through collaboration and information sharing, as compared with classical medical technologies developed behind closed doors. Consequently, the well-known stages of the creative process and related creativity promotion techniques have to be reconsidered and updated, to adequately help to deploy the mentioned potentials of OSMDs. Furthermore, the complete potential of such an open innovation scheme can only be obtained if individuals with heterogeneous backgrounds and abilities for mastering the whole innovation process can fully participate in the design. Issues including the collaboration of international teams, the sharing of information through online infrastructures (both with other developers and with end users), the involvement of larger sets of interest groups, the possibility of continuously evolving designs and products, the continuous improvement philosophy, linked to the product lifecycle management, and the employment of resources and methods of the maker movement, to cite some examples, can importantly foster creative problem-solving, if effectively managed. At the same time, the global development process has to be methodically mentored, so that compliance with regulations, risk-benefit ratio and device safety are not compromised by a desire for radical innovation and uniqueness. The application of standards should never be seen as a factor-limiting creativity but as a necessary companion to the innovation cycle. Accordingly, this chapter deals with creativity promotion in open-source projects, especially regarding the development of medical devices and healthcare technologies. The typical stages of the creative process are discussed in the light of the opensource software and hardware movements. Several methodologies and techniques for the generation, association and evaluation of ideas are presented and illustrated by different cases of application linked to OSMDs, many of which are part of the collection of UBORA devices, shared through the UBORA platform (https:// platform.ubora-biomedical.org/). Recently developed tools, also implemented

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

169

within UBORA in order to support medical device innovators throughout the innovation cycle, are illustrated. Some aspects of this chapter evolve from previous experiences and studies performed by the authors (i.e. Díaz Lantada and Munoz-Guijosa “Methods to promote creativity and technological transfer” in: Springer’s Handbook on Advanced Design and Manufacturing Technologies for Biomedical Devices). An adequate adaptation, so as to consider the special aspects of open-source medical devices, the potentials of international co-creation teams and environments and the possible impacts of e-infrastructures, among others, as innovative key resources to promote creativity and systematize the creative process, has been necessary.

8.2

Typical Stages of the Creative Process

Most experts consider that the creative process is the chain of activities conducted towards the resolution of complex, unexplored problems (Sternberg, 1999; Pahl & Beitz, 2013). There is a general agreement that the creative process can be divided into four main stages, with minor nuances depending on the references used. In the first stage, the problem is detected, and the decision to overcome the problem is taken. During the second stage, we prepare ourselves for solving the problem, which usually involves a precise definition of the problem and a specification for the required solution (see Chap. 3). Along the third stage, specific ideas for solving the problem are generated, associated, evaluated and finally implemented. In fact, the creative process does not finish when the idea is given birth. A very important fourth step must also be considered, related to the continuous improvement philosophy, which encourages the creators to contemplate the creation as just the starting point of new innovation cycles (Wongrassamee et al., 2003). The concrete formulation of these main stages, with their corresponding substages, leads to different product development methodologies, including the Pahl-Beitz engineering design approach (Pahl & Beitz, 2013), the CDIO (conceive-design-implement-operate) cycle (Crawley et al., 2011) or the Talgo “explore-ideate-investigate-design and test-implement” innovation scheme (https:// premiotalgoinnovacion.com/elciclo/). Due to the crucial importance of the creative process in the organizational management, international standards as the ISO 56000 family have been issued in order to spread the best practices and assessment methods among the interested organizations. It is important to distinguish between “creative process” and “creativity”. While the former is related to the whole chain of events leading to solve a complex problem, “creativity” focuses on the synthesis substages, where ingenious solutions are generated. Well-known methodologies traditionally associated to the whole creative process, as the TRIZ methodology (Altshuller, 2002), lateral thinking methods (De Bono & Zimbalist, 1970) or design-thinking methods (Rowe, 1987), mainly focus on the “creativity” and are just a subset of all the activities involved in the creative process.

170

A. Díaz Lantada and J. M. Munoz-Guijosa

Confrontation

Actual situation PROBLEM

Information

Ideal situation (vision, promise) Analysis / Formulation Iteration (Continuous) Improvement

Abstraction Measurement

Learning Creation (synthesis), including implementation

SOLUTION

Evaluation Teamwork

Fig. 8.1 Common substages in the creative process

Disregarding the approach employed, the stages of any creative process are composed by common substages, which are related with the cognitive processes occurring in the team members’ brains, as depicted in Fig. 8.1. During the first stage, which derives into the formulation of the problem, important attitudinal differences can be found between different individuals and personalities: whereas some individuals bravely face the problem and are able to outstandingly perform since its very beginning, other feel blocked by insecurity and need considerable mentoring before they can work autonomously. Surprisingly, these antagonist attitudes tend to correlate very well with the intellectual, interpersonal background and heuristic knowledge of each individual. In addition, the team must define the problem with sufficient accuracy, so that further steps can be carried out without any confusion related to scope, specifications or restraints. Questions as “What is the ideal situation we want to achieve?”, “Which are the constraints to be considered?”, and “Are there comparable problems that can be used as reference?” must be accurately solved. In order to give answers to these questions, the team must acquire the necessary amount of relevant information from a vast number of sources. Once the problem is correctly described and the system precisely specified, as quantitatively as possible, the conceptual design stage, to which creativity is inherent, starts. In most strategies for systematically and creatively solving engineering problems, this conceptual stage involves a formulation of the global function to be solved by the engineering system under development (in the case of medical technology the set of desired tasks that the device should perform for managing a specific pathology). Through deep analysis efforts, the global function is normally divided into easier-to-solve subfunctions, by direct application of the “divide and conquer” principle. For these subfunctions, several solutions may be synthesized and

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

171

combined for creating product ideas and finally comparatively assessed, so as to select the best concept, as main output of the conceptual design stage. In a way, the generation, association and evaluation of ideas, typical of the third stage of the creative process, mirror the mental process applied throughout the conceptual design stage in engineering projects, although generation-associationevaluation cycles can be applied to any stage of an engineering project and to most problems we may face. Well-trained engineers, aware of systematic engineering design methodologies and of these ideation-combination-assessment cycles, turn out to be especially creative persons and result ideal as creativity promotion mentors and innovation managers. This is direct consequence of understanding how creativity works and of being experienced with related methods and tools. Note that certain aspects are common along the substages of the creative process, as well as along the creative process as a whole: • Recursivity: the creative process is recursive (fractal) by nature. As described above, the endpoint of a creative process is the starting point for the next one. This recursive nature is also present at smaller scales in the process: information, analysis, synthesis and evaluation activities must be carried out during the problem formulation, concept creation and detailed design and implementation. • Multidisciplinarity: in order to avoid biased solutions, evaluation or solution procedures, the teams should be as multidisciplinary as possible, which in projects linked to OSMDs is of special relevance. • Positive thinking, combined with brave, emphatic and tolerant attitudes, is necessary in order to avoid such biases. • Teamwork plays a crucial role during the whole process in order to guarantee that correct attitude. As formidably described by De Bono, De Graff or Ballesteros (De Bono & Zimbalist, 1970; De Graff & Lawrence, 2002; Ballesteros-Sánchez et al., 2019), different attitudes and personalities must have self-consciousness about their individual characteristics and make efforts to align their behaviours towards the group’s common interest. • Quantification is necessary for guaranteeing adequate evaluation and decisionmaking processes. The team must create and validate adequate sets of metrics that allows for the classification of the different solution alternatives conceived during the synthesis substage. • Combination of intuitive and discursive techniques: while intuitive creativity is a powerful tool for disruptive innovation, the uncertainty in the incubation time makes the use of discursive (systematic) methods mandatory in order to guarantee the resolution of the problem in a given temporal framework, even though the level of innovation is lower. Considering all the above, and in accordance with the methodology for developing OSMDs that the UBORA community employs and that is used as reference throughout the handbook (see Chap. 2), Chap. 3 on “Systematic needs identification techniques for open-source medical technology projects” has already covered the previously mentioned first and second stages of the creative process.

172

A. Díaz Lantada and J. M. Munoz-Guijosa

Chapter 6 focuses more specially on the third stage and describes methods and techniques for creatively generating, associating, evaluating and implementing solutions, which are also illustrated with real development examples. However, it is also true that these approaches and tools can be also successfully applied to the product specification stage and effectively synergize with some of the strategies discussed in Chap. 3.

8.3

Methods and Techniques for the Generation, Association and Evaluation of Ideas

This section focuses on methods and techniques appropriate for the fostering of ideation, for the association of ideas and for the objective evaluation of solutions and illustrates some of the proposed tools by means of a case study. The application case is aimed at optimizing the weight of a robotic hand and at reaching an innovative usable, ergonomic and functional concept. The TRIZ methodology, developed by Altshuller (Altshuller, 2002), is also briefly introduced, as a support or companion methodology for innovating engineering systems, which nicely synergizes with other strategies including the Pahl-Beitz engineering design methodology (Pahl & Beitz, 2013), the CDIO (conceive-design-implement-operate) cycle (Crawley et al., 2011) and the UBORA design approach (http://ubora-biomedical.org/).

8.3.1

Techniques for Supporting Ideation and Debate

Successful product design relies on dividing the overall main problem or system in several easier to handle subproblems or subsystems, on assessing the best solution for each subproblem or subsystem and, finally, on integrating all of them for reaching the general solution for the global problem or system, as previously advanced. Consequently, the search for alternatives, for every subsystem of the device under development, is very relevant for promoting novel combinations and, hence, for achieving more innovative devices. The continuous asking of questions, including “Why?”, “Why not?”, and “What would happen if?”, typical of the inquisitive engineering mind, along the development life cycle and in collaboration with team members, is a key to success. However, there are several techniques that support ideation and debate, some more traditional or analytical (discursive), some more connected to “lateral thinking” (De Bono & Zimbalist, 1970) strategies, as presented below. In principle, the ideation stages are linked to generating as many possible solutions as possible, to challenging status quo, in many cases leaving the examination of technical or financial viability to subsequent elaboration, experimentation and assessment stages.

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

173

The sustained performance of ideation and elaboration cycles is a good practice for enlightening the development cycle. In this subsection, main techniques for supporting ideation and debate are presented, some of them are more analytical, some more inspirational. The elaboration, experimentation and assessment methods are detailed in Sects. 8.3.2 and 8.3.3 and in following chapters of the handbook. Regarding analytical (discursive) techniques for boosting innovation and problem-solving skills, it is necessary to highlight the fundamental importance of studying, resorting to the more classical methods: reading books, articles and catalogues, attending to conferences and professional fairs, solving theoretical problems, interacting with technologies in proof-of-concept practical activities, discussing with colleagues, conducting tests in scaled models, performing “forward steps” or “backward steps” (Ishikawa) (Best & Neuhauser, 2008) analyses, in which, given a solution, possible continuations in its development, or possible reasons why development arrived at that point, are inferred, and applying the Socratic methods. An ideal way to start with an OSMD project is arranging a team of developers with balanced knowledge and skills and expertise in their respective fields. Being true that incorporating members from disciplines far to those more specific to the project (e.g. a sociologist in a medical technology development project, a philosopher in a hospital process reengineering project or an artist in a prosthesis design project) promotes “lateral thinking” and tends to generate very innovative solutions, it is important to stress the relevance of understanding science fundamentals and mastering technology, for reaching to effective, efficient, viable and safe technological solutions. Creativity is not only about generating vast quantities of ideas; it is especially connected to reaching high-value solutions that must be continuously updated and improved, which can only be achieved following the motto “study, study, study and, then, keep on studying”. In fact, in one of its widest definitions, engineering can be contemplated as the art of ingenuously synthesizing novel solutions that fulfil the strong economic, normative and technical restraints present: while non-educated approaches may be able to find a vast number of sterile, partial or biased solutions, a scientific and/or engineering approach would be able to filter and adapt them to the existing, particular, dynamic restraints towards the obtaining of a feasible product. Regarding intuitive, “lateral” thinking tools or “out-of-the-box” techniques for creative problem-solving, some of the most commonly applied options for empowering the inspirational stages include: • Brainstorming (Bouchard Jr & Hare, 1970; Diehl & Stroebe, 1987; Dugosh et al., 2000; Mullen et al., 1991): A technique that consists of finding several alternatives, for solving a specific problem, just by gathering a list of ideas spontaneously contributed in public by the participants. Electronic brainstorming allows for mitigating the social anxiety problems in a non-negligible amount of individuals taking part of the procedure, by allowing for anonymous participation (Camacho & Paulus, 1995; Gallupe & Cooper, 1993). • Brainwriting: An alternative to brainstorming, in which the ideas are written down by the different participants and finally presented in public. This tries to

174

•

•

•

•

A. Díaz Lantada and J. M. Munoz-Guijosa

avoid the prevalence of more extroverted team members and to promote the participation of shier colleagues, whose ideas may be highly valuable for the project (Coskun, 2005; Heslin, 2009; Linsey & Becker, 2011). The activity can also be carried out electronically (Aiken et al., 2007). Phillips 6-6 (Watson, 1975; Summers & White, 1976) (and its variations, like brainwriting 6-3-5): A dynamic collaboration technique widely used in academia, consisting of dividing a group in subgroups of six persons for finding solutions during 6 min and finally putting them in common with the group. Lotus flower: An evolution of brainwriting, in which a central idea is written down, discussed for finding eight possible solutions, which are then written in eight surrounding circles for a second brainwriting round, based on each of the possible solutions. The process is iterative, and the number and quality of solutions is thus remarkably improved. SCAMPER (Serrat, 2017; Ozyaprak, 2016; Eberle, 1996): A technique using direct questions for promoting lateral thinking, by forcing to answer indirect questions, including Substitute? Combine? Adapt? Modify? Put to other purposes? Eliminate? Reverse? Mind maps (Anderson, 1993; Zampetakis et al., 2007): Conceptual schemes, either hand-written, employing post-it notes or drawn with specific software, which prove useful for the initial stages of a product development.

All the above techniques can also be applied as a starting point in the framework of a reverse engineering analysis of known products or components in which one or several functions are related to those to be solved. Other interesting options, for systematic creativity promotion along the development life cycle, include more comprehensive methodologies, including the CDIO approach, the use of design-thinking processes, Kansei engineering (Lévy et al., 2007; Nagamachi & Lokman, 2016) or the employment of TRIZ methodology, which is further explained in Sect. 8.3.4. To illustrate the use of some of the aforementioned techniques, an application case is proposed: the problem, for which creative solutions are sought and found, is linked to optimizing the weight of a robotic hand. It is one of the examples typically discussed in biomedical engineering courses and programmes at UPM, and some of the solving principles proposed here are a summary of such debates. The case is solved in this section by resorting to the combined application of a first brainstorming/Phillips 6-6 round, together with a subsequent brainstorming/ Phillips 6-6 cycle, performed upon the ideas of the first round. This second round, performed after the concepts generated during a first ideation step and represented in a fractal-like form, is sometimes referred to as lotus flower technique. Afterwards, in Sect. 8.3.4, the same problem is studied under the scope of TRIZ methodology, which more than an alternative to the option proposed here, can be seen as a complementary approach. Figure 8.2 presents, in the form of a mind map, the results from the described brainwriting combined with a lotus flower for solving the problem of “optimizing the weight of a robotic hand”.

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

175

Fig. 8.2 Result from a brainwriting combined with a lotus flower for solving the problem of “optimizing the weight of a robotic hand” and represented in form of mind map

8.3.2

Techniques for Association Towards Larger Sets of Products Ideas

Once a first set of ideas for solving a problem or for carrying out a project is generated, it is important to further work with them, so as to reach additional solutions or to obtain more transformative engineering systems in general. Association is fostered by some quite universal operations: addition (or incorporation of a new functionality), extraction (or purposely elimination of a part or component) or merging (or combination of two ideas and their possibilities) are just some examples. These techniques have been applied to all sorts of fields since the ancient history (i.e. a minotaur is a creature from the Greek mythology obtained by merging a bull and a man; a sidecar motorcycle is achieved by the addition of a one-wheeled device to a conventional motorbike). Apart from these straightforward techniques, other alternative tools prove also interesting, especially in connection with the conceptual design stage in product development processes, which also applies to the development of OSMDs. In this stage, once different solving principles are obtained for the different subsystems, association and integration leads to potential global solutions. This association step, part of the elaboration stage of the creative cycle, can be supported by interesting techniques: • Morphological box (Geschka, 1996; Deckert, 2015): A technique consisting of listing, in the form of matrix, the different viable solutions (i.e. in lines) for each product subsystem (i.e. in columns), thus helping to generate multiple “paths” or combinations for the global solution, before an additional final validation stage. • Mind maps: As previously advanced, are graphical representations that represent relationships between concepts, for instance, the subsystems of a medical device and the most adequate solving principles for the different subsystems, in order to organize information in a visually appealing way and support decision-making.

176

A. Díaz Lantada and J. M. Munoz-Guijosa

• PERT (program evaluation review technique) chart: Once the different subsystems of a new medical device, with particular solutions and mutual relationships, are represented in graphical form, a derived graphic can be also implemented (PERT chart) for helping with project management and preliminary cost and time evaluations. The PERT diagram consists of a number of nodes (events or tasks) connected by lines for depicting interdependence relationships. Apart from supporting project management, this graphical information may be used to compare alternative solutions and to further detail the subsystems of a device and generate high-value solutions. • Affinity diagrams – KJ method (Scupin, 1997; Kawakita, 1967; Kawakita, 1970; Iba et al., 2018; Munoz-Guijosa et al., 2009): this Japanese classification tool, created by Jiro Kawakita, encourages participants to classify the set of solutions obtained according to the emotions they induce. A surprising consensus is often achieved, as participants tend to agree in the classification of the whole set in just three to four categories. A subsequent creativity effort is then conducted, focused on those categories. Following with the case of the robotic hand prosthesis, in order to obtain different product ideas and hence promote creativity in the conceptual design stage, the morphological box technique is applied here, as example of these techniques. First, the different subfunctions of the device (structure, intelligence, actuation, detection, energy, etc.) are listed as the rows of a matrix. Then, for each subfunction, different solving principles are proposed (i.e. battery may be a solving principle for the energetic subfunction). Finally, the matrix or morphological box is employed to combine solving principles for different subfunctions and, by integration, reach possible product ideas. Figure 8.3 presents the application of a morphological box to generate three product ideas for the design of innovative robotic hand prostheses. To further progress with the engineering stage, the best product idea should be selected, after an objective multi-criteria evaluation that leads to the designated concept. Techniques from Sect. 8.3.3 can be applied for selection purposes.

8.3.3

Techniques for Evaluation Towards the Final Concept or Solution

The generation and association steps, through which several ideas are generated, should be followed by a focusing step to select the best idea and progress working upon it. This is done by comparing the ideas or solutions generated and, hence, selecting the most appropriate one. In many cases, it may be also interesting to carry on working with a couple of selected alternatives in parallel. In any case, after ideation and association, evaluation should follow. Regarding this evaluation stage, there are also advisable procedures for the methodic assessment of potentially valid concepts. Some of them include:

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

177

Fig. 8.3 Morphological box applied to generating product ideas for robotic hand prostheses

• Originality and value: An initial evaluation of the proposed solutions and ideas may be in form of a simple two-variable analysis, comparing solutions by “originality”, linked to the degree of novelty, and by “value”, linked to its actual viability. • Radar chart: In a second more exhaustive analysis, a radar chart can be constructed for providing a multivariable assessment of the proposed ideas and solutions. Typical features of interest for comparing engineering systems in general and medical device projects in particular include functionality, viability, safety, usability, aesthetics, cost, eco-impacts and degree of innovation, among others. • Analytic hierarchy process (AHP): Is a structured technique for helping with complex decisions, in which the evaluation is converted into numerical values for easier comparison and systematization. If needed, the different features of the system can be weighted, for establishing priorities (i.e. a factor of 0.1 for cost and a factor of 0.4 for safety would lead to an assessment, in which cost is four times more important than ergonomics). Results are typically represented in form of matrix, with the solutions or concepts in columns and the different factors for assessment in rows. Final row shows the global numerical results for helping with final decision. Continuing with the example of the robotic hand prosthesis, the different product ideas generated in Sect. 8.3.2 are further analysed and compared, by using both AHP and radar charts, whose results are presented in Fig. 8.4.

178

A. Díaz Lantada and J. M. Munoz-Guijosa

Fig. 8.4 Quantitative evaluation of product ideas and radar chart representation, so as to help with the selection of the best concept, for a hand prosthesis, among different product ideas. (product ideas according to results from morphological box application, as shown in Fig. 8.2)

8.3.4

The TRIZ Methodology for Innovative Problem-Solving

Among methodologies for creativity promotion in engineering disciplines, TRIZ (Russian acronym for “theory of the resolution of invention-related tasks” or simply “theory for inventive problem-solving”) stands out for being systematic and for its capability of bringing designers, in a quite straightforward way, to both original and reliable solutions. TRIZ was developed by Genrich Altshuller and his colleagues, by analysing a vast quantity of patents and solutions for engineering problems, along the second half of the twentieth century (Altshuller, 2002; Altshuller, 1984; Altshuller, 1994; Altshuller, 1999) and is currently further expanded, improved and disseminated by the Altshuller Foundation (www.altshuller.ru, https://www. altshuller.ru/world/eng/index.asp), by the TRIZ journal (https://triz-journal.com/) and by several researchers and institutes worldwide concerned with innovation in engineering design methods. Briefly speaking, TRIZ is based on the perception that radical innovation rarely occurs (unless when technological breakthroughs appear), on the understanding that problems and solutions from different industries and sciences share common patterns and on the verification that the employment of proven solutions (already working in some fields) into other scientific-technological areas may well bring to innovative and feasible solutions. In short, the TRIZ methodology works upon two basic assumptions, which are very well supported by Mr. Altshuller’s research: firstly, the fact that engineers must

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

179

face complex trade-offs (strictly speaking, Pareto fronts) when pursuing innovations in engineering products, processes and systems, being forced to face contradictions upon the system’s features to achieve and those that can be affected in a negative way and that the engineers want to preserve unaffected. Typically engineers work with around 39 system features (i.e. weight, strength, durability, productivity, power, reliability, etc.). An example of common contradiction in engineering problems may be that improving a product’s strength typically leads to undesired weight increases. The second assumption is that, for surpassing or handling these contradictions, engineers tend to use the same inventive principles. Mr. Altshuller was able to classify the vast amount of inventions he analysed during his professional work as patent examiner in just 40 “standard” inventive principles. For instance, for the aforementioned strength-weight trade-off, the “use of composite materials”, which was classified by Mr. Altshuller as inventive principle n 40, may lead to innovative and reliable solutions capable of overcoming such contradiction. The total of 39 trade-offs and 40 inventive principles were then coupled in form of matrix (the TRIZ matrix) that helps engineers to select the adequate inventive principles for a given design trade-off. Examples of inventive principles include the already mentioned “use of composite materials”, the “change of colour”, the “use of porous materials”, the “modification of local quality” and the “use of feedback”, among others. Summarizing, particular engineering problems can be expressed in a standardized way by means of defining the contradictions between the features of the problem. The standardized problems are almost directly (by using the TRIZ matrix) assigned standard solutions verified in many similar problems. Finally, the inventive principles of the standard solution are applied to the particular problem. It is important to note that the processes used within the TRIZ methodology are very systematic, try to take out any randomness and subjectivity from the creative process, promote a shift towards algorithmic approaches and can be easily programmed, even in form of online versions for supporting researchers (http:// www.triz40.com/TRIZ_GB.php). They have been successfully applied to the development of several innovative medical devices, whose geometries, materials or manufacturing processes have been reinvented thanks to TRIZ. It may be also a supportive methodology for fostering the area of OSMDs, considering that many tools of the methodology, including the set of 39 parameters of engineering systems, the 40 inventive principles and the TRIZ matrix, are well described and illustrated by application examples in several open-source publications and online resources. Going back to our case study from Sect. 8.3.1, linked to “optimizing the weight of a robotic hand”, TRIZ is now applied to generate alternative or complementary solutions in a systematic and very straightforward way. Summarizing, the specific problem can be expressed in the form of a more general contradiction, between a parameter to improve (“weight of moving object”) and other parameters, which should not be damaged or affected (“strength” and “reliability”), among other possible options. The parameters, according to TRIZ methodology, are coded as 1, 14 and 27 for weight, strength and reliability, respectively. The TRIZ matrix provides the following inventive principles for the combination 1–14: “mechanical vibration”, “cheap short-lived objects”, “mechanics substitution/change of field” and

180

A. Díaz Lantada and J. M. Munoz-Guijosa

Table 8.1 TRIZ applied to optimizing the weight of a robotic hand prosthesis Parameters to improve 1. Weight of moving object

Parameters not to damage 14. Strength

27. Reliability

Inventive principles (general solutions) 28. Mechanics substitution 27. Cheap shortlived objects 18. Mechanical vibration 40. Composite materials 1. Segmentation 3. Local quality

11. Beforehand cushion 27. Cheap shortlived objects

Particular applications Replace prosthesis by hologram for cosmetic purposes Thinner fingers, lightweight connections, sets of replacements Explore fingers actuated by resonance Use different reinforced polymers for creating hand components Place battery on other position, employ gadgets Change local density, use porous structure, employ additive manufacturing tools Sets of replacements for breakable components Thinner fingers, lightweight connections, sets of replacements

Fig. 8.5 Result from applying TRIZ methodology for solving the problem of “optimizing the weight of a robotic hand” and represented in form of mind map

“composite materials” (with codes 18, 27, 28 and 40). For the combination 1-27, the following inventive principles may apply: “segmentation”, “local quality”, “beforehand cushion” and “cheap short-lived objects” (coded as 1, 3, 11 and 27). These inventive principles have proven useful as general solutions for similar design contradictions in several engineering systems. After the finding of potentially useful inventive principles, it is necessary to particularize them to the specific problem. In this case study, as summarized in Table 8.1 and schematically presented in Fig. 8.5 as mind map, some particular applications of the inventive principles to the hand prosthesis include replacing the

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

181

prosthesis by a hologram; employing thinner (and probably short-lived) fingers and structures, while providing sets of replacements for broken components; modifying the inner structure of the different components and resorting to additive manufacturing, for achieving lightweight structures; or dividing the prosthesis into a set of connectable gadgets with an external battery, among others.

8.4

Additional Methods, Techniques and Supporting Tools in Open-Source Projects

Recent years have seen a reformulation of the innovation process, due to the emergence of open science and open innovation, to the progressive introduction of online co-creation environments and to the expansion of the makers’ movement. All this is clearly reshaping the way products are developed and affecting most industrial fields, including the biomedical industry and the medical technology development pathway. This handbook provides several examples of how users, patients and healthcare professionals are now being taken more into account, along the development life cycle, not just in the final pre-market approval tests but also from the very beginning of the specification and ideation stages (see Chap. 3). In the open-source arena, the development of OSMDs is importantly supported by the fruitful interactions among all players involved in the co-creation process. The fresh arrangement of international communities specifically focused on OSMDs can be a creativity multiplier, if the crucial aspects of these communities (international component, multidisciplinary profiles and massive collaboration) are adequately considered, managed and promoted. Different approaches, methods, techniques and tools, which complement those described in precedent sections, are presented in this section in connection with OSMDs and with communities of makers devoted to medical technology.

8.4.1

International Competitions and “Hackathons”

In order to rapidly achieve large sets of ideas and concepts for solving all kinds of challenges, the potential of international competitions and of the more recent “hackathons” (word inspired on the words “hack” and “marathon”, as the concept originated in the software realm) should be always taken into account. Even if these competitions and intense “hackfests” are in many cases guided by educational purposes, their potential industrial/entrepreneurial and social impact should not be discarded. Several start-up and spin-off creation programmes worldwide have helped to demonstrate that these challenges are very adequate for generating valuable business ideas, which can successfully reach society, if adequately mentored through the innovation funnel. More recently, international competitions combined

182

A. Díaz Lantada and J. M. Munoz-Guijosa

Fig. 8.6 Selected concepts proposed as solutions for dealing with pandemic diseases in the framework of the “UBORA 2020 Design Competition”. (a) Adjusted snorkel mask for being used as oxygenator. (b) Smart hospital bed for managing pronation. (c) Intelligent system for monitoring patient’s health state. (d) Mask with interchangeable filters. (e) Affordable ventilator. (f) Improved CPAP-type mask. Additional details: https://platform.ubora-biomedical.org/

with design schools, focused on OSMDs and arranged in form of intense “hackathons”, have also shown interesting approaches to reinventing the way medical devices are developed (Ahluwalia et al., 2018). To illustrate these benefits with real examples, Fig. 8.6 shows six selected concepts (from a total of 30) proposed as solutions for dealing with pandemic diseases in the framework of the “UBORA 2020 Design Competition”. The competition, sponsored by Università di Pisa and by Universidad Politécnica de Madrid, has been launched after the SARS-CoV-2 outbreak and related COVID-19 pandemic, so as to generate concepts, designs and prototypes of medical technologies for dealing with infectious diseases emergencies. An interesting aspect of this international design competition is the fact that proposed solutions should be completely documented and shared as OSMDs online. Besides, the designs and related documentation should be oriented to achieving devices compliant with the EU MDR 745/2017 and developed following internationally recognized standards. This is a singular feature of UBORA’s approach, and, once again, it is important to highlight that the future success of OSMDs relies on compliance with regulations and on guaranteeing patients safety, an aspect which unfortunately is not always taken into account in developments from communities of makers. Regarding the aforementioned competition results, 30 highly innovative OSMDs linked to safe diagnoses, monitoring systems, life-supporting technologies and affective medical devices have been proposed along April 2020 and are currently under development.

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

8.4.2

183

Tools for Supporting Innovation Through Co-creation and International Cooperation

Some additional resources, conceived in many cases for supporting online/cloudbased interactions among members of delocalized or international product development teams, also support innovation, dynamic cooperation and co-creation activities. Sharing of files and technologies, as linked to open-science and innovation trends, also add to the plethora of methods, techniques and tools adequate for promoting creativity along the innovation cycle. It is interesting to point out the following resources, which prove very appropriate for open-source projects in general and for developing OSMDs in particular: • Online project management tools: Software for creating and managing collaborative online environments and teams are becoming popular, both as open-source solutions and as closed-IP options, with interesting free or very affordable options in most cases. To cite some examples, tools including MS Teams, Asana, Trello, Ayoa and OpenProject are gaining popularity for carrying out projects through interconnected online teams. These resources enable working 24/7 and simultaneous documentation. Most professionals who have experienced remote collaboration in real time, through an online text editor, will agree that it constitutes an exciting and efficient way for preparing a project report, writing a proposal or documenting results. • Online design resources: In connection with the abovementioned tools, software solutions from the computer-aided design, engineering and manufacturing sectors are shifting to cloud-based operation and offering users innovative options for working and sharing information with colleagues online. • Open research databases infrastructures: Initiatives such as the European Open Science Cloud (ESOC), the European Research Infrastructure Consortium (ERIC) Forum and the EU Open Science Infrastructure (Open AIRE) are fostering the sharing of data, tools and technologies, as part of a global EU strategy to promote open science. These initiatives provide researchers with straightforward connections to key stakeholders and to colleagues with complementary skills and, in many cases, with avant-garde technologies resulting from research projects that gain visibility in this way. Another interesting example is the Karlsruhe Nano Micro Facility, a Helmholtz Research Infrastructure operating with an open innovation approach. This open research infrastructure helps to establish sustainable connections between researchers, with needs linked to the development of micro- and nano-systems for varied industrial applications, and some of the most relevant technology developers in the micro-/nano-field, located at the Karlsruhe Institute of Technology. • Websites sharing projects and files: Whenever starting a product development project, it proves interesting to analyse existing solutions, both in the field of the project and in related areas, which may help to stimulate creativity and inspire the

184

A. Díaz Lantada and J. M. Munoz-Guijosa

team of developers. In parallel to expansion of the makers’ movement, several websites sharing project and design files have appeared. These websites typically provide blueprints of the designs (i.e. CAD files in standard or common formats, such as .stl or .igs) and a brief explanation about the designs and the procedures to achieve prototypes. Among these, it is important to highlight Thingiverse, GrabCAD, TraceParts and TurboSquid, which are reformulating traditional product development and the way designers work and interact with users and potential clients. GitHub, devoted to sharing software, is also a remarkable environment. • Co-creation platforms and makers’ hubs: In other cases, not just a design, but completely documented projects and procedures are shared, as in the inspiring case of the Prusa 3D printers and related community of makers, which have popularized and democratized manufacturing tools. In the field of OSMDs, several co-creation platforms and communities of makers are emerging (see Chap. 1), among which Patients Innovation, Enabling the Future, Proto Central, Open Bionics and UBORA constitute remarkable examples.

8.4.3

Features of the UBORA e-Infrastructure Aimed at Creativity Promotion

Apart from helping to articulate an international community of medical device designers and constituting a sort of universal “Wikipedia” for OSMDs, the UBORA e-infrastructure is also conceived for fostering creativity throughout the development process of innovative medical technologies. In this e-infrastructure or online platform, each of the posted OSMDs has a common project meta-structure, which guides developers through the need identification, project planning, device specification, conceptual design, engineering, prototyping and pre-commercial validation stages (see Chap. 2). Ideally, each project ends with a well-documented medical device design, which includes detailed information, blueprints (i.e. CAD files for prototyping, design schemes), a dossier describing device’s compliance with the EU MDR 745/2017, so as to start interactions with notified bodies towards CE marking, and a filled canvas dealing with the go-to-market plans. Besides, specific features of UBORA are aimed at creativity promotion. First of all, the use of FAIR (findable, accessible, interoperable, reusable) data (Wilkinson et al., 2016) for the implementation of UBORA and as model for interactions among its community is noteworthy. It helps to better find and use the information of the platform, as inspiration for new projects, and to arrange project teams with members skilled in different disciplines in a straightforward way. In addition, the second work package (WP2) of UBORA devices, which is devoted to conceptual design, incorporates interesting tools to promote debate among project developers and to objectively select the best concept among candidate solutions, as further described. The following examples help to illustrate the application of these tools. In the first case (Fig. 8.7), different candidates for a medical passport are shown, together with prototyping and implementation results. Figure 8.8 presents examples of voting

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

185

Fig. 8.7 Candidates for UBORA’s Medical Passport: (a) low-memory NFC tag; (b) NFC tag with bigger memory yet not reliable translation; (c) NFC tag, structured, translatable, supports an offline mode. Design and prototyping results: (d) prototype of NFC tag embedded in PDMS protective structure; (e) dialogue screens of UBORA’s Medical Passport app. (Additional details: A. Quero Martín & A. Díaz Lantada, in SoftwareX, 2020 (QueroMartín & Diaz Lantada, 2020))

results, for a couple of UBORA OSMDs, aimed at selecting the most appropriate concept among different product ideas or candidates. In the case of the Medical Passport, the features of UBORA’s WP2, which enable the creation of a different post for each product idea or candidate, helped to describe (and illustrate with a design sketch) three different options: (a) passport based on a low-memory NFC tag, in which only short information, such as a website URL or pin code can be recorded; (b) passport employing a NFC tag with bigger memory available, in which information is recorded without proper structuration or normalization, which makes translation not reliable; and (c) passport relying on a NFC tag with bigger memory, in which user's data can be recorded and accessed without internet connection. After selection of concept “c”, prototyping is accomplished by placing the NFC tag within a stereolithographic mould and subsequent casting of white PDMS to embed and protect the tag. User-friendly storage and reading of data is achieved with the support of an open Android app. In the examples of Fig. 8.8, voting results for candidates for UBORA devices (helmets for correcting positional plagiocephaly and personalized back brace) are schematically shown. The platform incorporates, in WP2, options for allowing all members of a project to cast their own votes. The voting system is simple; each project member can vote for all candidate product ideas and give 1 to 5 points to the following features: feasibility, performance, usability and safety. All votes are summed and normalized, and this helps to straightforwardly and quite democratically select the best candidate or device concept, among the different ideas, with which the conceptual design stage is completed.

186

A. Díaz Lantada and J. M. Munoz-Guijosa

Fig. 8.8 Examples of voting results, for a couple of UBORA OSMDs, aimed at selecting the most appropriate concept among different product ideas or candidates. (a) Device for correcting positional plagiocephaly through redirecting skull growth. (b) Personalized 3D printable back brace for treating vertebral fracture or disc herniation. Additional details: https://platform.ubora-biomedical. org/

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

8.5

187

Methods, Techniques and Resources for Creativity Promotion Along the Life Cycle

Even though the methods, techniques and resources explained in previous sections support innovation along the whole development life cycle of novel products and all types of medical technologies, some of them are especially appropriate for specific moments and well suited to concrete tasks. Accordingly, Table 8.2 summarizes a personal proposal for application of different creativity promotion techniques to varied sets of tasks, linked to the more conceptual product development stages (UBORA’s WP1 and WP2 following description from Chap. 2). Table 8.2 Useful creativity promotion techniques for the more conceptual stages of the life cycle Stage of the development cycle Product planning and specification

Concrete tasks within the different stages Finding a relevant need

Studying existing solutions Selecting an objective market

Analysing economic viability Specifying the product Conceptual design

Enunciating the main function Describing functional structure Analysing solving principles Generating product ideas

Evaluating product ideas and reaching the selected concept

Most useful creativity promotion techniques Ideation cycles employing brain-(storming/ writing), lotus, Phillips 6-6, focus groups with experts. Focus groups surveys. Social networks mining, DELPHI Employment of co-creation platforms and open-innovation approaches for finding needs and matching technological demands and technological offers Documenting state-of-the-art and current challenges in the form of white papers Classifying, describing and evaluating existing solutions with the support of matrices and evaluation sheets, to find weak points and market niches Morphological boxes or matrix-like representations for comparing among existing and potential solutions for new challenges Templates of specification sheets Conceptual maps and mind maps.

Morphological boxes or matrix-like representations for comparing among existing and potential solutions for new challenges Additional ideation cycles employing brain(storming / writing), lotus, Phillips 6-6, focus groups with experts TRIZ as a systematic complement to other “out-of-the-box” tools Radar charts, assessment matrices, evaluation sheets

188

8.6

A. Díaz Lantada and J. M. Munoz-Guijosa

Conclusions

Creativity deals with efficient problem-solving and is deeply rooted in engineering practice and principles. Creativity promotion tools have been developed along the centuries, together with a progressive approach to understanding the basic mechanisms of innovative thinking for finding unconventional, effective and efficient solutions to complex problems. In the last decades, the innovation cycle in engineering fields, as connected to the development of processes, products and systems, has been importantly reformulated, through the growing internationalization and multidisciplinary ambience in project teams. Besides, emergent open-source approaches to technology development, which benefit in many cases from co-creation among global communities interacting online, constitute a relevant driver of change challenging the status quo. All these aspects have implications for the ideation, design, manufacture and use of open-source medical technologies. In consequence, this chapter has described the innovation cycle, presented different creativity promotion techniques and illustrated them by case studies linked to medical device projects. The potentials of open-source and collaborative approaches for the development of innovative medical devices have been also discussed and showcased with selected examples of recent development from the UBORA e-infrastructure and community. Acknowledgements

References Ahluwalia, A., De Maria, C., Diaz Lantada, A., et al. (2018). Biomedical engineering project based learning: Euro-African design school focused on medical devices. International Journal of Engineering Education, 34, 1709–1722.

8 Creativity Promotion in Open-Source Projects: Application to Open-Source. . .

189

Aiken, M., Rebman, C., & Vanjani, M. (2007). Comment generation with three electronic brainwriting techniques. Academic Information Management Science Journal, 10(1), 11–30. Altshuller, G. S. (1984). Creativity as an exact science: The theory of the solution of inventive problems. Gordon and Breach. Altshuller, H. (1994). And suddenly the inventor appeared: The art of inventing: How to invent and solve technical problems. Technical Innovation Center. Altshuller, G. S. (1999). The innovation algorithm: TRIZ, systematic innovation and technical creativity. Technical Innovation Center. Altshuller, G. (2002). 40 principles: TRIZ keys to innovation (Vol. 1). Technical Innovation Center. Anderson, J. V. (1993). Mind mapping: A tool for creative thinking. Business Horizons, 36(1), 41–47. Ballesteros-Sánchez, L., Ortiz-Marcos, I., & Rodríguez-Rivero, R. (2019). The impact of executive coaching on project managers’ personal competencies. Project Management Journal, 50(3), 306–321. https://doi.org/10.1177/8756972819832191 Best, M., & Neuhauser, D. (2008). Kaoru Ishikawa: From fishbones to world peace. BMJ Quality & Safety, 17(2), 150–152. Bouchard, T. J., Jr., & Hare, M. (1970). Size, performance, and potential in brainstorming groups. Journal of Applied Psychology, 54(1), 51. Camacho, L. M., & Paulus, P. B. (1995). The role of social anxiousness in group brainstorming. Journal of Personality and Social Psychology, 68(6), 1071. Coskun, H. (2005). Cognitive stimulation with convergent and divergent thinking exercises in brainwriting: Incubation, sequence priming, and group context. Small Group Research, 36(4), 466–498. Crawley, E. F., Malmqvist, J., Lucas, W. A., & Brodeur, D. R. (2011). The CDIO syllabus v2. 0. An updated statement of goals for engineering education. In Proceedings of 7th international CDIO conference, Copenhagen, Denmark, 2011. De Bono, E., & Zimbalist, E. (1970). Lateral thinking. Penguin. De Graff, J., & Lawrence, K. A. (2002). Creativity at work: Developing the right practices to make innovation happen (Vol. 28). Wiley. Deckert, C. (2015). Tensions in creativity-using the value square to model individual creativity. Diehl, M., & Stroebe, W. (1987). Productivity loss in brainstorming groups: Toward the solution of a riddle. Journal of Personality and Social Psychology, 53(3), 497. Dugosh, K. L., Paulus, P. B., Roland, E. J., & Yang, H.-C. (2000). Cognitive stimulation in brainstorming. Journal of Personality and Social Psychology, 79(5), 722. Eberle, B. (1996). Scamper on: Games for imagination development. Prufrock Press Inc.. English | World | Официальный Фонд Г.С. Альтшуллера (автора ТРИЗ-РТВ-ТРТЛ) | www. altshuller.ru. https://www.altshuller.ru/world/eng/index.asp. Accessed 12 May 2020. Gallupe, R. B., & Cooper, W. H. (1993). Brainstorming electronically. MIT Sloan Management Review, 35(1), 27. Geschka, H. (1996). Creativity techniques in Germany. Creativity and Innovation Management, 5 (2), 87–92. Heslin, P. A. (2009). Better than brainstorming? Potential contextual boundary conditions to brainwriting for idea generation in organizations. Journal of Occupational and Organizational Psychology, 82(1), 129–145. Iba, T., Yoshikawa, A., & Munakata, K. (2018). Philosophy and Methodology of clustering in pattern mining: Japanese Anthropologist Jiro Kawakita’s KJ Method. Philosophy and Methodology of Clustering in Pattern Minin, 1–11. Kawakita, J. (1967). Hassoho. [Abduction]. Tokyo Chuokoron-Shinsha. Kawakita, J. (1970). Zoku Hassoho. [Abduction: Part two]. Tokyo Chuokoron-Shinsha. Lévy, P., Lee, S., & Yamanaka, T. (2007). On kansei and kansei design a description of Japanese design approach. Linsey, J. S., & Becker, B. (2011). Effectiveness of brainwriting techniques: comparing nominal groups to real teams. In Design creativity 2010 (pp. 165–171). Springer.

190

A. Díaz Lantada and J. M. Munoz-Guijosa

Mullen, B., Johnson, C., & Salas, E. (1991). Productivity loss in brainstorming groups: A metaanalytic integration. Basic and Applied Social Psychology, 12(1), 3–23. Munoz-Guijosa, J. M., Bautista Paz, E., Verdú-Ríos, M. F., Díaz Lantada, A., et al. (2009). Application of process re-engineering methods to enhance the teaching–learning process in a Mechanical Engineering Department. International Journal of Engineering Education, 25. Nagamachi, M., & Lokman, A. M. (2016). Innovations of Kansei engineering. CRC Press. Ozyaprak, M. (2016). The effectiveness of SCAMPER technique on creative thinking skills. Journal for the Education of Gifted Young Scientists, 4(1), 31–40. Pahl, G., & Beitz, W. (2013). Engineering design: A systematic approach. Springer Science & Business Media. Premio Talgo a la Innovación Tecnológica | El Ciclo. https://premiotalgoinnovacion.com/elciclo/. Accessed 12 May 2020. QueroMartín, A., & Diaz Lantada, A. (2020). An open source medical passport based on an Android mobile application and near-field communication. SoftwareX, 11, 100492. Rowe, P. G. (1987). Design thinking. MIT press. Scupin, R. (1997). The KJ Method: A technique for analyzing data derived from Japanese Ethnology. Human Organs, 56(2), 233–237. https://doi.org/10.17730/humo.56.2. x335923511444655 Serrat, O. (2017). The SCAMPER technique. In Knowledge solutions (pp. 311–314). Springer. Sternberg, R. J. (1999). Handbook of creativity. Cambridge University Press. Summers, I., & White, D. E. (1976). Creativity techniques: Toward improvement of the decision process. Academy of Management Review, 1(2), 99–108. The Triz Journal – TRIZ Methodology, Tools, Articles and Case Studies. The Triz Journal. https:// triz-journal.com/. Accessed 12 May 2020. Transforming our world: The 2030 Agenda for Sustainable Development. Sustainable Development Knowledge Platform. https://sustainabledevelopment.un.org/post2015/ transformingourworld. Accessed 12 May 2020. TRIZ Matrix/40 principles/TRIZ contradictions table. http://www.triz40.com/TRIZ_GB.php. Accessed 12 May 2020. Ubora – Open Biomedical Engineering e-Platform for Innovation through Education. http://uborabiomedical.org/. Accessed 12 May 2020. Vérin, H. (1984). Le mot: ingénieur. Cultural Technology. Watson, C. E. (1975). Generating creativity, ideas and inventions: Developing creative people. Research Management, 18(3), 14–18. Wilkinson, M. D., et al. (2016). The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3, 160018. Wongrassamee, S., Simmons, J. E. L., & Gardiner, P. D. (2003). Performance measurement tools: The Balanced Scorecard and the EFQM Excellence Model. Measuring Business Excellence, 7 (1), 14–29. https://doi.org/10.1108/13683040310466690 Zampetakis, L. A., Tsironis, L., & Moustakis, V. (2007). Creativity development in engineering education: The case of mind mapping. Journal of Management Development.

Chapter 9

Methods and Technologies for the Personalized Design of Open-Source Medical Devices Andrés Díaz Lantada, William Solórzano, Adrián Martínez Cendrero, Rodrigo Zapata Martínez, Carlos Ojeda, and Juan Manuel Munoz-Guijosa

9.1

Introduction

Open-source medical devices (OSMDs) have emerged thanks to progresses in computer-aided design, simulation, and manufacturing (CAD-CAE-CAM) software resources, to advances in manufacturing technologies, including the advent of flexible production systems and additive manufacturing, and to improvements in medical imaging technologies and in related processing software. All these parallel innovations have enabled the personalization of medical devices and its affordable and rapid manufacturing, in many cases employing toolless approaches, which makes the production of complex-shaped geometries, like those required in many cases for interacting with the human body, technically and economically viable. Although mass-produced medical devices still rely on injection molding, casting, stamping, forging, high-speed CNC machining, turning and powder metallurgyrelated processes, to cite some relevant production methods, these options require from expensive systems and supporting tools, which only pay off for large series. These cost-related limitations prevent a more generalized personalization of healthcare technologies and even affect the equitable access to medical devices, especially in low- and middle-income settings or when rare pathologies or unconventional anatomical geometries or malformations are involved. The lack of A. Díaz Lantada (*) · A. Martínez Cendrero · R. Zapata Martínez · J. M. Munoz-Guijosa Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] W. Solórzano Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain Mechanical and Electrical Department, Universidad de Piura, Piura, Perú C. Ojeda Mechanical and Electrical Department, Universidad de Piura, Piura, Perú © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_9

191

192

A. Díaz Lantada et al.

personalization worsens performance, ergonomics, and even aesthetics. Besides, the use of traditional manufacturing processes leads to biomedical devices whose biomechanical performance is suboptimal, while advanced additive manufacturing technologies pursue a “3D matter made to order” paradigm (https:// www.3dmattermadetoorder.kit.edu/) and help to achieve biodevices with biomimetic inner structures, lightweight designs, and enhanced biological and biomechanical responses. Fortunately for the nascent field of OSMDs, the last couple of decades have seen a progressive diffusion of computational modeling tools and the rise of a plethora of quite reasonably priced design resources, imaging tools, and manufacturing systems, including many types of additive manufacturing technologies (3D printers), which allow for a straightforward design and production with a remarkable variety of polymers, ceramics, alloys, and biomaterials, in many cases apt for medical purposes. The makers’ movement has been also underpinned by communities of designers sharing, through open-source schemes, their designs, documents, and blueprints (i.e., CAD files), all of which promotes the replication of successful designs and, to some extent, their personalization, if interoperable formats are employed. Considering that innovative, open-source and even free software tools for processing medical images have been recently developed, the heyday of personalized medical devices may be just around the corner. The initial chapters of the handbook have focused on introducing the field of OSMDs and concentrated on describing and exemplifying synergic methods for their successful development, by following systematic approaches, taking regulations into account, employing relevant standards, minimizing risks, involving users, and fostering creativity. This chapter enters a more technical realm and addresses the more relevant technologies and methods for personalizing OSMDs, paying special attention to available open-source and freely accessible design software, computational modeling tools, medical imaging hardware and software, and related affordable manufacturing technologies, which are further analyzed in the following chapter. Two case studies illustrate the use of these resources before discussing current trends and challenges, linked to promoting the personalization of healthcare through OSMDs.

9.2 9.2.1

Overview of Technologies and Methods for the Personalized Design of OSMDs Open-Source Hardware for Promoting Personalized Designs

Since the dawn of additive manufacturing, the complex geometries achievable through layer-by-layer production processes proved extremely adequate for the development of medical applications and devices, from surgical training and

9 Methods and Technologies for the Personalized Design of Open-Source. . .

193

planning models, through surgical guides and personalized tools for surgery, to patient specific implants for tissue and organ repair or reconstruction (Díaz Lantada & Lafont Morgado, 2012). Consequently, procedures and technologies for bridging the gap between patients and CAD tools were developed. These involved both hardware advances, detailed here, and software innovations, covered in the following subsection. In terms of hardware, medical imaging technologies, like nuclear magnetic resonance and computed tomography, are among the most relevant medical advances of the twentieth century and were available well before the industrial development of additive manufacturing. The possibility of obtaining 3D reconstructions of the inner structures, organs, and tissues of patients and parallel software developments, enabling the use of patients’ information, as input for CAD programs, reformulated medical device design and promoted personalization. Even if such technologies were (and still are) too expensive for their straightforward incorporation to the product design field on a daily basis, now it is possible to hire CT or NMR services for product development purposes, and more affordable systems for varied engineering fields, like materials science, have been developed. Recently, some open-source hardware developments have been linked to “opening” complex and expensive medical imaging technologies, which will prove transformative, not only for increasing access to precise diagnoses worldwide but also for expanding the impact of personalized medical devices (Niezen et al., 2016). Less challenging, than working with patients’ inner structures, is the design of patientspecific orthoses and ergonomic supports, adapted in a personalized way to the external geometries of the human body (articular splints, face protecting devices, ergonomic handles, helmet therapies for babies, to cite just a few). To this end, optical scanning systems prove an interesting option. Even if the professional systems for highly precise optical scanning are not cheap (more than 10.000€ in many cases), interesting low-cost and even open-source DIY options exist, as showcased through a case study in Sect. 9.3. Table 9.1 below includes some of the most interesting open-source hardware resources for obtaining medical images and data from the human body for promoting the personalized development of healthcare technologies. Some have been previously mentioned in the annex tables of Chap. 1,

Table 9.1 Open-source hardware for obtaining medical images and data from the human body for promoting the personalized development of healthcare technologies Hardware Open Kinect

Details Community of makers developing open resources for using MS Kinect as sensor

Open MRI projects from Open-Source Imaging Initiative

Open-source magnetic resonance imaging resources developed by a community focused on affordable medical imaging technology Open-source desktop CT scanner

Open CT

Link/reference https://openkinect. org/wiki/Main_ Page https://www. opensourceimaging. org/ https://github.com/ tricorderproject/ openct

194

A. Díaz Lantada et al.

as happens with some of the resources included in the tables from present chapter, and cited in review publications, like the open CT scanner (Jansen, 2014).

9.2.2

Open-Source Software for Processing Medical Images

Successful modern medical technologies, in many cases, apart from relying on welldesigned hardware, require software for their operation or for processing their results and obtaining reliable and useful information, for preventive, monitoring, diagnostic, or therapeutic purposes. The previously mentioned open imaging hardware technologies can better deploy their potentials for transforming healthcare towards more equitable paradigms, if adequately combined with open-source software resources. These combinations prove especially useful when imaging technologies are used for obtaining patients’ details, which can be further used for personalized medical device design strategies. Additionally, in many cases, closed and opensource tools can be combined for the development of innovative, personalized, and more affordable medical devices, as further explained. Considering that open-source CT scanners and MRI machines are not yet common and still may lack some relevant functionalities and precision of state-of-the-art commercial systems, the personalized design of a medical implant may start with an imaging procedure using a commercial device with closed IP for obtaining a . DICOM file (acronym for “digital communications in medicine”) with the threedimensional information of the patient. Subsequently, the .DICOM file can be processed, resorting to commercial, freemium, or open-source solutions, and further used as input for design tasks employing a computer-aided design software, which can be also closed or open-source. Commercial solutions for manipulating medical images and obtaining 3D files, which can be further processed by CAD tools, are typically expensive (i.e., with licenses above 10.000€/year). Recently, the developers of medical imaging technologies and related processing software are also incorporating “prototyping” modules to their software tools, for enabling the linkage with CAD modeling and manufacturing technologies. These advances are in connection with device personalization trends including solutions like virtual and physical prototypes for surgical training and planning, surgical guides for enhanced procedures, and even implantable devices. However, such modules, again, are typically closed and expensive, at least as expensive as ad hoc software focused just on converting .DICOM files into CAD-usable formats. Fortunately, communities of developers have designed different solutions for managing medical images and personalizing biodevices, especially prostheses, based on the information obtained using CT-scanners, MRI systems, and other imaging resources. Some relevant examples are included in Table 9.2, which also presents some options for personalization based on external images and photogrammetry, usable for designing external orthoses.

9 Methods and Technologies for the Personalized Design of Open-Source. . .

195

Table 9.2 Open-source software for processing medical images and photographs Software 3D Slicer ITK-SNAP InVesalius

FaceGen

AliceVision

9.2.3

Details Open-source software for managing medical images and personalizing biodevices Open-source software for viewing .DICOM files from medical images and segmenting them Open-source software for the 3D reconstruction of computed tomography and magnetic resonance medical images Software with open basic version for creating 3D reconstructions of faces, linkable with CAD software, using just frontal and lateral photos Collection of open 3D reconstruction resources

Link/reference https://www.slicer.org/ http://www.itksnap.org/ pmwiki/pmwiki.php https://invesalius. github.io https://facegen.com/

https://alicevision.org/

Open-Source Computer-Aided Design, Engineering, and Manufacturing Software

The relevance of CAD-CAE software resources for modern product design is evident, and virtual prototypes are, in many cases, replacing conceptual models and rapid prototypes, allowing even (in some special cases) certification and direct production, from design to product. This impacts many industries, including the biomedical one. As previously mentioned, some years ago these resources were quite exclusive and only relevant research centers and companies with remarkable turnover could afford hiring professional design resources. However, in the last decade, very affordable or in most cases free open-source options, both for modelling and multi-physical simulation, have appeared, hence giving designers and enterprises with getting started in the design field. Some of them are summarized in Table 9.3 and further illustrated by means of a case study in Sect. 9.4, in which the personalized design of a hip prosthesis is presented. In turn, this democratization of computational modelling has also made an important impact on traditional software development companies, which now offer very interesting educational options and partnerships. After presenting some relevant open-source technologies and methods for the affordable personalization of medical devices, two case studies of personalized healthcare technologies are included in Sects. 9.3 and 9.4. These cases are linked, respectively, to a procedure for personalizing face protecting splints for safe sport practice and to a method for personalizing hip implants according to femoral stem morphology. The use and potentials of different open-source medical imaging software and hardware resources, of open-source repositories of medical images, of open-source CAD tools, and of affordable manufacturing tools, of which opensource options are also available, are thus illustrated. Through these studies, some challenges and trends, which are additionally discussed in Sect. 9.5, are also straightforwardly introduced and understood.

196

A. Díaz Lantada et al.

Table 9.3 Open-source software for design, simulation, and manufacture Software TinkerCAD

FreeCAD OpenSCAD OpenLSTO FEBio Software Suite Elmer Mesh Mixer Blender Ultimaker Cura Slic3r

9.3

Details Free collection of development tools including user-friendly CAD modules, electronic simulation toolkits, and educational resources Open-source software for computer-aided design Open-source software for computer-aided design Open-source software for level set-based structural topology optimization Open-source software for nonlinear finite element analysis in biomechanics and biophysics Open-source finite element software for multiphysical problems Free software by Autodesk for working with triangle meshes and .stl files Open-source 3D creation suite for designing and working with meshes Open-source software for generating G-code for manufacturing Open-source software for generating G-code for manufacturing

Link/reference https://www.tinkercad.com/

https://freecadweb.org/ http://www.openscad.org/ https://github.com/M2 DOLab/OpenLSTO https://febio.org/ http://www.elmerfem.org/ blog/ https://www.meshmixer. com/ https://www.blender.org/ https://ultimaker.com/en/ products/ultimaker-curasoftware http://slic3r.org/

Case Study: Personalized Design of a Face Protecting Splint for Safe Sport Practice

Nasal fractures, in spite of being the most common types of facial fractures, are often unrecognized and untreated at the time of injury. Different masks have been designed to help patients with broken nose, during their recovery, to perform normal life involving sport practice. However, their contact-based personalization and manufacturing processes sometimes worsen the injury and typically they are expensive devices, normally aimed at professional sport players. It is on this note that authors have designed a modular multipurpose face mask for aligning and protecting broken nose. Innovative processes for linking contactless personalization and rapid tooling for achieving personalized splints in composite materials have been developed and shared through the UBORA e-infrastructure. Several Class I devices, as this face protecting splint, may benefit from these processes. Following sections detail the personalized design and affordable manufacturing processes of these splints, which can be applied to other orthoses. A design optimization of this case study linked to risk minimization is also presented in Chap. 4 to illustrate FMEA and FEM techniques.

9 Methods and Technologies for the Personalized Design of Open-Source. . .

9.3.1

197

Medical Need and Product Description

Face protecting splints and related face orthoses are relevant for helping patients recover from craniofacial surgeries or facial injuries and their personalization is fundamental for enhanced functionality, ergonomics, and aesthetics. Apart from their medical uses, related products are getting more and more common in sports for high-profile athletes but may also support children from low- and middle-income settings in performing their daily activities (especially playing with their companions at school), while recovering for instance from a broken nose. To this end, more affordable solutions are needed. Besides, available manufacturing processes for personalized orthoses typically rely on contact-based procedures (a mold made directly upon the body of the patient or a cast applied directly to the skin), which in the case of face fractures is especially harmful and may even worsen the situation. The proposed design and manufacturing strategies may support the affordable and contactless personalized development of a wide set of biomedical devices and help to delocalize the supply chain for involving local populations in the development of medical technology. Regarding existing solutions for safe sport practice involving protecting the face have been reviewed. Protective equipment may include helmets, protective eyewear, mouth guards, and face protection, among others. Helmets prove uncomfortable for daily activities and sports where special peripheral visibility is needed, mainly football and basketball, in which injured professional players resort to splints more than to helmets. The helmet option would prove also inadequate at school, due to comfort reasons and increased sweat. Low-cost face guards and protecting splints are commercially available, but their mass production by injection molding prevents personalization (users have to select among a range of sizes, which leads to a wrong adjustment in many cases and may promote recurrence). Besides, these affordable mass-produced solutions are made of thermoplastic polymers, whose mechanical performance is limited. Consequently, professional players resort to face guards and protecting splints made of composite materials (fiber-reinforced polymers, usually CFRP), which can be personalized and lead to an adequate mechanical endurance but are consequently expensive. Even without personalization, a mass-produced CFRP face protecting splint costs around 200€ depending on manufacturer. To overcome the mentioned limitations, it is interesting to explore synergies between additive manufacturing and composite materials, which may support the medical device industry to advance toward affordable and personalized production. With present device, authors demonstrate the potential of 3D printed rapid molds, adequately post-processed, upon which fiber reinforced polymers (FRPs) can be laminated and cured in a very straightforward and affordable way, for the personalized development of face protecting splints. The concept and preliminary 3D printed prototypes (PLA) were developed as cases of study for the first UBORA Design School of Nairobi, in which an alternative concept of protecting splint was also developed (Ahluwalia et al., 2018a). The

198

A. Díaz Lantada et al.

development of rapid tooling processes for manufacturing with different types of fiber-reinforced polymers has been carried out, led by Prof. Muñoz-Guijosa, at UPM along years 2018 and 2019 and is related to previous publications describing similar procedures for the design and manufacturing of articular orthoses (Munoz-Guijosa et al., 2020). Considering potential users, the development focuses on amateur and semiprofessional sport practitioners that would like to continue developing their training, while recovering from a broken nose or face surgery and wishing a protecting device which can be personalized but still be affordable. At the same time, children with broken nose due to accidents typically remain at home for 3–4 weeks, while they recover from the injury, for the fear of their parents that the fracture may be recurrent during normal playing with companions. In some cases, an adapted design may have cosmetic or even sun-/chemical-protective function. Regarding specifications, the face protecting splint must be a lightweight device for enabling comfortable sport practice, should be personalized for improved ergonomics, and has to provide structural integrity to withstand eventual impacts present in sport practice. Therefore, fiber-reinforced polymers are chosen for their excellent compromise between strength and lightness. Additional quantitative details are provided in the conceptual design stage. Quality control during manufacturing is required, to guarantee adequate lamination and to check the absence of cutting edges, typical of not post-processed composite parts. A post-processing is hence needed to provide a final polishing and enhance comfort. Materials employed should withstand humidity (i.e., sweat) during sport practice, and their mechanical performance should not be thereby affected. Final roughness should be minimized to avoid skin abrasion. Besides, the device and its materials must comply with ISO 10993, which deals with device biocompatibility, as short-term contact with skin is expected. More quantitative details concerning the specifications of the device can be found in the UBORA’s site of the “face protecting splints”, shared as open-source medical devices. According to the classification annexes of EU MDR 745/2017, implemented in the UBORA e-infrastructure to guide researchers in medical device classification, the “face protecting splint for safe sport practice” is, as expected, a Class I Medical Device, as it is non-invasive; does not challenge or store blood, body liquids, or tissues; is not active; does not incorporate medicinal products with ancillary function; is not used for contraception or for prevention of sexually transmitted diseases; is not intended for disinfecting, cleaning, or rinsing contact lenses; is it not intended for sterilizing medical devices; does not use human or animal tissues or cells and does not incorporate nanomaterials; is not intended to administer medicinal products; is not introduced into the organism; and does not incorporate a diagnostic function.

9 Methods and Technologies for the Personalized Design of Open-Source. . .

9.3.2

199

Methods for Personalized Design of the Face Protecting Splints

In order to reach the personalized creation of face protecting splints, an adequate and precise approach consists of starting with the 3D digitalization of the patient’s or user’s face, as input for the CAD procedure and as an alternative to more rudimentary or manual processes based on taking some key measurements upon the user’s or patient’s head or face (i.e., distance between cheekbones, size of eyes and eyebrows, etc.). Open-source alternatives – for non-commercial use – combining easily accessible and low-cost hardware with specialized software programs co-exist and may support the three-dimensional reconstruction of external corporal geometries, in this case the head of final user or patient. Among these, it is important to mention the combined use of Microsoft Kinect for Xbox 360 (Microsoft Corporation, Redmond, Washington, USA) with Skanect 3D scanning software (Occipital Inc., San Francisco, CA, USA), especially developed for working with systems such as Kinect, Asus Xtion, or Structure Sensor, which provided the possibility of exporting to different CAD file formats, including .stl, .obj, .ply, and .vrml, directly from the images obtained with the mentioned imaging systems. Other alternatives may resort to the use of photogrammetry techniques and related software for recovering the exact positions of surface points using photographs as input. Several options exist, many of them working as apps within smartphones, depending on the size and type of object to reconstruct and on the required precision (i.e., SnapChat and others). In this case, we rely on an extremely straightforward, versatile, and easy to use tool for three-dimensionally reconstructing faces and heads using just three photos (one frontal and two lateral ones) as input and only a few clicks for the reconstruction: the FaceGen 3D Print (by Singular Inversions Inc., Toronto, ON, Canada). This tool does not only reconstruct the 3D geometry of the face but also links it to additive manufacturing resources via .stl files. We used a free demo version of the software, which is enough for our validation purpose, although very affordable alternatives provide additional features, including hairstyles, beards, and stands. Once the .stl file is generated, Autodesk Meshmixer is employed as support software for improving mesh quality and obtaining a softer refined surface (with additional number of elements), which is then used as input for the computer-aided design of personalized splints and related molds, as detailed in the following subsection. Results from digitalization with FaceGen, following a photogrammetry strategy, is shown in Fig. 9.1, while Fig. 9.2 presents the alternative based on digital scanning previously mentioned (Kinect). The procedure from Fig. 9.1 is even adequate for nonprofessional designers, as it can be easily performed following a simple video tutorial, which could help to expand these design processes and incorporate them in the future in the point-ofcare for being used by healthcare professionals. The optical procedure from Fig. 9.2 is more complex, as it needs the support of at least one professional devoted to managing the scan (although investing in facilities such process could be automated)

200

A. Díaz Lantada et al.

Fig. 9.1 Face digitalization process combining FaceGen photogrammetry program, in which an ordered selection of key points upon frontal and lateral images leads to a 3D reconstruction of the face. Final post-processing with Meshmixer is applied to enhance surface quality

Fig. 9.2 Alternative face digitalization process to obtain a .stl file with the outer morphology of a healthy volunteer by combining Microsoft Kinect for Xbox 360 as optical scanner with the processing using Skanect software

9 Methods and Technologies for the Personalized Design of Open-Source. . .

201

and one for controlling the scanning process. In terms of fidelity to the original model, for complex geometries such as that of the face, the optical scanning provides worse results, due to shades near the nose, than the photogrammetry. However, for other orthoses, such as articular splints, the optical scanner proves a very adequate option, as previously reported (Munoz-Guijosa et al., 2020). In current case study, the face protecting splint is further designed by working on the digitalization of Fig. 9.1, as shown in the following pages. Regarding actual design of the medical device and its production tools, we opt for NX-11 (Siemens PLM Software, Plano, Texas, USA) as CAD software of choice, due to its very special features for generating surfaces and interacting with .stl models. After importing the .stl file of the three-dimensionally reconstructed head and adequately scaling the file to the real size of the volunteer involved in the study, the design of the personalized splint starts: first of all, a frontal plane is created, and a sketch with the desired frontal proportions of the protecting splint is drawn upon it. Then, the sketch is projected upon the surface of the .stl reconstruction to create the profile of the splint upon the real face surface. With the support of surface design tools of NX, the surface of the protecting mask is created, and the “thicken” tool is used for obtaining a solid object. Final “pocket” and “hole” operations are used upon the solid model to obtain the desired design (.prt file), and conversion to .stl file enables connection with low-cost 3D printing systems or in general with almost all additive manufacturing technologies. Taking the design of alternative mold options into account, following the coreshell approach typical of injection molding systems, the process starts with the surface of the personalized splint by closing the holes designed for the user’s eyes and by extending the boundaries of the surface to define a partition surface (using “sew” and different available surface design and modification tools). The generated partition surface is employed to cut a block of material in the two possible tools, which can be used for lamination: one with a concave geometry, the other with a convex one. Final fine-tuning design operations, like the incorporation of some extrusions, for helping to laminate in the area surrounding the eyes, or some small cavities for enabling the use of screwdrivers and tools for helping with the demolding process after lamination, lead to the final computer-aided designs of the personalized molds. The complete design processes, both for the splints and for the related molds, are presented schematically in Figs. 9.3 and 9.4: Fig. 9.3 presents schematically the already described process for generating a solid and personalized CAD file with the geometry of the desired face protecting splint, which mainly includes the import of the 3D head reconstruction, the creation of a planar sketch and its projection upon the face surface, the tessellation of a surface and its thickening for obtaining a final printable solid. Figure 9.4 shows the creation of CAD models of different mold alternatives, which include the features of the personalized protecting splints and may serve as input for additive manufacturing technologies (a low-cost 3D printer in our case), which lead to the physical construction of rapid molds, upon which the final masks can be laminated.

202

A. Díaz Lantada et al.

Fig. 9.3 Schematic illustration of the CAD process for the personalized face protecting splints. Conceptual prototypes obtained by 3D printing and tested by the authors

9 Methods and Technologies for the Personalized Design of Open-Source. . .

203

Fig. 9.4 Mold design process based on the personalized face protecting splint CAD model.

The purposely designed parts are prototyped for the first time using a BCN 3D Sigma printer using conventional PLA filament. Figure 9.3 includes also results from alternative conceptual 3D-printed prototypes and the first ergonomic evaluations of the personalized face protecting splints designed for a team of healthy volunteers. These first prototypes are meant only for concept evaluation, for analyzing geometries, and for checking the quality of the digitalization process toward personalized design tasks. The printed masks adapt adequately in terms of dimensions and morphology to the volunteers, but these prototypes are not functional. In fact, the thickness of around 1 mm leads to quite flexible and not so tough prototypes, which can be damaged just by bending them within the hands. In any case, the geometries and morphology are correct, and the process of linking these geometries with rapid tooling, for the creation of first prototypes in composite materials, is further explained in the following section.

204

A. Díaz Lantada et al.

Starting from the personalized splint, an extended partition surface is created and used to conform the mold surfaces, upon which the composite fabric plies can be laminated and further autoclaved. Convex and concave alternatives are presented and tested

9.3.3

Methods to Produce the Personalized Face Splints in High-Performance Materials

Molds for lamination and curing of FRPs must fulfill several geometrical, mechanical, and thermal specifications, related to the final properties required for the cured FRP splint. Firstly, the mold must be able to withstand the mechanical loads related to the contraction suffered by the laminate during the curing process, as well as those produced by the high curing pressure, up to 0.8MPa, if autoclave curing is used. In addition, the mold must maintain the required stiffness and strength at the curing temperature, which may be as high as 180 C if autoclave curing is carried out. The mold mean surface roughness Ra must be small (in the order of 5 microns) for obtaining adequate aesthetical and/or ergonomic properties at the splint visible and skin contact surfaces. These properties cannot be directly obtained with the materials usually employed in fused deposition modeling, which, on the other hand, would allow for a rapid and low-cost manufacture of the desired molds. Consequently, special design considerations must be taken into account. Firstly, the outer shells of the mold are printed (Fig. 9.5). A dual-extruder BCN3D Sigma machine (BCN 3D Technologies, c/ Esteve Terradas, 1, 08860, Castelldefels, Barcelona) is used again with an ABS filament from Proto-pasta, white in this case, as printing material. If vacuum or pressure-assisted curing is to be used, the shell containing the lamination surface must include additional flat areas for the subsequent attachment of the supplementary materials needed. Once the outer shells of the molds are printed, a coating of epoxy resin is applied on the lamination surface (see Fig. 9.5 again) to reach the required roughness. Subsequently, minor gaps between printed layers are closed with the help of adhesive tape, which is needed for the next process. After fine-tuning of the printed mold shells, they are filled with a plaster slurry (liquid), which is hardened and dehydrated by placement in oven at 50  C for 2 h. Heating and cooling slopes of 1  C/min and 0.5  C/min, respectively, are employed. This filling provides a relevant improvement in terms of mechanical endurance and of heat absorption capacity, as required for the final lamination of composite fibers, either inside autoclave or at ambient pressure. The thermal and mechanical properties rely on the plaster core and the surface roughness properties on the epoxy coating, being the ABS shell the means for shaping the desired geometry. For this reason, the thickness of the ABS shell must be as reduced as possible. A thickness of 1 mm in the mold shells is obtained with the 3D printer employed.

9 Methods and Technologies for the Personalized Design of Open-Source. . .

205

Fig. 9.5 Rapid mold manufacture procedure, presented step-by-step, including (a) 3D printing process, (b) 3D-printed mold, (c) clay filling, and (d) final mold after heat treatment

206

A. Díaz Lantada et al.

Carbon fiber/epoxy prepreg is used as lamination material, together with autoclave curing, for the demonstration of the process, as it is the most complete processing scheme. However, the presented methodology can be easily applied in less complex and more economical processes, as hand lay up of manually polyesterimpregnated long fiber glass laminates or chopped strand mats. The prepreg laminate used are Toray T700C/epoxy with a surface weight of 200 g/m2 and a νF of 60%. The lamina thickness is approximately 0.2 mm. Each lamina is manually cut by means of special fiber scissors, so that the approximate final shape required for lamination is achieved and carefully placed at the mold-free surface and compacted by means of a compaction roll. Scissors are again used to remove the excess material and conform the splint edges as approximate as possible to the final desired shape. Small patches are cut for achieving a good geometrical accuracy in the complex zones, as nose or eye cavities, which present double curvatures and sharp curvature changes (Fig. 9.6). Once lamination is finished, a peel-ply is positioned over the outer lamina. A release film and a breather ply are positioned afterward. Finally, a vacuum bag is attached using tacky tape at the surface edges. The vacuum bag is perforated for connecting the vacuum valve to the vacuum tube (Fig. 9.6). The whole set is introduced in the autoclave (Fig. 9.6) and conducted to a curing cycle with a curing temperature of 120 C and a curing pressure of 5 bar. After curing, the laminate is carefully separated from the mold, and the edges are cut for its fine-tuning to the desired shape (Fig. 9.6). Figure 9.5 includes schematically the rapid mold manufacture procedure, presented step-by-step, including the 3D printing process (Fig. 9.5a), the 3D-printed mold (Fig. 9.5b), an alternative concave mold shell (Fig. 9.5c), the clay filling process (Fig. 9.5d), and the final mold after heat treatment (Fig. 9.5e). Some cello-tape patches can be seen in the mold for covering minor holes and manufacturing defects of the 3D printing process and, hence, preventing from leakage during the clay filling process. The final mold is polished and coated with epoxy resin for reaching the stage shown in Fig. 9.6a, upon which the composite fibers are laminated and subject to vacuum as schematically presented in Fig. 9.6 (images b–d). The final mask structure in composite material is presented in Fig. 9.6e, together with the 3D-printed mold after the production process. It is important to highlight that the lateral grooves planned in the mold help with demolding of the laminated masks and that the mask structure, in spite of requiring a final polishing, stands out for its structural integrity, stiffness, and lightweight, when compared with the 3D printed proof of concept model in PLA. After adequately laminated structures, postprocessing such as polishing, coating with silicone, covering with a fabric or textile coat, and drilling of holes or milling of grooves, so as to incorporate the elastic bands that help to place the face protecting splint in stable position, is needed. Ergonomic evaluation of these composite material prototypes and systematic evaluation of risks leads to improved design and processes (see the optimization presented in Chap. 4). This case study illustrates innovative manufacturing processes, in which the advantages of low-cost and straightforward additive manufacturing and composite materials and combined in a synergic way, by providing a complete development case of study linked to the personalized development of face protecting splints. The

9 Methods and Technologies for the Personalized Design of Open-Source. . .

207

Fig. 9.6 Lamination of composite fibers upon 3D-printed mold and pressurization during curing toward personalized final splints in high-performance materials (CFRPs)

208

A. Díaz Lantada et al.

procedure stands out for its affordability and personalized focus, which opens new horizons in the biomedical industry towards the delocalization of supply chains and the progressive involvement of local populations in low- and middle-income settings, both during the decision-making steps and during the implementation and operation phases, which may bring not just personalized and cheap highperformance solutions but also improve the local economies.

9.4

Case Study: Personalized Design of an Innovative Hip Prosthesis

9.4.1

Medical Need and Product Description

Total hip arthroplasty (THP) is considered to be one of the most successful procedures in orthopedics along last decades and its use is rising particularly in active and young patients, consequence of the increased need for better quality of life (Toth & Sohar, 2013; Drosos & Touzopoulos, 2019). The decreasing average age of patients and the growing expectations about the implant durability (Bergmann et al., 2016) lead to great variations in terms of implant fixation (cemented or cementless), materials employed for the load-bearing surfaces (polyethylene, ceramic, or metal), and surgical procedures, with a growing interest in less invasive approaches and implants structures (with recent developments linked to innovative ultra-short or short stems) (Toth & Sohar, 2013). Santori et al. (Santori et al., 2006) began, in 1993, to develop a new femoral implant whose objective was to save a good bone stock, for a possible revision procedure, and to promote a physiological strain distribution on the proximal femur, achievable through a proximal load transfer from implant to the femoral bone, with complete absence of the diaphyseal portion of the stem. The use of this kind of stems is increasing in the last years, as a consequence of the increment of follow-up studies (Morales de Cano et al., 2018; Gombár et al., 2019; Rinaldi et al., 2018), whose results show that these implants preserve bone stock and enable placing a conventional stem, at the time of revision surgery. The case presented here shows the personalized design of a short stem implant using open-source technologies.

9.4.2

Materials and Methods

9.4.2.1

Geometric Model

Personalized stem design requires a virtual model of cortical and trabecular femur. In this study, these models are obtained by importing into 3D Slicer 4.10.2® a “Digital and Communications in Medicine” (.DICOM) file provided by the Cancer Imaging Archive, a virtual library that hosts a large archive of medical images accessible for

9 Methods and Technologies for the Personalized Design of Open-Source. . .

209

Fig. 9.7 (a) The segmentation of right femur using 3D Slicer®. (b) The process to obtain the trabecular bone using Boolean tools. (c) The modified cortical bone achieved using Meshmixer®. (d) The solid file of cortical bone in FreeCAD®

public download. For this example, reference TCGA-VP-A878 is used. The . DICOM file has a slice thickness of 2 mm on the axial plane and of 0.909 mm on the coronal and sagittal planes, being each one 512  512 pixels in size. For obtaining the virtual model of the bone, it is necessary to segment the right femur and its cortical region by using threshold, level tracing, paint, erase, and smoothing tools. Then, the trabecular part is obtained by employing logical/Boolean tools. Both bones are exported from 3D Slicer® as .stl files and imported into Meshmixer 3.5® for inspecting, smoothing, and remeshing. Stl. file is a mesh archive used mainly for additive manufacturing, but further design operations required solid files formats like IGES, STEP, or Parasolid. FreeCAD 0.18® allows converting a mesh into a solid file employing its part tools. The software creates a shape from a mesh and then converts that shape into a solid body. In this case study, FreeCAD® is employed to perform the morphological study and the stem design (Fig. 9.7).

9.4.2.2

Morphological Study

The correct study of the morphology of the proximal femur is essential because it is the region that typically suffers from long-term bone resorption in most state-of-theart implants (Solórzano et al., 2020). It is also relevant for preoperative planning, because its state helps the surgeon to select the best stem or type of prosthesis, which may guarantee a long-term success rate (Carter et al., 1995; Crooijmans et al., 2009). Surgeons and designers usually study the morphology of the femur using two-dimensional images, such as standard radiographs, but this method is imprecise, when compared to the application of other medical imaging techniques, including computed tomography (CT) or magnetic resonance images (Iguchi et al., 1996; Rubin et al., 1992). Three-dimensional femur models obtained from CT scans provide better information that normally allows for a personalized evaluation of each patient. A morphological study is essential for the development of personalized cementless femoral stems because accurate dimensions, adapted to patients’ morphologies, may ensure primary fixation stability, enhance the bone-implant interface, and prevent stress shielding (Baharuddin et al., 2014).

210

A. Díaz Lantada et al.

Fig. 9.8 (a) Femoral coordinate system. (b) Anteversion estimation technique. (c) Osteotomy level. (d) Datum planes to assess the morphology of femur. (e) Oblique femoral slices

Morphological studies to design conventional stems usually involve making orthogonal virtual cuts to take measurements in the mediolateral, anteroposterior, and neck-oriented planes, although this procedure may change for innovative hip stems. First, designers should find the anteversion, the angle between the femoral neck axis and the line connecting posterior condyles in transverse view (Noble et al., 1988), to place the neck-oriented plane. A novel empiric method is proposed to estimate this angle: first, a sphere concentric with the femoral head and with similar diameter, approximately 50 mm, is created (see Fig. 9.8a). Subsequently, the sphere is cut through the middle, using a transverse plane. Then the curve is projected onto the plane and two lines are drawn, one going through the center of the projection and the other horizontal and tangent to the curve. Finally, the fourth part of the angle between these two lines is the estimated anteversion (Fig. 9.8b). Proximal femur is cut considering the accepted neck shaft angle of 135 (Gilligan et al., 2013) taking as reference a plane below the lesser trochanter (Fig. 9.8c). This angle is used to cut the femoral neck, to insert the stem into the femoral cavity, and may also serve as a guide to analyze the morphology of the femur. In a different way to the conventional analysis, which relies on orthogonal slices to evaluate the shape of the femoral canal, authors suggest using oblique slices because they help to achieve the correct lateral adjustment of the stem and simplify the design. Supplementary neck shaft angle (SNSA), composed of I and VI lines (Fig. 9.8d), is divided into five equal parts. As consequence, the lines II, III, IV, and V are added, and all of them serve as benchmarks to create datum planes that allow obtaining slices to evaluate the femoral channel (Fig. 9.8e). SNSA can be divided into more equals

9 Methods and Technologies for the Personalized Design of Open-Source. . .

211

parts but five gives great results. Also, this technique can be used to design conventional stems and obtain natural metaphyseal shapes.

9.4.2.3

Round-the-Corner Technique

The presence of a pronounced lateral flare in the design requires from a new insertion method, so as to achieve femoral broaching and stem insertion with complete respect for the greater trochanter and the gluteus muscles. This technique has been named “round the corner” and is made possible thanks to the absence of the distal portion of the stem. This technique requires the broach to be, first, inserted and hammered down in varus and, then, progressively tilted in the correct alignment, while progressing down the femoral metaphysis. Round the corner facilitates the use of minimally invasive approaches but precludes the use of intramedullary guides and can also result in a varus position, in which the tip of the stem can touch the lateral cortex, contributing to a potential fracture; for that reason, the use of fluoroscopy is advisable (Santori et al., 2006; Gombár et al., 2019; Santori et al., 2018).

9.4.3

Personalized Design

Based on the morphological study, that uses oblique slices to measure the femoral channel, it is possible to design a personalized short stem using the same planes (Fig. 9.9a) and considering the estimated anteversion to define the cross-sections of the prosthesis (Fig. 9.9b). Trapezoidal, oval, elliptical, and circular cross-sections may be used for stem designs, but previous studies (Rinaldi et al., 2018; Sabatini & Goswami, 2008) suggest that the circular and elliptical cross-sections produce good stress distributions along the stem, allow its primary fixation, and enhance its adaptability to different sections with changes of shape and size. Consequently, authors employ elliptical cross-sections in the first five planes and a circular crosssection in the sixth plane for this personalized example (Fig. 9.8c–e). Cross-sections are interpolated employing the loft tool of FreeCAD®. Edge rounding is required in the last section (Fig. 9.9c) to facilitate the access of the stem, employing the round the corner technique, in connection to improved implantability. Designers should carefully evaluate the cross-sections, as in some cases the channel may not be large enough to allow the entry of the short stem. In some interventions, surgeons employ another kind of osteotomy to expand the existing cavity and prevent fractures during the procedure. The sphere used in the morphological evaluation helps to design stem neck, as it is located in the center of the femoral head and contains information about the offset, an important parameter in these designs. Supporting planes adapted to the bone help to sketch the crosssection of the prosthesis neck, whose first part has the form of a truncated cone, which ends in a cylinder. The loft tool is used for obtaining a solid body from different sections and Boolean tools are used to unite the stem of the prosthesis with

212

A. Díaz Lantada et al.

Fig. 9.9 (a) Morphological study of the proximal femur. (b) Cross-sections of the stem. (c) Innovate hip stem. (d) Sketch of the neck stem. (e) Hip prosthesis without the rounded neck edges. (f) Final design. (g) Round-the-corner technique for implantation. (h) Isometric view of implanted prosthesis. (i) Transverse view comparing the intact and implanted femur

its neck (Fig. 9.9e). Stress concentration is minimized by rounding the edges of the stem neck (Fig. 9.9f), and, as a final step, viability is verified by emulating the implantation of the prosthesis in varus position, which helps to prevent clinical problems associated with the design (Fig. 9.9g). Then to check that the estimated anteversion is adequate, the implant is placed in position and compared to the intact femur, which serves are inspiration for biomimetic design, by using isometric (Fig. 9.9h) and transverse (Fig. 9.9i) views. The proximity or distance of the stem from the cortical bone depends on the mechanical features of the trabecular bone. Indeed, if spongy bone has poor properties, the stem must be closer to the cortical bone; otherwise, it can lean on the trabecular part. Once a personalized design is achieved, in silico studies (i.e., FEM simulations) should be also performed to evaluate eventual stress shielding and to estimate the quality of the primary fixation. For these studies, it is necessary to obtain the mechanical properties of the bones, which helps designers to recreate and simulate the environment, where the stem will be implanted, and to consider the external conditions that may influence long-term success. Solórzano et al. (Solórzano et al., 2020) propose to use transverse isotropic properties, estimated by using the mean Hounsfield unit of the segmentation for each bone region. These may be obtained from the CT images by employing the “Segment Statistics” tool of 3DSlicer® (Fig. 9.10a). Once the geometries, materials, loading, and boundary conditions are defined, the model may be solved, by applying FEM. Post-processing using the Tsai-Wu criteria and the maximum principal stresses provides useful

9 Methods and Technologies for the Personalized Design of Open-Source. . .

213

Fig. 9.10 (a) Trabecular bone segment statistics obtained from 3DSlicer®. (b) Anisotropic mesh obtained from Bonemat®

information about over- and underload regions of the intact and implanted femur. This helps to analyze the stress shielding phenomena and potential failures. A more precise analysis may rely on considering bone as a completely anisotropic material. Bonemat 3.2®, a free software that provides finite element meshes for bone, with elastic properties derived from CT images (Fig. 9.10b), may be employed for these more realistic studies with bone as a material with different properties in each point of space. The obtained design is finally illustrated with a couple of conceptual prototypes manufactured by laser stereolithography (SLA-3500 System by 3D Systems), as presented in Fig. 9.11. Direct manufacturing of more mechanically reliable and potentially biocompatible personalized hip implants should be performed with other technologies, like powder-based laser melting employing Ti alloys, to cite an option based on additive approaches.

9.5

Challenges and Trends: Toward Open and Personalized Medical Devices

Previous sections have presented different open-source hardware and software resources for medical image acquisition, processing, and consequent design personalization, in connection with the development of open-source medical devices. Such software and hardware resources, just a couple of decades ago, could involve costs of ca. 5,000–25,000 €/year for advanced computer-aided design and engineering software, 20,000–50,000 €/year for medical imaging processing software, and more than 60,000 €/year for the operational costs and depreciation of medical imaging hardware. This led to extremely challenging economic barriers for innovators and start-ups entering the field of medical device design, especially if personalized approaches were to be promoted. The progressive rise of open-source computeraided design tools, open-source medical imaging processing software, and even

214

A. Díaz Lantada et al.

Fig. 9.11 Rapid prototypes for geometrical validation employing laser stereolithography

open-source hardware for medical imaging (Jansen, 2014; https://hackaday.io/ project/9281-murgen-open-source-ultrasound-imaging; https://www. opensourceimaging.org/), as previously discussed, have importantly reduced the mentioned barriers and contributed to the emergence of the makers’ movement, also supporting the nascent field of OSMDs. The cases of study presented help to illustrate the current design paradigm, which is characterized by a much more affordable and straightforward access to design and prototyping resources, through which OSMDs (and their methods) are emerging as an alternative to the traditional biomedical industry. However, now that affordable technologies are more widespread, accessible and GDPR-compliant data (see Chap. 7) probably constitute the quid of the matter, for value generation in the open-source paradigm. Indeed, data sharing following FAIR principles (Wilkinson, 2016), while respecting privacy rights, is an extremely powerful competitive advantage for open-source communities and initiatives, especially in the medical field. Through data sharing, users’ needs can be better considered, along the engineering design process of innovative healthcare technologies;

9 Methods and Technologies for the Personalized Design of Open-Source. . .

215

safety may be promoted, thanks to increased scrutiny by developers and end users; and innovative developments and methods can be rapidly transferred to society. Ideally, all these benefits may lead OSMDs to outperform medical devices developed behind closed doors, as outlined in the Kahawa Declaration (Ahluwalia et al., 2018b). However, the privacy and ethical issues, involved in patients’ data sharing, should be always considered and the open-source community should always act according to law: open-source medical devices should be compliant to the applicable medical device regulations and developed respecting the privacy of participants, in Europe following the GDPR. Such regulations (EU MDR 745/2017 and GDPR) may be also excellent references for developing countries progressively setting up medical device industries. Access to data like medical images or body tissues and organs reconstructed from such medical images may be critical for demonstrating the technical and economic viability of an innovative medical device benefiting from design personalization. However, in many cases, it may be challenging to access patients and their medical images or medical imaging resources may not be available for biomedical engineering purposes, especially in low- and middle-income countries. Important economical and human resources may be consumed in long-lasting procedures, if field studies involving the imaging of tens of patients are required, which may limit the speed of response, innovation capability, and societal impact of designers in emergent economies. Consequently, promoting access to FAIR data in biomedical engineering may be a transformative trend, as further described. According to the desired accessibility to medical images and corporal CAD geometries, for research, development, and innovation purposes, which is directly linked to the personalization of medical devices designs in biomedical engineering, several teams or researchers and developers, as well as communities of innovators, have undertaken remarkable initiatives, summarized in Table 9.4. Other studies and databases provide additional references (http://www.aylward.org/notes/open-accessmedical-image-repositories). Initiatives fostering open-source solutions for improved communication in the medical field, communities collaborating in the development of open-source solutions for medical imaging, open-source collections of medical images and open CAD libraries of body structures, together with the documents and files shared by other open CAD and open BME communities, are opening biomedical data, strictly complying with regulations, for the benefit of more rapid advances in healthcare technology. All of them complement, in an excellent way, the advances regarding open-source medical imaging processing software, open-source design and simulation programs, and open-source digitalization hardware, described before, and synergize with the open-source manufacturing resources detailed in the following chapter. As in the case study described in the previous section, the use of such open repositories helps to evaluate the potential benefits of a personalized design or the potentials of personalized healthcare technologies developed using innovative strategies, by designing according to a specific geometry, but without needing to involve patients and imaging tools, hence accelerating the discovery path and innovation cycle.

216

A. Díaz Lantada et al.

Table 9.4 Initiatives sharing open-source medical images, designs, and anatomical information Initiative Open MRS

Open-Source Imaging Initiative The Cancer Imaging Archive Sicas Medical Image Repository Thingiverse GrabCAD Body Parts 3D NIH 3D printing exchange Bioverse Kaggle

9.6

Details Open-source medical record system platform, opensource solutions for improved communication, open-source solutions for big-data health Community collaborating in the development of open-source solutions for medical imaging Open-source collection of medical images for cancer research and treatment Freely accessible repository containing medical research data including medical images, surface models, clinical data, genomics data, and statistical shape models Open CAD library of all sorts of industrial components and devices Open CAD library of all sorts of industrial components and devices Open CAD library of body structures generated from medical images Repository for sharing medical files and blueprints linked to medical 3D printing Open CAD library of biodevices for the biofabrication field Community of data scientists that publish data sets, explore, and build models to solve data science challenges such as disease diagnosis using medical imaging

Link/reference https://openmrs.org/

http://www. opensourceimaging. org/news/ https://www. cancerimagingarchive. net/ https://www.smir.ch

https://www. thingiverse.com/ https://grabcad.com/ http://lifesciencedb.jp/ bp3d/ https://3dprint.nih. gov/ http://bioverse.co/ https://www.kaggle. com

Conclusions

Relevant technologies and methods for personalizing OSMDs have been introduced in present chapter. Special attention has been given, among others, to available opensource and freely accessible design software, computational modeling tools, medical imaging hardware and software, and related affordable manufacturing technologies. The role of communities of engineers, healthcare practitioners, patients, associations, and citizens has been also highlighted, as fundamental communities of co-creators, and “makers” are already working together for opening the information, resources, and methods needed to underpin the personalization of healthcare through patient-specific designs of medical devices. Two case studies have helped to illustrate these methods and resources and supported authors for discussing current challenges and the relevance of some emergent trends. Taking into account that open-source medical devices developers should not only share the final design results as open CAD files (or blueprints) but should also precisely present all the relevant information about the whole development process, disclosing the materials,

9 Methods and Technologies for the Personalized Design of Open-Source. . .

217

methods, and technologies employed, authors believe that the case studies presented contribute to the open-source movement in the development of medical technology. 

Acknowledgments This research has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No. 731053: “UBORA: Euro-African Open Biomedical Engineering e-Platform for Innovation through Education,” which also funds the participation of Adrián Martínez and Rodrigo Zapata in this study. We acknowledge the continued support of Mr. Pedro Ortego García in rapid prototyping tasks and manufacturing processes. William Solórzano acknowledges the support of Fondo Nacional de Desarrollo Científico, Tecnológico y de Innovación Tecnológica (FONDECYT) of Perú that through the contract N 316-2019 financed an internship to Product Development Laboratory at Universidad Politécnica de Madrid for 3 months.

References 3D Matter Made to Order Cluster of Excellence, Karlsruhe Institute of Technology and Heidelberg University. https://www.3dmattermadetoorder.kit.edu/ Ahluwalia, A., De Maria, C., Madete, J., Díaz Lantada, A., Makobore, P. N., Ravizza, A., Di Pietro, L., Mridha, M., Munoz-Guijosa, J. M., Chacón Tanarro, E., et al. (2018a). Biomedical engineering project based learning: Euro-African design school focused on medical devices. International Journal of Engineering Eduction, 34, 1709–1722. Ahluwalia, A., De Maria, C., & Díaz Lantada, A. (2018b). The Kahawa Declaration: A manifesto for the democratization of medical technology. Global Health Innovation, 1, 1–4. Baharuddin, M. Y., Salleh, S.-H., Zulkifly, A. H., Lee, M. H., & Mohd Noor, A. (2014). Morphological study of the newly designed cementless femoral stem. BioMed Research International, 2014. https://doi.org/10.1155/2014/692328 Bergmann, G., Bender, A., Dymke, J., Duda, G., & Damm, P. (2016). Standardized loads acting in Hip implants. PLOS ONE, 11(5), 1–23. https://doi.org/10.1371/journal.pone.0155612 Carter, L. W., Stovall, D. O., & Young, T. R. (1995). Determination of accuracy of preoperative templating of noncemented femoral prostheses. Journal of Arthroplasty, 10(4), 507–513. https://doi.org/10.1016/s0883-5403(05)80153-6 Crooijmans, H. J. A., Laumen, A. M. R. P., van Pul, C., & van Mourik, J. B. A. (2009). A new digital preoperative planning method for total hip arthroplasties. Clinical Orthopaedics, 467(4), 909–916. https://doi.org/10.1007/s11999-008-0486-y Díaz Lantada, A., & Lafont Morgado, P. (2012). Rapid prototyping for biomedical engineering: current capabilities and challenges. Annual Review of Biomedical Engineering, 14, 73–96. Drosos, G. I., & Touzopoulos, P. (2019). Short stems in total hip replacement: evidence on primary stability according to the stem type. HIP International, 29(2), 118–127. https://doi.org/10.1177/ 1120700018811811 Gilligan, I., Chandraphak, S., & Mahakkanukrauh, P. (2013). Femoral neck-shaft angle in humans: variation relating to climate, clothing, lifestyle, sex, age and side. Journal of Anatomy, 223(2), 133–151. https://doi.org/10.1111/joa.12073 Gombár, C., Janositz, G., Friebert, G., & Sisák, K. (2019). The De Puy ProximaTM short stem for total hip arthroplasty – Excellent outcome at a minimum of 7 years. Journal of Orthopaedic Surgery, 27(2), 2309499019838668. https://doi.org/10.1177/2309499019838668 http://www.aylward.org/notes/open-access-medical-image-repositories

218

A. Díaz Lantada et al.

Iguchi, H., Hua, J., & Walker, P. S. (1996). Accuracy of using radiographs for custom hip stem design. Journal of Arthroplasty, 11(3), 312–321. https://doi.org/10.1016/S0883-5403(96) 80084-2 Jansen, P. (2014). Open Source CT scanner. Make, 38, 112. Morales de Cano, J., Vergara, P., Valero, J., & Clos, R. (2018). Utilización de los vástagos metafisarios «Próxima» DePuy: nuestra experiencia a más de cinco años. Acta Ortopédica Mex., 32, 88–92. Munoz-Guijosa, J. M., Zapata Martínez, R., Martínez Cendrero, A., & Díaz Lantada, A. (2020). Rapid prototyping of personalized articular orthoses by lamination of composite fibers upon 3D-printed molds. Materials, 13(4), 939. Murgen: open source ultrasound imaging: https://hackaday.io/project/9281-murgen-open-sourceultrasound-imaging Niezen, G., Eslambolchilar, P., & Thimbleby, H. (2016). Open-source hardware for medical devices. BMJ Innovations, 2, 78–83. Noble, P. C., Alexander, J. W., Lindahl, L. J., Yew, D. T., Granberry, W. M., & Tullos, H. S. (1988). The anatomic basis of femoral component design. Clinical Orthopaedics., 235, 148–165. Open source magnetic resonance imaging: https://www.opensourceimaging.org/ Rinaldi, G., Capitani, D., Maspero, F., & Scita, V. (2018). Mid-term results with a neck-preserving femoral stem for total hip arthroplasty. HIP International, 28(2), 28–34. https://doi.org/10.1177/ 1120700018813216 Rubin, P., Leyvraz, P., Aubaniac, J., Argenson, J., Esteve, P., & de Roguin, B. (1992). The morphology of the proximal femur. A three-dimensional radiographic analysis. Journal of Bone and Joint Surgery, 74-B(1), 28–32. https://doi.org/10.1302/0301-620X.74B1.1732260 Sabatini, A. L., & Goswami, T. (2008). Hip implants VII: Finite element analysis and optimization of cross-sections. Materials & Design, 29(7), 1438–1446. https://doi.org/10.1016/j.matdes. 2007.09.002 Santori, N., Lucidi, M., & Santori, F. S. (2006). Proximal load transfer with a stemless uncemented femoral implant. Journal of. Orthopaedics Traumatology, 7(3), 154–160. https://doi.org/10. 1007/s10195-006-0141-x Santori, N., Albanese, C. V., Learmonth, I. D., & Santori, F. S. (2018). Bone preservation with a conservative metaphyseal loading implant. HIP International. https://doi.org/10.1177/ 112070000601603S04 Solórzano, W., Ojeda, C., & Diaz Lantada, A. (2020). Biomechanical study of proximal femur for designing stems for total hip replacement. Applied Science, 10(12). https://doi.org/10.3390/ app10124208 Toth, K., & Sohar, G. (2013). Short-stem hip arthroplasty. In P. Kinov (Ed.), Arthroplasty - Update. InTech. Wilkinson, M. D. (2016). Comment: The FAIR Guiding Principles for scientific data management and stewardship. Scientific Data, 3(160018), 1–9.

Chapter 10

Open-Source Medical Devices as Tools for Teaching Design, Standards and Regulations of Medical Technologies Licia Di Pietro, Gabriele Maria Fortunato, Ermes Botte, Arti Ahluwalia, and Carmelo De Maria

10.1

Introduction

The design of medical devices (MDs) requires a multidisciplinary approach, where scientific and technical knowledge in mechanics, electronics and information technology has to be combined with competences in biology, anatomy and physiopathology in order to address the clinical needs of patients and healthcare providers. MD projects involve teams of professionals, composed among others by biomedical engineers, medical doctors, management and business experts, so that attitude to teamwork and communication skills are as important as problem-solving abilities. In such context, collaborative project/problem-based teaching-learning methods have been suggested as effective strategies for bridging technical competences with the development of transversal skills (Dochy et al., 2003; Mahendru & Mahindru, 2011; Brennan et al., 2013) and may be invaluable in the education of BME students, the point of contact between medicine and engineering (Ahluwalia et al., 2018). The project-based learning (PBL) approach has been declined in various formats (Davidson & Major, 2014; Brodie, 2009) and in different graduate and undergraduate courses of several disciplines (Stratton, 2010; Caspary & Wickstrom, 2017), and generally, these student-centred teaching-learning experiences have helped to promote participant motivation of both students and teachers, which is a basic element for successful formative outcomes (Marra et al., 2014; Rillero et al., 2017). Collaborative PBL is generally well suited for complex real-world problems (Ortega-Sánchez et al., 2018; Downey et al., 2006; Fink, 2002), where design specification often not listed as in a verification exam but are identified during the

L. Di Pietro · G. M. Fortunato · E. Botte · A. Ahluwalia (*) · C. De Maria Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy e-mail: [email protected] © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_10

219

220

L. Di Pietro et al.

analysis of the assigned challenge. In the BME field, constraints and specifications are defined by international standards on medical technologies, mainly published by the International Standards Organization (ISO) and by the International Electrotechnical Commission (IEC), which have the paramount role to ensure safety and efficacy of the devices (De Maria et al., 2018). Despite their importance, it is difficult to engage engineering students’ attention when teaching norms and legislations, especially if compared with innovative technologies as 3D printing or artificial intelligence. At the same time, it is extremely difficult to find quality teaching materials, which will present the development of a medical technology from cradle-to-crave, i.e., from the analysis of clinical needs to the actual fabrication of a device, its use and final disposal. From this point of view, open-source medical devices (OSMDs), according to the definition provided in Chap. 1, represents an enabling tool, providing access to blueprints which have been designed according to relevant standards for being compliant with regulations. In this context, a trusted repository of such open-source projects, such as the UBORA e-platform (https://platform.ubora-biomedical.org), could facilitate the vetting process that educators and students usually perform when prepare and study classes on medical device design (King & Fries, 2003). Here we present the step-by-step development of two OSMDs, a walking frame and a smartified manual breathing unit, collaborative developed by group of students of the Master Program in BME at the University of Pisa, during the academic years 2017/2018 and 2018/2019, respectively. Students went through the process of needs identification, risk class assessment, standard identification, design, prototyping and verification of compliance. Before going in the detail, a short overview on the organization of the course on medical device design at University of Pisa is provided, as a point of reference for the competences needed for carrying out these projects.

10.2

Background and Context: Course of Medical Device Design

The course of Laboratory of Biomedical Technologies has been designed to offer in-depth theoretical knowledge on product design and prototyping, coupled with the capstone project’s minimal criteria defined by ABET (King & Collins, 2006; Ahluwalia et al., 2015). A great emphasis is put on hands-on activities on the model of African Biomedical Engineering Consortium (ABEC) Design Schools (De Maria et al., 2015; Diez, 2012) and FAB Academy (Crawley et al., 2007), taking inspiration from Conceive-Design-Implement-Operate (CDIO) teaching/ learning methodology (Cheng, 2003). The programme can be divided into six main parts, namely, as following:

10

1. 2. 3. 4. 5. 6.

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

221

Introduction to medical devices Fundamentals of engineering drawing Fundamentals of prototyping and manufacturing technologies Electronic rapid prototyping Electromechanical prototyping Practical examples of design of MDs

The content of each part is described in detail in Table 10.1, which highlights also the selected free and open-source teaching resources and tools, which have increased sustainability and replicability of the course. The final examination consisted in the discussion of a group project, focused on a medical device, developed up to the prototyping phase trying to respect the constraints and the specification given by ISO and IEC standards and medical device regulation MDR 2017/745. Indeed, the final project was expected to be a pre-production device dossier, drafted according to standards on Quality Management of Medical Devices ISO 13485. At the beginning of the course, students were invited to form groups of maximum three members, on the basis of their personal preferences (e.g., friendship, common interest), and to identify an existing medical device they would like to redesign or improve, or to look for an unmet clinical need and work on an innovative device. Thus, inventiveness was promoted even if it was not mandatory. Groups started their project in parallel with theoretical lessons, identifying the risk class of the device and relevant standards. In agreement with the teachers, each member of the group selected a specific part or subassembly of the device to be designed in depth, up to prototyping planning. Teachers acted as tutors by monitoring the intermediary results and helping the students to identify alternative solutions (World Health Organization, 2015). Synchronous teachers-group communication was preferred, by scheduling specific appointment, respect to asynchronous email discussion. The first, experimental, class on Laboratory of Medical Devices was held in the first year of the Master’s Degree in Biomedical Engineering, in the second semester of the Academic Year 2016–2017 (spring 2017). A total 70 students (46 females and 24 males) assiduously attended the class and successfully took the final examination within 8 months after the end of the class (60 within 3 months). A total of 34 different projects, consisting in a sort of medical device dossier, were illustrated during an oral presentation, of about 1 hour per group. All the projects, ranging from surgical scalpel to insulin pump and hearing heads, were correctly classified, and at least one component of the device per team member was designed according to relevant standards and analysed in terms of prototyping and fabrication cycles. If the selected device did not include electronic components (e.g. a bi-leaflet heart valve or a crouch), students worked also on the design of a sensorized test bench or on the “smartification” of the devices by adding new sensors or actuators (e.g. designing a “smart” crouch for blind people). A total of 38 students provided a feedback responding to the anonymous questionnaire on voluntary basis at the end of the course before the final examination. Three students over four expressed a strong positive opinion on the course and its importance on their future career, as more than

222

L. Di Pietro et al.

Table 10.1 Detailed course programme Part (hours) Introduction to medical devices

Fundamentals of engineering drawing

Topics Pretotype, prototype and product in the biomedical field

Instruments

Classification of medical devices, international standards for designing medical devices and certification routes The role of engineering drawing; manual sketch and computer-aided design (CAD) Use of a CAD software: part design and assembling

EU Medical Device Regulation MDR 2017/ 745; ISO database; IEC database

Quotation and tolerances

Fundamentals of prototyping and manufacturing technologies

Electronic rapid prototyping

Introduction to machine elements (gear, bearing, screws, etc.) Subtractive technologies: conventional technologies (milling, turning, etc.) and non-conventional technologies (laser cutting, water jet cutting, etc.) Formative technologies (moulding) Additive technologies

Boards for electronic rapid prototyping

Link/Ref. http://www. pretotyping.org/ uploads/1/4/0/9/ 14099067/pretotype_ it_2nd_pretotype_edi tion-2.pdf (Council Directive, 1993; MDR Regulation, 2017; Lieu & Sorby, 2015)

Commercial (Fusion 360, free version for students) and opensource (FreeCAD) software Measurement tools (caliper, micrometre screw gauge) Mechanical components

https://www. freecadweb.org/

Modela Player 4 CAM software and SRM CNC Desktop Milling Machine (Roland DG); CAM modules of Fusion 360

https://www. rolanddga.com/prod ucts/3d/srm-20-smallmilling-machine/ specifications

Slicing software Slic3r and fused deposition modelling 3D printer Prusa I3 rework Arduino electronic board (open-source); Tinkercad simulation environment (proprietary); Fritzing circuit CAD (open-source)

(chu, 2010)

(Bowyer, 2014)

(Bowyer, 2014)

https://www.arduino. cc/ https://www. tinkercad.com/ https://fritzing.org/

(continued)

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

223

Table 10.1 (continued) Part (hours)

Electromechanical prototyping

Practical examples of design of medical devices

Topics Signal acquisition from sensors Communication between boards, towards users and computers Electrical motors and motor drivers Control of electromechanical actuators Dimensioning of motors and batteries UV sterilization device

Instruments Arduino kit, multimeter

Link/Ref. https://www.arduino. cc/

Arduino kit Arduino kit Arduino kit

Electrical Thermometer

http://www. centropiaggio.unipi.it/ course/laboratorio-ditecnologiebiomediche http://www. centropiaggio.unipi.it/ course/laboratorio-ditecnologiebiomediche

75% would follow the course even if it was not mandatory. Overall, the collaborative project as examination modality was considered as extremely positive, and most of the comments point out the importance of the opportunity to work in team on practical problems, which in the context of a collaborative PBL can be considered as a learning outcome (Prepregs). Similar outcomes have been registered in the successive years.

10.3

The Walking Frame Project

This section illustrates how the prototype of an innovative walking frame was designed and fabricated according to MDR 2017/745 by a group of students of the course of Laboratory of Biomedical Technologies in 2018. The project started from the identification of clinical needs, passed through the risk assessment, the finite element modelling (FEM) simulations, the consequent CAD modelling and material choices, and ended with the fabrication of each component and their final assembly. The fabrication of carbon fibre composite tubes (the structural elements of the walking frame, see further details later in the text) was carried out in collaboration with the ITS Maker, in Fornovo di Taro, an Emilia-Romagna’s higher institute for the capacity building competences in handling and processing of carbon fibre composite materials, and thus in the fabrication of carbon fibre structures.

224

L. Di Pietro et al.

A premise is necessary: although carbon fibre components are not cheap, the fabrication of prototype components was part of ITS educational courses in additive manufacturing and composite materials. Students usually practise with simple shapes, such as tubes, without a specific application, making often these parts useless and unrecyclable. This situation was included in the project as a design constraint (i.e. the use of high-performing structural components with a simple shape). The project is fully available on the UBORA platform at the link: https://platform. ubora-biomedical.org/projects/54902d2c-5fde-4434-be65-e76dac5d816b.

10.3.1 Needs Identification The project started from the consideration that world population is continuously ageing, due to the consistently low birth rate and higher life expectancy, and thus innovative solutions to provide assistance and improving quality of life for elderly and disabled people are needed. For this reason, walking aids such as walkers could give back to elders the ability to move, an essential aspect for an independent life (Petersen, 2016). The proposed walking frame is a totally static aid (i.e. no wheels), which requires the user’s residual ability to pick it up step by step. However, employing low-weight composite materials reduces the physical efforts requested to the users, thus a greater number of people with different levels of disability could be able to get benefits.

10.3.2 Identification Risk Class and Relevant Standards According to the 2017/745 Medical Device Regulation, a walking frame is a Class I medical device because it is not invasive, does not get in contact with blood or other human fluids and does not exchange any kind of energy with the body. Consequently, it is associated to a low-risk level, and its compliance to international regulation can be self-certificated. If designed and produced taking into account the standards listed in Table 10.2, its conformity to norms is automatically satisfied. A risk analysis was carried out according to ISO 14971:2012, in order to identify possible critical situations/issues that may arise during the use of the proposed device. A list of possible hazards with related hazard situations and harms are shown in Table 10.3. For each harm, a risk has been reported. The list of the risks with their related probability of occurrence of harm and severity of harm is presented in Table 10.4. In the end, the final qualitative risk analysis is shown on Table 10.5. According to the table, risk number 3 has been considered by students as an unacceptable risk. For this reason, a correct maintenance is suggested to the final user as an appropriate risk control measure. In particular, after the implementation of this measure, all risks have been considered acceptable.

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

225

Table 10.2 International standards applicable to the walking frame Standard ISO 10993-1: 2018

EN ISO 13485: 2016

EN ISO 14971: 2012

MEDDEV 2.7.1 rev 4 Clinical Evaluation IEC 62366-1: 2015

EN ISO 15223-1: 2016

EN 12182:2012

ISO 9999:2016

EN ISO 11199-1: 2005

Description Biological evaluation of medical devices – Part 1: Evaluation and testing within a risk management process It has 18 chapters: Chap. 1 will guide the choice of the applicable additional chapters Medical devices: Quality management systems – Requirements for regulatory purposes This standard specifies requirements for all entities involved in medical devices, in all stages of the product life cycle: from design to manufacture to installation and disposal Medical devices – Application of risk management to medical devices This standard specifies requirements for designers and manufacturers of medical devices, in order to minimize the risk of the device itself. There is no “risk zero” device, but many activities can be implemented to reduce and manage risk. This standard provides useful checklists and also guidance on the most widespread risk management techniques such as FMEA A Guide for Manufacturers and Notified Bodies under Directives 93/42/EEC and 90/385/EEC This guideline provides information on methods used to assess the clinical performance and the clinical benefit of a medical device. It is provided for free by the European Commission Medical devices – Part 1: Application of usability engineering to medical devices This standard provides guidance on how to manage the human factors while designing a medical device (usability engineering) Medical devices — Symbols to be used with medical device labels, labelling and information to be supplied — Part 1: General requirements This standard lists a series of symbols that may be applicable in labels of medical devices Assistive products for persons with disability – General requirements and test methods This European Standard specifies general requirements and test methods for assistive products for persons with a disability Assistive products for person with disability – Classification and terminology It establishes classification and terminology of assistive products, especially produced or generally available, for persons with disability. Assistive products used by a person with disability, but which require the assistance of another person for their operation, are included in the classification. Walking aids manipulated by both arms – Requirements and test methods – Walking frames The first part specifically concerns walking frames (i.e. no wheels or other accessories). Technical names of the device components and the range of acceptable values for characteristic dimensions and user’s weight are defined. Walker requirements, concerning static load mechanical strength for the whole device and for legs only, fatigue strength and stability (forward, rear and lateral), in standard exercise conditions declared by the producer are specified. Moreover, it presents a detailed description of test methods and their relative settings in terms of load geometry, test pipeline and results registration for a correct integration in the self-certification of compliance (continued)

226

L. Di Pietro et al.

Table 10.2 (continued) Standard EN ISO 24415: 2009

EN 1041

Description Tips for assistive products for walking The first part is specifically about tips friction properties for assistive products for walking. Technical requirements for tips in standard exercise conditions declared by the producer are here defined. A detailed description of the simulation settings needed to measure friction force between tip and support surface is presented together with the requested characteristics for the measuring system and the post-processing to be carried out to get results for the self-certification of compliance Information supplied by the manufacture of medical devices This European Standard specifies requirements for information to be supplied by a manufacturer for medical devices

Table 10.3 List of hazards, hazard situation, harm and related risks Hazard Biological Biocompatibility Function

Hazard situation Bacterial released during the use Allergenicity/irritation Deterioration of tips

Labelling Mechanical Mechanical

Incomplete instructions for use Presence of sharp edges Deterioration of the device’s performance

Harm Bacterial infection Allergic reactions Loss of support function for users Incorrect use of the device Damage for users Fall of patients

Risk number R1 R2 R3 R4 R5 R6

Table 10.4 List of aforementioned risks with related probability of occurrence of harm and severity of the harm Risk number R1 R2 R3 R4 R5 R6

Probability of occurrence of harm Low Medium High Low Low Low

Severity of the harm Moderate Negligible Moderate Negligible Moderate Moderate

10.3.3 Finite Element Modelling FEM simulations of structural mechanics were carried out to define the section of the tubes, the angles between incident tubes, the optimal position for handles, the orientation of fibres and the stratigraphy for the composite material. Simulations were performed using COMSOL Multiphysics – structural mechanics module with the 3D Euler beam application mode. Indeed, being the radial dimension of the tubes negligible with respect to the axial one and according to the Euler-Bernoulli assumption, each tube was simply represented by its axis, obtaining a 3D beam model symmetric with respect to the lateral plane (i.e. the YZ plane in Fig. 10.1). The

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

227

Table 10.5 Qualitative risk matrix (3
3) with three levels of probability of occurrence of harm and three levels of severity of harm Severity levels Negligible High Probability levels

Moderate

Significant

Key

R3

Medium

R2

Low

R4

Unacceptable risk Acceptable risk

R1, R5, R6

Fig. 10.1 Boundary settings for the FEM simulation

Table 10.6 Domain settings for the FEM simulation

Property Density Elastic modulus Poisson’s ratio

Value 1800 kg m3 100 GPa

References “Prepregs.” [Online]

0.3

Estimated from (Smith et al., 1974, Krucinska & Stypka, 1991)

“Prepregs.” [Online]

beams corresponding to the four legs of the walker were orthogonal to the floor, and handles were placed at 1/5 of the relative tube length (from the rear extremity) to ensure stability. Moreover, the slope of the upper oblique tube was the result of trigonometric considerations dealing with the constraint that the direction of the user’s weight force has to cross the floor plane within the frame base to avoid overturning. The simulation was performed in static load and conditions, according to ISO 11199:2005, and each tube was assumed to have a ring-shaped section with external diameter of 30 mm and 2 mm in thickness, and to be almost infinitely stiff. Carbon fibre composite was indeed approximated as an elastic, linear, homogeneous and isotropic material with an elastic modulus of the order of magnitude of the fibre axial modulus (Table 10.6). Loads and constrains were imposed as boundary settings (Fig. 10.1).

228

L. Di Pietro et al.

Fig. 10.2 Maximum equivalent normal stress. Arrows indicate the most stressed points

The load given by the half user’s weight was considered to be applied in correspondence of handles as a 3D force (configuration for a standard user: 30 angle with respect to the z axis and 15 angle with respect to the y axis). Tips were constrained on the XY plane because of static friction force, which was previously verified to be sufficient to avoid any movement along both axes. For this evaluation, the normal component of the weight force and a static friction coefficient of 0.65 – a typical value for rubber-like materials onto standard surfaces – were considered. The geometry was then discretized by generating the mesh, and the simulation was run. Results are shown in Fig. 10.2, which represents the maximum equivalent normal stress (σ eq) calculated from tangential and normal stresses using the Von Mises’ approach. Note that the reported values of σ eq were cautiously overestimated, since the maxima of both tangential and normal stress distributions were assumed to coincide in the same point in each section of the structure. Critical points (highlighted by arrows in Fig. 10.2) show a σ eq of about 315 MPa. Knowing the mechanical strength of the available types of carbon fibre composite [30] and setting a minimum desirable safety factor equal to 2, the structure was verified for every possible material stratigraphy. Those observations suggested that the chosen design was suitable for the prototype production: it allows flexibility in terms of material composition and, consequently, of fabrication processes. Finally, displacements resulted coherently negligible and, even if approximated due to the hypotheses

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

229

about the linear, elastic, isotropic and homogeneous mechanical properties, the stability of the frame and the physical meaning of the model (i.e. no displacements beneath the floor level) were verified.

10.3.4 CAD Modelling In order to produce the physical prototype, a 3D model of the device was drawn by the 3D CAD software SolidWorks. Each tube of the structure was modelled according to the above-defined dimensions, according to the FEM geometry. The only exception is represented by the tubes inserted into the tips, which were smaller (external diameter of 26 mm) to form a telescopic coupling with the rest of the frame and presented four vertically equidistant holes on their lateral surface, in order to adjust the walker height. CAD design is presented in Fig. 10.3. Joints (a) were designed using relevant angles according to the results of the FEM simulation. Also tips (b), characterized by a no-slip surface (c), were modelled: the no-slip feature was obtained by designing grooves, which increase the contact surface under pressure, in order to make them able to provide enough friction force. The height-regulation mechanism is based on a flexible ring (d) and a pin, which is inserted in the hole present on the leg-tubes of the walker. Handles were designed in an ergonomic way as in (e) and were modelled to be fabricated in two half pieces, connected by bolts, for simplifying their assembly on the relative tube. The CAD models were optimized to be fabricated using the fused deposition modelling (FDM) additive manufacturing technology. Relevant components and dimensions of the device are provided in Fig. 10.4 and Table 10.7, respectively.

Fig. 10.3 CAD model of (a) three entries-joint designed with ad hoc angles, according to the results of the FEM simulation; (b) tip with grooves; (c) non-slip terminal part for the tips, characterized by grooves to increase the contact surface under pressure; (d) height-regulation mechanism; and (e) half of a handle, with the UBORA logo

230

L. Di Pietro et al.

Fig. 10.4 CAD model of the walking frame assembly with labelled components

10.3.5 Prototyping Process As already stated, the skeleton of the walker is made of tubes linked together with specific joints obtained by additive manufacturing. Each tube was laminated by assembling three different kinds of pre-impregnated layers (called skins) of carbon fibre composite (i.e. sheets with a specific carbon fibre pattern into which a pre-catalysed epoxy resin has been impregnated) [30]. In particular (from the outer to the inner layer), the choice was 200 g plain 0 , 300 g unidirectional 0 and 414 g twill 45 , which is a combination able to optimize the mechanical strength of the component according to the predicted stress distribution. Laminating the carbon fibre requires a mould, which is exactly the negative of the piece of interest and, in the present work, it was also made of carbon fibre composite. After lamination, each component was cured at controlled temperature and pressure in an autoclave to ensure its stabilization. Joints (Fig. 10.5a), the core part of the tip (Fig. 10.5b) and the height-regulation mechanism (Fig. 10.5c) were 3D printed in ABSplus – guaranteeing suitable mechanical strength – using a Stratasys Fortus 3D printer, while the terminal parts of the tips (Fig. 10.5) and the handles (Fig. 10.5e) were 3D printed in thermoplastic polyurethane (TPU) – guaranteeing the desired friction properties – using the Creality cr10 3D printer. All the printed parts were linked together with the tubes using a photocurable resin-based glue. The final result of the fabrication steps is shown in Fig. 10.6.

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

231

Table 10.7 Relevant dimensions for walking frame components shown in Fig. 10.4 Component Tube 1

Quantity (#) 1

Dimensions (mm) – D (diameter), L (length), H (height) Dext ¼ 33, L ¼ 340

Tube 2

1

Dext ¼ 33, L ¼ 362

Tube 3

2

Dext ¼ 33, L ¼ 488

Tube 4

2

Dext ¼ 33, L ¼ 480

Tube 5

2

Dext ¼ 33, L ¼ 520

Tube 6

2

Dext ¼ 33, L ¼ 430

Tube 7

4

Dext ¼ 33, L ¼ 132

Tube 8

4

Dext ¼ 32, L ¼ 155

Joint A (3 entrances) Joint B (2 entrances) Joint C (4 entrances) Joint D (3 entrances) Height adjustingmechanism Tip Handle

2

Dint ¼ 33.5, Dext ¼ 38.5

Material Uniaxial composite (IMC) Uniaxial composite (IMC) Uniaxial composite (IMC) Uniaxial composite (IMC) Uniaxial composite (IMC) Uniaxial composite (IMC) Uniaxial composite (IMC) Uniaxial composite (IMC) ABS

2 2 2

Dint ¼ 33.5, Dext ¼ 38.5 Dint ¼ 33.5, Dext ¼ 38.5 Dint ¼ 33.5, Dext ¼ 38.5

ABS ABS ABS

4

Flexible ring with a pin

ABS

4 2

Dint ¼ 32.5, Dext ¼ 45, H ¼ 35 Dint ¼ 33, L ¼ 150

TPU TPU

IMC intermediate modulus carbon fibre composite, ABS acrylonitrile-butadiene-styrene, TPU thermoplastic polyurethane

Fig. 10.5 3D-printed components of the walking frame prototype: (a) tube joints; (b) the core part of tips; (c) no-slip terminal part of tips; (d) pin for the height-regulation mechanism; and (e) half of a handle. (a), (b) and (c) were printed in ABSplus using a Stratasys Fortus 3D printer; (d) and (e) were printed in TPU using a Creality cr10 3D printer

10.3.6 Verification of ISO Compliance and Safety Assessment The device satisfies the requirements of the standard EN ISO 11199:2005 – Walking aids manipulated by both arms as indicated in Table 10.8. Furthermore, the used

232

L. Di Pietro et al.

Fig. 10.6 The final prototype of the walking frame Table 10.8 Compliance with the EN ISO 11199:2015. Numerical values of the specification are not provided for avoiding any infringement of the copyright

Requirement Standard range for user’s weight Dimensional requirements

Mechanical requirements (for maximum standard weight)

Specification Range of weight Rear legs distance Base tips diameter Handles diameter No damage after a specific fatigue load No damage to any component after a specific static load Limited permanent displacement for each leg’s distal extremity After a specific static load Forward stability until at least a specific angle of inclination of the Support plane with a specific static load Rear stability until at least a specific angle of inclination of the Support plane with a specific static load Lateral stability until at least a specific angle of inclination of the Support plane with a specific static load

Compliance with EN ISO 11199: 2015 YES YES YES YES YES YES YES YES

YES

YES

material shall comply with ISO 10993 Biological evaluation of medical devices (Table 10.9).

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

233

Table 10.9 Observations on compliance with ISO 10993 TPU

ABSplus

Carbon fibre

The handles of the prototype were 3D printed with TPU, a biocompatible material, already used in several biomedical applications and compliant with the ISO 10993 The tips and the height-regulation mechanism were 3D printed with ABSplus that was not tested by the producer according to ISO 10993. Thus, an appropriate test for skin sensitization should be developed. However, the final product can be printed with ABS-M30i a biocompatible material, already used in several biomedical applications The skeleton of the walker was made up by carbon fibre, a biocompatible material, already used in several biomedical applications

https://solutions.covestro.com/en/high lights/articles/stories/2020/new-medi cal-grade-tpu-materials

https://www.stratasys.com/it/materials/ search/absplus

Petersen, 2016

10.3.7 The AMBU+ Project This section illustrates how the prototype of an innovative AMBU was designed according to MDR 2017/745 by a group of students of the course of Laboratory of Biomedical Technologies. The project started from the identification of clinical needs, passed through the risk assessment, the electronic prototyping and materials choices, and ended with the fabrication of each component and the final assembly. The final assembly was carried out at the Research Centre “Enrico Piaggio” at the University of Pisa. The project is available on the UBORA platform at the following link: https://platform.ubora-biomedicalorg/projects/54f880bb-993342e3-88c9-a7b1ecb09e5b.

10.3.8 Needs Identification When a patient cannot breathe properly, or after a cardiac arrest, a combination of basic and advanced airway and ventilation techniques is used to deliver supplemental oxygenation. Manual ventilation is a vital procedure and is regularly necessary in medical emergencies when the patient’s breathing is insufficient (respiratory failure) or has ceased completely (respiratory arrest). The bag valve mask (BVM) enables healthcare providers, which are regularly trained, to operate within almost any environment or situation to deliver life-saving oxygen to the patient’s lungs. BVM, sometimes known by the proprietary name AMBU, is a handheld device commonly used to provide positive pressure ventilation to patients who are not

234

L. Di Pietro et al.

Fig. 10.7 Common parts for bag-valve-mask devices, In this case a self-inflating style bag. The reservoir bag, when present, can be filled with pure oxygen from an external source, allowing to deliver to the patient nearly 100% oxygen

breathing or not breathing adequately. It is used to provide or assist pulmonary resuscitation in adults and children with a body weight >15 kg. The BVM device consists of a flexible bag that attaches to either a ventilation mask or endotracheal tube via some form of pressure control valve. The main parts of the device are presented in Fig. 10.7. Considering the emergency situations in which this device is used, the idea of this project was to provide a commercial AMBU of an electronic part with the aim of improving/simplifying the use of the AMBU, by emitting a signal to coordinate the inflation manoeuvres of a trained operator. This AMBU+ will help the healthcare professional to perform a more accurate ventilation procedure, by indicating the correct frequency with which the operator needs to ventilate the patient. The frequency is indicated using a visual and an audio signal. The operator can choose between two different modes of use: the first one for a normal ventilation procedure (eupnoeic) and the second one for cardiopulmonary resuscitation, which requires different frequency of ventilation.

10.3.9 Risk Assessment and Standards Identification According to the 2017/745 Medical Device Regulation, an AMBU is a Class IIa medical device, according to Rule 2, because it is not invasive, does not get in contact with blood or other human fluids and does not exchange any kind of energy with the body. It is intended for channelling oxygen into the body. Consequently, it is associated to a medium-low risk level, and its compliance to international regulation must be certified by a notified body. If designed and produced taking into account the standards listed in Table 10.10, its conformity to norms is automatically satisfied. Since the proposed device is connected to a IIa medical device, it is also classified as IIa.

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

235

Table 10.10 International standards applicable to the smartified bag valve mask, AMBU+ Standard ISO 10993-1:2018

EN ISO 13485: 2016

EN ISO 14971: 2012

MEDDEV 2.7.1 rev 4 Clinical Evaluation

IEC 62366-1:2015

EN ISO 15223-1: 2016

ISO 10651-4:2002

IEC 60601 EN 62304:2006 +A1:2015 EN 1789 ISO 5356-1

Description Biological evaluation of medical devices – Part 1: Evaluation and testing within a risk management process It has 18 chapters: Chap. 1 will guide the choice of the applicable additional chapters Medical devices – Quality management systems – Requirements for regulatory purposes This standard specifies requirements for all entities involved in medical devices, in all stages of the product life cycle: from design to manufacture to installation and disposal Medical devices – Application of risk management to medical devices This standard specifies requirements for designers and manufacturers of medical devices, in order to minimize the risk of the device itself. There is no “risk zero” device, but many activities can be implemented to reduce and manage risk. This standard provides useful checklists and also guidance on the most widespread risk management techniques such as FMEA A Guide for Manufacturers and Notified Bodies under Directives 93/42/ EEC and 90/385/EEC This guideline provides information on methods used to assess the clinical performance and the clinical benefit of a medical device. It is provided for free by the European Commission Medical devices – Part 1: Application of usability engineering to medical devices This standard provides guidance on how to manage the human factors while designing a medical device (usability engineering) Medical devices — Symbols to be used with medical device labels, labelling and information to be supplied — Part 1: General requirements This standard lists a series of symbols that may be applicable in labels of medical devices Lung ventilators – Particular requirements for operator-powered resuscitators (or equivalent national or regional standard compliance). It specifies requirements for operator-powered resuscitators intended for use with all age groups and which are portable and intended to provide lung ventilation to individuals whose breathing is inadequate. Operatorpowered resuscitators for infants and children are designated according to body mass range and approximate age equivalent Medical electrical equipment – Part 1: General requirements for basic safety and essential performance Medical device software – How to design and code software for medical devices and requirements for SW change control Medical vehicles and their equipment – Road ambulances Anaesthetic and respiratory equipment – Conical connectors – Part 1: Cones and sockets This part of ISO 5356 specifies dimensional and gauging requirements for cones and sockets intended for connecting anaesthetic and respiratory equipment, e.g. in breathing systems, anaesthetic-gas scavenging systems and vaporizers

236

L. Di Pietro et al.

Table 10.11 List of hazards, hazard situation, harm and related risks Hazard classification Electromagnetic voltage (Line voltage) Electromagnetic voltage Electromagnetic voltage Electromagnetic voltage Thermal energy Operating instructions

Risk number R1

Hazard situation Possible sparks that start a fire

Harm Burns for patient and user

Body fluids or water that wet the electronic circuit Damage/wear to wires and electronic components Battery wear

Interruption of device operation Interruption of device operation Interruption of device operation Burns for patient and user Incorrect patient ventilation

R2

Incorrect patient ventilation Plastic splinters/fragments that hit the patient’s face

R7

High temperature

Labelling

Incorrect choice of operating mode in the emergency situation, due to unclear instructions for use Incomplete instruction for use

Labelling

Packaging breakage during use

R3 R4 R5 R6

R8

Table 10.12 List of aforementioned risks with related probability of occurrence of harm and severity of the harm Risk number R1 R2 R3 R4 R5 R6 R7 R8

Probability of occurrence of harm Low Medium Low Medium Low Low High Low

Severity of the harm Significant Negligible Moderate Significant Moderate Moderate Moderate Negligible

A risk analysis was carried out according to ISO 14971:2012, in order to identify possible critical situations/issues that may arise during the use of the proposed device. A list of possible hazards with related hazard situations and harms is shown in Table 10.11. For each harm, a risk has been reported. The list of the risks with their related probability of occurrence of harm and severity of harm is presented in Table 10.12. In the end, the final qualitative risk analysis is shown on Table 10.13. According to the table, risk numbers 4 and 7 are considered unacceptable. For this reason, appropriate risk control measures have been taken into consideration. In particular, for risk number 4, the students proposed a correct maintenance to be suggested to the final user, and for risk number 7, a correct user training before the first use has been

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

237

Table 10.13 Qualitative risk matrix (3
3) with three levels of probability of occurrence of harm and three levels of severity of harms Severity levels Negligible High Probability levels

Moderate

Significant

R7

Medium

R2

Low

R8

Unacceptable risk R4

R3, R5, R6

Key

Acceptable risk

R1

Fig. 10.8 External case: (a) is the top part; (b) is bottom part; (c) represents the assembly of (a) and (b) secured with four M3x25 bolts. The top part presents three holes for the two buttons and the seven-segments display (1) and an opening for the buzzer (2). In (b), two holes on the lateral panels accommodate the battery charger port (3) and the on/off button (4), while two slots in the bottom part (5) allow the case to be fixed to the BVM through an elastic band

considered. After the implementation of theserisk control measures, all risks have been considered acceptable.

10.3.10

CAD Modelling

In order to produce the physical prototype, a 3D model of the device was obtained by the 3D CAD software Autodesk® Fusion 360, and an appropriate external case was designed to contain all the electronic components. The CAD model is presented in Fig. 10.8. The commercially AMBU used for the project is presented in Fig. 10.9. It has an average cost of 50$. The external case presents two holes where appropriate buttons are allocated to select the right inflation therapy: 1. Normal ventilation procedure (eupnoeic) – this procedure ensures the normal respiratory rate (15 breaths per minute (bpm). It means that it is necessary to deliver 1 breath every 4 s)

238

L. Di Pietro et al.

Fig. 10.9 Bag valve mask commercially available. Life/form® BVM

Fig. 10.10 AMBU+ electronic scheme on protoboard. The electronic prototype is composed of a battery, charger circuit, MT3608 DCDC, seven-segments display, a buzzer and two push buttons

2. Cardiopulmonary resuscitation (CPR) procedure – this procedure uses the bag mask ventilation with supplemental oxygen. During CPR, the bag mask is used to give two breaths after every 30 compressions of the chest

10.3.11

Electronic Prototyping

The electronic prototyping has been based on commercial components in order to design the final electronic system. The commercial components are listed in Table 10.13. The electronic system, based on Arduino Nano board, is shown in Fig. 10.10.

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

239

The MT3608 DCDC step up (boost) converter takes input voltages as low as 2 V and step up the output up to 28 V. The MT3608 features automatic shifting to pulse frequency modulation mode at light loads. It includes under-voltage lockout, current limiting and thermal overload protection. This module has a multi-turn potentiometer to adjust the output voltage. In this device, the MT3608 DCDC step up converter takes input voltage of 3.7 V and step up the output to 5 V. A buzzer and a sevensegments display provide an auditory and visual feedback, respectively. The sevensegments display is used to show the operator how many seconds are left until the next insufflation. The moment when the insufflation has to happen, a “P” is displayed. During the design process, a lithium polymer (LiPo) battery was chosen to power the Arduino, since it is considered safer than lithium-ion (Li-ion) battery. Moreover, it presents the advantage to be rechargeable. The charger battery unit has been appropriately designed in order to prevent the battery failure. In order to ensure the safety of battery use, an appropriate overdischarge protection has been considered to prevent the discharging battery below 2.4 V. At the same time, the proposed system presents also an overcharge protection to ensure a voltage under 4.2 V. A firmware, designed with Arduino IDE, was uploaded on the Arduino nano board. The code is available in the Repository of the UBORA platform at the link https://platform.ubora-biomedical.org/projects/54f880bb-9933-42e3-88c9a7b1ecb09e5b/Repository. The code flowchart is shown in Fig. 10.11.

10.3.11.1

Prototyping Process

All the designed parts were 3D printed in ABSplus, which guarantees mechanical strength, using a Stratasys Fortus 250mc 3D printer. The electronic components (reported in Table 10.14) were all soldered on a matrix board secured to the case with four M2
12 bolts, except for the battery, located in an ad hoc housing underneath the board. The case was then fixed to the commercial AMBU with an elastic band. Final result of the fabrication phases is shown in Fig. 10.12.

10.4

Conclusions

Here we described two paradigmatic open-source projects of medical devices, to be used as teaching tool in collaborative PBL experience for teaching/learning standards and regulation in the field of MDs. A walking frame and a smartified AMBU were selected, as they allow to focus on mechanical performances and on electronic prototyping, respectively. Interestingly, despite both of them representing well-known devices, students were able to include innovative features, using advanced lightweight materials and electronic components.

240

L. Di Pietro et al.

Fig. 10.11 Code flowchart of AMBU+ firmware Table 10.14 AMBU+ commercial electronic components Component Arduino Nano microcontroller Seven-segments display Buttons Buzzer Resistors Switch button Lithium-ion battery LiPoLy charger single cell MT3608 DCDC step-up converter Wires M2x12 bolts

Quantity 1 1 2 1 3 1 1 1 1 – 4

Note https://store.arduino.cc/arduino-nano

1
220 Ω; 2
10 kΩ

https://www.sparkfun.com/products/12711 Different size For assembling the electronic board

10

Open-Source Medical Devices as Tools for Teaching Design, Standards and. . .

241

Fig. 10.12 Bag valve mask commercially available plus the external case purposely designed

Finally, both projects include relevant standards and a risk analysis as starting points for their design.

References Ahluwalia, A., et al. (2015). Open Biomedical Engineering education in Africa. Proceedings of the Annual International Conference on IEEE Engineering Medical Biological Society EMBS, 2015, 3687–3690. Ahluwalia, A., et al. (2018). Biomedical engineering project based learning: Euro-African design school focused on medical devices. International Journal of Engineering Education, 34(5), 1709–1722. Bowyer, A. (2014). 3D printing and humanity’s first imperfect replicator. 3D Printing and Additive Manufacturing, 1(1), 4–5. Brennan, R. W., Hugo, R. J., & Gu, P. (2013). Reinforcing skills and building student confidence through a multicultural project-based learning experience. Australasian Journal of Engineering Education, 19(1), 75–85. Brodie, L. M. (2009). eProblem-based learning: Problem-based learning using virtual teams. European Journal of Engineering Education, 34(6), 497–509. Caspary, M., & Wickstrom, C. (2017). Multicultural problem-based learning approaches facilitate ESP language acquisition. IJLTER. ORG, 16(3), 1–14. Cheng, M. (2003). Medical device regulations: Global overview and guiding principles. World Health Organisation. chu Yeh, Y. (2010). Integrating collaborative PBL with blended learning to explore preservice teachers’ development of online learning communities. Teaching and Teacher Education, 26(8), 1630–1640. Council Directive. (1993). 93/42/EEC of 14 June 1993 concerning medical devices. Crawley, E. et al. (2007). Rethinking engineering education. The CDIO Approach, 60–62. Davidson, N., & Major, C. H. (2014). Boundary crossings: Cooperative learning, collaborative learning, and problem-based learning. Journal on Excellence in College Teaching, 25(3&4), 7–55. De Maria, C., Mazzei, D., & Ahluwalia, A. (2015). Improving African healthcare through open source biomedical engineering. International Journal of Advance Life Sciences, 7(1–2), 10–19.

242

L. Di Pietro et al.

De Maria, C., et al. (2018). Safe innovation: On medical device legislation in Europe and Africa. Health Policy Technology, 7(2), 156–165. Diez, T. (2012). Personal fabrication: Fab labs as platforms for citizen-based innovation, from microcontrollers to cities. Nexus Network Journal, 14(3), 457–468. Dochy, F., Segers, M., Van den Bossche, P., & Gijbels, D. (2003). Effects of problem-based learning: A meta-analysis. Learning and Instruction, 13(5), 533–568. Downey, G. L., et al. (2006). The globally competent engineer: Working effectively with people who define problems differently. Journal of Engineering Education, 95(2), 107–122. Fink, F. K. (2002). Problem-based learning in engineering education: A catalyst for regional industrial development. World Transactions on Engineering Technological Education, 1(1), 29–32. King, P. H., & Collins, J. C. (2006). Ethical and professional training of biomedical engineers. International Journal of Engineering Education, 22(6), 1173–1181. King, P. H., & Fries, R. (2003). Designing biomedical engineering design courses. International Journal of Engineering Education, 19(2), 346–353. Krucinska, I. & Stypka, T. (1991). Direct measurement of the axial poisson’s ratio of single carbon fibres. Composites Science and Technology, 41(1), 1–12. Lieu, D. K., & Sorby, S. A. (2015). Visualization, modeling, and graphics for engineering design. Nelson Education. Mahendru, P., & Mahindru, D. V. (2011). Problem-based learning: Influence on students’ learning in an Electronic & Communication Engineering course. Global Journal of Reserach in Engineering Electrical and Electronics Engineering, 11(8), 1–9. Marra, R., Jonassen, D., Palmer, B., & Luft, S. (2014). Why problem-based learning works: Theoretical foundations. Journal on Excellence in College Teaching, 25, 221–238. MDR Regulation. (2017). Regulation (EU) 2017/745 of the European Parliament and of the Council of 5 April 2017 on medical devices, amending Directive 2001/83/EC, Bibliography 205 Regulation (EC) No 178/2002 and Regulation (EC) No 1223/2009 and repealing Council Directives 90/385. Ortega-Sánchez, M., et al. (2018). Confronting learning challenges in the field of maritime and coastal engineering: Towards an educational methodology for sustainable development. Journal of Cleaner Production, 171, 733–742. Petersen, R. (2016). Carbon fiber biocompatibility for implants. Fibers, 4(1), 1–13. “Prepregs.” [Online]. Available: https://www.gurit.com/Our-Business/Composite-Materials/Pre pregs. Accessed 27 Jan 2021. Rillero, P., Koerner, M., Jimenez-Silva, M., Merritt, J., & Farr, W. J. (2017). Developing teacher competencies for problem-based learning pedagogy and for supporting learning in languageminority students. Interdisciplinary Journal of Problem Based Learning, 11(2). Stratton, B. J. (2010). The practice of problem-based learning: A guide to implementing PBL in the college classroom. Teaching Theology and Religion, 13(1), 79–80. Smith, A. Wilkinson, S. J. & Reynolds, W. N. (1974). The elastic constants of some epoxy resins. Journal of Material Science, 9(4), 547–550. World Health Organization. (2015). World report on ageing and health.

Chapter 11

On the Sustainable Growth of the Biomedical Industry Reinvented Through Innovative Open-Source Medical Devices Andrés Díaz Lantada, Rocío Rodríguez-Rivero, Ana Moreno Romero, Rafael Borge García, Luis Ignacio Ballesteros Sánchez, Licia Di Pietro, Carmelo De Maria, and Arti Ahluwalia

11.1

Open-Source Medical Devices and Healthcare Technology Equity

Open-source medical devices (OSMDs) and collaborative developments in biomedical engineering (BME) are challenging the status quo of the biomedical industry. The more traditional approaches to the development of medical technology may soon be outperformed by these emerging trends, which pursue healthcare technology equity. This equity can be achieved by additionally involving end users (i.e. patients, healthcare professionals, associations, citizens), as co-creators throughout the innovation chain, and by sharing the final blueprints of open-source devices, ideally with a “free as in freedom” scheme. As discussed in this chapter, OSMDs may convey several social, economic and environmental benefits, when compared with more traditional medical technologies, and their expansion, if adequately supported and mentored, is aligned with several Sustainable Development Goals (SDGs) of the 2030 Agenda (see Sect. 11.2). Accordingly, the collaboration of international communities through online e-infrastructures and the promotion of basic common principles for the co-creation of safe and competitive OSMDs are necessary. Besides, an understanding of the objective benefits of shifting to open-source and collaborative approaches, when

A. Díaz Lantada (*) · R. Rodríguez-Rivero · A. Moreno Romero · R. Borge García · L. I. Ballesteros Sánchez Mechanical Engineering Department, Universidad Politécnica de Madrid, Madrid, Spain e-mail: [email protected] L. Di Pietro · C. De Maria · A. Ahluwalia Research Center “E. Piaggio” and Department of Information Engineering, University of Pisa, Pisa, Italy © Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0_11

243

244

A. Díaz Lantada et al.

engineering innovative medical technologies for global health concerns, is required. The illustration of such benefits, by means of concrete examples, is also needed for promoting this innovative field. Furthermore, adequate policies are also important for a sustainable growth of the open-source biomedical industry, whose connections with the SDGs put forward the urgency for this change of trend.

11.2

Social Impacts: Open-Source Medical Devices and the SDGs

11.2.1 Open-Source Medical Devices and the SDGs: Opportunities, Challenges and Threats On the 25th of September 2015, worldwide leaders formulated and adopted in General Assembly of the United Nations a set of global objectives to eradicate poverty, protect the planet and assure prosperity for all, as part of a new agenda for sustainable development. These global objectives were named “Sustainable Development Goals” and evolved from the Millennium Development Objectives, although reformulated with the collaboration of 193 countries and taking civil society into account more than ever before. These 17 SDGs, using abbreviated descriptors, include no poverty, zero hunger, good health and well-being, quality education, gender equality, clean water and sanitation, affordable and clean energy, decent work and economic growth, industry, innovation and infrastructure, reduced inequality, sustainable cities and communities, responsible consumption and production, climate action, life below water, life on land, peace and justice strong institutions and global partnerships for achieving the goals. Each goal was defined with specific objectives and concrete indicators for measuring progress along the 2015–2030 period (United Nations, 2015; https:// sustainabledevelopment.un.org/). Our future as global society relies on the achievement of such goals or, at least, on honestly working towards their fulfilment by developing change strategies to the extent of our capabilities. Engineers must play a fundamental role in the ideation, developing and mentoring of such strategies. OSMDs pursue reduced inequalities and promote good health and well-being, while clearly linked to innovation and economic growth. At the same time, they constitute an emergent set of technologies, articulated through original good practices, and are deeply interwoven with several SDGs and with their specific objectives, as well as with a handful of the concrete indicators for each related SDGs. Table 11.1 presents a summary of connections between the OSMD paradigm and the SDGs focusing especially on potential benefits and opportunities of resorting to OSMDs, towards achieving the different SDGs, and on threats and challenges regarding OSMDs and the SDGs. The analysis serves as basis for proposing actions for solving current challenges and for mitigating risks in the OSMD field, which is

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

245

Table 11.1 Open-source medical devices and the Sustainable Development Goals *(Davidson et al., 2011) SDGs

Opportunities of employing OSMDs towards achieving the different SDGs The OSMD field emerges as a model for transforming the biomedical industry towards healthcare technology equity, aiming at bringing medical technology to patients regardless of their income level, living place or type of pathology Key initiatives in the OSMD field are aligned with the concept of “innovation through education”, and several case studies of OSMDs are shared both for medical purposes and for supporting BME education worldwide Women and girls have specific health needs, and health gender inequality still persists*. OSMDs can contribute to reducing inequalities through thematic challenges, co-creation hackathons and focused international working groups Thematic challenges, co-creation hackathons and focused international working groups, aimed at developing and sharing devices for water filtering, sterilization and sanitation, can improve water quality and sanitation in poorer regions The life cycle of OSMDs is well suited for involving local populations in the design, production and maintenance and may support the creation of economic growth in developing regions where OSMDs may be needed most Supporting medical device manufacturers to shift to OSMDs can lead to a more socially responsible biomedical industry and accelerate innovation, through open innovation, for solving global health concerns (see Sect. 11.3) As already explained in SDGs 5 and 8, OSMDs may support economic growth, decent work and gender equality, hence contributing to reducing inequalities, especially in low- and middle- income settings. OSMDs can support safer migrations

Threats and challenges regarding OSMDs and the SDGs OSMDs developed and delivered without following international standards and fulfilling regulations may be harmful for patients

Without the adequate capacity building in terms of higher education worldwide, the educational impact of OSMDs will be much reduced

OSMDs developed and delivered without following international standards and fulfilling regulations may be especially harmful for women

Safety is also an issue in devices aimed at enhancing water quality, and internationally admitted testing procedures for validating effectiveness should be applied

There is a threat that medical companies shifting to OSMDs and open innovation may fire their R&D teams and use “OSMD-makers” as low-cost engineers Open-innovation is not necessarily linked to innovation without protection. If the “free as in freedom” concept is not applied to OSMDs, their impact may be low Quality and safety issues in OSMDs, if not adequately taken into account, may minimize the potential of these technologies as regards inequality reduction

(continued)

246

A. Díaz Lantada et al.

Table 11.1 (continued) SDGs

Opportunities of employing OSMDs towards achieving the different SDGs OSMDs can support resource-efficient point-of-care production of medical devices (even with personalized approaches), following the “doing more and better with less” concept. The life cycle of OSMDs is well suited to support this goal The life cycle of OSMDs may turn out to be more efficient due to improved use of resources and reduced production, supply and usability-related environmental impacts, if the adequate methods are followed (see Sect. 11.4) Engineering OSMDs relies on collaboration, and their international impact can only be achieved through adequate partnerships. UBORA and ABEC communities are pioneering examples showing that healthcare can be transformed

Threats and challenges regarding OSMDs and the SDGs Disposable OSMDs may constitute a new source of waste in the communities, in which they are applied, if the end of life is not adequately planned

Again, disposable OSMDs may constitute a new source of waste in the communities, in which they are employed, if the end of life is not correctly considered Examples in various economic sectors (tourism, transport) show that “sharing economy” models may lead to monopolies and negatively affect SDGs

outlined in the following subsection and further expanded in remaining sections focusing on economic, environmental and social impacts of OSMDs.

11.2.2 Proposed Actions for Solving Challenges and Mitigating Risks A recent call for action, the Kahawa Declaration (Ahluwalia et al., 2018), proposed a list of five key activities, for the systematic promotion of the OSMD field and for making universal health coverage possible in the next decades. These activities, importantly associated to the aforementioned opportunities, threats and challenges of OSMDs, include: 1. The promotion of collaborative biomedical engineering design methodologies for approaching and solving global health concerns. 2. The development of open-access e-infrastructures for enabling global action through the collaboration of international communities. 3. The promotion of biomedical engineering education for all, as a way for capacity building and for the application of common approaches for safe OSMDs. 4. The progressive harmonization of medical device directives and the development of a new generation of accessible standards.

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

247

5. The fostering of international partnerships, involving all relevant stakeholders, for achieving universal healthcare. Indeed, the promotion of innovative and systematic engineering design methodologies, oriented to adequately considering the special features of collaboratively developed and OSMDs along their whole development life cycle, can be a key for deploying the potential of open innovation, towards safer, more efficient and highquality solutions. The dissemination of these innovative methodologies relies on devoted engineering educators, capable of involving their students in the co-creation of medical technology through global communities, for training the next generation of biomedical engineers, which necessarily should focus on global health coverage and work towards the 2030 Agenda. These engineers of the future should be capable of analysing and minimizing the environmental impacts of their developments, while maximizing social benefit, and should be able to justify the overall benefits of shifting to OSMDs and to collaborative BME approaches (see Sect. 11.5 for a concrete example). Furthermore, the impact of OSMDs depends on a straightforward development life cycle capable of more directly connecting the ideation process with the prototyping, validation and commercialization stages, for which adequately considering regulations and standards from the very beginning of the development is fundamental. In addition, new routes for the safer devices may be explored, and the elaboration of standards, regulations and policies should be boosted, as they should keep up pace with technological advances, so as not to constitute barriers to progress. Quality and safety management in these open-source and collaboratively developed devices should be reinvented as well.

11.3

Economic Sustainability of Open-Source Medical Devices

According to Donaldson, medical devices, including those developed for low- and middle-income settings, should be designed to be (a) world-class work on par with the most advanced available solutions, (b) user-centred and (c) economically viable to promote a sustainable expansion and increased social benefits (Donaldson, 2013). All this applies to OSMDs as well. Regarding world class, it is important to highlight that affordable should never mean low quality: the challenge of OSMDs is delivering high quality and value, while keeping the development process open and the final solution cost-effective. As for the user-centred designs, open-source and collaborative developments are especially well-suited to achieving user-centred solutions, thanks to their bottom-up development strategies and to involving final healthcare professionals and patients along the whole development cycle. Economic viability and sustainable growth options are analysed below, starting with different options for commercialization. This approach means a redesign of the supply chain that it is not easy. The players of

248

A. Díaz Lantada et al.

the traditional profit-oriented supply chain would have to move to a new framework with distributed value. One of the main drivers for the new value system is the recognition of the contribution of individuals to the innovation ecosystem (Mazzucato, 2018).

11.3.1 Business Models for Developers of Open-Source Medical Devices (a) Selling Open-Source Medical Devices and Related Components OSMDs can be commercialized following different pathways. A remarkable option is making the projects available for download (Pearce, 2015), in a similar way as mobile phone apps are downloaded in online stores, such as Apple Store, in some cases for free and in others on a cost basis. In fact, examples of OSMDs are linked to open-source medical apps and to open-source software. There may be cases, in which downloading a software, app or CAD design may not be enough: the final user may need a physical device, whose manufacturing services can be also provided through the online store or community. In many cases, the user may just need to download a computer-aided design (CAD) file for printing at home, which again can be free, as in open collections of CAD files, such as Thingiverse or GrabCAD, or may be downloadable after paying a fee or buying some tokens from the e-infrastructure or online store, through which the files are shared. Prototyping is often accomplished as proof of concept in “fablabs” but not yet in an organized fashion. The true potential of OSMDs will be accomplished when these communities of designers or “makers” and fablabs are also methodically interconnected for collaborating in joint missions by sharing physical infrastructures. The concept of global manufacturing networks, capable of delivering OSMDs as near as possible to the point-of-care and fulfilling harmonized regulations, quality criteria and standards, should be further constructed. In fact, efficiency may be promoted not by replicating the same fablabs in different locations but by using common and almost “plug and play” manufacturing resources near to the patients (plugging the local production chain) and by sharing other more sophisticated and expensive technologies on an open-access approach throughout the online communities. Inspiring open-access manufacturing infrastructures, such as the Karlsruhe Nano Micro Facility, accessible worldwide through six-monthly calls, show the path to a future, in which both design and manufacturing will be settled upon open-source and open-access foundations. To cite an example of commercial OSMDs, the Enabling the Future initiative, focused on open-source customized hand prostheses for children, is already selling kits of their more standard designs, as well as components for replacing or updating parts of worn out or damaged prostheses. Customization is also offered (i.e. special

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

249

colour) and design personalization services, in connection with the business models explained below. (b) Selling Services for Open-Source Medical Devices: Customization and Updates The OSMDs shared through co-creation online environments (i.e. UBORA, Enabling the Future, Open Prostheses, GitHub and Autofabricantes, among others) may catch the attention of patients, patient associations, medical professionals, etc. that can benefit from an already existing open-source development after some slight update or minor customization. These modifications can be offered by open-source communities, usually be promoting contact between the original designers and the final users. In this way, the original designers may increase the impacts of their developments, while obtaining some sponsorship for ongoing or future projects. (c) Developing Open-Source Medical Devices for Specific Needs Patients and their families and patient associations may well desire to sponsor the development of medical technologies for specific purposes, including devices for rare diseases and special health conditions, tools for self-diagnoses and monitoring in remote locations and technical aids for injured patients, among others. Healthcare professionals and hospitals may also like to fund the support of designers for developing special tools for surgical and medical practice, devices –including software – for helping to better follow the evolution of their patients, technologies for enhanced decision-making, etc. These innovations typically share a special degree of personalization or correspond to uncommon but still socially relevant needs, which are often discarded along cold profit-based decision processes. In addition, a special need for creativity promotion and for involving end users, both in the approach to such medical needs and during the technology development process, may be needed. With these boundary conditions, OSMDs may result especially adequate and both technically and economically viable. To start up these sponsored open-source developments for specific medical needs, there are some interesting options: contacting a community of makers dedicated to biomedical technology seems a good approach. Then the sponsoring may be articulated in form of contest open to the whole community for the generation of a wide set of product ideas, among which the concept to develop further on can be selected. A good alternative may be to consult collections of open-source medical technologies and select a specific designer, whose designs have proven successful, or arrange a group of developers with desired expertise. This is already well-established procedure in the software industry (developers are in many cases found through GitHub searchers) and may soon be common trend in hardware development. (d) Consulting Open-Source Solutions for Traditional Manufacturers International communities of OSMD developers working together through online platforms or infrastructures are a great source of talent. These communities may support and mentor traditional medical device developers, usually working on an IP-protected/patent-based way, to enter the realm of open-source software and hardware applied to health, by helping them understand the benefits of OSMDs.

250

A. Díaz Lantada et al.

Companies with interest about open-source approaches to technology development may well wish to incorporate a developer with experience in open-source methodologies to their staff. However, sometimes it may be interesting to count with the support of a larger community of makers, capable of providing the required consultancy. The sponsoring may be managed through non-profit associations or foundations, which are the more common legal entities working pro bono in the development of open-source technologies for different sectors. Success may well rely on adequately combining all the presented options and on profiting from some additional competitive advantages that OSMDs provide, discussed below.

11.3.2 Economic Advantages of Shifting to Open-Source Medical Devices (a) Open-Innovation Strategy: Expanding the Portfolio Several companies, even biomedical multinationals, already see open innovation as part of their strategy for developing new products and opening markets. Initiatives such as Medtronic’s Eureka or hiring the services of creativity boosters working on a contest-based approach to problem-solving (i.e. InnoCentive) are among common options. However, in almost all cases, these concepts proposed by external collaborators through the open-innovation paths end up belonging to the companies and developed in the usual closed way. Other well-established initiatives have proven very successful for bringing R&D results into practice through enterprise creation (i.e. Actúa UPM) with interesting examples of innovative biomedical companies created this way, but normally without attention to open-source options. In any case, there is an urgent need to further connect open innovation with opensource development processes, so that ideation, design, production and use can benefit from the different advantages of collaborative and open projects, products and processes. There are interesting examples, both in connection with open software and with open hardware, that demonstrate the remarkable benefits of open innovation linked to open-source approaches in several industrial sectors. The fact is that technology developers are not always very creative when it comes to finding additional applications and niches to the technologies developed. In many cases, being the development a direct consequence of a specific need, it is challenging to think out-of-the-box and find such new applications. Opening a technology and making it available through open-source schemes (like CC licenses) may turn out into new ideas, relevant joint ventures and more multidisciplinary impacts. The product portfolio can sustainably increase in this fashion. In the OSMD field, it is interesting to cite the Bitalino example, whose opensource technologies for managing biomedical signals find applications in extremely varied projects (both medical and beyond), which are also shared through their

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

251

community of users interacting online (Bitalino, 2020). The UBORA case is also interesting: in the first 3 years of operation, the UBORA platform (UBORA, 2020) has led to more than 300 innovative concepts of medical technologies, with around 20 of them reaching prototyping stage and technical validation. Such creative potential can support medical device manufacturers, healthcare professionals and patients in finding technological solutions to medical needs, and the only requisite is that the achieved solutions are open. Such solutions, if developed in collaboration with experienced and varied professionals and by involving patients, can result in more effective and safe (through increased review and monitoring intrinsic to the open paradigm) than closed devices. Besides, few companies can afford a R&D department capable of challenging an international community of “makers” working for developing the medical technologies of the future, which should be affordable and reach all those needing them. There is indeed much to win by shifting to open-source approaches, as further analysed. (b) R&D and Tax Savings Companies investing in R&D collaborative engineering design methods and related open-source developments can benefit from public- and private-funded calls supporting research and innovation activities (Pearce, 2015), which may cover personnel and other direct and indirect costs and convey fiscal benefits. Partnering through online e-infrastructures devoted to open-source and collaborative biomedical engineering may support the straightforward arrangement of consortia for such research and innovation projects and related proposals. Chances of success should be high due to the enhanced social impacts of these solutions, to their reduced environmental impacts (see following sections) and to the intrinsic novelty of OSMDs. Furthermore, being the OSMD field a so recent emerging trend, most resources allocated to OSMD development can be considered R&D investments, which may be subject to tax savings in many countries. (c) Savings in IP Protection and Litigation Patent-protected intellectual property is the classic approach taken by companies to defend from competitors. A patent provides a legal monopoly of two decades and constitutes also a negotiation tool and a source of income when licensed. However, the protection provided by patents is in fact not that high. Developers get their patented products and processes copied more often than not. Sometimes litigation with unfair competitors does not pay off, especially if the patent is illegally used in a remote country or by too powerful competitors. Too long and expensive juridical process linked to IP may suppose start-ups become bankrupt (Wohlsen, 2013). Considering that patents do not always provide the expected competitive advantage (due to being often infringed) and that maintaining a patent over the years is not so cheap, either with or without including litigation costs, companies shifting to open-source options could spare important quantities of money. In fact, the patent document should explain the invention in detail, so that any expert in the field could replicate it. This may be closer to the open-source concept than one could initially

252

A. Díaz Lantada et al.

think. It should be also noted that the best strategy for keeping the product portfolio competitive and protected is by means of a continuous innovation strategy that does not allow competitors to keep up pace. (d) Open-Source Social Responsibility A long-term compromise with open-source approaches and communities should be expected from biomedical companies benefiting from already developed OSMDs or from makers and platforms of the open-source BME field, as is happening in other open-source software and hardware sectors (Ornbo, 2013). According to Ornbo (Ornbo, 2013), open-source social responsibility should be considered on par with corporate social responsibility (CSR). During the last 10 years, companies have move from responsive CSR to strategic CSR, and many large companies are key players in the transformation of their sector towards sustainability agenda (electricity, food production, etc.). Biomedical companies are in the core of the digital-health transformation, and the role of private companies to push the sustainability agenda is an opportunity and a duty. Options for biomedical companies wishing to contribute to the open-source movement, in connection with the social corporate responsibility, include sponsoring specific projects and developers, supporting and participating in events focused on these approaches (i.e. International Conferences on Open-Source and Collaborative Biomedical Engineering*) and their solutions as open-source projects, progressively opening their portfolios and changing from a closed to a more open model. Moreover, activities in connection with open-source software and hardware and interactions with open-source communities should be reported, as CSR reporting helps to increase the commitment of organizations with ethically and socially relevant missions, supports the making of strategic decisions and may even constitute a competitive advantage. According to authors’ opinion, universities should also start analysing and reporting their open-source projects, as part of the R&D and technology transfer results, which currently focus mainly on published papers in impact journals, patents, created spin-off companies and overall incomes through projects. And perhaps the biomedical sector can find inspiration in the natural ecosystems to build bridges between solutions in nature with zero waste, and health technologies following the Blue Economy principles, also in connection with nature’s adaptive intelligence methods (Pauli, 2010). Summarizing all the above, OSMDs are not only a socially relevant option but may also provide several economic and competitive advantages to companies following or shifting to open-source and collaborative approaches, including an honourable mission to engage potential customers and investors. *http://ubora-biomedical.org/first-international-conference-on-open-source-medi cal-technologies/

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

11.4

253

Environmental Sustainability of Open-Source Medical Devices:

11.4.1 The Life Cycle of OSMDs Compared to That of Conventional Medical Devices (a) Conception and Design Impacts Open innovation through collaborative co-creation and design environments is arguably more efficient than traditional alternatives, both in terms of cost and time, as well as regarding environmental impacts (Pearce, 2015). Communities working online or in pop-up innovation hubs and hackathons prove more adequate for medical technology innovation and for reaching out-of-the-box solutions than entire R&D departments from large multinationals conditioned by their already existing product portfolio, as already analysed. Besides, there are countless examples of more traditional innovation in the medical field, in which spin-off or start-up companies developing technological breakthroughs are absorbed by larger multinationals and then closed down, purely to avoid concurrence and for simple financial reasons. In those cases, patients keep on receiving state-of-the-art solutions, instead of the more advanced and technologically feasible. Such stories of lost human capital and of technological advances kept under key or killed by partnership pilots (Barad, 2019) are also examples of purposeless generated efforts and waste. The open-source model proposes precisely the opposite: continued innovation and sharing of ideas and solutions for the benefit of patients first. (b) Production Impacts Many successful OSMDs rely on additive manufacturing (AM) technologies for their production. This not only enables a shift from mass production to mass personalization but also helps to reduce the required tooling and the debris generated in the production stage. In fact, AM technologies are seen as environmentally sustainable technologies, when compared to traditional machining and forming processes (Díaz Lantada et al., 2017), especially now that they are starting to be also competitive in terms of production speed and of number of parts per machine hour created. (c) Supply Chain Impacts Traditional medical device development and delivery to patients involves at least two relevant transport stages, that of raw materials, components and spare parts to the production site and the actual transport of the manufactured medical device to the local supplier and then to the point of care. The production of OSMDs can minimize these transport impacts by in situ production (e.g. in hospital-hosted machine shop)

254

A. Díaz Lantada et al.

even in a personalized and patient-by-patient approach. In this fashion, only the transport of raw materials and components to the actual point of care, where the devices are assembled and delivered to the patients, is necessary, which may also reduce the need for several storage places (see quantitative study of Sect. 11.5). Occupying less space means also that the floor can be used for other productive or socially relevant tasks. (d) Use-Related Impacts Local populations are not only involved in the conception, design and production of open-source devices, but also in their maintenance, which reduces the impacts associated to sending medical devices and especially medical equipment for repair purposes to the production point. In the open-source and collaborative paradigm, the whole life cycle can be carried out near the final users, and impacts derived from use, such as maintenance, repair or update tasks, are more controlled. Maintenance is also supported by well described and public documents and by collaboration among communities of developers. In fact, among the most promising market niches for 3D-printed open-source components in low- and middle- income settings, it is important to stand out the repair and update of medical equipment, in many cases donated, which without the adequate refurbishment can turn out to be useless (Miesen, 2013; Douglas, 2017). (e) End-of-Life Impacts OSMDs differ from the traditional medical devices designed and produced following a “black box” strategy, in many cases with built-in programmed obsolescence. In traditional devices (not just in the biomedical industry), the failure of a single element or subsystem leads to the whole “black box” being thrown away as rubbish and replaced by a completely new device. In the case of open-source designs, reusability or reconditioning practices are importantly promoted. Indeed, being the whole design open, replacement of components and subsystems is much easier, as the whole manufacturing process is well and openly documented and because modularity is also intrinsic to most successful OSMDs. Again the connections between AM and the expansion of the OSMD field are relevant to highlight for their implications for a more sustainable medical industry: 3D-printed components (by fused deposition modelling of thermoplastic filaments) have a remarkable recyclability, as they can be converted into printing filament quite straightforwardly (Pakkanen et al., 2017). In the case of medical devices, recyclability must also consider the original use of the device or component, and risks must be taken into account (contact with body fluids, contact with skin, duration of contact, etc.). The use of standards to this end is essential (ISO 10993, sterilization and others). The sterilization potential of the 3D printing processes should be also considered, as well as its limitations (Neches et al., 2016).

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

11.5

255

Case of Study: Life Cycle Analysis of an Open-Source Medical Device

Among pioneering and most inspiring cases of success in the open-source medical device arena, it is important to mention Enabling the Future, a team of designers of open-source prostheses for children; BITalino and ProtoCentral, companies devoted to open-source solutions for managing medical signals; Patient Innovation, a community of patients, healthcare professionals and innovators sharing ideas and cases of success for solving healthcare problems; or UBORA, a community of co-creators aimed at developing the largest online repository of open-source medical devices projects and at transforming BME education, among others reviewed in Chap. 1. In order to provide a case study for qualitatively comparing an open-source medical device with its more traditional closed counterpart, the development cycle of a hand prosthesis is chosen as functional unit. The case could be more quantitatively illustrated, for instance, by comparing the well-known 3D-printed and opensource solutions of the aforementioned Enabling the Future (http:// enablingthefuture.org/) with other commercially available options developed following closed IP approaches (i.e. hand prosthesis from medical companies). However, such quantitative study is beyond the purposes of present chapter. Here, a focus is placed on highlighting differential aspects, along the life cycle, between open-source and closed solutions. We expect that this case may help developers to approach quantification, when comparing both alternatives (open vs. closed) to medical device development and when trying to justify the environmental, economic and social benefits (and eventual drawbacks) of opensource solutions, as related to more traditional and common closed options. The different product life-cycle stages (conception and design, production, commercialization and supply, use and end-of-life), required for the development and application of a hand prosthesis, are analysed. The impacts of traditional methods from the medical industry are compared to those typical from the emergent field of open-source medical devices. In this new paradigm, co-creation, distributed production and collaboration with local populations play very special roles and lead to more dynamic and tunable solutions. Tables 11.2 and 11.3 summarize the results of this qualitative comparative study, including both environmental (Table 11.2) and social (Table 11.3) impacts. Both approaches (closed and open) have their benefits and drawbacks, but it is important to highlight that the life cycle of open-source medical devices can turn out to be extremely competitive, even as compared with well-established solutions, if their positive environmental and social impacts are considered and promotes along the whole life cycle. Risks minimization methods, for managing the more critical impacts of OSMDs (i.e. potential generation of rubbish in remote regions, challenges linked to the transport of raw materials), should be applied for solving the drawbacks and taking benefit of the positive aspects mentioned in Tables 11.2 and 11.3. In order to present a quantitative evaluation of both closed (traditional) and opensource alternatives, according to the qualitative considerations of Tables 11.2 and

256

A. Díaz Lantada et al.

Table 11.2 Qualitative analysis of environmental impacts of different life cycle stages: comparative study between conventional and open-source hand prosthesis Life cycle stages a) Conception/design impacts

b) Production impacts

c) Supply chain impacts

d) Use-related impacts

e) End-oflife impacts

Closed solution (conventional hand prosthesis) - International teams of designers, in traditional engineering design approaches, tend to meet physically in kick-off, monthly control or weekly recap or bilateral meetings, along the development stage, with the important environmental impacts of transport associated - Design teams wishing to understand the needs of local populations, typically displace to end users’ locations, which has also transport environmental impacts - Production is typically carried out in large production sites or cities (i.e. China and South East Asia). - Assembly is usually performed in the design and selling site, after receiving the components back from the production site, for quality control, packaging and marking. The environmental impacts of transport should be considered - Complex transport networks, with related workforce, are needed for reaching users. The environmental impacts of such networks cannot be discarded - Maintenance in traditional industry relies on sending the original component back to the manufacturer and on waiting for a response. In LMI settings, this is sometimes too timely, expensive and unsustainable - Well-established recycling networks and waste management processes help to minimize harmful environmental impacts of medical device waste

Open-source medical device (open-source hand prosthesis) - Co-creation through online environments (i.e. UBORA) and with participation of multidisciplinary and international communities may lead to more versatile, creative, user-centred and sustainable design processes - Online meetings and e-based co-creation and reporting minimizes the environmental impacts of transport, while helping to continuously update the project’s information and the design files under development - The creation of hubs of medical fablabs with a delocalized production strategy brings devices closer to remote and scarcely connected regions - Devices (i.e. hand prostheses) are manufactured closer to patients just by downloading the adequate blueprints from an online repository. Transport of raw materials is needed, but transport impacts are minimized, as compared with traditional options - Point-of-care production minimizes supply chain environmental impacts, enabling access to healthcare technologies in remote and scarcely connected regions and LMI* settings - Presence of design and manufacturing hubs close to users may help with maintenance and minimize maintenance transport impacts

- If recycling and collection sites are not well planned, prostheses may end up as rubbish in medical device graveyards - Related environmental impacts must be taken into account

*LMI low- and middle-income settings

11.3, a case study in a refugee camp is proposed. The refugee camps are settlements that, beyond being places of transit, more than once become actual stable and lasting settlements (see the Tindouf refugee camps in Algeria) where hundreds of thousands

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

257

Table 11.3 Qualitative analysis of social impacts of different life cycle stages: comparative study between conventional and open-source hand prosthesis Life cycle stages a) Conception/design impacts

Closed solution (conventional hand prosthesis) - Typical orientation to a specific market (i.e. EU, USA, Japan) - Customization usually relies on parametric/modular solutions - Final users are not so usually involved in the conceptual and design stages

b) Production impacts

- Mass production in large production sites or cities (i.e. China and South East Asia) and related creation of jobs, many of them precarious - Assembly close to design and selling site and related creation of jobs, usually in highly developed regions - Creation of complex transport networks, with related workforce, for reaching users - In many cases, devices cannot reach those patients needing them most

c) Supply chain impacts

d) Use-related impacts

e) End-oflife impacts

- Lack of designers and producers close to users limits versatility and personalization of the prostheses - Creation of jobs associated to medical device maintenance close to large design or production sites - In remote bad communicated regions, maintenance tasks of a defected prosthesis last long and may lead to users discarding the use of such prostheses (or other medical devices) - Well-established recycling networks and waste management processes help to minimize harmful social impacts of medical device waste

*LMI low- and middle-income settings

Open-source medical device (open-source hand prosthesis) - Involvement of patients and associations for enhanced usability - Involvement of local populations for user-centred solutions - Possibility of personalizing designs while creating (design) jobs in LMI* settings - Enabling point-of-care production and adjustment for design personalization - Creation of jobs associated to production in LMI settings

- Counting with a hub of medical fablabs and with a delocalized production strategy brings devices closer to remote and scarcely connected regions - Social impact is generated in such remote regions, both in terms of improved healthcare, as well as in terms of decent work, economic growth and industrialization - Presence of design and manufacturing hubs close to users may help with maintenance and refurbishments for an improved usability - Creation of a distributed network of jobs associated to medical device usability in LMI settings - Additional interactions between users, designers and producers accelerate innovation and help to improve the prostheses - If recycling and collection sites are not well planned, prostheses may end up as rubbish in medical device graveyards - Associating production, maintenance and recycling, as key tasks of the delocalized design and manufacturing hubs, close to users, may be an adequate option

258

A. Díaz Lantada et al.

Fig. 11.1 Dadaab refugee camp in Kenya. (Source: Africa-facts.org)

of people live together in very precarious conditions. According to the United Nations data (UNHCR), almost 26 million people in the world are officially refugees (UNHCR, 2020), and most of them live in refugee camps in countries close to their places of origin. This population had to leave their countries mainly because of civil and political strife, and many of them have lost not only their homes and family members but many times also some limbs. Being amputees in refugee camps prevents them from integrating into the host communities and from developing work that will allow them to overcome the condition of vulnerability. This case study is proposed with the idea of leaving none behind and comparing the social, environmental and economic impact of providing 1000 active hand prostheses in the context of the Dadaab refugee camp in Kenya. Dadaab is considered the largest refugee camp in Africa, located less than 500 km west of the capital, Nairobi. The organization responsible for the purchase and distribution of the prostheses will be the International Committee of the Red Cross (ICRC), based in Nairobi. From this headquarters, are not only the needs of Kenya met but also those of nearby countries such as Sudan, Somalia and the Great Lakes region (Fig. 11.1). Alternative A: The Traditional Closed-IP Solution The ICRC buys the prostheses. They will be manufactured in a company located in Shenzhen, which makes the moulds and obtains the 1000 replicas of the hand prostheses components by injection moulding (upper part of the hand, lower part of the hand, fingers and connection to forearm). Four injection moulding machines and moulds are considered necessary for this approach, and polypropylene (PP) pellets are employed as raw material (400 g per prosthesis). Each injection moulding machine consumes 100kW and the four

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

259

moulding machines working together help to obtain one prosthesis per minute. After injection moulding, the parts are packaged and send by boat to the Port of Mombasa, from where is driven to the ICRC headquarters in Nairobi for distribution to the Dadaab refugee camp. Used devices may be recycled following usual processes for thermoplastics. Minor connecting elements, batteries and conductive wires, similar for both alternatives, are not considered for the study. Alternative B: The Open-Source Solution The ICRC is partnering with the Kenyatta University, a public research university with one of its main campuses in Nairobi. This university is also part of the African Biomedical Engineering Consortium (ABEC), a collaborating partner of the UBORA initiative. In this alternative, production is performed by conventional 3D printing (fused deposition modelling) in the specialized labs from the Kenyatta University. During printing, the printers consume around 200 W, and printing time of a whole prosthesis is 6 h. Polylactic acid (PLA) is used as printing material (again 400 g of material per prosthesis). The designs are shared online and printed and assembled by technicians in the labs. The mentioned labs may act as recycling points and, in fact, convert the used components into new filament for its use in the creation of new medical devices, after adequate sterilization. Both alternatives are studied through the stages analysed in previous tables. This allows performing a preliminary approach to quantitatively compare the proposed alternatives. It is important to note that a full life-cycle analysis is beyond the purpose of the present chapter. Extraction of Raw Materials Impacts At this stage, the extraction of raw materials is also analysed, thanks to IDEMAT dataset from Delft University of Technology (DUT), and available at: www. ecocostsvalue.com (Vogtländer, 2020; IDEMAT App, 2020). The IDEMAT dataset is a set of life cycle inventories (LCI) of more than 1000 materials, services, production processes and end-of-life scenarios. The eco-costs system was introduced at the beginning of the twenty-first century and has been made operational, thanks to general databases such as the DUT used here. The eco-cost method is based on the sum of the marginal costs of preventing toxic emissions related to human health and ecosystems, emissions that cause global warming and the depletion of natural resources. The practical use of eco-costs is to compare in standardized economic terms (€) the sustainability of various types of products with the same functionality. Regarding the raw materials used in the case of study, the following consideration can be applied. Firstly, PP is a petroleum derivative, as opposed to PLA which is obtained from corn or cassava starch, or sugar cane. In addition to the fact that the latter material is biodegradable, it is easy to get in countries such as Kenya. Although the price of both filaments, PP and PLA, is very similar (around 20€/kg), substantial differences in eco-costs related to their use are identified. The eco-cost of PP is 0.54€/kg, as opposed to the PLA, which is 0.40€/kg. As 400 kg are required, in terms of eco-cost, alternative A is 56€ more expensive than alternative B.

260

A. Díaz Lantada et al.

Production Impacts Considering the energy required to mould the polymers, the value for the PP is 0.2541, and for the PLA is 0.1919, both are obtained in €/kg. This means manufacturing 1000 prostheses would have a cost in terms of energy required for moulding of 101.64€ for the alternative A and 76.76€ for the alternative B. It means alternative A is almost 25€ more expensive than alternative B. It is also necessary to consider the cost of energy consumption during the fabrication process. For that, the duration of the process is a crucial element. The electricity prices are obtained from the data of the works led by economist Neven Valev and can be consulted in additional detail in www.GlobalPetrolPrices.com (Valev, 2020). They are 0.09€/kWh in the case of China and 0.17€/kWh in the case of Kenya. In alternative A, the process to produce 1000 prostheses is 16.67 h, with the four moulding machines working at the time, which means a total cost of 600€. In alternative B, 6000 h are required to produce the 1000 prostheses, but as the energy consumption of the machine is low (200 W), the energy demand is 1200 kW, so the total cost of printing the prostheses is 204€. However, given the significant difference in hours required for production in both alternatives, the costs associated with the personnel responsible for supervising them must also be considered. For this purpose, the average net salaries of both countries, China and Kenya, provided by a database such as www.preciosmundi.com, have been taken into account. In the case of China, this salary is 859.26€/month, compared to 347.77€/month in Kenya. Thus, in terms of supervisory workforce, for the case of the alternative A, it would be around 43€, compared to 4,300€ in the case of Kenya (alternative B). At this point, it is worth remembering that alternative A is working with four machines, as opposed to one in the case of alternative B. And that the difference in the price of both machines is very high, being the price of an injection machine around 100,000€, compared to a 3D printer of around 1,000€. This aspect will not be assessed, but it should be noted that with the investment of 1000€ in an additional 3D printer, the price of labour dedicated to supervision would be reduced by half, that is, by 2150€. Additional investments leading to the creation of “3D printing farms” may further balance the situation in favour of the open-source alternative. In any case, the related creation of decent work in the case of alternative B positively contributes to SDG8, which is noteworthy. Supply Chain Impacts This section considers the costs associated with the supply chain from the time the prostheses are manufactured until they are delivered to ICRC headquarters. The transfer from the Nairobi ICRC headquarters to the refugee camp is not considered since it is the same for both alternatives. In this case, for alternative B, the costs associated with the transport of the prostheses from the university to the ICRC headquarters are negligible.

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

261

Fig. 11.2 Shipping route from Shenzhen port to Mombasa Port. (Source: searates.com)

In contrast, for alternative A, the supply chain costs are considered as follows, boat transport from the port of Shenzhen to the port of Mombasa and, from there, road transport to Nairobi. The calculations are made in a simplified way, taking into account only the main costs. The measures of each package of a prosthesis are assumed to be 0.4  0.4  0.2 m3, so the total space of a thousand boxes is 32 m3, which would fit in a single 20 feet container. The freight of a 20 feet container has a price of approximately 700 € for this load, according to the data obtained from www.worldfreightrates.com. Although more costs associated would have to be considered for customs and insurance, as well as transport from the factory to the port, these are not taken into account in this simplified model. The shipping time is 18 days according to www.searates.com. It should be remembered that the distance between both ports is more than 10,000 km. From the Mombasa port to ICRC headquarters in Nairobi, the transport is made by road. Taking into account the distance (488 km), 9 h of travel are assumed, for which only the cost of a day’s work at Kenya’s average net wage is considered, and the cost of petrol (0.823€/l), considering average consumption of 30 litres every 100 km. In this way, road transport represents a total of 137€ (Fig. 11.2). In environmental terms, the impacts are also quantified economically through the eco-costs and carbon footprint. The values obtained for both transports are presented in Table 11.4 Use-Related Impacts In terms of the use of the prosthesis, impacts can be related to the possibility of customizing and adjusting the designs offered by alternative B, which are impossible

262

A. Díaz Lantada et al.

Table 11.4 Supply chain transport impacts for alternative “A” Eco-cost Total: 477€

Carbon footprint Total: 1064€

Shipping transport Data: 0.01€/(m3
10 km) Load: 32 m3 Distance: 10,032 km Eco-cost: 321€ Data: 0.02 kg CO2/(m3
10 km) Load: 32m3 Distance: 10,032 km Eco-cost: 642€

On road transport Data: 0.01€/(m3
km) Load: 32 m3 Distance: 488 km Eco-cost: 156€ Data: 0.027 kg CO2/(m3
km) Load: 32 m3 Distance: 488 km Eco-cost: 422€

for alternative A. These impacts can be quantified in terms of the cost of people being unable to work because their prostheses do not fit their needs, considering the minimum wage in Kenya (320€). It is estimated that 5% of prostheses cannot be used from the beginning by amputees, which means a total of 50 prostheses, that is, 50 people who cannot work according to their expectations, which results in an economic impact of 16,000€ per month. End-of-Life Impacts Finally, for the last phase of the life cycle of the prosthesis, the energy required for the recycling of the materials used will be taken into account. These costs are 0.58€/ kg of material used, in the case of polypropylene, according to the data provided by the DUT. As 400 kg were needed, this cost amounts to 232€ (alternative a). The costs of this process for PLA are considered zero, according to the same database. For this study, the price of the prosthesis itself has not been included, but it is estimated that it can be around 1000–2500€ per prosthesis. This cost could be saved entirely with the open-source alternative (alternative b), and could be invested in the purchase of more 3D printers to distribute with more technology centres and universities in the countries of the area of action of ICRC Kenya, as well as leaving one in the refugee camps located in Kenya. In addition, it could be invested in training for its use, subsequently employing refugees in this task. This preliminary evaluation may help to support some of the presented considerations about the potential benefits of open-source medical devices and innovative approaches to delocalized production using 3D printing and supply chain management. As demonstrated through this case, the economic, social and environmental impacts can in some cases decline in favour of OSMDs, allowing also a priceless accessibility to the most vulnerable places, such as refugee camps. A summary of the analysis is presented in Table 11.5.

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

263

Table 11.5 Summary of the impacts during the different life cycle stages: comparative study between conventional and open-source hand prosthesis

a) Conception/design impacts

Alternative A (the traditional closed-IP solution) • Social: no impact assessed • Environmental: PP is a petroleum derivative and its eco-cost is 216€ • Economical: price of PP filaments is around 20€/kg

b) Production impacts

• Social: no impact assessed • Environmental: no impact assessed • Economical: cost of the energy consumption for moulding is 101.64€ and for manufacturing is 600€. Supervision is valued in 43€. Cost of one injection machine is estimated in 100,000€

c) Supply chain impacts

• Social: source of employment for transport (only 1-day job for transport in Kenya is valued: 16.5€, but other jobs could be valued during the transport) • Environmental: eco-cost is valued in 477€ and carbon footprint in 1060€ • Economical: 837€ invested in freight and road transport (excluding taxes) • Social: about 50 people without being able to work (losses of 320€/month for each person/family) • Environmental: no impact assessed • Economical: losses of 16,000€/month of economic development in the area • Social: no impact assessed • Environmental: no impact assessed • Economical: cost of the energy consumption for the recycling is estimated in 232€. Price estimated per prosthesis is 1000€

d) Use-related impacts

e) End-oflife impacts

11.6

Alternative B (the open-source solution) • Social: PLA is obtained from local resources (source of employment but not valued in this analysis) • Environmental: eco-cost of PLA is 160€, and it is biodegradable • Economical: price of PP filaments is around 20€/kg • Social: Employment for required supervision is valued in 4300€ (average annual salary of a person in Kenya) • Environmental: no impact assessed • Economical: cost of the energy consumption for moulding is 76.76€ and for manufacturing is 204€. Supervision is valued in 4300€. Cost of one 3D printer is estimated in 1000€ • Social: no impact assessed • Environmental: no impact assessed • Economical: no impact assessed

• Social: no impact assessed • Environmental: no impact assessed • Economical: no impact assessed

• Social: no impact assessed • Environmental: no impact assessed • Economical: cost of the energy consumption for the recycling is considered zero

Policymaking for a Sustainable Open-Source Biomedical Industry

11.6.1 Sustainable Development Policies Focusing on Equitable Healthcare Policymaking is an extremely multifaceted and complex set of processes and counts with important mechanisms for enabling bottom-up processes that take into account the needs of specific groups of citizens and the views of experts. This applies to research and innovation policies and to healthcare politics.

264

A. Díaz Lantada et al.

Open-source communities should work together for the promotion of policymaking for supporting open-source software and hardware. Acting as ethical lobbies and synergizing with key actors – World Health Organization, EU Commission and Parliament, international associations on medical technology, notified bodies – is a desirable approach to harmonize medical device regulations and to develop policies supporting OSMDs. The accountability public processes are nowadays well suited for organizations based on strategic planning, hierarchical organizations structures and quality control in processes. But in the digital society, new organizational models are emerging, based on trust and purpose. The evolutive “teal” organizations (Laloux, 2016) follow new principles, mainly: 1. Self-management 2. Wholeness 3. Evolutionary purpose The requirements of public assurance processes should be aligned with these breakthroughs. The special risks of OSMDs should be also considered and minimized by using standardized development methodologies capable of tackling current challenges. Among these, it is important to mention managing data and privacy in online co-creation environments, adequately tracking and registering modifications in open-source and collaboratively developed projects, traceability of design and manufacturing processes in international networks and innovation infrastructures, and quality controls in the point of manufacture, which may be the point of care in many cases, among others. On the other hand, policies supporting the emergence of OSMDs may regulate public contests, through which hospitals acquire medical technology and equipment, so that the evaluation procedures consider in a positive way the technologies developed as open-source solutions, for instance, by leaving a 10–15% of the evaluation to these issues or by fixing a minimum % of budget to acquiring open-source solutions.

11.6.2 Specific Calls and Topics of Public Research Programmes To this end, it is essential that recently created groups from international organizations (IFMBE, EAMBES) focusing on improving healthcare equity through innovative technologies put forward the relevance of OSMDs as transformative technologies. This can be done through their groups focusing on low- and middleincome settings by preparing white papers, conferences, workshops and seminars and meetings with the EU Commission, the EIT Health and other entities deciding on fund allocation for R&D tasks. A relevant result in the short term may be the incorporation of at least a couple of specific project calls to support open-source

11

On the Sustainable Growth of the Biomedical Industry Reinvented Through. . .

265

solutions for global healthcare concerns, as part of the health and technology research topics of the EU. Around 5M€ of budget per call for funding collaborative proposals of 100–200k€ would prove adequate for a start: with such budget, it would be feasible to conceive, design and industrialize more than 50 varied medical technologies, as open-source solutions, covering a wide set of healthcare specializations. The expected missionbased and more bottom-up implementation of the forthcoming EU framework programme (FP9 / Horizon Europe) can also contribute to funding specific collaborations among open-source communities and their stakeholders to deliver concrete open-source solutions to health issues. Non-profit foundations already making worldwide impact, in connection with the development of healthcare technologies (i.e. The Bill and Melinda Gates Foundation), could also play a very relevant role for the field of open-source medical technologies, just by proposing in their calls for ideas and in their funded projects the sharing of results as open-source solutions. This could be also considered for the results of several calls of EU-funded projects, as has recently been done with the support of the EU to open-access publishing, which is already changing the sharing of knowledge. Acting in a similar way with public-funded software and hardware developments could prove transformative.

11.7

Conclusions

OSMDs have emerged with transformative ambition as a new trend in BME. Their progressive expansion and long-term impact depend on the efforts of researchers in the field capable of demonstrating their social, economic and environmental advantages, through the successful open-source medical technologies. The role of international communities and non-profit associations focused on promoting and mentoring the expansion of OSMDs and on interacting with all relevant stakeholders, for the construction of the open-source and collaborative biomedical engineering field, can prove fundamental in these transformations. Adequate policymaking to keep up pace with these innovations and to support a straightforward deployment of their beneficial potentials, while tightly controlling safety and quality in these novel co-creation environments, is required. Possibly the introduction of new medical device classes, including those of “open-source medical device” or “open-source assistive technology”, may be needed, as well as the introduction of harmonized regulations and standards taking account of the special characteristics of OSMDs and considering the processes of co-creation environments, “makers” communities and open-source platforms with potential solutions to healthcare problems.

266

A. Díaz Lantada et al.

References Ahluwalia, A., De Maria, C., & Díaz Lantada, A. (2018). The Kahawa Declaration: A manifesto for the democratization of medical technology. Global Health Innovation, 1(1), 1–4. Barad, J. (2019). Healthcare startups struggle to navigate a business world that’s set up for them to fail. Tech Crunch. Bitalino: Community projects. https://bitalino.com/en/community/projects [On line, verified on May 2020]. Davidson, P. M., et al. (2011). The health of women and girls determines the health and well-being of our modern world: A white paper from the International Council on Women’s Health Issues. Health Care Women International, 32(10), 870–886. Díaz Lantada, A., De Blas Romero, A., Sánchez Isasi, A., & Garrido Bellido, D. (2017). Design and performance assessment of innovative eco-efficient support structures for additive manufacturing by photopolymerization. Journal of Industrial Ecology, 21(1), 179–190. Donaldson, K. (2013). The 80-dollar prosthetic knee that is changing lives”. TED Talk, TED Women. Douglas, T. (2017). Biomedical engineers from across Africa are collaborating to build medical devices. Quartz Africa, November 12th, 2017. IDEMAT App. http://idematapp.com/ [On line, verified on May 2020]. Laloux, F. (2016). Reinventing organizations. Nelson Parker. Mazzucato, M. (2018). The value of everything: making and taking in the global economy. Penguin. Miesen, M.. (2013). The inadequacy of donating medical equipment to Africa. The Atlantic, September 20th, 2013. Neches, R., Flynn, K., Zaman, L., Tung, E., & Pudlo, N. (2016). On the intrinsic sterility of 3D printing. https://doi.org/10.7287/PEERJ.PREPRINTS.542V2. Ornbo, G. (2013). Open source social responsibility. Post in: https://shapeshed.com [On line, verified on May 2020]. Pakkanen, J., Manfredi, D., Minetola, P., & Iuliano, L. (2017). About the use of recyclable biodegradable filaments for sustainability of 3D printing: State of the art and research opportunities. In: G. Campana et al. (Eds.), Sustainable design and manufacturing (Smart innovation, systems and technologies 68). Pauli, G. (2010). Blue economy: 10 years – 100 innovations – 100 million jobs. Paradigm Publications. Pearce, J. (2015). Quantifying the value of open source hardware development. Modern Economy, 6, 1–11. UBORA: Open biomedical engineering e-platform for innovation through education. https:// platform.ubora-biomedical.org/ [On line, verified on May 2020]. UNHCR. https://www.unhcr.org/figures-at-a-glance.html [On line, verified on May 2020]. United Nations. (2015). Transforming our world: The 2030 Agenda for Sustainable Development, Resolution: A/RES/70/1 of September 25th, 2015. United Nations. Sustainable Development Goals. https://sustainabledevelopment.un.org/. Valev, N.. www.GlobalPetrolPrices.com [On line, verified on May 2020]. Vogtländer, J. G. The model of the eco-cost/value ratio. www.ecocostsvalue.com [On line, verified on May 2020]. Wohlsen, M. (2013). Patent trolls are killing startups, except when they are saving them. Wired, October 9th, 2013.

Index

A Accountability, 153 Active devices, 133 Active therapeutic device, 133 Affinity diagrams, 176 African Biomedical Engineering Consortium (ABEC) Design Schools, 220 2030 Agenda for Sustainable Development, 25 AI Regulation, 158–160 Algorithm Change Protocol (ACP), 160 Analytic hierarchy process (AHP), 177

B Bag valve mask (BVM), 233 Biodesign method, 23 Biomedical engineering (BME), 5, 15, 16, 25 abnormal use, 63 healthcare technology development, 60 medical technology, 59 open-source medical devices (OSMDs), 59 regulations, 62, 63 risk minimization techniques, 60 standards, 62, 63 usability, 60, 61 usability assessment (see Usability assessment) use error, 64 users’ needs, 60, 61 Biomedical field, 25 Biomedical industry, 2 Boolean tools, 211 Bottom-up strategies biomedical industry, 52 co-creation and collaborative design, 41

collaborative product development, 41 e-infrastructures, 41 industrial innovation, 51 innovation process, 40 innovation through education, 42 Lazarus Biotech Standing Frame, 53 medical device innovation, 51 medical needs, 52 motivational questionnaires, 41 needs assessment blood transfusion, 45 children suffering, 44 clinical and medical device gaps, 47 clinical status, 43 computed tomography (CT), 46 disease burden/clinical condition, 44, 45 ethical review, 43 Gulu RRH Workshop, 48 health facility, 45 maternal health, 46 medical equipment, 46, 47 medical imaging, 46 medical instruments, 47 methodology, 43 neonates, 44 paediatric and maternal departments, 46 PNW, 47, 50 sickle cell anaemia, 44 Sub-Saharan Africa, 43 tuberculosis, 44 UBORA project, 43 Uganda’s healthcare system, 43 online infrastructures, 42 open-source and collaborative BME, 39, 40 project-based service learning activities, 51

© Springer Nature Switzerland AG 2022 A. Ahluwalia et al. (eds.), Engineering Open-Source Medical Devices, https://doi.org/10.1007/978-3-030-79363-0

267

268 Bottom-up strategies (cont.) quality improvement cycles, 52 service-learning model, 51 transforming education, 52 UBORA community, 51 UBORA e-infrastructure, 53–55 Brainstorming, 173 Brainwriting, 174

C CAD modelling, 229, 231, 237 Carbon fibre reinforced polymer (CFRP), 92 Cardiopulmonary resuscitation (CPR) procedure, 238 Computer-aided design, simulation, and manufacturing (CAD-CAE-CAM) software, 191 Conceive-design-implement-operate (CDIO), 22, 220

D Data concerning health, 151 Data minimisation, 152 Decision-making process, 39 Design for Social Impact, 103

E East Meets West Foundation, 107 Electronic prototyping, 238, 239 Engineering design, 27 Equitable healthcare, 15, 25 European Commission, 158 European Medical Device Regulation, 2, 72

F FAB Academy, 220 Face protecting splints in high-performance materials, 204–206 medical need and product description, 197, 198 personalized design methods, 199, 201, 203 Failure modes and effects analysis (FMEA) technique, 80, 90 FAIR data, 215 Finite element method (FEM), 90 Finite element modelling, 226, 228 First International Conference on Collaborative Biomedical Engineering for OpenSource Medical Technologies, 3

Index FreeCAD®, 211 Frugal Biodesign approach, 24

G General Data Protection Regulation (EU) data subject, 152 FAIR Guiding Principles, 157 in OSMDs, 156 personal data, 151 principles for data protection, 152–156 processing of data, 151 General safety and performance requirements, 134, 135 Geometrical validation, 214 Good health and well-being, 6

H Harmonized methodology biomedical industry, 34 challenges, 23–25 healthcare technologies, 34 medical devices, 23–25, 34, 35 modern product development, 21–23 open-source medical devices, 23–25 PBL, 25–27 systematic design methodologies, 21–23 Healthcare industry, 2 High-precision medical imaging, 1 Human-centered design (HCD) strategies background research, 107, 108 benefits, 102 companies, 104 detailed design process, 118–120 diagnostics, 104, 106, 123 healthcare technologies, 102 high-impact technology, 103 ideation and prototyping, 113–115 impact, 104, 105 industry practices, 103 interactive systems development, 101 interviews, 108–110 medical devices, 101, 102 multiple design traditions, 103 privacy, 102 process, 105, 106 reliability, 102 stakeholder observations, 108–110 stakeholders, 103 supply chain, 102 synergies, 120–122 synthesis, 110, 113

Index team formation, 107, 108 user and stakeholder feedback, 115–117 Human factors, 61, 70, 79

I International Electrotechnical Commission (IEC), 220 International Organization for Standardization (ISO), 22 Invasive devices, 133 ISO Compliance and Safety Assessment, 231

J Japanese International Cooperation Agency (JICA), 46

K Kahawa Declaration, 215 KJ method, 176

L Laser stereolithography, 213, 214 Lazarus Biotech Standing Frame, 52, 53 Life cycle analysis, 255, 256, 258–261 Lotus flower, 174

M MD legislation, 2 Medical devices (MDs), 127, 219 classification, 134 custom-made medical devices, 137 defined, 132 general safety and performance requirements, 134, 135 guarantee safety and efficacy, 128 legislation on, 127 on-market regulation, 135–137 post-market control, 139, 140 regulations in China, 142 in Japan, 142 in U.S, 143 software in, 140, 141 Medical Technology Transfer and Services (MTTS), 107 Mind maps, 174, 176 Monopolization, 146 Morphological box, 175 Multidisciplinary teams, 40

269 N Needs identification, 41, 47, 51–53, 55 New Engineering Education Transformation model (NEET), 26 Non-invasive devices, 133 Normal ventilation procedure, 237

O Open innovation, 2 Open-Source Hardware Association (OSHWA), 4 Open-source medical devices (OSMDs) analogous networks, 7 biomedical industry, 5 challenges and 5-year view, 15–17 collaboration and information sharing, 3 collaborative biomedical engineering, 5 collaborative project/problem-based teaching-learning methods, 219 creativity promotion co-creation and international cooperation, 183, 184 international competitions and hackathons, 181, 182 lateral thinking methods, 169 methods, techniques and resources, 187 organizational management, 169 problem solving, 169 products ideas, 175, 176 substages in, 170 supporting ideation and debate, 172, 173 techniques for evaluation, 176, 178 TRIZ methodology, 169, 178–180 UBORA devices, 168 UBORA features, 184–186 development, 7 distributing costs, 161–163 economic sustainability of business models for, 248, 250 IP protection and litigation, 251 open-innovation strategy, 250, 251 open-source social responsibility, 252 R&D and tax savings, 251 efficient healthcare system, 6 engineering methods, 6 environmental sustainability of conception and design impacts, 253 end-of-life impacts, 254 production impacts, 253 supply chain impacts, 253 use-related impacts, 254 European Regulation on Medical Devices, 130, 131

270 Open-source medical devices (OSMDs) (cont.) development phase, 131 pre-market control, 131–134 FabLab and Shapeways networks, 7 field, 7 GrabCAD, 7 growth and international projection, 7 hardware, 4, 5 hardware and software resources, 8 harmonization, 5 healthcare, 7 and healthcare technology equity, 243 innovative walking frame project, 223, 224, 228–230, 233, 234, 237, 239 life cycle analysis, 255, 256, 258–261 medical device, 4 medical device regulations, 128 medical industry, 3 medical signals, 7 modern medical technologies, 1, 2 neurology and cardiology, 7 personalized design of cross-sections, 211 innovative hip prosthesis, 208, 209, 211 open-source hardware, 192 open-source software, 194 safe sport practice, 196, 199, 201, 202, 204, 205, 208 software for design, simulation, and manufacture, 195, 196 quality and reliability, 147 remote, low- and middle-income settings, 128 safety and innovation, 147 safety criteria and performance data, 129 and SDGs, 244, 245 social product development, 1, 2 software, 4, 148 in European Union, 149, 150 IMDRF definition, 148 in U.S., 149 sustainable development policies, 263, 264 Thingiverse, 7 UBORA community, 5 UBORA e-infrastructure, 8, 10, 15 Open-source software, 4, 7

P Perception-cognition-action (PCA) technique, 81 Personal data, 151 Phase change material (PCM), 50

Index Phillips 6-6, 174 Point-of-care (POC), 106 Policymakers, 41 Portable neonate warmer (PNW), 47, 50 Post-market control, 139, 140 Pre-market control, 131–134 Privacy by default, 153 Privacy by design, 153 Processing of data, 151 Product planning, 39–41, 52 Program evaluation review technique, 176 Project-based learning (PBL) model, 25–27, 42 Prototyping process, 230, 239 Proximal femur, 210, 212 Pseudonymization, 154, 155 Public research programmes, 264, 265

R Radar chart, 177 Research and development (R&D), 23 Reusable surgical instrument, 133 Right to data portability, 156 Risk mitigation techniques design and processing, 91 design methods, 90 device validation, 88, 89 FEM analysis, 94–97 FEM simulations, 91 FMEA analysis, 91, 92 methodology, 91 post-processing, 92, 94 pre-processing step, 93 priority risk index (PRI), 91 risk control measures, 86, 87 Round-the-corner technique, 211

S SaMD pre-specifications (SPS), 160 SCAMPER, 174 Sickle cell anaemia, 44 Smartphones, 23 Sodium acetate, 50 Software as a Medical Device (SaMD), 76 Spiral innovation approach, 23 Supplementary neck shaft angle (SNSA), 210 Supply chain, 260, 261 Surgically invasive device, 133 Sustainable Development Goals (SDGs), 6, 25, 35 Synchronous teachers-group communication, 221

Index T Total hip arthroplasty (THP), 208 Total Product Life Cycle (TPLC) approach, 159 TRIZ methodology, 174, 178–180 Tsai-Wu criteria, 212

U UBORA e-platform biomedical engineering resources, 28 collaboration, 27 collaborative methods, 33 and community, 28 degree of multidisciplinary and international component, 34 hybridization, 34 industrial innovation purposes, 27 long-term impact, 34 medical devices, 28 methodology, 28 open-source devices, 33 project management metastructure, 29–31, 33 regulations and standards, 34 web-based infrastructure, 27 United Nations 2030 Agenda, 35 Usability assessment engineering workflow, 65

271 environment profile, 70, 71 evaluation techniques, 74, 76, 77 formative assessments, 74 formative evaluation, 65, 66 human error analysis, 71 human error reduction and mitigation, 72, 73 international standard, 65 medical devices, 74, 79 methodology, 79 OSMDs projects, 81, 84, 85 prototypes, 74 quick-and-dirty approach, 73 risk identification, 80–83 SaMD, 76 software design, 74 summative evaluation, 67, 69 task analysis, 71 technique, 73 usability tests, 80 user profile, 70, 71 use specification, 69

V Verein Deutscher Ingenieure (VDI), 22