Biomedical Visualisation: Volume 10 [1334, 1 ed.] 303076950X, 9783030769505

Citation preview

Advances in Experimental Medicine and Biology 1334

Paul M. Rea   Editor

Biomedical Visualisation Volume 10

Advances in Experimental Medicine and Biology Volume 1334 Series Editors Wim E. Crusio, Institut de Neurosciences Cognitives et Intégratives d’Aquitaine, CNRS and University of Bordeaux, Pessac Cedex, France Haidong Dong, Departments of Urology and Immunology, Mayo Clinic, Rochester, MN, USA Heinfried H. Radeke, Institute of Pharmacology & Toxicology, Clinic of the Goethe University Frankfurt Main, Frankfurt am Main, Hessen, Germany Nima Rezaei, Research Center for Immunodeficiencies, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran Ortrud Steinlein, Institute of Human Genetics, LMU University Hospital, Munich, Germany Junjie Xiao, Cardiac Regeneration and Ageing Lab, Institute of Cardiovascular Science, School of Life Science, Shanghai University, Shanghai, China

Advances in Experimental Medicine and Biology provides a platform for scientific contributions in the main disciplines of the biomedicine and the life sciences. This series publishes thematic volumes on contemporary research in the areas of microbiology, immunology, neurosciences, biochemistry, biomedical engineering, genetics, physiology, and cancer research. Covering emerging topics and techniques in basic and clinical science, it brings together clinicians and researchers from various fields. Advances in Experimental Medicine and Biology has been publishing exceptional works in the field for over 40 years, and is indexed in SCOPUS, Medline (PubMed), Journal Citation Reports/Science Edition, Science Citation Index Expanded (SciSearch, Web of Science), EMBASE, BIOSIS, Reaxys, EMBiology, the Chemical Abstracts Service (CAS), and Pathway Studio. 2020 Impact Factor: 2.622 More information about this series at http://www.springer.com/series/5584

Paul M. Rea Editor

Biomedical Visualisation Volume 10

Editor Paul M. Rea Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences University of Glasgow Glasgow, UK

ISSN 0065-2598 ISSN 2214-8019 (electronic) Advances in Experimental Medicine and Biology ISBN 978-3-030-76950-5 ISBN 978-3-030-76951-2 (eBook) https://doi.org/10.1007/978-3-030-76951-2 # The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 This work is subject to copyright. All rights are solely and exclusively licensed 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

The utilisation of technologies in the biomedical and life sciences, medicine, dentistry, surgery, veterinary medicine and surgery, and the allied health professions has grown at an exponential rate over recent years. The way we view and examine data now is significantly different to what has been done over recent years. With the growth, development, and improvement of imaging and data visualisation techniques, the way we are able to interact with data is much more engaging than it has ever been. These technologies have been used to enable improved visualisation in the biomedical fields, but also how we engage our future generations of practitioners when they are students within our educational environment. Never before have we had such a wide range of tools and technologies available to engage our end-stage user. Therefore, it is a perfect time to bring this together to showcase and highlight the great investigative works that are going on globally. This book will truly showcase the amazing work that our global colleagues are investigating, and researching, ultimately to improve student and patient education, understanding, and engagement. By sharing best practice and innovation we can truly aid our global development in understanding how best to use technology for the benefit of society as a whole. Glasgow, UK

Paul M. Rea

v

Acknowledgements

I would like to truly thank every author who has contributed to the tenth edition of Biomedical Visualisation. By sharing our innovative approaches we can truly benefit students, faculty, researchers, industry, and beyond, in our quest for the best uses of technologies and computers in the field of life sciences, medicine, the allied health professions, and beyond. In doing so, we can truly improve our global engagement and understanding about best practice in the use of these technologies for everyone. Thank you! I would also like to extend a personal note of thanks to the team at Springer Nature who have helped make this possible. The team I have been working with have been so incredibly kind and supportive, and without you, this would not have been possible. Thank you kindly!

vii

About the Book

Following on from the success of the first nine volumes, Biomedical Visualisation, Volume 10, will demonstrate the numerous options we have in using technology to enhance, support, and challenge education, clinical settings, and professional training. The chapters presented here highlight the wide use of tools, techniques, and methodologies we have at out disposal in the digital age. These can be used to image the human body, to educate patients, the public, faculty, and students in the plethora of how to use cutting-edge technologies in visualising the human body and its processes, to create and integrate platforms for teaching and education, to visualise biological structures and pathological processes, and to aid visualisation of the historical arenas. The chapters presented in this volume cover such a diverse range of topics, with something for everyone. We present here chapters on technologyenhanced learning in neuroanatomy, 3D printing and surgical planning, changes in higher education utilising technology, decolonising the curriculum, and visual representations of the human body in education. We also showcase how not to use protective personal equipment inspired by the pandemic, anatomical and historical visualisation of obstetrics and gynaecology, and 3D modelling of carpal bones and augmented reality for arachnid phobias for public engagement. In addition, we also present face modelling for surgical education in a multidisciplinary setting, military medical museum 3D digitising of historical pathology specimens, and finally computational fluid dynamics.

ix

Contents

1

2

3

4

Evaluating the Efficacy and Optimisation of the Peer-Led Flipped Model Using TEL Resources Within Neuroanatomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deepika Anbu, Alistair Robson, Octavia Kurn, Charles Taylor, Oliver Dean, December Payne, Eva Nagy, Charlotte Harrison, Samuel Hall, and Scott Border Observation of Patients’ 3D Printed Anatomical Features and 3D Visualisation Technologies Improve Spatial Awareness for Surgical Planning and in-Theatre Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Toby M. Branson, Leonard Shapiro, and Rudolph G. Venter Pandemics, Protests, and Pronouns: The Changing Landscape of Biomedical Visualisation and Education . . . . . . Gabrielle M. Finn, Rebecca Quinn, Katherine Sanders, William Ballard, Abisola Balogun-Katung, and Angelique N. Dueñas What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol . . . . . . . . . . . . . . . . . . . . Eve Gibbons, Matthieu Poyade, Paul M. Rea, and David Fitzpatrick

1

23

39

55

5

The Embryonic re-Development of an Anatomy Museum . . . Catherine MacRobbie

81

6

Visualising the Link Between Carpal Bones and Their Etymologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Kaitlin Nasrala, Matthieu Poyade, and Eilidh Ferguson

7

Augmented Reality Application of Schizocosa ocreata: A Tool for Reducing Fear of Arachnids Through Public Outreach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 Rohne Nyberg, Matthieu Poyade, Paul M. Rea, and Jeremy Gibson

xi

xii

Contents

8

The Surgical Art Face#: Developing a Bespoke Multimodal Face Model for Reconstructive Surgical Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 Partha Vaiude, Mark Roughley, Amy Redman, and Aenone Harper-Machin

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 Kristen E. Pearlstein, Terrie Simmons-Ehrhardt, Brian F. Spatola, Bernard K. Means, and Mary R. Mani

10

An Introduction to Biomedical Computational Fluid Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 Luke Reid

Editors and Contributors

About the Editor Paul M. Rea is a Professor of Digital and Anatomical Education at the University of Glasgow. He is qualified with a medical degree (MBChB), an MSc (by research) in craniofacial anatomy/surgery, a PhD in neuroscience, the Diploma in Forensic Medical Science (DipFMS), and an MEd with Merit (Learning and Teaching in Higher Education). He is an elected Fellow of the Royal Society for the encouragement of Arts, Manufactures and Commerce (FRSA), elected Fellow of the Royal Society of Biology (FRSB), Senior Fellow of the Higher Education Academy, professional member of the Institute of Medical Illustrators (MIMI), and a registered medical illustrator with the Academy for Healthcare Science. Paul has published widely and presented at many national and international meetings, including invited talks. He sits on the Executive Editorial Committee for the Journal of Visual Communication in Medicine, is Associate Editor for the European Journal of Anatomy, and reviews for 25 different journals/ publishers. He is the Public Engagement and Outreach lead for anatomy coordinating collaborative projects with the Glasgow Science Centre, NHS, and Royal College of Physicians and Surgeons of Glasgow. Paul is also a STEM ambassador and has visited numerous schools to undertake outreach work. His research involves a long-standing strategic partnership with the School of Simulation and Visualisation, the Glasgow School of Art. This has led to multi-million pound investment in creating world-leading 3D digital datasets to be used in undergraduate and postgraduate teaching to enhance learning and assessment. This successful collaboration resulted in the creation of the world’s first taught MSc Medical Visualisation and Human Anatomy combining anatomy and digital technologies. The Institute of Medical Illustrators also accredits it. It has created college-wide, industry, multi-institutional, and NHS research linked projects for students. Paul is the Programme Director for this degree.

xiii

xiv

Editors and Contributors

Contributors Deepika Anbu Faculty of Medicine, University of Southampton, Southampton, UK William Ballard Hull York Medical School, University of Hull, Hull, UK Abisola Balogun-Katung Hull York Medical School, University of York, York, UK Scott Border Faculty Southampton, UK

of

Medicine,

University

of

Southampton,

Toby M. Branson Adelaide Medical School, The University of Adelaide, Australia Oliver Dean Faculty Southampton, UK

of

Medicine,

University

of

Southampton,

Angelique N. Dueñas Hull York Medical School, University of York, York, UK Eilidh Ferguson Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK Gabrielle M. Finn Faculty of Biology, Medicine and Health, University of Manchester, Manchester, UK David Fitzpatrick Scottish Ambulance Service and University of Stirling, Stirling, UK Eve Gibbons Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK School of Simulation and Visualisation, The Glasgow School of Art, Glasgow, UK Jeremy Gibson Zoology, Kentucky Wesleyan College, Owensboro, USA Samuel Hall University Hospital Southampton, NHS Foundation Trust, Southampton, UK Aenone Harper-Machin Department of Plastic Surgery, Whiston Hospital, Merseyside, UK Charlotte Harrison Oxford University Hospitals, NHS Foundation Trust, Oxford, UK Octavia Kurn Faculty Southampton, UK

of

Medicine,

University

of

Southampton,

Catherine MacRobbie Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK

Editors and Contributors

xv

Mary R. Mani Federal Bureau of Investigation, Laboratory Division, Virginia, USA Bernard K. Means Virginia Commonwealth University, Virginia, USA Eva Nagy Faculty of Medicine, University of Southampton, Southampton, UK Kaitlin Nasrala School of Simulation and Visualisation, Glasgow School of Art, Glasgow, UK Rohne Nyberg Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK School of Simulation and Visualisation, Glasgow School of Art, Glasgow, UK December Payne Faculty of Medicine, University of Southampton, Southampton, UK Kristen E. Pearlstein National Museum of Health and Medicine, Defense Health Agency, Silver Spring, Maryland, USA Matthieu Poyade School of Simulation and Visualisation, The Glasgow School of Art, Glasgow, UK Rebecca Quinn Hull York Medical School, University of Hull, Hull, UK Paul M. Rea Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK Amy Redman Surgical Art, Douglas, Isle of Man, UK Luke Reid Centre for Anatomy and Human Identification, University of Dundee, Dundee, Scotland, UK Alistair Robson Faculty of Medicine, University of Southampton, Southampton, UK Mark Roughley Liverpool School of Art and Design, Liverpool John Moores University, Liverpool, UK Katherine Sanders Centre for Anatomical and Human Sciences, Hull York Medical School, University of Hull, Hull, UK Leonard Shapiro Department of Human Biology, University of Cape Town, Cape Town, South Africa Terrie Simmons-Ehrhardt Virginia Commonwealth University, Virginia, USA Brian F. Spatola National Museum of Health and Medicine, Defense Health Agency, Silver Spring, Maryland, USA Charles Taylor Faculty of Medicine, University of Southampton, Southampton, UK

xvi

Partha Vaiude Department of Plastic Surgery, Noble’s Hospital, Douglas, Isle of Man, UK Liverpool School of Art and Design, Liverpool John Moores University, Liverpool, UK Surgical Art, Douglas, Isle of Man, UK Rudolph G. Venter Division of Orthopaedic Surgery, Stellenbosch University, Stellenbosch, South Africa

Editors and Contributors

1

Evaluating the Efficacy and Optimisation of the Peer-Led Flipped Model Using TEL Resources Within Neuroanatomy Deepika Anbu, Alistair Robson, Octavia Kurn, Charles Taylor, Oliver Dean, December Payne, Eva Nagy, Charlotte Harrison, Samuel Hall, and Scott Border

Abstract

The flipped classroom (where students prepare before and then develop understanding during class) and technology-enhanced learning (audio-visual learning tools) are increasingly used to supplement anatomy teaching. However, the supporting literature lacks robust methodology and is conflicting in demonstrating efficacy outcomes. Contrastingly, near-peer teaching (where senior students teach juniors on the same academic programme) is well researched and reported to be both effective and versatile. This provides an ideal vehicle in which to investigate and potentially optimise these approaches. This study aims to assess educational impact of the peer-led flipped model and student engagement and perceptions regarding traditional and TEL resources. A quasi-randomised, cross-sectional study was conducted with 281 second-year

D. Anbu · A. Robson · O. Kurn · C. Taylor · O. Dean · D. Payne · E. Nagy · S. Border (*) Faculty of Medicine, University of Southampton, Southampton, UK e-mail: [email protected] C. Harrison Oxford University Hospitals, NHS Foundation Trust, Oxford, UK S. Hall University Hospital Southampton, NHS Foundation Trust, Southampton, UK

University of Southampton medical students. Students were randomly allocated to 3 groups: traditional lecture (control), flipped text resource, or flipped video resource. The first group received no pre-teaching material, but the flipped groups received a text or video pre-teaching resource. Objective outcomes measured were: – Knowledge gain and retention via multiplechoice questionnaires and formative exams – Student perceptions and engagement using questionnaires and 2 focus groups All groups demonstrated significant knowledge gain post-teaching ( p < 0.0001). However, regardless of engagement with pre-teaching material, no significant difference was found in knowledge gain or retention between the groups. Students engaged 21.1% more with the text rather than video resource ( p ¼ 0.0019), but spent equal time using both ( p ¼ 0.0948). All resources and teaching approaches were perceived ‘very useful’ with no significant differences found between groups. A qualitative approach utilising thematic analysis of focus groups identified 4 themes, including ‘Attitudes towards flipped classroom’, which revealed mixed reviews and perceptions from participants. This study has found the peer-led flipped model is of no detriment to educational impact compared to peer-led traditional approaches in

# The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. M. Rea (ed.), Biomedical Visualisation, Advances in Experimental Medicine and Biology 1334, https://doi.org/10.1007/978-3-030-76951-2_1

1

2

D. Anbu et al.

a well-established peer teaching programme in undergraduate medicine at the University of Southampton. Students value traditional and video resources but engage with them differently. Additionally, it was reported that in this experiment, NPT did not seem well suited to the flipped classroom, suggesting a rare limitation of the utility of NPT application within an anatomy curriculum. Keywords

Flipped classroom · Technology-enhanced learning · Multimedia learning · Near-peer teaching · Anatomy · Medical education

1.1 1.1.1

Introduction Climate of Anatomy Education

In the UK, timetabled anatomy education has been in decline over a number of years to prioritise professionalism-based themes along with early clinical exposure (Gogalniceanu et al. 2009; Turney 2007). One particular challenge has been that there is less time to teach difficult subjects that students find the most challenging, such as neuroanatomy (Gogalniceanu et al. 2009). Nevertheless, graduates should begin their fulltime clinical training with a good level of anatomical understanding and competencies—a view that remains largely unchanged in modern reappraisals of core syllabi (C. F. Smith et al. 2016). Consequently however, new doctors have often attributed their unpreparedness for daily clinical procedures to an insufficient knowledge of anatomy (Fitzgerald et al. 2008; Gogalniceanu et al. 2009). Thus, for both students and future patients, it is important that anatomy educators can adapt and innovate to deliver the same content in a pressured and restricted timeframe. The application of innovative pedagogical approaches, including those that fall under ubiquitous terms such as technology-enhanced learning (TEL), the flipped classroom (FC), and near-peer teaching (NPT), is increasingly being utilised by modern educators,

with TEL and FC growing in popularity for their potential to both optimise and supplement anatomy education (Clunie et al. 2018).

1.1.2

Technology-Enhanced Learning (TEL) in Anatomy Education

TEL is an umbrella term to describe the design and application of a variety of audio-visual resources for educational purposes (Kirkwood and Price 2014): one of the most popular being educational videos and screencasts. These are especially favoured in anatomy education, with 78–85% of students using videos in this discipline (Barry et al. 2016; Reverón 2016), due to how videos complement anatomy education. Anatomy is a visual and information-rich subject and the impact of reduced teaching has decreased the opportunity for students to develop confidence in skills such as spatial awareness—a vital skill for surgery and radiology (Rochford 1985). In using anatomy videos, students see key dissections and orientations alongside diagrams and narrations and are able to develop these skills remotely as part of a blended curriculum (Garg et al. 2001). Videos allow pausing, replaying, and slowing, which enables students to learn at ‘their own pace’ with reduced stress and anxiety, making TEL ideal for difficult subjects like neuroanatomy (Clark 2002; Sotgiu et al. 2020). It has been suggested that the flexibility of these features can facilitate information rehearsal, thus improving consolidation to long-term memory, as described by the multi-store model of memory (Fig. 1.1) (Atkinson and Shiffrin 1968; Topping 2014). The cognitive theory of multimedia learning elaborates on this basic principle to describe how presenting information, such as images alongside audio (as videos do), navigates learners to vital areas of each image. This has potential to increase information processing for long-term memory storage, by reducing cognitive load—in essence, by utilising auditory information to avoid visual overload (Fig. 1.1) (Mayer and Moreno 2003). Incorporating instructional design principles

1

Evaluating the Efficacy and Optimisation of the Peer-Led Flipped Model. . .

3

Fig. 1.1 Diagram to show the formation of long-term memory

such as these could be ideal for content heavy disciplines such as anatomy. Given these benefits, educators have increasingly integrated videos into practice, with its application within the FC being particularly popular (Sotgiu et al. 2020).

1.1.3

Flipped Classroom (FC)

The FC is an educational approach that inverts traditional learning so students complete introductory teaching material (at home), before engaging in active study in the classroom with tutors (Fig. 1.2) (Cheng et al. 2017).

Fig. 1.2 Diagram to show stages of learning in both the traditional and flipped classrooms

It has been suggested that prior engagement with pre-teaching material develops students’ responsibility for their own learning (Taylor and Hamdy 2013). Subsequently, students attend classes more prepared to engage in discussion with their peers and instructors, encouraging effective communication with colleagues of varying seniority (Tazijan et al. 2018). Additionally, it can facilitate the development of other important transferable skills required of medical professionals (such as teamwork and independence), which can then be utilised in multidisciplinary settings and therefore help prepare students for their professional life (Gogalniceanu et al. 2009). There is evidence to suggest that more effective class time communication better enables educators to address specific learning needs and offer a more personalised learning environment— this is because tutors are able to better recognise learners’ deficits of knowledge and to develop strategies to overcome these obstacles (Pickering and Roberts 2018). Evidence also alludes to the idea that the FC can support learning by promoting student engagement with difficult subject matter and concepts—something which is likely to be optimised for challenging topics such as neuroanatomy (Clark 2002; Sotgiu et al. 2020). The application of the FC has witnessed a rise in popularity alongside TEL approaches (Clunie et al. 2018). However, evidence in the field of anatomy supporting any such improvements in the student experience or impact on knowledge

4

D. Anbu et al.

is both limited and conflicting (Clunie et al. 2018).

1.1.4

Lack of Evidence in Favour of Combining TEL and the FC

Whilst the body of TEL literature is mostly positive, the majority of reports predominantly investigate outcomes of student satisfaction (Adamczyk et al. 2009) and are set within a live curriculum with uncontrollable variables (Choudhury et al. 2009), or do not explore educational impact (Clunie et al. 2018). Those who do undertake more robust studies state conflicting findings. Saxena et al. and Topping et al. have both independently implemented gross anatomy videos for first-year undergraduate medical students and compared exam performance scores. Despite similar methodologies, Saxena et al. discovered a 3.4% improvement, whilst Topping et al. concluded a non-favourable conclusion (Saxena et al. 2008; Topping 2014). There are further examples of contradictory findings amongst the TEL and the FC literature. For instance, the knowledge-acquisition phase is fundamental to the FC, yet the literature supporting the use of pre-teaching material compliance is somewhat unconvincing—with some reports indicating significant student engagement and others reporting very little (Bouwmeester et al. 2016; Gilliland 2017). Furthermore, some studies investigating FC efficacy differ significantly in their methodology, rendering meaningful comparisons of their findings difficult (O’Flaherty and Phillips 2015). Of these reports, very few are centred within medical education, and even fewer in the subject area of anatomy (O’Flaherty and Phillips 2015). Whilst TEL and the FC are increasingly being utilised in combination with one another there really is no educational validation for their pairing (Clunie et al. 2018; Pickering and Roberts 2018). Further robust, hypothesis-driven investigations using control groups for comparison are required before implementation within a high stakes live curriculum (Pickering and Roberts 2018). One

application which has potential to trial this approach in a less restrictive and formal way is near-peer teaching (Hall et al. 2018). These programmes are mostly ancillary to formal curriculum sessions and pose lower levels of ethical risk in terms of compromising educational quality (particularly for those in experimental teaching settings).

1.1.5

Near-Peer Teaching (NPT)

NPT is defined as a student learner being taught by a student teacher more senior to them within the same training pathway (Hall et al. 2018). There is ample literature to support NPT in both medical education and anatomy teaching (Harrison et al. 2018; Leong et al. 2012), the benefits of which are largely attributed to optimised (small) distances in social and cognitive congruence (Lockspeiser et al. 2008). Social congruence describes NPTs' ‘informal and empathetic’ communication with students due to age proximity, making learners more comfortable with classroom participation (Lockspeiser et al. 2008). Cognitive congruence on the other hand describes the effect of similar knowledge levels between NPT and learner, allowing content delivery in a relatable and digestible manner via informal explanations and language (Schmidt and Moust 1995). Consequently, NPT has been implemented with high efficacy and student satisfaction in many institutions (Sotgiu et al. 2020), across multiple disciplines (Leong et al. 2012), using various teaching approaches (Ten Cate and Durning 2007). Combining the FC and TEL in a peer-led flipped model offers potential to reliably investigate and compare the use of TEL to the use of more traditional resources in this setting. Although the available evidence is yet to offer insight into the true educational value of the FC, educators are still choosing to adopt it and with the rapid conversion to online teaching strategies and models of blended delivery being supported as good practice—combining the use FC with TEL resources is only likely to increase.

1

Evaluating the Efficacy and Optimisation of the Peer-Led Flipped Model. . .

Furthermore, the peer-led flipped model presents an exciting opportunity to evaluate the versatility of NPT programmes. NPTs have only basic training in teaching, and flipping the classroom may well require a range of skills and judgement associated with experienced educators. Therefore, in this study the primary aims were to: 1. Investigate the educational impact of the peer-led flipped model via: a. Knowledge gain b. Knowledge retention 2. Investigate student’s degree of engagement and student perceptions regarding traditional and TEL educational resources in the peer-led flipped model

1.2

Methods

University of Southampton (UoS) Ethics and Research Governance, ID: 23736.

1.2.1

Study Setting and Population

Recruiting Participants Participants were 281 second-year medical students spanning 2 cohorts, starting the nervous system (NS) module between October 2018 and October 2019. Both attended a 60-min NPT-led session, with 2018/19 cohort being the control, and 2019/20 cohort the experimental group. All participants were blinded to study aims and informed that their participation would not affect summative assessments or progression. Consent was obtained via a consent tick-box after teaching sessions with lack of consent forming the exclusion criteria. As sessions were timetabled as compulsory, further recruitment was unnecessary. Recruiting NPTs Yearly, students above second year enter the training programme consisting of 4 full-day training sessions, after which they are selected by a panel of senior members of the NPT team and faculty members.

5

Recruiting for Focus Groups Via social media, poster campaigns, and word of mouth 12 students from the 2019/20 cohort attended 2 focus groups, each with 6 participants. Students were given participant information sheets, briefed, and signed a consent form before each focus group.

1.2.2

Resource Development

Teaching Content ‘Introduction to Cranial Nerves’ was selected as the learning outcomes for the study as students had not previously been taught this topic, and therefore, baseline knowledge levels were fairly consistent amongst participants. This is a topic introduced early in the module and reduced the risk of student’s self-study time interfering with the measurement of learning gain. Formulating Learning Objectives Six basic and six advanced learning objectives (LOs) were created using the UoS Medical Curriculum. Basic LOs covered identification and basic functional knowledge and were used to develop the pre-teaching resources for the FC group and the TLS group session (control group). The more advanced LOs covered knowledge application (such as pathologies). They were used to develop the FC teaching session, and given to the control cohort after teaching to guide their independent study. Pre-teaching Resources The 2019/20 cohort were split into the text-based resource group (TG) and video resource Group (VG), and 2 pre-teaching resources were created accordingly. To ensure validity of the independent variable measurement, efforts were made to ensure the only difference between the TG and the VG was the modality of the pre-teaching resource. The content included was standardised for all teaching groups (Fig. 1.3). The learning outcomes for the topic were used to find relevant images and prepare a script.

6

D. Anbu et al.

Fig. 1.3 A schematic to demonstrate the development and standardisation of pre-teaching and teaching resources for both control and FC cohorts

– To create the text-based resource (TBR) the script and images were combined and formatted into a 4-page PDF. – To create the 20-minute video resource (VR), using Blue®, digital Spark USB condenser microphone the script was narrated and recorded. Images were redrawn and labelled using Adobe Illustrator Draw, on the iPad Pro (Apple Inc.). Using the screen record feature on the iPad Pro (Apple Inc.), animations were created from the images and combined with recorded narration using video editing software iMovie (version 10.1.9 for macOS). All resources created were made by the principal investigator, with input and quality control and assurance by UoS Faculty of Medicine academic staff. Teaching Session Resources For the TLS control cohort, the script and images guided development of a 12-slide PowerPoint Presentation, 1 slide dedicated to each cranial nerve (CN). For the FC cohort, this was adapted

according to the advanced LOs (Fig. 1.3). Each slide was followed by a slide containing further clinical pathologies building upon pre-teaching material.

1.2.3

Teaching Sessions

Control Group: 2018/19 Cohort UoS randomly allocates students into groups (of 15) for clinical placements, which were utilised for this study. In week 2 of NS module, each group was taught by a NPT using the TLS presentation. NPTs were briefed by the principal investigator to teach according to the TLS. After teaching, students were provided advanced LOs to guide their independent learning. Flipped Group: 2019/20 Cohort Existing faculty-derived clinical placement groups were used to allocate students to either TG or VG via quasi-randomisation. In week 1 of the NS module, TG students were emailed with instructions to access the TBR on

1

Evaluating the Efficacy and Optimisation of the Peer-Led Flipped Model. . .

the university’s virtual learning environment (Blackboard) to prepare for their teaching session in week 2. In week 3, VG students were emailed with instructions to prepare for their teaching session in week 4 using a private YouTube link to the VR. In the teaching sessions, each group was taught by a NPT using the flipped presentation. NPTs used were briefed by the lead researcher on how to utilise the flipped classroom approach. To ensure only modality of pre-teaching resource differed between both groups, the same NPTs were used for both TG and VG students. To ensure no students were disadvantaged, both resources were made available to all students in week 5 so all students received equal opportunity to utilise available resources.

1.2.4

Assessment of Knowledge Gain

Pre-teaching Assessments All students completed a 12-question cranial nerves multiple-choice questionnaire (MCQ) as a pre-teaching MCQ. Additionally, the flipped cohort also completed a pre-teaching 5-point Likert style questionnaire with yes/no questions to assess engagement. Post-teaching Assessments All students completed a 12-question post-teaching MCQ; additionally the flipped cohort completed a post-teaching perception questionnaire. Original study plan aimed for all groups to complete a 12-question retention MCQ after completing the NS module (8 weeks long). Instead due to COVID, 10 questions that aligned to this study’s LOs were extracted from the 40-question formative spotter examination (which all groups completed). Paired pre- and post-teaching MCQs were used to assess knowledge gain, whilst formative spotter examinations were used to assess knowledge retention. Unique candidate numbers (given by the UoS) were written on each test and

7

questionnaire for confidentiality and to maintain linked anonymity. This allowed pairing of teaching session scores with formative exam scores. The post-teaching questionnaire given to the flipped cohort utilised 5-point Likert scales to investigate student engagement and perceptions regarding the teaching session. After completing the NS module, 12 students from the flipped cohort who expressed interest in discussing their experiences were recruited to partake in 2 focus groups. These were conducted using a pre-prepared semi-structured framework designed to assess student opinions regarding educational resources and the peer-led flipped model. They were recorded using a digital MP3 voice recorder (Olympus WS 750M) and transcribed externally using Olympus DDS Player Standard Transcript Module (Release 2). See Fig. 1.4 for summary of data collection points.

1.2.5

Analysis of Data

Calculating and Comparing Knowledge Gain To calculate knowledge gain, normalised learning gain (LG) was used instead of absolute LG (Fig. 1.5). With absolute LG, a high pre-test score results in a misleadingly low absolute gain figure. By normalising the true change in knowledge independent of pre-test scores can be calculated. All the following tests were chosen as Shapiro–Wilk’s tests indicated all data were non-parametric. To test effectivity of teaching session in all groups, Wilcoxon signed rank tests using preand post-test scores were used. To compare knowledge gain between groups, learning gains were calculated and compared using Kruskal– Wallis ANOVA with post hoc comparisons via Dunn test. Similarly, to compare knowledge retention between groups, exam scores were compared using Kruskal–Wallis ANOVA with post hoc comparisons via Dunn test.

8

D. Anbu et al.

Fig. 1.4 A schematic to display the different data collection points for each group

Analysing Engagement and Student Perceptions Perception questionnaire responses were used to investigate engagement with both pre-teaching resources, including perceptions and attitudes towards each flipped session. For responses to Fig. 1.5 Displaying two equations that can be used to quantify learning gain

yes/no questions, descriptive statistics were calculated along with comparative analysis via Chi squared test (with Fisher’s exact test). For Likertscale data, both descriptive statistics and Mann Whitney U analysis were performed to compare between groups.

1

Evaluating the Efficacy and Optimisation of the Peer-Led Flipped Model. . .

9

To further analyse student perceptions, focus groups were coded and grouped on Microsoft Word 2015 (Microsoft Corp., Redmond, WA) via inductive thematic analysis by the principal investigator. For all statistical tests α < 0.05 was used. Preliminary quantitative data organisation was performed using Microsoft Excel 2019 version 16.32 (Microsoft Corp., Redmond, WA) with all inferential statistical analysis and graph production generated in GraphPad Prism® (GraphPad Software, LLC).

1.3

Results

All error bars shown represent upper and lower quartiles.

1.3.1

Student Engagement

The pre-teaching questionnaire assessed student engagement with pre-teaching resources in the flipped cohort. 79% (79/100) of TG students answered ‘yes’ when asked if they used the resource, compared to 57.9% (55/95) of VG students (Fig. 1.6). The 21.1% difference in engagement between groups was statistically significant ( p ¼ 0.0019). Participants who answered ‘yes’ were asked how long they spent using the resource. Despite modality differences, both TG and VG students spent a median of 31–60 mins on the pre-teaching material, with no significant difference between groups ( p ¼ 0.109) (Fig. 1.7).

1.3.2

Fig. 1.6 A graph to show student engagement with resources of different modalities. Bar labels represent the number of students, with total n ¼ 195

gain between approaches. Figure 1.9 shows median post-teaching normalised learning gains for all groups: control—75% (n ¼ 88), TG— 83.3% (n ¼ 79), and VG—75% (n ¼ 55). Comparing students who did and did not engage with pre-teaching material revealed no significant difference in knowledge gain in both TG ( p ¼ 0.607) and VG ( p ¼ 0.841) groups (Fig. 1.10).

Educational Impact

Knowledge Gain Comparing knowledge test scores between control cohort and those who engaged in the flipped cohort, all groups showed a significant increase in knowledge, (p reduced to 1787

Patient

Patient being treated by the paramedic.

12,231

Surgical mask

Mask to be used during the mask contamination scenario.

41,000 > reduced to 13,756

subconsciously while using PPE. Therefore, it was decided that the app would aim to target this by using interactive 3D animations to complement the existing material. The narrative of the app was developed: the user will be faced with three different situations in the form of a 3D animation, each presenting a different type of PPE protocol violation, such as touching the patient without gloves on. The user can interact with the environment and must then select the area, i.e. the gloves, with the problem. They are able to toggle a semi-transparent layer showing where infection transmission could occur in this case, with an information box reminding the user of the correct protocol in this situation (Fig. 4.5).

4.3.3

Digital Design

For the user interface (UI) design inspiration, a search for e-healthcare and medical app design was conducted on Pinterest, Google Images and

Creator 3D Adult Man Uniform Surgeon—Deep3dstore (2020) Price: $35 (Turbosquid) Available at: https://www.turbosquid.com/3dmodels/3d-adult-man-uniform-surgeon1521762 Ambulance—Silaspaige1998 (2019) Price: free (Turbosquid) Available at: https://sketchfab.com/3d-models/ ambulance41574a23a9ae4c9d990ffe672af6281e Hospital Trolley/Gurney—Johnny83 (2012) Price: free (Turbosquid) Available at: https://www.turbosquid.com/3dmodels/free-hospital-trolley-gurney-3d-model/ 710965 3D Casual Man Body/Face Rigged Free 3D Model—Amaugendre (2018) Price: free (Turbosquid) TurboSquid 1273007, Turbosquid.com. 3D Medical Mask model—db0013 (2020) Price: free (Turbosquid) Available at: https://www.turbosquid.com/3dmodels/3d-medical-mask-model-1535141

Dribbble. It was immediately clear from the results that the running theme of most apps within this industry had a colour palette consisting of calming blues, greens and pastels and very clean UI assets to ensure clear comprehension from the user while still remaining attractive. As the focus on many of these apps is to convey important information without the need for decorative embellishments, which is also the case for this research, it was decided that the UI design would maintain a clean and cohesive theme incorporating similar calming colours from a colour palette of blue, greys, and a contrasting red for the transmission layer. The blues used in the app’s colour palette would be similar hues to the NHS and SAS logos to maintain cohesion. The app’s unique user interface was designed on Adobe Illustrator. The buttons, such as a the ‘home’ button, benefitted from simple outlines in order to be consistent with generic designs used in other apps for the user to navigate easily (Fig. 4.6).

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

Fig. 4.5 Storyboard for the ‘What not to do with PPE’ app

4

63

64

E. Gibbons et al.

Fig. 4.6 UI Kit developed for the ‘What not to do with PPE’ app, displaying the colour palette used and the buttons created

NASA’s guidance on designing flightdeck procedures and checklists was consulted, which followed existing literature on checklists, not limited to aviation, where following the incorrect steps is most detrimental (Barshi et al. 2016). The design of these steps follows strict rules, such as the format of text and font use, elements of which were taken into account when designing the app’s UI.

4.3.4

3D Model Development

4.3.4.1 Identification of PPE Violations It was important that the animations of the chosen PPE violations were accurate in terms of importance in order for the app to be a relevant addition to current guidelines. A list of common scenarios where violations might occur were identified from the literature (Krein et al. 2018) and sent to an experienced ambulance paramedic in order to heuristically rank them from most risky to least risky. The top three were identified as follows: (1) Touching/repositioning the front of the mask with/without gloves on while carrying out procedure; (2) Using a mobile device with contaminated gloves still on; and (3) Removing/

putting goggles on with gloves on. This list was taken into account, in addition to the need for the animations to look realistic given the context and long enough for the user to identify the violations when deciding the definitive animation sequences for the app. The identified ‘problems’ to be in the app are as follows, however, given time constraints, it was decided that the first two would be essential and the third if time allowed: 1. Scenario 1 (Contamination of face mask): Scene shows a paramedic using his stethoscope on a patient lying on a stretcher. Mid procedure, the paramedic adjusts the front of his face mask before continuing on. (a) The user will have to select the mask, or the hand used to touch the mask. 2. Scenario 2 (Contamination of phone): Scene shows a paramedic sitting down in the ambulance. They reach into their pocket to pick up a mobile phone and proceeds to use it with gloved hands. They then put the phone up to their ear as if on the phone. (a) The user will have to select the mobile phone or the gloved hands as the ‘problem area’. 3. Scenario 3 (Contamination of goggles): Scene shows paramedic entering the

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

ambulance and proceeds to don eye protection while already wearing gloves. (a) As this is the incorrect donning order, the user will have to select the eye protection and gloved hands as the ‘problem area’.

4.3.4.2 Modelling Various 3D models for the scenes were required: a main avatar as the paramedic to demonstrate the various PPE misuse scenarios, equipment to be used incorrectly, a patient, the ambulance and the objects within the ambulance environment. As the rigging and animation of the avatars would take up a significant amount of time, it was decided that the simpler models, e.g. components of the ambulance such as cabinets, would be developed and the rest purchased from a 3D model marketplace. The 3D models were developed in Maya 2019. Initially, the total polygon count within the ambulance environment was approximately 70,000. The majority of these polys were from the imported stretcher asset and the ambulance skeleton itself. To combat this, each model was split into its individual components using the ‘Separate’ tool and, using the ‘Reduce’ tool, the polycount was reduced to its lowest possible level while maintaining the original shape. Given that the main focus within the animations is the paramedic and to reduce cognitive load, it was not as important for the models within the ambulance environment to be detailed, only to provide some context. Using this same method for all models within the ambulance environment, the total poly count was reduced to 27,441. This reduced poly count allowed for more detail when it came to the paramedic, which had a high poly count of 30 k+, and ensured system performance was not compromised once imported into the Unity engine. 4.3.4.3 Animation For all three animations outlined in 3.3.1, a general workflow was developed. It was decided that the paramedic model would be isolated and animated in Maya, then placed into the scene within Unity. The workflow consisted of first blocking

65

out and keyframing the major poses—for the Phone Contamination scenario, this involved the paramedic (1) sat down in an idle pose; (2) appearing to touch his phone as if keying in a phone number; and (3) talking on the phone. This is done by using the Rotation tool on the relevant joints on the character’s pre-rigged skeleton. Once these poses were keyed into the time editor, they were then refined into a smoother animation using the graph editor. In order to animate the transfer of the phone in this scenario from the paramedic’s pocket to the paramedic’s phone, it was constrained using a Parent constraint twice, first to a locator in the paramedic’s pocket and then to a locator in hand. In the attribute editor, only one of these constraints was on at any one time and animated to be switched between the two (Figs. 4.7, 4.8, 4.9, and 4.10).

4.3.5

App Development

In order to ensure the user’s experience made sense chronologically and ensure clarity before the development of the app began, a schematic map of the scene flow was developed. Upon opening the app, the user begins with the main menu, followed by a scene with instructions to select a scenario of their choice. They are then taken to the 3D animation of this scenario, from which they must identify the area where there is a failure to follow PPE protocol. The user is then taken back to a menu to complete the other scenarios.

4.3.5.1 User Interface Set-Up Within the UI of the animated scenes, there is a control bar at the top of the screen, allowing the user to control the animation and return to the main menu or quit the app at any time while using it. This was done by having clickable replay and pause buttons on the left side and clickable home and quit buttons on the right side. In the spot test panel, it was decided that the rendered image would be the textured version of the paramedic model, to put what the user has seen in the monochrome animation into a more

66

E. Gibbons et al.

Fig. 4.7 Blocking out the main poses to be animated, shown in the graph editor

familiar context. The texture originally downloaded with the model was altered in Adobe Photoshop using a hue/saturation layer to change the paramedic’s scrubs from purple to green, standardised uniforms across all UK

Fig. 4.8 Animation of the mask contamination scenario

Ambulance Services. Using transparent buttons was deemed to be the most effective way in allowing the user to simply click on the problem areas. After clicking, the user sees the same image but with a semi-transparent, red layer to

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

67

Fig. 4.9 Modifying the animation in the Graph Editor

consolidate the information they have identified (Fig. 4.11). This image, Fig. 4.12, was modified in Photoshop using the Pen tool to outline the areas and filled in with a red brush with lower opacity.

4.3.5.2 Interactive Components The interactive functions of the app were programmed using C# scripts. The animations of the 3D models were controlled by the Animator Controller within Unity and then further

Fig. 4.10 Final animation graph for the phone contamination scene

68

Fig. 4.11 Rendering a still image of the paramedic animation in Maya

Fig. 4.12 Image shown to user once they have successfully identified problem area

E. Gibbons et al.

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

controlled by the player using pace control UI buttons. It was initially decided that the user would be presented with a ‘transmission toggle’ whereby they would be able to toggle a layer of visible contamination on the paramedic on and off, using Particle Systems. This was originally going to be done by using colliders on the contaminating objects to instantiate the particles, however, it appeared to be quite taxing on the computer to keep instantiating whenever the user replayed the animation. It was therefore decided that the user would have the option at the end of the first animation to watch it again with the particle systems turned on or go straight through to the spot test if they had already identified the violation, thus giving the user complete control of their experience. For the visible contamination animations, the particle systems were parented to the object they must follow. For example one particle system representing contamination of the phone is parented to the hand holding the phone. Instead of using colliders, it was much simpler to delay the start of the particles for the areas, such as the phone, that became contaminated as a result of the violation in PPE protocol. Considering the outcomes of the AMRSim project and the visual implementation of contamination, it was decided that particle systems would be explored in place of projectors. It is hypothesised to provide a more reliable visualisation of contaminated areas on deformable meshes, as well as being more familiar to the user. In addition, it is expected that the slight blinking effect of the particles would bring further attention to this visualisation.

4.3.5.3 Build to Android To build the app for an Android phone, the Android Studio SDK and JDK were installed and found within Unity preferences. The app was built on a Samsung S7 using the Android OS v8.0.0 system with an aspect ratio of 2560x1440.

4.4

69

Results: Application Development Outcome

The developed app is called ‘What Not to Do With PPE’, and the app icon shows two gloves in the same style as the rest of the UI, designed in Adobe Photoshop.

4.4.1

Main Menu

The first initial scene is the main menu or the ‘landing page’ of the app. It displays an ambulance with flashing sirens on a looping animation, the title of the app, and the play button. It also includes the logos for The Glasgow School of Art and the University of Glasgow. The font used here and throughout the entire app is ‘Axiforma’, externally sourced from the website Fontsfree. pro. The play button takes the user through to the Instructions scene (Fig. 4.13).

4.4.2

Instructions

The second scene gives the user a brief overview of what the app will consist of and instructions on how to control the animation. As some of the buttons may be difficult to understand at face value, such as the spot test button and the button to watch the animation with a layer of visible contamination, they were explained in this section too. The user is then directed to the Scenario Selection scene via a button below the instructions panel (Fig. 4.14).

4.4.3

Scenario Selection

The scenario selection presents the user with two options—scenario 1 or scenario 2, each represented by a virus icon. Initially, these icons were to represent the type of PPE misused, such as gloves, but as the user has to spot this in the following scenario, it was decided that a simple

70

E. Gibbons et al.

Fig. 4.13 Main Menu Scene

numbering system would work better to not give the user an idea of what is to come in the animation. The user also has the option to navigate back to the main menu or exit the app in this scene (Fig. 4.15).

Fig. 4.14 Instructions Scene

4.4.4

Scenario 1: Phone Contamination

When the user selects the first scenario, they are taken to the first animation, showing the

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

71

Fig. 4.15 Scenario Selection Scene

paramedic using his phone in the ambulance. The animation plays immediately as the scene loads, and the user is able to control it by pausing and replaying or using the camera view panel to look at the animation from a birdseye view, from the side, or back to the front view. As there is no visible contamination in this scene, this is intended for users that may be well versed in PPE protocol and recognise the misuse of PPE easily. After the animation has ended, the user has the option to either go through to the spot test straight away or watch the animation again with a visible layer of contamination. In the case that the user selects the spot test, a panel pops up showing a still frame of the animation. The user must identify the problem area: in this case, this is the gloved hands. The user is able to try multiple times until they get this right. To reinforce the correct protocol, a layer of red at a 50% opacity is shown highlighting these areas. When they are ready, the user can then go back to the Scenario Selection menu and try out the next scenario (Figs. 4.16, 4.17, 4.18, and 4.19).

4.4.5

Scenario 1: Phone Contamination with Visible Transmission

In the case that the user chooses to see the animation again with contamination showing, they are taken through to the same animation again, except this time Unity’s Particle Systems show where cross-contamination may occur as a result of the PPE violation. This is intended to facilitate the idea that when there is instant visual feedback, it is much easier to understand how your actions can be harmful. At the end of the animation, the spot test panel pops up, and they can select the problem area (Fig. 4.20).

4.4.6

Scenario 2: Mask Contamination

If the user selects Scenario 2 in the Scene Selection menu, they are shown the second animation using the same approach as the first

72

E. Gibbons et al.

Fig. 4.16 Scenario 1, Phone Contamination

scenario. This particular animation shows a paramedic tending to a patient and palpating the chest before touching the front of the mask (Fig. 4.21).

4.4.7

Scenario 2: Mask Contamination with Visible Transmission

The user is able to see the animation again with contamination visible (Fig. 4.22). Figure 4.23

Fig. 4.17 Option for the user to select the spot test or to watch the animation again with visible contamination

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

Fig. 4.18 Spot test panel for Scenario 1, user selected incorrect area

Fig. 4.19 Spot test panel for Scenario 1, user selected correct area

73

74

Fig. 4.20 Scenario 1, Phone Contamination, with visible contamination

Fig. 4.21 Scenario 2, Mask Contamination

E. Gibbons et al.

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

Fig. 4.22 Scenario 2, Mask Contamination, with visible contamination

Fig. 4.23 Spot test panel for Scenario 2, user selected correct area

75

76

E. Gibbons et al.

shows the outcome of the spot test development for Scenario 2.

4.5 4.5.1

Discussion Reflection on the Design Process

On the whole, the design process for the application went well. The final application hits the targets of wanting to create an educational tool about PPE protocol for paramedics, building upon principles from the cognitive load theory (van Merriënboer and Sweller 2010) as a basis for instructional design. One particular aspect that was a success was the additional scenes depicting cross-contamination to replace the initial plan of the transmission toggle. This approach has had similar success in raising awareness of the spread of bacteria in a veterinary setting (Macdonald et al. 2019). Although the toggle may have provided an additional interactive feature, the ability for the user to select a different scene, and therefore level of difficulty, depending on their level of experience and knowledge in the area, is intended to facilitate the integrated to non-integrated strategy suggested by van Merriënboer and Sweller (2010) to combat expertise-reversal effect. Whether donning and doffing PPE is a relatively new procedure or one that the user is well versed in, the varying difficulty allows the user to use this educational tool with more autonomy.

4.5.2

Limitations

4.5.2.1 Animations Another difficulty evident in the final application was the challenge of learning how to develop complicated animated sequences from scratch. As the paramedic character downloaded from Turbosquid (Table 4.2) came with a basic skeleton but was not completely rigged, i.e. did not have rig controls, it made the animation process more time consuming. It also meant that realistic movements of the character became more easily

compromised by manipulation errors—for example up-scaling an arm by accident when keying poses would not happen if the scale attributes had been locked. More time would have been required to anticipate such issues through additional previous training. The initial plan was to have the paramedic performing a simple procedure, using his stethoscope on the patient, however, it quickly became apparent how difficult animating a dynamic stethoscope would be, so this was abandoned and replaced by a more ‘general’ animation of the paramedic tending to a patient. However, feedback from supervisors suggested this could potentially be distracting from the actual breach in PPE protocol. The time taken to alter an animation in the later stages of development was difficult, and ideally, given more time, this would have been refined and improved upon. Despite the problems with this stage, the model’s movements appear to be well executed, and the action of violating PPE protocol comes across in them.

4.5.2.2 Models The use of downloaded models was not the initial intention; it would have been preferable to model from scratch all assets within the app to allow more creative freedom. This, again, was a result of the short window of time given to design the app and the need to prioritise certain aspects, such as the animation, over others. The paramedic model could have been improved, specifically with it being clearer that the character had gloves on. Due to the decision to make the app monochrome, this might not be particularly obvious— this then affects the user’s experience with this scenario. Additionally, it may have been effective to use a green shader on the PPE, i.e. the gloves and mask in these scenarios, to further signify. This would have required more work on the paramedic’s model to separate the gloves from the main mesh. 4.5.2.3 Future Directions of Work Unfortunately, despite putting out various calls for participants among ambulance staff to test the app, the global pandemic limited the availability of the application target user, and there were

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

no sign-ups for testing, which caused difficulty in confirming the effectiveness of the app. Evaluation will be needed in the future to verify the validity of the application to educate and raise awareness about the correct use of PPE. The lack of ambulance guidance demonstrated in the literature may be due to the fact that guidance is generalised to all healthcare providers, and therefore paramedic-specific PPE guidance is not necessary. However, as providers of emergency healthcare, it may be more common for paramedics to subconsciously breech PPE protocol in the more time-pressured, pre-hospital environment; the true frequency of possible breeches is unknown as there is a dearth of published research on paramedic behaviour in the context of PPE. Again, this is why feedback after testing the app on paramedics would be particularly beneficial to determine the app’s eventual use. Depending on the outcome of the testing investigation, it may appear to be more useful for the app to have a wider-reaching audience for not only those in the paramedic service, but for people that use PPE regularly. This technology has the potential to be widely distributed, specifically in order to tackle the impact the current pandemic in the UK has had on care homes, whereby staff may not have the skillset and materials to use PPE properly. Additionally, the app could have potential in the midst of continually developing government guidelines around the public use of PPE, notably the mandatory use of face masks in some spaces to stem the spread of the virus. From observations in public and in the media, it is evident that the use of these masks is often incorrect, with the unfamiliarity leading to people frequently touching the front of the mask or only covering the mouth, whether this can be attributed to misinformation or subconscious slip-ups. Following successful testing of the app and positive feedback, there could be some scope for this app to extend to the public in the event that the compulsory use of PPE becomes long-term.

4.6

77

Conclusion

Since completing the research for this project on August 2020, the number of confirmed cases of COVID19 worldwide had reached 20.2 million (Worldometer 2020). Since then, this number has increased five-fold. More than ever, the need for novel, emerging technologies combining visualisation and healthcare is paramount. In this chapter, a prototype of an educational, interactive tool for ambulance clinicians has been developed as a proof of concept to investigate the potential for raising awareness of proper PPE protocol and the resulting contamination if not followed correctly. Although the effectiveness of the app is yet to be determined through testing, it is clear that What Not to Do with PPE was created from a well-defined concept, and this was carried through to the finished app. The research would benefit hugely from the further study; firstly, user evaluation, and secondly, the potential to broaden the content to a wider audience, whether this is the public or less specialised HCP. The results, therefore, presented in this dissertation uphold the use of an interactive application in the education of HCP through the novel integration of 3D animation and interactive content.

References Adobe (2020) https://www.adobe.com/uk/ Amaugendre (2018) https://www.turbosquid.com/Search/ Artists/amaugendre Autodesk (2020) https://www.autodesk.co.uk/free-trials Barshi I, Mauro R, Degani A, Loukopoulou L (2016) Designing Flightdeck Procedures 81 BBC News (2020) Coronavirus: ambulance staff “feeling unprotected.” BBC News Blackburn B, Ferryman JT, Alattas A (2020) Impact of COVID-19 on UK Foundation doctors: training and safety (preprint). https://doi.org/10.21203/rs.3.rs30834/v1 Clock SA, Cohen B, Behta M, Ross B, Larson EL (2010) Contact precautions for multidrug-resistant organisms (MDROs): current recommendations and actual

78 practice. Am. J Infect Control 38:105–111. https://doi. org/10.1016/j.ajic.2009.08.008 Fischer WA, Hynes NA, Perl TM (2014) Protecting health care workers from Ebola: personal protective equipment is critical but is not enough. Ann Intern Med 161:753. https://doi.org/10.7326/M14-1953 GOV.UK (2021) PHE and LSHTM: the contribution of nosocomial infections to the first wave, 28 January 2021. [online]. Available at https://www.gov.uk/ government/publications/phe-and-lshtm-the-contribu tion-of-nosocomial-infections-to-the-first-wave-28-jan uary-2021. Accessed 28 February 2021 Health Protection Scotland (2020a) Assessing the evidence base for medical procedures which create a higher risk of respiratory infection transmission from patient to healthcare worker 15 Health Protection Scotland (2020b) HPS website COVID-19 - the correct order for donning, doffing and disposal of personal protective equipment (PPE) for healthcare workers (HCWs) in a primary care setting [WWW document]. https://www.hps.scot.nhs. uk/web-resourcescontainer/covid-19-the-correctorder-for-donning-doffing-and-disposal-ofpersonalprotective-equipment-ppe-for-healthcareworkers-hcws-in-a-primary-care-setting/. Accessed 6 April 2020 Huertas PS, León EA, Tarabla HD (2019) Zoonosis and veterinary waste disposal in rural practice. Rev Argent Microbiol 51:251–254. https://doi.org/10.1016/j.ram. 2018.08.004 John A, Tomas ME, Hari A, Wilson BM, Donskey CJ (2017) Do medical students receive training in correct use of personal protective equipment? Med Educ Online 22:1264125. https://doi.org/10.1080/ 10872981.2017.1264125 Katanami Y, Hayakawa K, Shimazaki T, Sugiki Y, Takaya S, Yamamoto K, Kutsuna S, Kato Y, Ohmagari N (2018) Adherence to contact precautions by different types of healthcare workers through video monitoring in a tertiary hospital. J Hosp Infect 100(1):70–75. https://doi.org/10.1016/j.jhin.2018.01.001.Epub Krein SL, Mayer J, Harrod M, Weston LE, Gregory L, Petersen L, Samore MH, Drews FA (2018) Identification and characterization of failures in infectious agent transmission precaution practices in hospitals: a qualitative study. JAMA Intern Med 178:1016–1022. https://doi.org/10.1001/jamainternmed.2018.1898 Liu Y, Yan L-M, Wan L, Xiang T-X, Le A, Liu J-M, Peiris M, Poon LLM, Zhang W (2020) Viral dynamics in mild and severe cases of COVID-19. Lancet Infect Dis. https://doi.org/10.1016/S1473-3099(20)30232-2 Macdonald A, Poyade M, Noble S, Laragione R, Watson F, Chambers M, Chambers M (2019) Resistance movements: combatting infection risk in veterinary practice through visualisation methods. WHO CARES 4 Mann CM, Wood A (2006) How much do medical students know about infection control? J Hosp Infect

E. Gibbons et al. 64:366–370. https://doi.org/10.1016/j.jhin.2006.06. 030 Northington WE, Mahoney GM, Hahn ME, Suyama J, Hostler D (2007) Training retention of level C personal protective equipment use by emergency medical services personnel. Acad Emerg Med 14:846–849. https://doi.org/10.1197/j.aem.2007.06.034 Oracle (2020) https://www.oracle.com/uk/index.html Phin NF, Rylands AJ, Allan J, Edwards C, Enstone JE, Nguyen-Van-Tam JS (2009) Personal protective equipment in an influenza pandemic: a UK simulation exercise. J Hosp Infect 71:15–21. https://doi.org/10. 1016/j.jhin.2008.09.005 Poller B, Hall S, Bailey C, Gregory S, Clark R, Roberts P, Tunbridge A, Poran V, Crook B, Evans C (2018) ‘VIOLET’: a fluorescence-based simulation exercise for training healthcare workers in the use of personal protective equipment. J Hosp Infect 99:229–235. https://doi.org/10.1016/j.jhin.2018.01.021 Public Health England (2020a) COVID-19 personal protective equipment (PPE) [WWW Document]. GOV. UK. https://www.gov.uk/government/publications/ wuhan-novel-coronavirusinfection-prevention-andcontrol/covid-19-personal-protective-equipment-ppe. Accessed 5 February 2020 Public Health England (2020b) Recommended PPE for ambulance staff, paramedics, other patient transport services and pharmacy staff 1 Public Health England (2020c) Guide to donning and doffing standard Personal Protective Equipment (PPE) Samsung (2020) https://www.samsung.com/uk/ Savage Interactive (2020) https://savage.si The BMJ (2021) Up the line to death: covid-19 has revealed a mortal betrayal of the world’s healthcare workers - The BMJ. [online] Available https://blogs. bmj.com/bmj/2021/01/29/up-the-line-to-death-covid19-has-revealed-a-mortal-betrayal-of-the-worldshealthcare-workers/ Accessed 28 February 2021 Tomas ME, Kundrapu S, Thota P, Sunkesula VCK, Cadnum JL, Mana TSC, Jencson A, O’Donnell M, Zabarsky TF, Hecker MT, Ray AJ, Wilson BM, Donskey CJ (2015) Contamination of health care personnel during removal of personal protective equipment. JAMA Intern Med 175:1904–1910. https://doi. org/10.1001/jamainternmed.2015.4535 Tomas ME, Cadnum JL, Mana TSC, Jencson AL, Koganti S, Alhmidi H, Kundrapu S, Sunkesula VCK, Donskey CJ (2016) Utility of a novel reflective marker visualized by flash photography for assessment of personnel contamination during removal of personal protective equipment. Infect Control Hosp Epidemiol 37:711–713. https://doi.org/10.1017/ice.2016.44 UnityTechnologies (2020) https://unity.com van Merriënboer JJG, Sweller J (2010) Cognitive load theory in health professional education: design principles and strategies. Med Educ 44:85–93. https:// doi.org/10.1111/j.1365-2923.2009.03498.x Verbeek JH, Rajamaki B, Ijaz S, Sauni R, Toomey E, Blackwood B, Tikka C, Ruotsalainen JH, Balci FSK

4

What Not to Do with PPE: A Digital Application to Raise Awareness of Proper PPE Protocol

(2020) Personal protective equipment for preventing highly infectious diseases due to exposure to contaminated body fluids in healthcare staff. Cochrane Database Syst Rev. https://doi.org/10.1002/14651858. CD011621.pub4 Worldometer (2020) Coronavirus update (Live): 3,507,789 cases and 245,258 deaths from COVID19 Virus Pandemic - Worldometer [WWW Document]. https://www.worldometers.info/coronavirus/. Accessed 5 March 2020

79

Worldometer (2021) https://www.worldometers.info/coro navirus/?utm_campaign=homeAdUOA?Si% 23countries Zamora JE, Murdoch J, Simchison B, Day AG (2006) Contamination: a comparison of 2 personal protective systems. CMAJ Can Med Assoc J 175:249–254. https://doi.org/10.1503/cmaj.060094 Zellmer C, Van Hoof S, Safdar N (2015) Variation in health care worker removal of personal protective equipment. Am J Infect Control 43:750–751. https:// doi.org/10.1016/j.ajic.2015.02.005

5

The Embryonic re-Development of an Anatomy Museum Visualising Anatomical Collections: Human Female Reproductive Anatomy and Biology and Foetal Development Catherine MacRobbie Abstract

This chapter discusses the history of the Museum of Anatomy at the University of Glasgow in the context of a planned themed display on obstetrics and pregnancy, centred around human female reproductive anatomy, to support a showcase of Plaster Casts made and used by William Hunter. This exhibition aims to enhance the audience’s experience with an educational display of historical specimens as well as anatomical artwork and medical models. It is anticipated that the resultant exhibition will include a series of visualisations and diagrams for use within the collection display to support the audience’s understanding of the biological processes involved in reproduction, foetal development and women’s experiences of childbirth. The chapter considers historical and contemporary methods of visualising embryos, as well as the developing discourse around menstruation, the gendered body and the lack of diverse representation in gynaecological images, and reflects on some of the historical, scientific, situational and societal considerations needed to achieve an inclusive and accessible exhibition. It also reflects on the artist’s role in

C. MacRobbie (*) Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK e-mail: [email protected]

the embryonic development of this exhibition. The artworks in this chapter and the more that are planned should guide viewers with intentionally inclusive visual content. The project requires considerable further development and discussion with the team of experts involved. It is hoped that this intervention will broaden the impact of the collections in this space and provide opportunities to improve audience engagement by creating content that reflects and includes the voices of society in its creation. Keywords

William Hunter · Reproductive and developmental biology · Anatomy museum collections · Gravid uterus · Medical visualisation · Embryology · Gendered body · Diversity in medical images · Museum learning · Access and inclusion This chapter discusses the development of a productive partnership project that aims to investigate how the impact of anatomical collections can be enhanced by creating visualisations for specimen displays for diverse audiences, using human female reproductive anatomy and biology and foetal development as a case study. An initial proposal for this work was generated by a team of University of Glasgow and Hunterian staff in response to working with the collections in the Hunterian Museums; it is this proposal that has

# The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. M. Rea (ed.), Biomedical Visualisation, Advances in Experimental Medicine and Biology 1334, https://doi.org/10.1007/978-3-030-76951-2_5

81

82

inspired the wider investigation explored in this chapter. The specimens within the Hunterian collection are of enormous historical relevance and anatomical significance and provide countless opportunities to apply this method of creative support/digital visualisation. This will involve the development of feasible and deliverable educational resources and learning tools for using anatomy collections and online learning spaces, encouraging users to become involved in activities related to exhibitions, fostering a culture of participation and connecting the community to their museum. The early stages of an exhibition development project are being realised in collaboration with a team of staff in the University of Glasgow and the Hunterian and with experts in human fertility and reproduction. This will ensure wide-reaching support, comprehensive consideration of factors and successful implementation, as well as unfettered access to the unique collection of dedicated material, clear recommendations and access to facilities and additional resources. This exhibition pilots a themed display on obstetrics and pregnancy in the Museum to support the phased return of the Hunter Plaster Casts. This aims to enhance the audience’s experience and provide a professional and appropriate educational display of historical specimens. The artist’s role is to generate visualisations and diagrams of foetal and reproductive specimens and reproductive biological processes for use within the collection display. This chapter will provide illustrated examples of draft artwork and work-in-progress visuals that demonstrate the diversity of images, models and visualisation that can be produced to improve knowledge of pregnancy, foetal development and medical history. The exhibition will incorporate specimens, surgical instruments and anatomical models along with illustrations to tell the story of the human female reproductive anatomy and biology and embryonic/foetal development. This chapter explores various methods of visualisation that could be taken when approaching illustration for a modern anatomy museum and showcases them alongside examples of traditional methods of visualising female reproductive health.

C. MacRobbie

This visualisation project considers the role and function of the artist and the connections between the artwork and museum display of anatomical preparations and recognises the importance attached to these objects. The broad intentions of this project are to cultivate public awareness and appreciation of the broader themes of anatomy and medicine in museum collections. I will begin with a brief history of the museum to illustrate the context for the exhibition and for the visualisation of female reproductive health to date. After this, I will discuss advances in embryology and visualisation in pregnancy, leading onto menstrual health and addressing the lack of diverse representations in gynaecological images. Finally, I will discuss the role of the illustrator in visualising reproductive health with a focus on the future of the museum and visualisations based on developments and advances in medicine and visual technology and on changing demographics of audiences.

5.1

History and Context

The Museum of Anatomy (Hunterian), University of Glasgow, is home to one of the world’s most significant anatomical displays. An understanding of how the Museum of Anatomy has reached its present form will help determine the direction taken in and the assets that are created for the exhibition. For the purposes of this chapter, I focus on the Museum’s significance for obstetrics and reproductive medicine. A rapid growth of medical teaching in the first half of the eighteenth century established the study of anatomy at the core of teaching at the University of Glasgow from 1751 (University of Glasgow GB 248 GUA MED n.d.). William Hunter (1718–1783) is arguably the most eminent figure associated with anatomy at the University of Glasgow. Hunter undertook an apprenticeship under Dr. William Cullen, which facilitated an illustrious career in London in midwifery, surgery and anatomy. He became a personal obstetric consultant to Queen Charlotte and was elected as a Fellow of the Royal Society in 1767 and Professor of Anatomy to the Royal Academy in

5

The Embryonic re-Development of an Anatomy Museum

1768. Hunter’s private obstetric practice afforded him the opportunity to assemble his own significant private anatomical collections, which he used in teaching from 1746 until his death in 1783. Following William Hunter’s death, his magnificent collection of anatomical and pathological specimens was bequeathed to the University of Glasgow under the condition that Matthew Baillie acts as a custodian until his retirement from teaching (University of Glasgow GB 247 MS Hunter n.d.). The pathology preparations, particularly in the areas of birth defects that were poorly understood, presented a challenge to the medical profession at that time. Prior to his death, Hunter had written to Dr. Cullen stating that his collections were ‘to be well and carefully packed up and safely conveyed to Glasgow. . . . I give and bequeath the same to be kept and preserved by them and their successors for ever. . . in such sort, way, manner and form as . . . shall seem most fit and most conducive to the improvement of the students of the said University of Glasgow’ (Thomson 1850). Hunter also bequeathed funds for a building to house and display his collections. The Hunterian Museum opened in 1807, as Scotland’s first public museum. The Hunterian collection hosts 2600 anatomy and pathology specimens, the core of which are permanently displayed in the Thomson

Fig. 5.1 Museum of Anatomy. Digital illustration. Catherine MacRobbie 2021

83

Building, named in honour of Allen Thomson, Regius Professor of Anatomy from 1848 to 1877 and a draughtsman of textbooks of anatomy and physiology. In addition to the Anatomy Museum (Fig. 5.1), the Thomson Building (Fig. 5.2) houses the Anatomy Facility, part of the School of Life Sciences of the College of Medical, Veterinary and Life Sciences. The Museum of Anatomy contains a series of themed bays and an upper gallery. The northern facade has decorative timber columns around a central door. The opulent detail is in the Baroque Glasgow style. The Museum was designed purposefully to house this important teaching collection and the anatomical component of the Hunterian collection. Meanwhile, John Hunter’s collection of specimens forms the basis of the Hunterian Museum in London. The original Hunterian collections are of international importance. The Cleland Collection was a further significant addition to the Museum of Anatomy, with over 500 specimens of comparative anatomy. John Cleland (1835–1924) was Regius Professor of Anatomy at the University, 1877 to 1909. The lower walls of the museum are lined with original cabinets that contribute to the library character of the space. The expanded anatomical collections have surpassed the capacity of the room, and a curatorial decision is needed to

84

C. MacRobbie

Fig. 5.2 Thomson Building. Digital illustration. Catherine MacRobbie 2021

design and display the collection material in such a way as to allow full appreciation of both the architecture and the collection. A major impulse behind the planned exhibition has been the anticipated return of Hunter’s plaster cast gestation models following the major Tercentenary Exhibition, William Hunter and the Anatomy of the Modern Museum, which ran at the Hunterian Art Gallery, University of Glasgow, from September 2018 to January 2019 and then at the Yale Center for British Art (Yale University, New Haven, Connecticut, USA) from 14 February to 20 May 2019. The Tercentenary Exhibition offered a wider understanding of Hunter’s collections. The University of Glasgow Technical Art History Group, in collaboration with The Hunterian, conducted a technical examination of 5 of the 11 life-size plaster cast gestation models that entered the collection with William Hunter’s bequest in 1807 and which were used in Hunter’s lectures on midwifery. This series shows the dissected uterus in different stages of pregnancy and illustrates a range of complications including breech presentation, placenta previa, umbilical strangulation and

obstructed labour with distended bladder and colon. The casts correspond to illustrations in Hunter’s ‘The Anatomy of the Gravid Uterus Exhibited in Figures’, published in 1774, with exquisite plates from drawings by the renowned anatomical artist Jan van Rymsdyk. All casts are made of plaster except one composite model, which is made of lead and plaster (Richter and Sanchez-Jauregui n.d.). The accessibility of Hunter’s material is enriched by the scientific examination and conservation of the plaster casts: this treatment constitutes one of the key elements in informing knowledge of their production in progressive stages of dissection of the same specimen. Ten of the life-size plaster casts of dissections showing the uterus in late pregnancy previously formed a display in the south upper gallery of the Anatomy Museum. The gestational models show anatomical relationships in situation, showing the shape, size and position of the uterus in relation to the pelvic structures and the abdominal organs, and presenting the uterus opened with membranes reflected to reveal the foetus (McCulloch et al. 2001a, b). The returning casts will be displayed supine on purpose-built

5

The Embryonic re-Development of an Anatomy Museum

covered plinths, taking centre stage between each of the inset bays in the Museum. Hunter’s discovery that the umbilical vessels insert on the placental disc provided a wider understanding of the anatomy of the placenta and description of the separate circulatory systems of the mother and the foetus. By injecting the vasculature with red and blue wax, Hunter definitively demonstrated the maternal and foetal circulations to be separate as the umbilical vessels on the foetal placental surface were found not to contain wax. This led to an understanding of transplacental exchange between the foetal and maternal circulation. After nearly 240 years, Hunter’s collection continues to be relevant and to attract widespread interest. It is crucial to celebrate its legacy and to honour public accessibility in the museum (Longo and Reynolds 2016). Hunter’s remarkable ‘The Anatomy of the Human Gravid Uterus Exhibited in Figures’ is an astonishing obstetric atlas that disseminated anatomically accurate illustrations and showcased developments in the practice of midwifery and scientific discoveries on the causes of maternal death in childbirth, saving the lives of thousands of women and children (McCulloch et al. 2001a, b). Created from direct observations examining the process of pregnancy and foetal development, with a focus on the uterus, placenta and embryos from three weeks gestation to near full term, the atlas visually and medically transformed the field of obstetrics and supported a generation of obstetricians to better understand childbirth and pregnancy. This acceleration of research meant that there was a constant need for illustrations, and this moulded a synergy between the artwork and the forum in which it was created. It has been argued that William Hunter and Jan Van Rymsdyk reduced female anatomy into constituent parts, resulting in the woman being objectified as a teaching tool (McCulloch et al. 2001a, b). The images portrayed in ‘The Anatomy of the Human Gravid Uterus Exhibited in Figures’, similarly to the plaster casts, generally contain only the trunk and partial thigh of the cadavers, allowing for a focus on the anatomical components of the uterus.

85

These preparations were designed to communicate the findings of William and his brother John Hunter; they are beautiful, but also graphic and challenging. The foetus shown in utero appears intact and alive, preparing to be born, but the mother is dehumanised, anatomised. There is a powerful disparity between the humanity of the baby and the inhumanity of the dissected mother. Ruth Richardson argues that ‘Hunter and his contemporaries form the medical authority which privileges the interests of a dehumanised medical science’ (Richardson 2000). Reproductive anatomy displayed as specimens inspires criticism from feminist literature, which argues that the unnatural positioning and flayed aesthetic of the historical specimens and the resulting illustrations are barbaric and unnatural (Richardson 2000). The museum contains cabinets of preserved reproductive specimens, suspended by threads in preservative fluid, in glass jars known as pots. This positions the (female) body as contained and observed (the very opposite of Bakhtin’s leaking and open women): the unreachable, unknown, mysterious nature of the internal working is laid bare and on display (Bakhtin 1984). The pots, the faded colouration of the specimen and their deteriorated labels, as well as the interior of the museum, invite a sense of the time in which the specimens were collected. The objects are obscured by distortion, and this creates limitations in viewing the specimens behind glass, which exacerbates the inaccessibility of the specimen to current audiences. Before advances in medical imaging and dissection, the developing foetus was concealed and knowledge of the functions and processes of the pregnant uterus and developing foetus was sought for the purposes of avoiding miscarriage. Illustrating the foetus in the womb served as memento mori for the precautions that pregnancy might necessitate. With the growing use of dissection in the eighteenth century, the unobserved aspects of pregnancy could be explored in more detail. Hunter’s study and dissection of women in varying stages of pregnancy enabled the foetus to be tracked from the earliest stages. Figure 5.3

86

C. MacRobbie

Fig. 5.3 Plate VI 1774 Anatomia uteri humani gravidi tabulis illustrata. The anatomy of the human gravid uterus exhibited in figures—Plate VI 1774 Anatomia uteri humani gravidi tabulis illustrata. William Hunter Published: 1774. Wellcome Library, London. Wellcome Images http://wellcomeimages.org. Copyrighted work available under Creative Commons Attribution only licence CC BY 4.0. http:// creativecommons.org/ licenses/by/4.0/

shows Plate VI 1774 Anatomia uteri humani gravidi tabulis illustrata; this work places the foetus and uterus within the body of the woman. The visualisation contribution to the museum project should facilitate the lay viewers’ understanding of the female reproductive system. Introducing the viewer to the anatomy of this bodily system is important to assist in their understanding of the body in pregnancy. ‘The anatomy of the female pelvis is complex, multi layered, and its three-dimensional organisation is conceptually difficult to understand’(Sergovich et al. 2010 p: 127–33). There are multiple anatomical structures—as well as multiple

non-genital peripheral anatomic structures— which comprise the female reproductive system and which it will be important to explain to visitors of the museum. In a pregnant woman, the foetus develops and grows in the uterus. Figure 5.4, ‘Plate IX 1774 Anatomia uteri humani gravidi tabulis illustrata’ by Jan Van Rymsdyk, visualises the pregnant uterus and surrounding structures in cross section. A pregnant uterus may measure around 22–26 cm. We can describe the healthy nulliparous, or non-pregnant uterus, as a pear-shaped organ around 7.5 cm in length, comprised of a fundus, body, isthmus and cervix. The fallopian tubes and ovarian ligaments are

5

The Embryonic re-Development of an Anatomy Museum

87

Fig. 5.4 Plate IX 1774 Anatomia uteri humani gravidi tabulis illustrata. The anatomy of the human gravid uterus exhibited in figures—Plate IXI 1774 Anatomia uteri humani gravidi tabulis illustrata. William Hunter Published: 1774. Wellcome Library, London. Wellcome Images http://wellcomeimages.org. Copyrighted work available under Creative Commons Attribution only licence CC BY 4.0. http:// creativecommons.org/ licenses/by/4.0/

attached to the body of the uterus. These major organs will form the basis of the visualisation of this topic; this can also include the supporting structures such as the broad ligament that contains the fallopian tubes, ovaries, ovarian arteries, uterine arteries and round, suspensory and ovarian ligaments. The cervix is embraced by the vagina to create the vaginal structures. The vagina is lined with epithelium and a sub-dermal layer rich in capillaries, supported by a thick smooth muscle layer. This epithelium experiences hormone-related cyclical changes during the menstrual cycle. The cavity of the uterus communicates with the cervical canal via the

internal os, and the canal opens into the external os at the vaginal vault (the expanded region of the vaginal canal at the internal end of the vagina). The uterine body flexes at the cervix (anteflexion), and the uterus is tipped forward (anteversion). There are variations in normal anatomy as well as under pathological circumstances (Ellis 2011; p: 99–101). The pelvis is highly vasculature and served by the pelvic venous plexus. Connecting the internal and external anatomy for the viewer is an important route to reading the function of the bodily system. The external anatomy known collectively as the vulva includes the labia minora, labia majora, the

88

C. MacRobbie

Fig. 5.5 Internal affairs. #Jamie McCartney 2012, reproduced with permission from the artist

clitoris, the urinary meatus, the vaginal opening and the paired corpus spongiosum erectile tissue. This clitoral body projects into the fat of the mons pubis (O’Connell and DeLancey 2005; p: 173). Contemporary sculptor Jamie McCartney has produced a series of casts taken from the internal vagina displaying what is typically unseen. Constructed as one continuous cast, the outside situates the otherwise amorphous shape in context. The interior object makes the unseen visible; his sculpture ‘Internal Affairs’, shown in Fig. 5.5, uses glass to represent the potential of the internal space, rather than conveying only the solid structures. The vagina tangibly as well as conceptually expands our view of the narrow tubular representations often viewed in diagrams and outspreads to accommodate the casting material (alginate). This method of production has informed the project, in construction, conception and display. Observed are the depression of the cervix OS, the urinary meatus and anus and the texture and folds in the epithelium and the Grafenberg mass. McCartney is also responsible for the artwork ‘The Great Wall of Vagina’ and describes his work as curious, educational, sensational and beautiful (McCartney 2012). Conversely, the Gynaeplaque model in the Science Museum Group Collection presents the external openings and hidden structures of the reproductive system contained within a black leather box/carry case, a spongey rubber model that opens in two halves to show a section through the female reproductive organs. The

Wellcome Museum ‘Institute of Sexology’ exhibition featured this model in an attempt to open up access to an understanding of the female body. This type of model was used in the c. 1930s in instructing women at the American Birth Control Clinical Research Bureau in the insertion of the cervical cap as a method of contraceptive, although the maker is unknown (Science Museum Group 2007). This public display of antiquated medical models in a contemporary gallery environment highlights their relevance for the public interest. Medical instruments and models provide the viewer with contextual information that relates the body to the clinical procedures the body may undergo and to antiquated procedures that are no longer practiced. It is important to ensure that the body of women is described with agency, an active voice and with clarity to ensure that women’s bodies are viewed in a modern museum context by themselves and others with the power they deserve. Visualisation of the organs and structures of the pelvic cavity, surrounding musculature and neurovascular structures should be presented to reveal organisation and relationships between structures and develop spatial perception; there should be foregrounding of the major organs involved in reproduction. Understanding the normal anatomy of the uterus can assist with the appreciation of the changes in its development, in pregnancy and after the menopause and to recognise an anomaly of development or a normal anatomical variation. Further changes take place

5

The Embryonic re-Development of an Anatomy Museum

89

Fig. 5.6 Cross section of the anatomy of the female pelvis. Digital illustration. Catherine MacRobbie 2021

in the uterus during the menstrual cycle and with events such as pregnancy and pelvic surgery such as hysterectomy. Complications such as pelvic inflammatory disease, fibroids, endometriosis or a pelvic tumour can often be symptomless, and accurate clinical information would be needed to support visualisations on these topics. A possible mode of illustration for outlining the reproductive system is indicated in Fig. 5.6, in a draft artwork proposed for inclusion in this visualisation project. This colourful and near diagrammatic illustration suits visualising for a modern museum with a public audience and, juxtaposed with the historical material and the original Rymsdyk illustrations, could be well placed to demonstrate the variety of methods of visualising the same topics for different

audiences. This could support the new functions of museums—which focus on culture rather than primarily on teaching—that apply humanities and museology techniques to reach out to new, twenty-first-century audiences with different intentions and objectives within the shared culture of anatomical knowledge.

5.2

Visualising Embryos

Developing embryos were first illustrated during the eighteenth-century Enlightenment. Our understanding of reproduction has been enhanced through modern medicine, image production, visualisation, imaging technology and models. Diagrams from the late medieval/early-modern

90

period incorrectly depict a homunculus in the head of a sperm, breeding dispute between ovists, who believe that the complete embryo is contained and preformed within the ovum, and the spermists, who believed that the complete human being was contained in the sperm. Eventually, the commonly accepted explanation of reproduction became recognised as the growth and development and ensuing differentiation of specialised tissues and organs, as well as the maintenance of an increasingly complex organisation of structures. This gave rise to embryology as the study of human development. This understanding is often contextualised with images of embryos and the developing foetus as the dominant representations of pregnancy (Buklijas and Hopwood 2008–2010). Towards the end of the Middle Ages, anatomists were beginning to portray bodies increasingly realistically, allowing for more accurate illustrations of

C. MacRobbie

embryonic development, such as Samuel Thomas von Soemmerring’s representations ‘Icones embryonum humanorum’ – shown in Figs. 5.7 and 5.8—which were published by Varrentrapp u. Wenner, Frankfurt a. M. in 1799 and later published in Geburtshülflicher Atlas in 48 Tafeln und erklärendem 1835–1844 (Wellcome Collection). The viable pregnancy became increasingly understood, and consequently, the progress of the embryo in both size and emerging form came to be recognised, showing pregnancy as a developmental process with a known duration, its stages and progress evaluated by clinical examination. Embryos were being visualised more realistically following changing attitudes towards science around 1800. Anatomists began creating complete and accurate drawings and models and visualising the changing form in its developmental stages. Romanticism inspired connections

Fig. 5.7 Geburtshülflicher Atlas in 48 Tafeln und erklärendem Texte [Hermann Friedrich Kilian]. Credit: Wellcome Collection. Attribution 4.0 International (CC BY 4.0)

5

The Embryonic re-Development of an Anatomy Museum

91

Fig. 5.8 Geburtshülflicher Atlas in 48 Tafeln und erklärendem Texte [Hermann Friedrich Kilian]. Credit: Wellcome Collection. Attribution 4.0 International (CC BY 4.0)

between histories of natural sciences and human development which in turn inspired important embryological work. Microscopists now analysed the internal environments, and Darwinism made space for the successes of embryology. This facilitated collaborations between biologists and publishers, artists and historians and the media benefitted from the flourishing print industry and prominence of museum exhibitions. In the early 1900s, obstetricians assumed more power in an attempt to reduce further the maternal mortality rate, and they eliminated the distinction between ‘normal’ and ‘abnormal’ births. Every pregnancy became potentially pathological and so became increasingly medicalised. A method of illustrating the comparable relationship between embryos across various species was developed in 1874 by Ernst Haeckel (1834–1919). These controversial comparative plates were extensively criticised, and he was

accused of speculation and deception. Meanwhile, fellow embryologist Wilhelm His (1831–1904) insisted that human embryology should focus on human embryos, which restricted the research to anatomical experts. He produced a standard developmental series, or normal plate, in 1855. This provided a framework for further research of normal stages, related normal plates and tables that define standard divisions of developmental biology and produce methods for assigning an embryo to a stage within a staging system. Using a large microtome that allowed a series of slices in sections and stained-glass slides to be examined under the microscope, he was able to prepare drawings that illustrated normal development activating a new understanding of embryology (Hopwood 2009). This revolutionised human embryology and established a standard approach by the early 1900s. His’s collected embryos and foetuses derived from miscarriage

92

and prenatal deaths, in a similar manner to William Hunter, through private medical practice. Today developmental biologists work to improve and extend stage series and to take individual variation into account (Hopwood 2007). In 1914, the Johns Hopkins University of Washington Department of Embryology at the Carnegie Institute gave birth to a network of human embryologists who initiated the collecting and classifying embryos and were instrumental in the conception of the most extensive system of embryo categorisation. These influential scientists collected thousands of embryos from which an established classification system of human development was born. Franklin P. Mall (1862–1917) and Wilhelm His prepared and preserved serial sections of the embryos and collaborated with artists to make hundreds of three-dimensional models at different stages of growth (Buettner 2007). The embryological models produced using the collections are significant, and they will be consulted during the development of the Hunterian Museum of Anatomy exhibition and discussed in more detail within this chapter. These embryos continue to shape our view of development in textbooks and as digital animations. The Carnegie stages are based on the morphological development of the embryo. Criteria can also include ranges of age in days, number of somites present, and embryonic crown-rump lengths. The Carnegie system marks the division of human development into embryonic and foetal periods. The human embryonic period is divided into 23 Carnegie stages taking place in the first 8 weeks post-ovulation (Gestational Age week 10). The system proposes that 90% of identifiably human structures are most probably present by stage 23. The beginning of the foetal period is indicated by bone marrow having replaced the humerus cartilage. The Carnegie embryo collection is still accessible for use by researchers, and this system still remains the international standard by which human embryos are described and classified. Carnegie Stage Table (Hill 2021).

C. MacRobbie

2

Days (approx) 1 (week 1) 2–3

3

4–5

4 5

5–6 7–12 (week 2) 13–15

Stage 1

6 7

15–17 (week 3)

Events Fertilised oocyte, zygote, pronuclei Morula cell division with reduction in cytoplasmic volume, blastocyst formation of inner and outer cell mass Loss of zona pellucida, free blastocyst Attaching blastocyst Implantation Extraembryonic mesoderm, primitive streak, gastrulation Gastrulation, notochordal process

Weeks shown in the table are embryonic post-ovulation age; for clinical gestational age (GA) measured from the last menstrual period, add 2 weeks

Figures 5.9 and 5.10 visualise the early Carnegie stages and do so in two formats: the first indicates the diagrammatic approach to illustrating and the second shows a more conceptual or representational approach. Each image provides information on the series of events taking place in embryo development and appeals to different levels of engagement with the material. The preliminary digital sketches of embryo development weeks 1–4 shown in Figs. 5.9 and 5.10 are produced using different techniques and software and different approaches to illustrating the same subject. Applying different image styles or visualisation techniques can enhance or reduce the viewer’s understanding of a subject. Clarifying images for the purpose of explanation can be a useful tool when visualising for the purposes of education, and the diagrammatic illustrations in Fig. 5.9 utilise symbols to guide the viewer through a series of clearly coded visuals in sequence to extract information about this usually unseen minuscular process. Including visualisations such as those represented by Fig. 5.10, which do more than simply explain, clarify, illuminate and visually represent concepts, can invite the viewer to engage their

5

The Embryonic re-Development of an Anatomy Museum

93

Fig. 5.9 Weeks 1–4. Digital illustration. Catherine MacRobbie 2021

imagination and infer metaphor and attach meaning. These visualisations and the approach taken in each will be reviewed with the team of staff involved in the museum exhibition and developed further based on the critique by embryology specialists. While the science of developmental biology was evolving during the 1800s, the need grew for models and illustrations to demonstrate the intricate details of embryological development. Professional model makers were employed to create specialist models that were used to aid standard anatomy teaching and would allow visualisation of the biological process of embryology in three dimensions. The Hunterian Museum of Anatomy at Glasgow displays and utilises some such models for teaching, and it would be hoped that these models could be included to reinforce the narrative and

descriptions of the process as part of the planned museum exhibition. Sculptors such as Adolf Ziegler (1820–1889) and his son Friedrich set out to fulfil the demand for these models. Adolf Ziegler began crafting hundreds of anatomical models in the 1850s and worked in collaboration with leading embryologists to produce models of embryos (Hopwood 2002), signifying a change in how embryos were viewed and purposed. The embryologists would provide drawings, and the Zieglers produced anatomically correct wax models. Ziegler and Embryologist Alexander Ecker dissected, illustrated and made models of embryos by hand. Ziegler teaching models became the authoritative standard. Figure 5.11 illustrates a stage seven Ziegler human embryological model produced by Friedrich Ziegler. Friedrich joined the studio and begun using Gustav Jacob Born’s modelling technique in the

Fig. 5.10 Weeks 1
4. Digital illustration. Catherine MacRobbie 2021

94

C. MacRobbie

Fig. 5.11 Ziegler, Friedrich, 1860–1936, “Stage seven Ziegler human embryological model,” OnView: Digital Collections & Exhibits, available at: https:// collections.countway. harvard.edu/onview/items/ show/5878. Accessed on March 11, 2021

1880s to produce a series of human embryo plate models (Wellner 2009a, b). Sections of the embryo are projected from the microscope and enlarged onto sheets of wax and then stacked to produce a 3D layered model. Demand for models also fell in line with the rise of photography, and in 1936, the Ziegler modelling studio ceased (Irving 2020). In the early 1900s, renowned embryo modeler O. Heard built upon Ziegler’s techniques and produced more detailed models than their earlier embryo models (Hopwood 2009). Heard created over 700 wax models over a 42-year period at the Carnegie Institute, mainly using lost-wax casting, which is produced by cutting the structures of interest out of wax plates corresponding to highly magnified serial sections, stacked to produce a

mould for a plastic solution, and then removing the wax when it had set hard (Miyazaki et al. 2017). The decline of model making took hold following the success of the microtome. Examples of contemporary embryological medical models produced within arts and culture can be seen in Damien Hirst’s The Miraculous Journey (2005–2013), installed at Sidra Medicine in Doha, Qatar women’s and children’s health clinic, in which 14 bronze sculptures chronicle the gestation of a human foetus. Each sculpture shows a different stage of the foetus’ development, and the work was based on Hirst’s drawings of prenatal and natal development from egg and sperm to foetus to newborn (Vogel 2013).

5

The Embryonic re-Development of an Anatomy Museum

The success of the microscopic examination of bodies rendered cells as a source of medical knowledge to be analysed exclusively by the medical gaze (Lie 2012). Contemporary medical imaging technologies enable the production of images of human cells that are magnified to make them visible to the eye. These images have now moved out of Lab, and they have become available for the public. As such, these images influence the cultural perception of reproductive techniques, conception and early stages of embryonic development and the foetus during later stages of development. Assisted reproductive technologies are becoming more integrated into contemporary practices of human reproduction, and medical imaging techniques have been of vital importance to research, which has resulted in new fertility techniques. Understanding the three-dimensionality of the developing foetus requires an ability to interpret the educational resources on embryology and the series of folding movements. By including interpretations of the complex formation of embryos, the artwork can also act as a teaching resource for anatomy and medical students, increasing the value of the visual material produced to support the collections and the potential

95

for its integration into the curriculum. Conceptually challenging and complex, embryology is reserved for individuals with the enthusiasm to comprehend the fundamentals of how our biological structure is formed. The flat trilaminar embryonic disk becomes more cylindrical because of the longitudinal and transverse folding that results from embryonic growth, particularly of the neural tube. The folding occurs simultaneously, not sequentially. In a process of curving, the embryo is transformed into a tube attached to the umbilical cord. The embryo increases rapidly in its long axis due to central growth being greater than peripheral growth, resulting in the embryo curving itself around the umbilical region (Pansky n.d.). Figure 5.12 illustrates the stages of folding movements in embryo development; visualising the staging of embryos and their appearance at these stages supports scientific, individual and cultural perceptions of the growing foetus. The discovery of x-ray in 1895 provided an alternative means of viewing inside the human body and offered a better way to illustrate and understand the structure and functions of the hidden body and to observe the living foetus. The incredible medical imaging technologies

Fig. 5.12 Sketches of folding movements in embryology. Digital illustration. Catherine MacRobbie 2021

96

such as CT, MRI and 3D printing produce an image that allows internal structures to be seen without disturbing the developing foetus or specimen. Imagery has assisted in the formulation and dissemination of delivery techniques. Ultrasound images enable the observation of early and advancing pregnancy, and they are used by obstetricians for diagnostic purposes, as well as for understanding the intrauterine environment and other impacts on foetal health (McIntosh 2020). These technologies are also contributing to the development of new methods of medical illustration. Obstetric ultrasound has been established as a safe imaging mechanism for diagnosing, monitoring and observing the stages of pregnancy, and sonography in obstetrics and gynaecology dates from the first clinical use of ultrasound in Glasgow in 1956 (Campbell 2013: 213–29). In 2021, images of cells, embryos and the female reproductive system are consumed expansively, often being utilised to inform patients about their health and fertility treatments and in advertising for menstrual management products and even on phone apps. Contemporary clinicians exercise the now increased limits of current knowledge to produce positive outcomes in assisted fertility treatments. Pregnant women view images of their developing foetus at crucial stages during their pregnancy. Retaining keepsake scans and 3D prints are now commonplace in some private clinics; with the rise of social media, scan keepsakes are being utilised in artworks and pregnancy announcements. The general public has access to view medical animations showing embryo development and stages of pregnancy, as well as interactive virtual reality experiences and human life cycle immersive museum exhibits such as ‘IVF: 6 Million Babies Later’ in the Science Museum in London. Understanding what might go wrong in pregnancy and developing interventions to help will ultimately improve the health of society. Significant complications in pregnancy such as preterm birth, pre-eclampsia and foetal growth limitation are major causes of mortality in the early years as well as life-long morbidity. Research in

C. MacRobbie

reproductive health and physiology, pregnancy and birth, as well as stem cell biology, has facilitated collaborations between scientists and clinicians to translate research to improve outcomes for women and their children. There is an emerging cultural awareness of the possibilities of reproduction, expanding previous limits of human biology such as infertility. New imaging technologies, particularly ultrasound technology in pregnancy, have contributed to changing the cultural perception of the foetus as integral to a woman’s body into a separate individual. Medical imaging technologies provide new material for cultural interpretations of human bodies and can transform human cells into astonishing, aesthetically appealing and independently identifiable objects. What information is conveyed in these new images and does visualisation contribute to changing relationships with reproduction? The images created represent the potential contained with the cells or are used to illustrate politicised health. For example, the legalisation of abortion prominently positioned images of embryos as diverse. The UK Abortion Act 1967 provided a safe and legal option for ending a pregnancy in England, Scotland and Wales up to 23 weeks and 6 days of pregnancy. Public perception of embryonic development was changed by Lennart Nilsson’s photography of abortus/foetus material, and these images attracted significant attention when used by antiabortion activists and counter-movements (Lie 2012). Nilsson’s photos created a radical shift in the cultural discourse of the foetus as detached from the pregnant woman, and the foetus becomes recognised as an individual being. Figures 5.13 and 5.14 represent dualistic methods of visualising abortus material for differing audiences and with different purposes, but both are shown in estrangement from the mother. This notion of the developing foetus as separate from the mother is further reinforced with the development of assisted reproductive technologies that allowed for the first ‘test-tube baby’ to be born via in vitro fertilisation (IVF) in Manchester, UK, in July 1978. Now over 8 million babies have been conceived via IVF, and the number of

5

The Embryonic re-Development of an Anatomy Museum

97

Fig. 5.13 Graphic illustrations of abortion and the diseases of menstruation/consisting of 12 plates from drawings engraved on stone, and coloured by Mr. J. Perry, and two copperplates from the Philosophical Transactions, coloured by the same artist. The whole representing 45 specimens of aborted ova and adventitious productions of the uterus, with preliminary observations, explanations of the figures, and remarks, anatomical physiological. By A. B. Granville. Credit: Wellcome Collection. Attribution 4.0 International (CC BY 4.0)

couples requiring assisted reproductive technology is steadily increasing (Science Museum 2018). Visualisations of reproductive medicine and developmental biology have traditionally been restricted to reflect the scientific lens guiding the narrative. As a result, accessible, educational and sensitive visualisations that reflect the diversity of womanhood and female experience have been restricted to experts and have not always been made available in public museum contexts. The planned exhibition at the Hunterian Museum of Anatomy and the visualisations therein aim to continue broadening public access to reproductive science. Insufficient communication in healthcare can be linked to this lack of accessibility in visualisation. There is noticeably poor quality of some health information and support for patients. Information given to patients is important, along with support to use that information effectively

and for it to be converted into knowledge and understanding. The evidence shows that the quality of health information has an impact on patients’ experience of healthcare and patients’ health behaviour and status (McCartney 2013). When trying to explain complex ideas to non-experts, simplifying the information may not be the most effective strategy. Public health translates scientific evidence for public communication, and successful communication acknowledges how difficult it can be for laypeople to understand terminology or data. As a result, it includes rather than selects information to convey an accurate understanding of the findings. Science literacy of the general public is qualitatively different from that of the expert, and the translation is as much about creating meaning as it is about relaying information. The goal is training scientists to communicate plainly to help the public understand a complex scientific concept clearly. With politicised subjects, our ability to

98

C. MacRobbie

Fig. 5.14 Normal pregnancy. Credit: Michael Frank. Attribution 4.0 International (CC BY 4.0)

understand is often overwhelmed by our inability to hear. Regardless of how clear the communication or scientifically literate the audience, if the information poses any kind of threat to an individual or collective identity, then communication will break down (Saffran 2017). The media has an influence over how women engage with childbirth, often depicting birth as particularly traumatic and painful. Luce suggests that the media has a negative effect on first-time pregnant women and childbirth in society, and the media’s portrayal of birth as dramatic may bolster the medicalisation of birthing practice. There is a perceived lack of description of a normal childbirth experience even in today’s media, and it has been suggested that improvement is needed in the

representation of childbirth in the public domain (Luce et al. 2016). Institutionalised birth practise can create an imbalanced perception of risk and safety, suggesting that medical settings can be detrimental to standard birth practices and outcomes (Newnham et al. 2017). It is important to understand the influence and relationship between the body, power and technology and the social constructs of illness and health and the medicalisation of health when communicating with public audiences. We must consider the impact of social health outcomes and health outcomes based on disparity of gender, socioeconomic status, race and ethnicity when imaging women’s experiences of healthcare and childbirth.

5

The Embryonic re-Development of an Anatomy Museum

5.3

Visualising Discourse around Menstruation

Women and girls receive surprisingly little information about the experience of having a uterus and about what is normal to expect. It is hugely important for women to be able to access information and services and to be able to increase their awareness and knowledge of medical conditions and reproductive disorders, dysmenorrhea and unresolved pelvic pain, as well as for them to be able to prepare themselves for cervical smears, internal transvaginal ultrasound and clinical events such as miscarriage and pregnancy loss. By addressing health education and reproductive health topics with sensitivity, the planned exhibition at the Hunterian Museum of Anatomy hopes that museum visitors would be better able to understand the biology of menstruation and the menstrual cycle and to integrate this knowledge with their cultural experience. The female reproductive system and menstrual cycle literacy is a key part of work in this area. A visualisation of the menstrual cycle containing schematic representations from a variety of different perspectives is presented in Fig. 5.15. Understanding more about how our bodies work is essential for our physical and emotional health, for potential or future fertility and for reasons of contraception as well as appropriate individual menstrual management strategies. It is important to understand one’s own cycle and to recognise problems when they arise. Many cultural practices impact women and girls across the world, and education can be used to empower. Access to concrete tools and dialogue can minimise the shame and embarrassment associated with menstruation (Venkatesan 2019). Worldwide, girls, women, transgender and intersex people experience the stigma of menstruation through discrimination as well as experiencing the indignity of period poverty. This discussion recognises recent moves in Scotland for equality, dignity and rights for those who menstruate in commitment to gender equality in the provision of free sanitary products

99

(BRIA. 2018). Sex education in schools has only recently evolved to include menstrual health and the menstrual cycle, LGBT+ inclusive relationships education, relationships and sex education in the health education curriculum (UK Gov 2019). Almost one in four women do not understand their menstrual cycle, a third of girls are not told about periods by their parents, and 10% receive no preparation before their first period. A lack of awareness of what a normal period looks like could result in delayed diagnosis of medical conditions such as endometriosis, which affects 1 in 10 women in the UK (Hey Girls 2019). When creating visual content on these subjects, it is important to recognise that not everyone who gets their period is a woman and not every woman gets their period. Persistent gendered messages and visualisations can be challenging and even harmful. It is hugely important to include the marginalised voices of trans menstruators whose experience has been left out of the conversation and to promote sensitive and inclusive communication that deeply impacts a person’s body, regardless of whether or not that person is a woman (Rydström 2020). The discourse around menstruation, visual culture and medical practice draws from a diverse range of voices and practices in a socio-cultural context and scientific framework. Visualising the norm only provides a norm with a voice. The universalised understanding is menstrual normativity and conformity rather than embracing plurality and diversity. This creates broader rejection of women who don’t menstruate or no longer menstruate. There is an important impulse to celebrate and represent a challenge that is rooted in diversity and embraces issues not represented in the broader normal culture. Women’s bodies have the improbable reproductive capacity and to bleed for days without dying. Women’s bodies can act as a mechanical metaphor with success or failure in the process/goal of pregnancy. The conversation often excludes those who do not have this goal, which leads to exclusion and status value judgements. Pervasive approaches spread misinformation and often fail

100

C. MacRobbie

Fig. 5.15 Menstrual cycle and ovarian follicle stimulation. Digital illustration. Catherine MacRobbie 2021

to address the social determinates of health and singular experiences. The Vagina Museum in London acts as an exemplar for the exhibition of female reproductive health: their museum mission is to spread knowledge and raise awareness of the gynaecological anatomy and health, to give people confidence to talk about issues surrounding these areas, to erase the stigma around the body and gynaecological anatomy and to act as a forum for feminism, women’s rights, the LGBT+ community and the intersex community, and for challenging heteronormative and cis normative

behaviour and promoting intersectional, feminist and trans-inclusive values (Vagina Museum n.d.). Information that educates everyone about periods is still needed, with wider only recent representation in period advertising that contains words such as ‘vagina’, ‘bleeding’ or ‘stain’. This erasure is rooted in a cultural legacy, treating bleeding as if it’s something to be ashamed of (Parker 2018). Period poverty impacts our choices and further biases the notion of shame associated with bleeding (Tull 2019). The wider understanding of what women experience, feel and think to understand the emotional complexity

5

The Embryonic re-Development of an Anatomy Museum

of having a uterus at different stages in their lives, from getting their first period to the menopause and various stages in between, is often underrepresented. Normalising the diversity of women’s experience of sanitary products, menstruation, sex, contraception, infertility, miscarriage, abortion, menopause, conditions such as endometriosis, LGBTQI+ relationships, pleasure, pain, emotional distress, relief, first periods, fertility treatment, hysterectomy and childbirth is essential. Problems arise in the assumptions that women’s reactions to gynaecological experiences are uniform, despite differing social factors such as age and menopausal status, desire for (more) children, significant emotional and sexual relationships, and individual experience of gender identity (Elson 2005). Queer experiences of conception, pregnancy, birth and parenting are under-recorded, underresearched, and under-heard. Research on pregnancy continues to be centred within a heteronormative framework and fails to address the invisibility of lesbian, gay, bisexual, transgender and queer (LGBTQ) people (Charter et al. 2018). In the UK, data from fertility clinics (HFEA, 2019) and birth registrations (via the Office of National Statistics) identify that lesbian couples are one of the fastest-growing groups within maternity services, with fertility treatment and live births increasing by 15–20% in this group, every year for the last 10 years. The enduring heteronormative approach risks conflating gender and role. There is not yet a developed shared language in research or practice to adequately describe reproductive histories outside of cis mothers’ histories. The research does not adequately address the diversity of families and any circumstances where the gender of at least one of the parents differs from societal expectations. My visualisations for the museum project hope to expand on the experiences and constructions of pregnancy and parenthood in its illustrations (Darwin and Greenfield 2019). It is important to keep public health initiatives such as adherence to smear testing and barriers to informed uptake in the wider discussion about female reproductive health in mind when creating visual health information. Visualising women’s

101

lived experiences and reflecting contemporary attitudes towards reproductive health can support viewers to better understand the process and eliminate a certain amount of fear or apprehension around gynaecological procedures. As uptake of cervical screening continues to decline, the variation in individuals’ understanding and perceptions of cervical screening suggests that there is a need for more support for women in their decision to be tested (Chorley et al. 2017;26 (2):161–172). Fear of the speculum and feelings of vulnerability during the gynaecological exams are two of the biggest barriers to cervical cancer screening (Asiedu et al. 2020). Sensitivity issues around clinical events in pregnancy and childbirth should be addressed in the exhibition that is being planned. Individual experiences of infertility can differ quite substantially. The damage to an individual’s self-concept caused by infertility can be significant. Changing societal attitudes towards gender, sexuality and reproduction are changing to ensure that women experiencing infertility are no longer stigmatised and traumatised. Communication of these issues can be exceptionally difficult (Clarke et al. 2008). It is important to recognise the loss of life of the mother and child that occurred to create the historical collections within the Museum of Anatomy, and it is rare that a viewer remains unaffected in their viewing of sensitive foetal material. Clinical miscarriage is both a common and distressing complication of early pregnancy (Larsen et al. 2013). Around one in eight pregnancies, where the person knows they were pregnant, ended in miscarriage, more before a woman was aware that she had become pregnant. Early miscarriage is defined as pregnancy loss before 12 weeks and occurs in one in five pregnancies. Recurrent miscarriage losing three or more pregnancies in a row affects around 1 in 100 women (NHS n.d.). Miscarriage is the most common complication of pregnancy with one in four women experiencing at least one miscarriage during their reproductive lifetime, meaning a quarter of all families are affected by baby loss (Tommy’s. 2020). There is a cultural silence surrounding miscarriage and a woman’s relationship to reproduction and the meanings and

102

materials of miscarriage (Withycombe 2018 p. 220). In the time before medical imaging and prenatal care, physicians were unaware of the properties and causes of miscarriage. With the growing self-regulation of fertility, the dominant narrative remains that miscarriage is a loss. The Hunterian collections were made possible because of the increased physician involvement in birthing practice. As the study of embryology developed, so too did the scientific interest in pregnancy loss and physicians sought to consensually retain every miscarried foetus. Physicians were interested in what the products of miscarriage could reveal about the process of pregnancy and causes of miscarriage. Physicians and surgeons have supported women to overcome difficult pregnancies for centuries, but routine childbirth remained at home under the supervision of midwives until the mid-1900s. Midwifery, historically female lead, in the early nineteenth century increasingly involved men in childbirth practices. Doctors became increasingly involved in live birth and miscarriage experiences. Quite simply, ‘Anything and everything that came out of a woman’s vagina was important for understanding the science of human anatomy’ (Stolz 1866). The foetal specimens from miscarriage cases were studied, preserved and shared. As many nineteenth-century women and physicians understood, grieving a miscarriage can embody a wide range of meanings: losing an opportunity, an idea, a fear, or locus of anxiety. The safe diagnosis of miscarriage has only recently been understood. New criteria have been established to diagnose miscarriage and improve prediction of pregnancy viability and adverse outcomes, as well as the mechanisms underlying miscarriage (Bourne n.d.). Contemporary reproductive literature, medical texts, images and public health initiatives view these experiences differently, and they are working towards becoming more perceptive or representative of women’s experiences. The planned exhibition at the Museum of Anatomy aims to contribute to this trend in an accessible museum context.

C. MacRobbie

5.4

The Gendered Body and the Lack of Diverse Representation in Gynaecological Images

Medical images play a significant role in how we are conditioned to view bodies. When particular kinds of bodies become established as normal, others are marked as abnormal and excluded. The creation of medical images then carries the potential to harm the lives of those who fall outside those norms, as social bias can be reflected and amplified by images we choose to create. This in turn reinforces normative relationships and representations in medical illustration (Smith 2019). How the female body is visualised reflects how the society understands the purposes and functions of the human body as well as the role of women in the society. By considering the social and political framework that influences these illustrations of the female body, we learn how political gendering affects the choices of contemporary medical illustrators. Illustrations in medical texts can be assumed to be scientific, objective and neutral, in fact revealing explicitly gendered roles (Brasseur and Thompson Torri 1995). The rise of the bourgeois in the eleventh century reinforced and internalised the medical discipline of a closed and clean body by using it as a ritual of demarcation: while the body of the underclass and of women is leaky, flowing and unproductive, the male bourgeois body of the Homoeconomicus is divided into closed economic units and therefore bodies are arranged into a hierarchy positioning the male body as superior to the female body (Bakhtin 1984). The implications of representing women as ‘others’ result in communication that obscures, excludes and conceals information. Cultural and social imperatives affect what is illustrated or written, how a body or instrument is shown and what text accompanies it, revealing important information about how it is used to create otherness (Bakhtin 1984). The anatomy depicted by Rymsdyk in 1774 was reflective of an exclusively male practice by male artists as anatomy teaching was restricted to

5

The Embryonic re-Development of an Anatomy Museum

only male students. Medical illustration is now a primarily female-led profession. In the early 1940s, women practitioners were able to partially circumvent gendered norms through the unconservative efforts of the women involved, but this inclusion was still limited mainly to White and upper-middle-class women, and many earlier achievements often remain largely unrecognised (Levine 2019). The work attracted women who were interested in both anatomy and art despite institutional and societal barriers to success for women at the time, and women continue to dominate the field of medical illustration today. These women have helped to change the way the body was understood by medical professionals all over the world. This profession embodies this ability to take in very complex information and translate that into a form that other people can understand. Over time, the field of medical illustration has changed and expanded to include research, new technologies and a deeper exploration of the body from a cellular level largely in part due to these strong visual communicators and visual problem solvers (Levine 2019). The imagery in traditional medical literature affirms the lack of female representation in anatomical texts and atlases, but even more notable is the distinct lack of non-White bodies. These textbooks present a normalised human body that reduces variation for the purposes of standardised teaching. The ideological narrative in anatomical illustration centres the Caucasian male as the normative universal body and to the consequent subordination of all other bodies. Contemporary anatomical literature reveals that a recent understanding of race as socially constructed rather than biologically determined has not changed the fact that illustrations of the universal Caucasian male are still favoured overwhelmingly to the exclusion of all other bodies. The continual exclusion of non-White bodies in anatomical literature is an inherently racist practice. Deliberate inclusion is therefore required of the author, artist and publisher in an attempt to dismantle the concept of the ‘universal body’. An anti-racist rendering of the human body in the biomedical sciences will hopefully help move bodies traditionally

103

relegated to the margins to the centre of the scholarship (Cober 2015). There is evidence that the narrative has been misrepresented in the historical record that questions the impact historical practices were having on the care given to women from Black and Asian backgrounds. The medical ethics of Dr. J Marion Sims, an Alabama surgeon, has been associated with the manipulation of the institution of slavery to perform unethical experiments on non-consenting primarily Black women between 1845 and 1849 (Wall 2006). While Sims is recognised for his work on Vaginal Fistula and the invention of the speculum, history did not record the voices of the enslaved women he performed his experimental gynaecological techniques on. Systemic racism permeates every level of society, including maternity care. Black women have a 5 times higher maternal mortality rate in the UK. In 2020, being a White woman is 5 times safer than a Black woman (MBRRACE-UK 2019). Disparity exists in maternal and infant birth outcomes of Black, Asian and minority ethnic women giving birth in the UK, compared to the majority, requiring maternity healthcare providers to modify their current services to include ‘culturally competent’ service provision and meeting the diverse needs of evolving demographic profiles (Garcia et al. 2015). A recent review of maternal morbidity by Oxford University in 2019 found that Black women are five times more likely to die in pregnancy, childbirth or in the postpartum period, and Asian women were also twice as likely to die compared to White women. It was previously reported that the death rate could be explained by pre-existing conditions amongst Black women such as high blood pressure, or the higher prevalence of complications such as pre-eclampsia (Okereke et al. 2019). Underlying prejudice among midwives is a crucial factor in the deaths of Black mothers. The ‘My Midwives Initiative’ is engaged in restorative methods to raise awareness amongst midwives, address disparities in care, and explore current understanding of racism, prejudice and the lived experiences of midwives and students and how this may impact the care (My Midwives UK 2020). There is a recognised

104

disproportionately high mortality rate for Black babies. Although Black babies are three times as likely to die as White babies, when Black babies are delivered by Black doctors, their mortality rate was more than halved. Improvements are noted when hospitals deliver more Black babies. The disparity in care requires further scrutiny to generate reproductive health equity, furthering diversity initiatives, and re-evaluating organisational routines. Cumulative experience of racism and sexism, throughout the lifetime, can prompt a chain of biological processes that undermine Black women’s physical and mental health. The health of babies can be improved by addressing the needs of Black mothers (Greenwood et al. 2019). It is important to consider normativity and diversity in healthcare imagery as well as the demographic of the content creators. This includes the social inequalities developing from the dominance of able-bodied, thin, young, cisgender White male bodies as standard in medical visualisations (Belsky 2019). Medical artists have to make a commitment to reflect the culture in which the images were created. Increasing diversity and inclusion in the field could directly result in a diversification of perspectives and of representations in medical illustration. Criticism of medical images of women’s bodies, despite the prevalence of female medical illustrators, recognises the focus and power in medical history on the role of men—men who established their body as the standard from which all others differ (Belsky 2019). The persistence of standardised images and lack of representational diversity in medical images play a key role in how medical professionals and public learn to see the human body. ‘It’s rare to see women and people of colour in medical textbooks’ (Parker et al. 2017 p. 106–113). As Figure 5.16 demonstrates black and non-White skin tones contain rich colours and a variety of undertones and concentrations of pigmentations that are often overlooked or underused by illustrators unfamiliar with illustrating the vibrancy of black skin. There is a recognised absence of women in almost all medical textbooks, which are heavily biased towards depicting male bodies. The

C. MacRobbie

Fig. 5.16 Pregnant woman. Digital illustration. Catherine MacRobbie 2021

5

The Embryonic re-Development of an Anatomy Museum

artwork may not intentionally discriminate, but while artists continue to create representations in their own image, the result will always be biased (Eveleth 2019a, b); when it comes to medical illustration around reproduction, men outnumber women in textbooks 2.5 to 1 (Lawrence and Bendixen 1992 p 925–34). Where the body could be depicted as either male or female, it was also found that bodies were noticeably white, slim, and young. Of the images of women, 86% of them were white (compared to 76% of identifiably male bodies). Male bodies are almost always muscular, while female bodies are drawn thin. The absence of diversity in the images was noted as significant. Visibly disabled bodies were shown in 2.7% of textbook images analysed. Elderly patients were seen in only 2.2% of images. Whether there is implicit or suggestive attitudes of bias, there can be a dangerous impact on medical images. Medical bias has real, negative impacts on patients, which reduces medical help-seeking behaviour. Research and scholarship on weight-based stigma reveals the disproportionate degree of bias experienced by overweight women (Fikkan, J.L. 575–592 (Fikkan and Rothblum 2012). Older people are rarely illustrated with diseases that typically affect their age group; diverse skin tones are rarely depicted, where there should be a variation in clinical presentation illustrated to aid with diagnosis. Diversity in illustration should reflect our rich society. Every kind of person should be illustrated. The gender disparity is slowly changing. But the intersections with ethnicity and body type are getting worse. We must approach illustration by thinking about how to diversify the images we create (Eveleth 2019a, b). There are limited examples of inclusive design in the period products market. A menstrual blood management product has been specifically designed to allow a higher functionality for users with disabilities. This menstrual cup was an example of a successful inclusive product design that was produced by a diverse set of designers. Including people with varied ability and disabled people within a design process may generate a different workflow than with a fully able-bodied process (Eveleth 2019a, b). The success of design for all

105

encourages a change in the process to allow everyone to thrive.

5.5

The Role of the Illustrator

The current use of the Hunterian Anatomy Museum space and collection is as an educational and research resource for undergraduate and postgraduate students and for teaching staff at the University of Glasgow Anatomy Facility, Medical School and Dental School. Some of the collection is in storage, which requires requests for access to specific objects. The collection is a valuable asset. The planned exhibition, which is the inspiration for this chapter, aims to support the promotion and protection of the Museum as a valuable space of learning and to enable wider community access to the collection, in line with the aims of the School of Life Sciences, University of Glasgow and Hunterian Museums. A considered, well-designed and innovative anatomical museum has a vital role in the education of its visitors. Discoveries in the history of medicine, particularly the field of anatomy, can be used to reveal more about the historical specimens. The specimens can provide information about what was important to the anatomist and about the physical health of the person the material was removed from, as well as about the culture in which the specimen was created. The effectiveness of a museum is predicated on the value that viewing an object enhances recollection and retention far better than a written or a verbal description. The planned exhibition aims to be faithful to the sense of antiquity inherent in the Museum and will be designed in such a way that it is sympathetic to the historical themes and resources and the sensitive nature of the museum content whilst improving the functionality of existing resources to support use of the collections. The work will connect the original Hunterian material with the new resources, delicately balancing modernity with traditional practice to describe the human material. The anatomical specimens are in a variety of display containers (pots)—glass and plastic, square and cylindrical, each preserved in conservation fluid.

106

The museum currently has very few introduction labels, section labels or object labels, though there has been extensive cataloguing and object reference numbers are present. The exhibition project will involve creating meaningful documentation and presentation of the museum objects to provide context for the diverse users and audiences of this learning space to interpret both the anatomical and historical information. The exhibition will include a series of tailored illustrations alongside the historical specimens, and these, following a redesign of the female reproductive and foetal development collection in the Museum of Anatomy, will be featured in a currently unused area of display cabinets. These illustrations will provide new educational material and artworks, which will communicate themes relevant to the items in historical collection to a twenty-first century public audience and which will support the audience’s understanding of the anatomical structures and biological processes related to foetal development and female pelvic anatomy and reproductive health, as well as understanding of the historical objects and specimens. The visualisations that are planned to support the exhibition will include but will not be limited to • Basic female reproductive anatomy; uterus and ovary—utilising imaging and models of the ovary with oocytes at different stages. • Biology of reproduction; discussing the menstrual cycle showing the changes in the uterus. Utilising representations of histological samples to show how the uterine lining changes during the cycle. • Foetal development from oocyte to birth. This will start with a fertilised zygote and will involve the creation of models of pre-implantation development. • A clinical example such as endometriosis explaining the connection to lining of the uterus as well as demonstrating the anatomical relations between the female pelvic organs— this will include information about how this can affect fertility.

C. MacRobbie

• A public health narrative may involve the role of folate in preventing neural tube defects. This may be represented by a neurulation stage model. • Obstetrics; the journey then culminates with birth, and the introduction of historical obstetric instruments (e.g. forceps). • The function of the placenta, development of the placenta from the embryonic membranes, information on the role of the placenta and discussion of the clinical condition pre-eclampsia and the impact on foetal and maternal health. Demonstration of the differences in the maternal and foetal sides of the placenta. A digital representation of the blood flow will greatly enhance the user’s understanding of the placenta. The role of the illustrator in this exhibition is that of an interpreter and communicator who can communicate the messages of the anatomical or clinical information contained within the museum specimens as accurately and as accessibly as possible (Parker et al. 2017). The artwork produced for the exhibition should support public understanding, further connections between art and science and the community, and enhance the impact of the anatomical collection by pairing carefully considered images with information that is accessible to diverse audiences. With these aims in mind, the illustrator should consult traditional methods as well as current theory and practice and modern techniques used to visualise the body. Working within the context of the museum and with real human material enables the medical artist to create realistic and wellinformed models and illustrations and to provide the viewer with contextual historical information, which will expand their understanding not only of the female and foetal anatomy but also of the relationship between the new illustrations and the historical material. As discussed above, focussing on accessibility and on the support of lay understanding may suggest a need to sanitise these images for public audiences, to neutralise or normalise pathology, and to censor, restrict or even remove specimens that appear challenging

5

The Embryonic re-Development of an Anatomy Museum

by current ethical guidelines. By providing user-friendly but fully informative images as an alternative way of accessing these subjects, the illustrator can enable their audience to confront this discomfort in a supported way. After centuries of progression in science and technology, modern human anatomy has benefitted from the influence of generations of artists, anatomists and clinical practitioners. While the challenges in the field of medical illustration are the continual advancements in technology, and the benefits are the unlimited possibilities of imagination that these new technologies can support. As science has developed, so too has the ability of the artist to create a more realistic and immersive experience of learning. Advances in computer graphic capabilities and imaging are generating infinite new opportunities to visualise science. Advances in the field of game design have also had a positive influence on the scope of medical illustration, providing opportunities to develop 3D models or to animate digital illustrations and even generating virtual reality simulation. Artists are increasingly using Graphic Medicine to widen perspectives and adopting a comic illustration style or utilising gamification to engage audiences. In addition to the draft images presented in this chapter, the planned exhibition should support and meet audience expectations for immersive and interactive learning and explore this as a method for preserving the historical collections and for explaining embryology and related physiology in an accessible way. The law and ethics around visualising human remains restrict the artist’s ability to represent these subjects using the tools we might expect to encounter in the Anatomy Museum. Through text and illustrations, modelling and animation and utilising different materials and tools for different audiences and purposes ultimately for application in useful learning, artists can communicate the complex information contained in the specimens within the limits of legislation. The history and ethics of anatomy are an ongoing association. Human remains held in museum collections are the result of more than three centuries of collecting and scientific study. These collections

107

have been invaluable in piecing together our biological history. Of particular importance is the Human Tissue (Scotland) Act 2006. Scotland has its own distinctive cultural traditions on the treatment and use of human remains, which involves the responsible and respectful care of human remains within collections and the ethical framework of displaying human remains (Museums Galleries Scotland 2011). The exhibition actively pursues avoidance of the morbid curiosity by ‘images, or other artefacts produced from donations placed in the public domain’ (FICEM 2018). The appropriate educational purpose is required to ensure respectful contact with the dead who have a willed donation for scientific and/or educational purposes. Many visitors who see anatomical specimens for the first time will be affected emotionally to the degree that can reduce the acquisition of anatomical and medical knowledge. The ways we are permitted to visualise anatomical subjects have recently experienced new challenges. The legislation that governs the display of human remains prohibits unauthorised imaging in the Museum of Anatomy. New guidelines set out that govern the imaging of human remains in Scotland now include drawing. This restricts visitors and researchers from engaging in their own image taking, and this restricts both their own experience of learning and that of others by limiting the way in which visual information can be gathered and disseminated from these spaces and resources and prevents visitors from continuing to engage with the exhibits after their visitor experience. Learning provision in museums also remains in threat due to considerable challenges, divergences in capacity and resources and competing pressures on physical space and budgets. There is also importance placed on the absent objects to historians of anatomy, and the culturally determined ethics of collecting and the contemporary conservation or disposal of anatomical collections from the past (Claes and Deblon 2018, p 351–362). The subjects for these visualisations provide a challenge to interpret with a sensitivity, integrity and respect that also supports science

108

communication that raises awareness of the science-related topics. As well as responding to scientific advances, the planned exhibition should also respond to the societal challenges faced by and posed by contemporary, twenty-first-century audiences. Socially engaged museums are accessible to the audience the collection was intended for. Leading museums have more recently begun to implement socially engaged approaches. By examining the role of museums and practitioners in shaping society’s understandings of contemporary issues, upholding values of inclusivity and accessibility, we can ensure that museums benefit their communities (Museums Association 2020). Observing socially engaged practice in museums can influence the impact museums have on advocating for social justice and enable them to better explore and address the contemporary issues in our society. The Museums Change Lives campaign encouraged museums to improve their socially engaged practice, delivering a positive social impact with the communities in which they function (Museums Association 2020). Many museums are generating strategies that aim to positively impact audiences and communities. Addressing poverty, inequality, intolerance and discrimination, museums can help understand, debate and challenge these concerns and breaking down barriers to access and inclusion by working with communities and delivering positive social impact. Similarly, ‘Museums are Not Neutral’ (Museums Association 2020) is attempting to erase the myth of museum neutrality and to demand that institutions act as agents of change, changing how people think about museums, building community and questioning the traditional role of the museum and museum educators. This allows for a renewed focus on how visualisation can contribute to knowledge formation in science from the learners’ perspective. In this work, I have responded to these societal movements in museum education in line with the changing cultural epoch, in the images that I am choosing to make to be consciously inclusive and applicable for audiences who may not regularly see themselves and their experiences reflected sensitively in museum contexts. By creating content that

C. MacRobbie

reflects and includes the voices of society in its creation, this exhibition content should reflect that history and that story. The artworks in this chapter and the more that are planned should guide viewers with intentionally inclusive visual content that observes and addresses our changing societal frameworks and effectively challenges problems with inclusivity. This work is being used to advance understanding of current use and possibilities for the anatomical collections in line with an advancing awareness of its current insufficiencies. In light of recent COVID-19 response measures to temporarily close museums, plans for the exhibition have been halted. This has further necessitated the creation of resources to showcase collections and activities in other ways. Learning from the recent emergent need to move all access education online, having a supplementary online platform with which to host a digital exhibition could be a method of futureproofing museum access and supporting those with additional access needs. Further considerations on the development of such a resource will be needed. It is hoped that the project can respond with the development of feasible and deliverable education resources and learning tools for using anatomy collections encouraging users to become involved in activities related to the production of the exhibition, fostering a culture of participation and connecting the community to their anatomy museum. The project considers how museums are responding to a fast-changing educational landscape and will hope to offer the opportunity for input from the community in its creation.

References Asiedu MN, Agudogo JS, Dotson ME et al (2020) A novel speculum-free imaging strategy for visualization of the internal female lower reproductive system. Sci Rep 10:16570. https://doi.org/10.1038/s41598-020-722199 Bakhtin M (1984) Rabelais and his world. Translated by Hélène Iswolsky. Indiana University Press, Bloomington Belsky D (2019) Representing diverse bodies in medical illustration. Available from http://blog.castac.org/

5

The Embryonic re-Development of an Anatomy Museum

2019/10/representing-diverse-bodies-in-medicalillustration/ Bourne T Early pregnancy and miscarriage. http://www. imperial.ac.uk/metabolism-digestion-reproduction/ research/reproductive-developmental-biology/clinicalthemes/early-pregnancy/ Brasseur LE, Thompson Torri L (1995) Gendered ideologies: cultural manuals in renaissance England social contexts. Transactions on Professional Communication 38:4. Illustrated BRIA (2018) Access to free sanitary products. Available from https://www.parliament.scot/ parliamentarybusiness/CurrentCommittees/112914. aspx. Accessed 11 March 2021 Buettner KA (2017) “Franklin paine mall (1862–1917)”. Embryo project encyclopedia (2007-11-01). ISSN: 1940–5030 http://embryo.asu.edu/handle/10776/1682. Last modified: Wednesday, July 4, 2018–04:40 Accessed 14 March 2021 Buklijas T, Hopwood N (2008–2010) Images of human embryos. An online exhibition developed by the University of Cambridge with support from the Wellcome Trust. Last reviewed 2014. Available from http://www. sites.hps.cam.ac.uk/visibleembryos/index.html). Accessed 10 March 2021 Campbell S (2013) A short history of sonography in obstetrics and gynaecology. Facts, Views & Vision in ObGyn 5(3):213–229 Charter R, Ussher JM, Perz J, Robinson K (2018) The transgender parent: experiences and constructions of pregnancy and parenthood for transgender men in Australia. Int J Transgenderism 19(1):64–77. https:// doi.org/10.1080/15532739.2017.1399496 Chorley AJ, Marlow LA, Forster AS, Haddrell JB, Waller J (2017) Experiences of cervical screening and barriers to participation in the context of an organised programme: a systematic review and thematic synthesis. Psychooncology 26(2):161–172. https://doi.org/ 10.1002/pon.4126 Claes T, Deblon V (2018) When nothing remains: anatomical collections, the ethics of stewardship and the meanings of absence. J History of Collections 30 (2):351–362. https://doi.org/10.1093/jhc/fhx019 Clarke LH, Martin-Matthews A, Matthews R (2008) The continuity and discontinuity of the embodied self in infertility. Canadian Review of Sociology/Revue Canadienne De Sociologies 43(1):95–113. https://doi. org/10.1111/j.1755-618X.2006.tb00856.x Cober K (2015) Dissecting race: an examination of anatomical illustration and the absence of non-white bodies. University Halifax, Nova Scotia Darwin Z, Greenfield M (2019) Mothers and others: the invisibility of LGBTQ people in reproductive and infant psychology. J Reprod Infant Psychol 37 (4):341–343. https://doi.org/10.1080/02646838.2019. 1649919 Ellis H (2011) Obstetrics: anatomy of the uterus. Anaesthesia & Intensive Care Med 12(3):99–101. https://doi. org/10.1016/j.mpaic.2010.11.005

109

Elson J (2005) Am I still a woman? Hysterectomy and gender identity. Temple University Press, Philadelphia Eveleth R (2019a) This user-friendly menstrual cup is what happens when design is inclusive. https://www. vice.com/en_us/article/vbk5ky/keela-flex-menstrualcup-design-bias Eveleth R (2019b) Medical textbooks overwhelmingly use pictures of young white men https://www.vice.com/ en_us/article/3k3kkn/medical-textbooks-overwhelm ingly-use-pictures-of-young-white-men FICEM (2018). https://www.ifaa.net/committees/ethicsand-medical-humanities-ficem/ethical-and-medicalhumanities-perspectives-on-the-public-display-ofplastinated-human-bodies/ Fikkan JL, Rothblum ED (2012) Is fat a feminist issue? Exploring the Gendered Nature of Weight Bias Sex Roles 66:575–592. https://doi.org/10.1007/s11199011-0022-5 Garcia R, Ali N, Papadopoulos C et al (2015) Specific antenatal interventions for black, Asian and minority ethnic (BAME) pregnant women at high risk of poor birth outcomes in the United Kingdom: a scoping review. BMC Pregnancy Childbirth 15:226. https:// doi.org/10.1186/s12884-015-0657-2 Greenwood BN, Hardemanb RR, Huangc L, Sojournerd A (2019) Physician–patient racial concordance and disparities in birthing mortality for newborns. Proc Natl Acad Sci U S A 117(35):21194–21200 Hey Girls, Our approach to period education. 1st April 2019. Available from https://www.heygirls.co.uk/ourapproach-to-period-education/ Hill MA (2021) Embryology Carnegie stages. Available from https://embryology.med.unsw.edu.au/embryol ogy/index.php/Carnegie_Stages Hopwood N (2002) Embryos in wax: models from the Ziegler studio. Whipple Museum of the History of Science, Univ. of Cambridge, and Institute of the History of Medicine, Univ. of Bern, Cambridge and Bern Hopwood N (2007) A history of Normal plates, tables and stages in vertebrate embryology. Department of History and Philosophy of science, University of Cambridge, U.K. J Dev Biol 51:1–26. https://doi.org/ 10.1387/Ijdb.062189nh Hopwood N (2009) Embryology. In: Bowler P, Pickstone J (eds) The Cambridge history of science the Cambridge history of science. Cambridge University Press, Cambridge, pp 285–315. https://doi.org/10. 1017/CHOL9780521572019.017 Irving C (2020) Ziegler waxes: visualising the embryo. Surgeons hall the anatomy lab. Available from https:// surgeonshallmuseums.wordpress.com/tag/wax-model/ Accessed 01 December 2020 Larsen EC, Christiansen OB, Kolte AM et al (2013) New insights into mechanisms behind miscarriage. BMC Med 11:154. https://doi.org/10.1186/1741-7015-11154. https://bmcmedicine.biomedcentral.com/articles/ 10.1186/1741-7015-11-154 Lawrence SC, Bendixen K (1992 Oct) His and hers: male and female anatomy in anatomy texts for U.S. medical

110 students, 1890-1989. Soc Sci Med 35(7):925–934. https://doi.org/10.1016/0277-9536(92)90107-2 Levine R. (2019) Body of work: the pioneering women behind the ground breaking Grant’s Atlas of Anatomy, University of Toronto https://www.utoronto.ca/news/ body-work-pioneering-women-behind-groundbreak ing-grant-s-atlas-anatomy Lie M (2012) Reproductive images: the autonomous cell. Sci Cult 21(4):475–496. https://doi.org/10.1080/ 09505431.2012.679728 Longo D, Reynolds LP (2016) Wombs with a View. Illustrations of the Gravid Uterus from the Renaissance William Hunter Anatomia uteri humani gravidi tabulis Illustrata 12 May 2016 Luce A, Cash M, Hundley V, Cheyne H, Teijlingen EV, Angell C (2016) “Is it realistic?” the portrayal of pregnancy and childbirth in the media. BMC Pregnancy Childbirth 16:40 Mbrrace-UK (2019) Maternal, newborn and infant clinical outcome review programme. Available from https:// www.npeu.ox.ac.uk/assets/downloads/mbrrace-uk/ reports/MBRRACE-UK%20Maternal%20Report% 202019%20-%20WEB%20VERSION.pdf McCartney J (2012) Internal Affairs. [Lead Crystal Glass]. Available from https://jamiemccartney.com/portfolio/ internal-affairs. Accessed on 13Oct 2020 McCartney M (2013) Patient information leaflets: “a stupid system”. BMJ 2013:347. https://doi.org/10.1136/ bmj.f4748 McCulloch NA, Russell D, McDonald SW (2001a) William Hunter's casts of the gravid uterus at the university of Glasgow. Clin Anat 14(3):210–217. https:// doi.org/10.1002/ca.1035 McCulloch NA, Russell D, McDonald SW (2001b) William Hunter’s casts of the gravid uterus at the university of Glasgow. Clin Anat 14(3):210–217 McIntosh T (2020) Imagining the foetus https:// wellcomecollection.org/articles/ Xw7mQBAAACEAKRJ3. Accessed 20 December 2021 Miyazaki R, Makishima H, Männer J, Sydow H, Uwabe C, Takakuwa T, Viebahn C, Yamada S (2017) Blechschmidt collection: revisiting specimens from a historical collection of serially sectioned human embryos and foetuses using modern imaging techniques. Japanese Teratol Soc 58:152–157. First published: 03 November 2017. https://doi.org/10. 1111/cga.12261 Museums Association (2020) Accessed on December 20 2020. https://www.museumsassociation.org/ campaigns/museums-change-lives/measuringsocially-engaged-practice/ Museums Galleries Scotland (2011) Guidelines for the care of human remains in scottish museum collections. Available from https://www.museumsgalleriesscotland. org.uk/media/1089/guidelines-for-the-care-of-humanremains-in-scottish-museum-collections.pdf My Midwives UK (2020). https://www.mymidwives.co. uk/mymidwivesinitiative

C. MacRobbie Newnham E, McKellar LV, Pincombe JI (2017) Paradox of the institution: findings from a hospital labour ward ethnography. BMC Pregnancy and Childbirth 17:2 NHS (n.d.).https://www.nhs.uk/conditions/miscarriage/ O’Connell HE, DeLancey JO (2005) Clitoral anatomy in nulliparous, healthy, premenopausal volunteers using unenhanced magnetic resonance imaging. J Urol 173 (6):2060–2063. https://doi.org/10.1097/01.ju. 0000158446.21396.c0 Okereke E, Ishaku SM, Unumeri G, Mohammed B, Ahonsi B (2019) Reducing maternal and newborn mortality in Nigeria-a qualitative study of stakeholders' perceptions about the performance of community health workers and the introduction of community midwifery at primary healthcare level. Hum Resour Health 17(1):102. https://doi.org/10.1186/s12960019-0430-0 Pansky B (n.d.) Review of medical embryology. Embryonic Development & Stem Cell Compendium https:// discovery.lifemapsc.com/library/review-of-medicalembryology/chapter-26-embryonic-folding-and-flex ion-of-the-embryo Parker R (2018) The problem with period branding October 25 2018, Available from https://www. creativebloq.com/features/the-problem-with-periodproduct-branding. Accessed 12 November 2020 Parker R, Larkin T, Cockburn J (2017) A visual analysis of gender bias in contemporary anatomy textbooks. Social Science & Medicine 180:106–113. https://doi. org/10.1016/j.socscimed.2017.03.032 Richardson R (2000) A necessary inhumanity. J Med Ethics 26:104–106 Richter M, Sanchez-Jauregui L (n.d.) Technical art history group. Available from https://technicalarthistory.gla. ac.uk/gravid-uteri/. Accessed on: 10 December 2020 Rydström K (2020) Degendering Menstruation: Making Trans Menstruators Matter. In: Bobel C, Winkler IT, Fahs B, Hasson KA, Kissling EA, Roberts TA (eds) The Palgrave handbook of critical menstruation studies. Palgrave Macmillan, Singapore. https://doi.org/10. 1007/978-981-15-0614-7_68 Saffran L (2017) The art of translating science. https:// spark.adobe.com/page/Xf7mZ1hQVGOdY/?w¼0_ 1 9 2 1 & f b c l i d ¼I w A R 2 Z S C - 5 g 8 2 O _ boVxGhmNbLbvSQQVzDgGm9rcEIPFN5EvmIRhC e-vnU3MhM Science Museum Celebrates 40 Years of IVF with new E x hi b i t i o n ( 2 01 8 ) . Av a i l a bl e f r o m w ww . sciencemuseum.org.uk/ivf. Accessed: 20 December 2020 Science Museum Group (2007) ‘Gynaeplaque’ model, United States, 1925–1935. 2007–186 Science Museum Group Collection Online. Accessed October 31, 2020. https://collection.sciencemuseumgroup.org.uk/objects/ co8078370/gynaeplaque-model-united-states-19251935-model-representation Sergovich A, Johnson M, Wilson TD (2010) Explorable three-dimensional digital model of the female pelvis, pelvic contents, and perineum for anatomical

5

The Embryonic re-Development of an Anatomy Museum

education. Anat Sci Educ 3(3):127–133. https://doi. org/10.1002/ase.135 Smith C (2019) Dealing with bias in artificial intelligence. Available from https://www.nytimes.com/2019/11/19/ technology/artificial-intelligence-bias.html Stolz J (1866) Respiration and signs of life in a five months Foetus. Med Surg Rep 15(16):344–345 Thomson W (1850) Deeds instituting bursaries, scholarships, and other foundations, in the college and University of Glasgow. UK, Maitland club, p 248 Tommy’s (2020). Available from https://www.tommys. org/baby-loss-support/miscarriage-information-andsupport/miscarriage-statistics Tull K (2019) Period poverty impact on the economic empowerment of women. University of Leeds Nuffield Centre for International Health and Development. Available from https://assets.publishing.service.gov. uk/media/5c6e87b8ed915d4a32cf063a/period.pd University of Glasgow Archive GB 248 GUA MED (n.d.) 1856–1969 Michael Moss, Moira Rankin and Lesley Richmond. The history and constitution of the University of Glasgow. University of Glasgow, Glasgow, p 2001

111

Vagina Museum (n.d.). https://www.vaginamuseum.co. uk/about/story Venkatesan N (2019) Menstrual health education: reflections from the field. https://ecofemme.org/men strual-health-education/ Vogel C (2013) “Art, from conception to birth in Qatar”. The New York Times. Accessed 20 November 2018. https://www.nytimes.com/2013/10/08/arts/design/ damien-hirsts-anatomical-sculptures-have-their-debut. html Wall LL (2006) The medical ethics of Dr J Marion Sims: a fresh look at the historical record. J Med Ethics 32 (6):346–350. https://doi.org/10.1136/jme.2005. 012559 Wellner K (2009a) “Adolf Ziegler”. Embryo Project Encyclopedia (2009-06-10). http://embryo.asu.edu/handle/ 10776/1968 Wellner K (2009b) “Gustav Jacob Born (1851–1900)”. Embryo Project Encyclopedia (2009-07-22). http:// embryo.asu.edu/handle/10776/1967 Withycombe S (2018) Lost: miscarriage in nineteenthcentury America, Critical issues in health and medicine, vol vii. Rutgers University Press, New Brunswick, N.J, p 220

6

Visualising the Link Between Carpal Bones and Their Etymologies Kaitlin Nasrala, Matthieu Poyade, and Eilidh Ferguson

Abstract

It has been observed through published studies, as well as anecdotally, that medical students struggle with retention of anatomical knowledge. Studies have found that having an established understanding of classical Greek or Latin languages, which underpin medical terminology, can result in higher anatomy test scores by medical students. It has also been established that three-dimensional (3D) visualisation tools can aid in student learning. This chapter will examine the research conducted at the University of Glasgow, which focused on the creation of a mobile application that visualises the etymology of the carpal bones for the purpose of aiding medical students in their learning and retention of knowledge of anatomy. The chapter will first build a body of knowledge by reviewing previous studies in which a carpal bone test was used as a measure of medical students’ anatomy knowledge, as well as the relevance of etymology in medicine and its use in the study of anatomy, and the current K. Nasrala · M. Poyade School of Simulation and Visualisation, Glasgow School of Art, Glasgow, UK e-mail: [email protected]; [email protected] E. Ferguson (*) Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK e-mail: [email protected]

teaching methods of anatomy, with a focus on how 3D visualisation tools can aid learning. It then outlines a methodological and technical framework to create anatomically accurate 3D models of the carpal bones and develop the final mobile application. It also discusses the methodology used to carry out suitable user testing and collect user feedback. This chapter concludes by discussing the results of user testing, where feedback was analysed to improve the mobile application design for further use in anatomy teaching. Limitations and future outlooks of the study, along with the future of integrating 3D visualisation tools as teaching methods to aid in student learning of anatomy, are also explored. Keywords

Medical Visualisation · Carpal Bones · Etymology · 3D Models · Educational Application

6.1 6.1.1

Theoretical Background Introduction

The many years of education that medical students complete are largely focused on learning the thousands of terms and concepts of human anatomy, which can be quite a daunting task. Moreover, much of the anatomical terminology

# The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. M. Rea (ed.), Biomedical Visualisation, Advances in Experimental Medicine and Biology 1334, https://doi.org/10.1007/978-3-030-76951-2_6

113

114

K. Nasrala et al.

with which they need to be familiar is derived from ancient Greek and Latin, presenting additional challenges akin to learning a new language. Many students struggle to learn the terms through reliance on rote learning methods, a repetitionbased memorisation technique (Mayer 2002; Brown 2014; Ebenezer and Mohanraj 2020). Specifically, it has been observed through published studies, noting comments from senior doctors, as well as anecdotally from University lecturers, that medical students struggle with retention of anatomical knowledge, which could be caused by factors such as reduced teaching times and less access to physical dissection, and may be adding to the reliance on rote learning methods (Spielmann and Oliver 2005). A proposed solution for students to overcome the challenges presented from learning medical terminology through memorisation techniques is to establish a link between the anatomy and their respective etymologies, the study of the origins of words. Studies have found that having an understanding of a word’s etymology enhances students’ learning and recollection, as it is a form of meaningful learning—a type of learning connected to prior learning, more highly retainable and generalizable—making it superior to simple rote learning of vocabulary (Pierson 1989). This chapter will first explore the study of etymology and its use in medicine, the influence understanding etymology has on learning anatomy, and the benefits of the use of digital technology in learning.

the hands. A 2005 study conducted by Spielmann and Oliver provides some insight into medical students’ and junior doctors’ knowledge of anatomy by using a carpal bone test (Spielmann and Oliver 2005). Fifty questionnaires regarding the labelling of the eight carpal bones were administered to a cohort of 25 medical students, 15 pre-registration house officers (PRHOs) and 10 senior house officers (SHOs). Out of the 50 participants, only 15 could correctly name all eight carpal bones, seven of them being SHOs. Similarly, in a 2012 study by Valenza et al., a carpal bone test was used to assess the anatomical knowledge of third-year medical and physical therapy students, and the results held consistent with the Spielmann and Oliver study (Valenza et al. 2012). A group of 134 students were allotted 5 min to identify and label each of the eight carpal bones on the test. It was found that only 39 students (29%) correctly identified all eight carpal bones, and 36 physical therapy students (66%) correctly identified five or more carpal bones as compared with only 26 medical students (32.5%). The results of both studies have portrayed that the level of anatomical knowledge of entry-level medics is far from ideal. They also specifically emphasised the lack of carpal bone anatomical knowledge possessed by medical students, which further warranted the creation of an aid to help them learn and remember the bones to a higher degree of excellence.

6.1.3 6.1.2

Why Carpal Bones?

There have been multiple studies conducted to assess medical students’ knowledge of anatomy in recent years, a common method of assessment among them being the use of a carpal bone test. The reasoning behind the carpal bones being chosen as a benchmark of basic anatomical knowledge is that they have clinical relevance to a variety of different specialties, and they are easy to objectively examine. They also make up a significant component of the upper limbs, as they are important to the movement and use of

The Study of Etymology and Its Use in Medicine

6.1.3.1 The Study of Etymology Etymology is briefly defined as, the scientific study of the origins and history of the changing meanings and forms of words (Ross 1969). It is essentially the breakdown of words into their prefixes, suffixes, and root words and the process of linking them to their classical origins, in order to give meaning to the words. As an academic specialty, it is usually taught as part of postgraduate English literature studies; therefore, it is unlikely that many medical students have

6

Visualising the Link Between Carpal Bones and Their Etymologies

sufficient knowledge in etymology, or classical Greek and Latin languages (Pierson 1989).

6.1.3.2

Relevance of Etymology in the Medical Field The use of classical Greek and Latin language is extremely prevalent within medical terminology, the language used by medical professionals. However, an issue is presented when physicians use complex medical terminology when interacting with their patients, as it can cause a communication barrier (Sevinc et al. 2005). Medical terminology is useful as a standard form of communication between professionals in the field but may not be understood by patients. This disconnect in communication and understanding between patient and physician can cause the patient to feel intimidated and overwhelmed (Sevinc et al. 2005). As classical Greek and Latin courses are currently not required to fulfil a medical degree, many medical students and professionals in the field rely on rote learning methods, in conjunction with observation, dissection, and other hands-on learning techniques, when learning anatomy and medical terminology (Brown 2014; Lewis et al. 2014; Murgitroyd et al. 2015). As memorisation plays a major role in these typical learning methods, many of these individuals will go through their training and experience without having sufficient etymological knowledge of the anatomy being learned. One study found that no house officers or medical students (n ¼ 52) could translate common abbreviations used in medical practice (Drury et al. 2002). Another study found that in an emergency department, the percentage of patients who did not recognise analogous terms such as bleeding and haemorrhage was 79%, broken and fractured bone, 78%; at the same time, the percentage of patients who did not recognise non-analogous terms such as diarrhoea and loose stools was 37%, for example (Lerner et al. 2000). The patient–physician communication barrier occurs when physicians use terminology which the patient does not understand and the physician does not have the ability to translate or explain. When there is no depth in understanding of the classical origins of these terms, it makes it much

115

more difficult for communication to be had between people in the medical field and others not so well-versed, like patients. It is therefore considered useful for the physician to have background knowledge in anatomical and medical etymology, as it would allow them to explain in simple and relevant terms what the words being used with their patients actually mean.

6.1.4

The Link Between Knowledge of Etymology and Successful Learning of Anatomy in Medical Students

Numerous studies have been conducted to explore the influence of understanding etymology on learning human anatomy (Spielmann and Oliver 2005; Smith et al. 2007; Pampush and Petto 2011; Ebenezer and Mohanraj 2020). In order to understand this, one must first understand how the process of learning new information works.

6.1.4.1

How Do We Learn? Three Learning Outcomes The different ways in which we approach new information being presented can be categorised into three groups: no learning, rote learning, and meaningful learning (Mayer 2002). Each of these learning outcomes results in a difference in the way new knowledge is stored and used. Consider a scenario in which new information is skimmed over quickly, without taking time to analyse and understand the content. It would likely be found that little to no information could later be recalled, only a few key points may be remembered, and the individual would not be able to use the information that had been provided to solve problems (Mayer 2002). This would be categorised as “no learning”. Rote learning and meaningful learning are the second and third learning categories. They differ in the regard that individuals are able to recall most of the newly learned information; however, individuals who have achieved meaningful learning are also able to transfer and apply their knowledge to new learning scenarios and problems, unlike individuals who have only

116

achieved rote learning (Mayer 2002). In the case of learning the anatomy of carpal bones, the heavy reliance on repetition-based memorisation rote learning methods may be one of the reasons that medical students struggle. It could be helpful to use meaningful learning methods instead when it comes to the carpal bones, as students would be able to recall information about the structural anatomy but also transfer and apply their knowledge of the etymology of the bones in order to help them identify each bone correctly.

6.1.4.2

How Etymological Understanding Aids Anatomical Learning in Medical Students The influence of knowledge of etymology on how medical students learn human anatomy has been thoroughly studied over the years. One study looked at first-year medical students who were given a basic etymology pre-test before taking a gross anatomy course to evaluate students’ familiarity with the roots, prefixes, and suffixes of anatomical terms rather than the anatomy itself (Smith et al. 2007). The course then defined etymologies during lectures and dissection laboratories, and explanations of etymologies were provided for each section of anatomy. After completing the course, the students were given the same etymology test used as the pre-test. The results were in line with the hypothesis, in that students scored significantly higher on the post-test than the pre-test, by an average of 6.2%. They also found that previous exposure to medical terminology enhances scores; students with exposure to medical terminology scored significantly higher on the pre-test than students with no exposure to medical terminology, by an average of 8.7%. Finally, they found that knowledge of etymology enhances gross anatomy learning and enjoyment, based on the result of qualitative questionnaires administered to the students. Another study published in 2020 by Ebenezer and Mohanraj draws parallels to these results (Ebenezer and Mohanraj 2020). In this crosssectional study, 214 preclinical medical students were randomly separated into two groups. The first group was taught osteology of the skull with etymology, while the control group was

K. Nasrala et al.

taught the same topic without etymology, using the classical teaching method. The results mirrored the findings of Smith et al. (2007): a positive correlation between learning etymology and higher test scores was found. The students taught osteology of the skull with the addition of etymology scored twice as high as the students taught without etymology.

6.1.5

Use of Digital Technology in Learning

6.1.5.1 Current Teaching Methods Traditionally, teaching methods for anatomy are based on gross dissection and group lectures with the help of 2D atlas images, anatomical models, and clinical cases (Murgitroyd et al. 2015). Various studies have explored the views that students hold on the different teaching methods teachers use during lectures, and an almost universal conclusion has been found that students are not engaging with didactic lecture material as much as they should be (Gilbert 2004; Teoh and Neo 2007). In their 2007 study, Teoh and Neo reported that their respondents had claimed that it was boring to hear the lecturer talking in front of them and that an integration of technologies in their lectures would aid them in their learning process. Students generally find subjects in the field of science to be abstract and require a depth of prior understanding and visualisation skills (Gilbert 2004). A large problem with the current teaching method of students attending lectures arises when disengagement in conjunction with the learning of new, difficult, and abstract subject matter causes difficulty in understanding concepts well; this leads to the formation of misconceptions (Saidin et al. 2015). Misconception can then interfere with students’ learning of new concepts, as it goes on to cloud understanding and warp how new information is being retained. As students seem to struggle with carpal bone anatomy, it may be helpful to incorporate a digital learning aid into lessons and lectures to help engage students in the material to a higher degree.

6

Visualising the Link Between Carpal Bones and Their Etymologies

This, in turn, might yield higher retention of knowledge of the wrist anatomy.

6.1.5.2

How Visualisation Techniques Aid in Student Learning A study conducted in 2016 explored the effect of etymological elaboration, pictorial elucidation, and a combination of these two strategies on the learning of idioms, which are words or phrases that aren’t meant to be taken literally, like the expression “having cold feet,” which refers to feeling nervous or lacking confidence in a planned action (Haghshenas and Hashemian 2016). The researchers found that all three strategies yielded significantly higher post-test results than the control, and that combining etymology and their corresponding explanations with visuals was most effective in aiding idiom learning. This same principle can be applied to medical students in their efforts in learning the carpal bones. The use of 3D visualisation techniques in conjunction with anatomy learning has been explored in multiple studies, where the general consensus has been that they aid in student learning through the enhanced acquisition of knowledge and increased learning enjoyment (Nicholson et al. 2006; Brewer et al. 2012; Murgitroyd et al. 2015; Pujol et al. 2016). In 2012, a study tested whether the use of 3D visualisation techniques would improve trainee understanding of brain anatomy, orientation, visualisation, and navigation, in comparison with current teaching methods (Brewer et al. 2012). The study found that a 3D digital lab in addition to traditional dissection can improve learning for new students in the field of neuroanatomy (Murgitroyd et al. 2015). In 2006, Nicholson et al. carried out a study to test the educational effectiveness of a computer-generated 3D model of the middle and inner ear (Nicholson et al. 2006). The post-test scores of the intervention group were significantly higher than those of the control group (83% vs 65%), showing that the use of 3D visualisation techniques enhanced students’ anatomy learning. In a more recent study, experiments were conducted to demonstrate the feasibility and benefits of developing innovative

117

teaching modules for anatomy education of firstyear medical students by using 3D models of anatomical structures (Pujol et al. 2016). Written task material and qualitative review by students suggested that the use of the 3D models led to a better understanding of the shape and spatial relationships among anatomical structures and helped illustrate variations in structural anatomy from one body to another. The study showed that using 3D visualisation techniques, such as 3D models of anatomy, aids first-year medical students in learning human anatomy (Pujol et al. 2016).

6.1.5.3

Benefits of E-learning and Digital Technology Use in Learning In addition to 3D visualisation techniques enhancing student learning through improving knowledge acquisition and student enjoyment, digital technologies are playing a key role in the evolution of learning methods in the everadvancing world. With the arrival of COVID19, a global pandemic has rocked life as it has always been known. With the introduction of nation-wide lockdowns, educational facilities across the globe were urged to close, forcing many institutions to move their curriculum online in place of in-person learning. E-learning has now taken over as the main teaching method on a global scale; lessons are being taught by educators over digital meeting applications like Zoom, and whole curriculums have been converted into PDF, PowerPoint, and web-module formats. While there are inconveniences with not being able to learn in face-to-face settings, digital learning has brought many benefits as well. The current generation of students have been surrounded by technology at home and in the classroom from an early age and are more competent with technology use than previous generations (Bice et al. 2016). There is a plethora of digital learning tools, like mobile applications, which are now widely available and cater to enhancing student learning and engagement. A 2016 study looked at whether the use of a general learning assistance mobile application focusing

118

K. Nasrala et al.

on the skeletal system improved student performance on examinations (Bice et al. 2016). The results of the study found a positive correlation between mobile application use and high test scores; student test scores were significantly higher with the application use than test scores with no application use, by an average of 6.65%. These results demonstrate that the use of a form of digital learning technology in addition to normal teaching methods enhances student learning and their abilities to remember and recall correct anatomical information, more than current teaching methods alone.

6.1.6

6.2.1

Research Questions

The following questions were proposed prior to the start of testing, based on two main aspects: mobile application usability and usefulness. 1. Is the mobile application entertaining and easy to use? 2. Does the mobile application help medical students learn carpal bone anatomy?

Conclusion

It has been established by published studies, as well as observation from anatomy lecturers, that medical students struggle to learn the anatomy of carpal bones. However, with the supplementation of etymology and digital learning technologies, such as web-modules and mobile applications, learning, remembering, and recollection can be enhanced in students. There have not yet been studies conducted exploring whether a digital learning aid focused on visualising the link between carpal bones and their etymologies will aid medical students’ learning and retention of knowledge of wrist anatomy; this study will aim to answer this research question. It is hypothesised that the use of a digital learning aid that visualises the link between carpal bones and their etymologies will improve learning and retention of carpal bone anatomy in medical students.

6.2

positive impact on learning and retention of knowledge of carpal bone anatomy.

Aims and Hypothesis

This project aimed to effectively bridge the gap between the origins of the names of the carpal bones and their structural anatomies using a mobile application, and to test whether the application will improve learning in medical students. It was hypothesised that the creation of an application that provides a visual link between the carpal bones and their etymologies would have a

6.3 6.3.1

Materials and Methods Materials

Table 6.1 lists all the software, hardware, data, and resources used to create and develop the mobile application used in this research study. The purpose of use for each, and the publishers or source of each, are provided in the corresponding columns and rows.

6.3.2

Methods

6.3.2.1 Design and Development A flowchart was created to map the workflow for application design and development. The process was easily categorised into ten stages, as seen in Fig. 6.1. Concept With the research question in mind, an idea for a mobile application was conceived, and a mood board and storyboard (Fig. 6.2) were created. Medical students are the intended audience for the application, which guides each user through three different educational exercises: a guided educational module, 3D models of the carpal bones, and a flashcard game.

6

Visualising the Link Between Carpal Bones and Their Etymologies

119

Table 6.1 List of software, hardware, and data used for this research project Software Slicer

Purpose 3D visualisation and manual segmentation of medical imaging data

Publisher Open-source platform from BWH and 3D Slicer contributors (BWH 2019); available at www.slicer.org

Autodesk 3DS Max

3D modelling software used to cap holes in mesh after retopologising in Mudbox

Autodesk, Inc., New York, USA (Autodesk 2019)

Autodesk Mudbox

Organic 3D sculpting software used for separating imported bones from Slicer, model refinement, capping holes in mesh, and retopologising models

Autodesk, Inc., New York, USA (Autodesk 2019)

Unity

Games and app development engine

Unity Technologies, California, USA (Unity Technologies 2019); available at unity3d.com

C# scripting platform in collaboration with the Unity game engine

Microsoft, California, USA (Microsoft 2019); available at https://visualstudio. microsoft.com

Digitisation of illustrations and graphic design of the user interface components and mood board

Adobe Systems, Inc., California, USA (Adobe 2019)

Android Studio Software Development Kit (SDK) used to build an executable application to Android device

Google, Inc., California, USA (Google 2019); available at developer.Android. com/studio/index.html

Visual Studio Adobe Photoshop

Android SDK

Hardware

Use Tablet device to test the application during development and evaluation stages. Running Android OS, v6.0.1 (Marshmallow) operating system with an aspect ratio and screen resolution of 2048  1536

Samsung Galaxy Tablet S2 Data & Resources Visible Human Project CT Datasets All forms and quizzes

Use Used pelvic scans to segment wrist bones in order to create the carpal bone models Created and distributed quizzes, participant information and consent forms, and questionnaires, and collected data

Publisher Samsung Group, Seoul, South Korea (Samsung 2019)

Source Visible Human Project (https://mri. radiology.uiowa.edu/visible_human_ datasets.html) JISC Survey Platform (https://www. onlinesurveys.ac.uk/) (continued)

120

K. Nasrala et al.

Table 6.1 (continued) Data & Resources Microsoft Yi Baiti font Arial font 3D carpal bone models

Use Font used for all custom UI elements in mobile application Font used for all text in mobile application Used these 3D carpal bone models for reference when creating carpal bone models for the mobile application

Source Adobe Photoshop Unity Sketchfab, artist: Ajlieurance (https:// sketchfab.com/ajlieurance)

Fig. 6.1 Diagram displaying mobile application creation workflow

3D Bone Model Production Computerised tomography (CT) scans and 3D modelling software were used to produce the 3D bone models of the wrist, which was the first step in the application development process. Segmentation Using the software 3D Slicer, a dataset of DICOM CT scans of a wrist from an adult female patient1 were used to segment the eight carpals, the distal ulna, and the distal radius for the

creation of the 3D bone models. The CT scans were filtered using the Laplacian Sharpening Image Filter to enhance the contrast between the bones and the surrounding tissues, easing the anatomical structure edge recognition during the segmentation of the bones, as seen compared to the original scans in Fig. 6.3. The Crop Volume

1

Patient data from the Visible Human Project (https:// www.nlm.nih.gov/research/visible/applications.html) were used for research purposes.

6

Visualising the Link Between Carpal Bones and Their Etymologies

121

Fig. 6.2 Storyboard created to plan the scenes and UI element placements for the mobile application

Fig. 6.3 CT DICOM dataset viewed in the axial plane before and after Laplacian Sharpening Image Filter was applied

module was then used to crop the scans to only include the hand and wrist, using a region of interest (ROI) bounding box, displayed in Fig. 6.4. The ThresholdEffect tool in Slicer uses the Hounsfield unit (HU) of tissue density to inform the selection of pixels from the scans,

which makes it the ideal tool for the segmentation of bones, as they have high densities, especially compared to surrounding tissues. As indirect volume rendering techniques in Slicer favour the contrast between high and low densities, it was the ideal method of segmentation of the carpal

122

K. Nasrala et al.

Fig. 6.4 CT DICOM dataset viewed in the axial plane, showing the ROI bounding box encasing the patient’s right wrist

bones. A threshold of 200.00 to 1530.00 was set for use by the PaintEffect tool to paint the bones slide by slide, while only including the values that fell inside the boundaries of the threshold. Once the bones were painted, the threshold was expanded to include all values, ranging from 1024.00 to 1530.00 to allow for the holes in the paint to be filled. This would make sure that the bone models built from the segmentation would be solid and contain very few holes in the structures, if any. The EraseLabel tool was then used to erase any paint connecting one bone to another; this would allow for each bone to be separate when imported into the 3D modelling software and would save time from having to delete faces, edges, and vertices from each of the bone models in order to manually separate them from each other. The before and after of this process can be seen in Fig. 6.5. Once segmentation was complete, the Merge and Build tool was used to build the bone models, portrayed in Fig. 6.6. After examination of the models for satisfaction with the general shapes of the bones, they were exported as an obj file to be later imported into the 3D modelling software for refinement, including sculpting, smoothing, and retopology.

Refinement of 3D Bone Models in Adobe Mudbox The obj file from Slicer was imported into Mudbox, where the raw models created from Slicer segmentation were refined and retopologised. Free-floating objects that came in with the eight carpal bones, distal ulna, and distal radius were deleted from the file, leaving only ten bone models in the scene. Each of these bones was first inspected for holes in the meshes that were patched to ensure continuity, which would allow for functionality in real-time engines like Unity during application development. An example of this process can be seen in Fig. 6.7. Once all the holes were patched, each bone was refined using the sculpting tools, harsh edges were smoothed, and articulating surfaces were defined. Photos of the carpal bones, found in Chap. 13 of the Human Bone Manual (White and Folkens 2005), along with 3D models of the carpal bones found on Sketchfab, were used as references when refining the bone models, as access to real bones were not available. A before and after of the modelling process in Mudbox can be seen in Fig. 6.8. When the models were finally satisfactory, the “Reduce Mesh” tool was used to lower the polycounts of each bone model by

6

Visualising the Link Between Carpal Bones and Their Etymologies

123

Fig. 6.5 Sagittal plane: segmentation of the carpal bones and distal radius using (left) a threshold of 200–1253 and (right) a threshold of 1024–1530, and EraseLabel tool was applied

95–98%, deeming them suitable for import into Unity (Fig. 6.9). After mesh reduction, the bone models were exported to 3DS Max for a final check of the meshes and polycounts. Application Development In total, the application holds 12 interactive scenes: the first being the main menu, followed by nine contributing to the educational module, one being the 3D bones scene, and one being the flashcard test. A schematic map of the scene flow is given in Fig. 6.10. C# scripts were coded for interactive use of the application in each scene, and animations of 3D models and UI elements were created using the Animation and Animator Controller tools in Unity and were triggered by C# scripts or event triggers with box or mesh colliders. The first scene built was the main menu (Fig. 6.11). This included the addition of a canvas and panel, and the placement and anchoring of four UI buttons, each directing the user to either a new scene or the information window. The buttons were placed from top to bottom in the intended order of play by the user. It would be ideal for the user to start off their application experience by first going through the educational module (top button), followed by the interaction

with the 3D bone models (middle button), and finally finish with the testing of their knowledge with the flashcard game (bottom button). An information button is at the very bottom of the main menu screen if direction for use of the application should be needed. The goal of the educational module was to introduce each carpal bone and visualise how the etymology of its name is linked to the shape of the bone. Each bone module begins showing the entire wrist and distal ulna and radius bones, which then disappear to reveal the placement of the bone in the context of all the others. The bone is then enlarged and rotated while a script of the anatomy and articulations of the bone is sounded via voice over, and finally, an illustration of the etymology visualisation is shown beside the bone for comparison. The interactive features of these scenes are the home button to return to the main menu, scrolling script for accessibility purposes, and the next button to move on to the next bone. The Bones scene allows the user to explore the carpal bones’ anatomies and etymologies at their own pace and in the context of their placement relating to their surrounding structures. The user can read about each carpal bone’s anatomy and etymology by selecting from the tabs in the information panel, which will appear once a bone is

124

K. Nasrala et al.

Fig. 6.6 3D bone models created from segmented carpal bones, distal ulna, and distal radius

selected. The selected bone will be highlighted in blue, allowing the user to identify which bone is being studied, and where that bone is placed in the context of the surrounding bones. An example of this can be seen in Fig. 6.12, where the lunate bone is selected. The main goal of the application is to improve medical students’ learning and retention of carpal bone anatomy, and the flashcards allow them to test their knowledge in-app after going through

the educational module and 3D bones scene. The scene consists of eight flashcards displaying the 3D models of the carpal bones on the fronts of the cards (Fig. 6.13). When the “begin” and “next” buttons are tapped, a new flashcard will appear on the screen. The user can manipulate each bone by rotating it in place to view its different angles. When the “flip” button is selected, it reveals the name of the bone, as well as its etymology.

6

Visualising the Link Between Carpal Bones and Their Etymologies

125

Fig. 6.7 Process of finding holes in a model’s mesh and patching them. First, a hole is identified in the model’s mesh (left). Then, the border of the hole is selected using

the border selection tool (middle). Finally, the patch function is used to cap the hole (right)

6.4

implemented into the application to make final improvements.

Evaluation

Medical students tested the mobile application and completed pre- and post-tests, as well as a participant questionnaire at the end of the testing sessions. Testing was conducted to assess the educational aspect of the application and to validate the design and development process. User feedback was collected and analysed and later

6.4.1

Research Evaluation Methods

6.4.1.1 Materials and Methods Access to the participant group was granted by the University of Glasgow Undergraduate Medical School. Students in the first and second year of the MBChB degree at the University of

Fig. 6.8 Before and after refining of bone models imported from Slicer in Mudbox

126 Fig. 6.9 Bone model obj files exported from Mudbox into 3DS Max to do a final check of the meshes and polycounts

Fig. 6.10 Map of mobile application scene flow

K. Nasrala et al.

6

Visualising the Link Between Carpal Bones and Their Etymologies

127

of any previous anatomical or etymological knowledge. The results collected from the testing sessions, specifically the results from the pre-test, post-test, and participant questionnaire, were analysed using Microsoft Excel.

6.4.1.2

Experimental Protocol

Carpal Bone Pre-test and Post-test The participants were given a link to an online carpal bone test created by the researcher. They were tasked with identifying each of the eight carpal bones from individual photographs. The students were given 5 min to complete and submit the test and were not allowed to use additional resources outside of their own knowledge of anatomy. After testing the application, participants completed the same carpal bone test as a posttest to assess their knowledge of the bones. The same parameters were applied; they were given 5 min and were not allowed any additional resources. Fig. 6.11 User interface for the main menu scene of mobile application in Unity

Glasgow were invited to participate in the study. All participants were required to have access to an Android device and complete a participant consent form before taking part in the study. Testing sessions were hosted remotely via video meetings on Zoom. In the testing sessions, each participant was first given a carpal bone pre-test and then asked to use the mobile application on their own Android device and to finish a carpal bone posttest and participant questionnaire. The mobile application was developed for and tested on different models of Android devices. JISC Online Survey service was used to create and collect data from the participant information and consent form, pre and post-test carpal bone test, and the participant questionnaire. A USE-type questionnaire allowed participants to rate their experience and their thoughts on the mobile application on a scale of one to seven; these scores are defined in Table 6.2. Two participant screening questions were given at the start, concerning the establishment

Mobile Application Use After the pre-test, the participants began using the mobile application on their mobile device. They went through each of the scenes, including “Watch,” “Bones,” and “Flashcards” while still on the video call with the researcher. Usability Questionnaire At the end of the testing session, they completed a usability questionnaire, which was inspired by the usefulness, satisfaction, and ease of use (USE) questionnaire (Lund 2001), consisting of 20 questions focused on the user’s experience and thoughts on the mobile application. After this questionnaire was completed and submitted, the researcher asked if the student had any more questions or concerns before the video call was ended.

6.4.1.3 Ethics Approval Ethics approval for this study was granted by the Glasgow School of Art. All data collected in this study were collected and kept anonymous by the researcher.

128

K. Nasrala et al.

Fig. 6.12 User Interface and interactions with the 3D bone models. The lunate bone is selected and highlighted in the colour blue. The left image shows the anatomy tab

6.5 6.5.1

Results Participants

Two participants accessed the application using a Samsung device, and one used a Huawei device.

Fig. 6.13 Set of all eight carpal bone flashcards in Unity

of the lunate information panel selected. The right image shows the etymology tab of the lunate information panel selected

All three participants used mobile phones instead of tablets. One of the participants, a graduate teaching assistant (GTA), tested the application and provided qualitative feedback, which was only used to implement changes to the mobile application for the final version. They did not

6

Visualising the Link Between Carpal Bones and Their Etymologies

Table 6.2 Participant questionnaire number scores and their corresponding meanings

Score 1 2 3 4 5 6 7

complete the pre- and post-test quizzes, due to their background in anatomy. It is important to note that the testing pool of medical students was therefore extremely small (n ¼ 2). Two testing sessions were held in total, each hosting one medical student and the lead researcher of the study. The GTA did not attend a testing session.

6.5.2

129 Score meaning Strongly disagree Disagree Somewhat disagree Neither agree nor disagree Somewhat agree Agree Strongly agree

application, the post-test results showed that both participants had correctly identified all eight carpal bones (100%), resulting in a 68.75% increase from the average pre-test results. The results of the post-test compared to the pre-test are portrayed in the double bar graph in Fig. 6.14. The frequency of correct identification of each carpal bone in the post-test compared to the pre-test is displayed in Fig. 6.15.

Carpal Bone Pre-test and Post-test Results 6.5.3

Two carpal bone pre-tests were completed prior to mobile application use. Participants correctly identified two and three bones, respectively. Both participants correctly identified the lunate, with one participant also identifying scaphoid and the other identifying hamate and pisiform. On average, the medical students were only 31.25% successful in correctly identifying the eight carpal bones in the pre-test. After using the mobile Fig. 6.14 Double bar graph displaying the medical students’ pre-test and post-test results

Application Use

Although the researcher was present on the video calls to answer any questions or provide guidance, in both cases, the students did not need to consult with the researcher during their use of the application. The duration of the application experience took approximately 12 minutes for each participant.

130

K. Nasrala et al.

Fig. 6.15 Double bar graph displaying the frequencies of correct identification of the carpal bones in the pre-test and post-test

6.5.4

Participant Questionnaire Results

The results of the questionnaire are divided into four categories: usefulness, ease of use, ease of learning, and satisfaction. In these results, there are three participants included, as the GTA had completed the qualitative questionnaire in order to give feedback about the application.

6.5.4.1 Screening Questions Two out of three participants stated that they had previous knowledge of anatomy; one had three years of experience, while the second had four years of experience. All three participants indicated they had previous knowledge of medical terminology, while one also indicated an established knowledge of Latin. One participant had one semester or less of experience, while two participants had four years of experience. Usefulness This section included four questions regarding the usefulness of the mobile application, the results of which are displayed in Fig. 6.16. In general, the results for this section were high, with all statements being at least somewhat agreed with.

The final statement had the most variance; one somewhat agreed, another agreed, and the final participant strongly agreed that the application helped them remember carpal bone anatomy by using etymology. Ease of Use This section included four statements regarding the ease of use of the mobile application, the results of which are displayed in Fig. 6.17. All participants somewhat agreed to strongly agreed that the app was useful. Ease of Learning This section consisted of three statements regarding the ease of learning of the application, and the results are displayed in Fig. 6.18. Participants agreed or strongly agreed that the app was easy to learn to use. Satisfaction Four statements regarding user satisfaction were included in this section, and the results are displayed in Fig. 6.19. The participant scores were generally high, and all agreed or strongly agreed that the application made learning anatomy easy and enjoyable.

6

Visualising the Link Between Carpal Bones and Their Etymologies

131

Fig. 6.16 Bar graph showing the average questionnaire results for the usefulness section. Error bars represent standard deviations

Fig. 6.17 Bar graph displaying the average results of the ease of use section of the participant questionnaire. Error bars represent standard deviations

132

K. Nasrala et al.

Fig. 6.18 Bar graph displaying the average results of ease of learning section of the participant questionnaire. Error bars represent standard deviations

6.5.4.2 Qualitative Comments The final sections of the participant questionnaire included three prompts for any additional opentext comments from the participants. These included the questions:

elements such as arrows to the animations would better indicate which parts of the bones were being discussed during the modules.

1. What did you like most about the app? 2. What did you like least about the app? 3. Any additional comments?

6.6

The comments left by the participants were generally positive; a general consensus was that the application made distinguishing between the carpal bones easy and aided them in their learning. The participants suggested that adding

6.6.1

Discussion Summary of Findings

The sample size of the study was very limited due to the application only being available for Android devices. Many medical students had expressed interest in participating in the study and had signed up, but only two who had signed

Fig. 6.19 Bar graph displaying the average results from the satisfaction section of the participant questionnaire. Error bars represent standard deviations

6

Visualising the Link Between Carpal Bones and Their Etymologies

up had access to an Android device were therefore able to participate in the study, while the rest had IOS (Apple) devices. Therefore, adapting the application for use on IOS devices would benefit future studies by increasing the scope of possible participants. Although limited in number, the analysis of student pre- and post-test scores provides support for the project hypothesis. It was found that the mobile application aided medical students in their learning and retention of knowledge of carpal bones. They were able to understand and retain the information given in each educational scene in the application and managed to easily navigate the application without help, stating that the app was easy to use and made learning carpal bone anatomy simple and entertaining. The graphic design of the application was commended for its simple and minimalistic aesthetic, which contributed to the ease of use and learning. When comparing the pre-test and post-test results, a clear upwards trend can be recognised between the test scores. In the pre-test, medical students were only able to correctly identify an average of 31.25% of the carpal bones, with their individual quiz scores being 25% and 37.5%. Previous studies that have also used the carpal bone test to establish anatomical knowledge in medical students have yielded similar results (see sect. 6.1.2). An average increase of 68.75% in test scores was observed in the post-test, where both participants were 100% successful in correctly identifying the eight carpal bones. These results suggest that the mobile application does, in fact, aid medical students in their learning and retention of knowledge of the carpal bones. The participant questionnaire to collect feedback used a Likert scale to gauge participant responses on a scale of 1 to 7, allowing the participants to record their responses to a high degree of accuracy. There was a general consensus of user satisfaction with the mobile application and no negative feelings towards the validity of the application as an aid to anatomical learning. However, one must note that there is a possibility of bias that could have influenced the participants’ questionnaire answers. For example, if the participant knew the desired outcome of the

133

study, this could have skewed their results, either knowingly or unknowingly. Taking the results of the participant questionnaire as an indication of the success of the application, it can be assumed that the application has carried out its intended purpose and can be used as evidence to support the hypothesis of the study.

6.6.2

Limitations

The process of designing and developing the mobile application followed a logical and pragmatic approach, outlined in Fig. 6.1, allowing the timeline of the project to be well managed, while avoiding any unnecessary time loss, despite the many challenges. Due to obstacles presented by the COVID-19 pandemic, time constraints and limited access to real specimens caused a delay in the development and design process. Access to carpal bones held within the University of Glasgow Anatomy Facility was not possible due to COVID-19-related restrictions, so photos of the carpal bones were used from textbooks, specifically, Chap. 13 of the Human Bone Manual and 3D models from Sketchfab (White and Folkens 2005). Because of the lack of real reference bones, modelling the carpal bones proved to be time consuming, as reference for the shapes of each had to be taken from photographs. Another limitation of the study was the inaccessibility of the university campus, where all the specialty equipment for developing applications, like computers, mobile devices, and so on, are kept. Because there was no access to a Macintosh computer, the application was only able to be developed for Android. This caused a large decrease in the participants eligible to test the application on their own mobile devices, as most of the medical students who had signed up only owned Apple products. The limitations proposed by small participant numbers are the insignificance of the results obtained from the study. Although promising, the results cannot be considered significant without having tested on a larger participant pool.

134

6.6.3

K. Nasrala et al.

Post-evaluation Modifications

The results from participant questionnaires were analysed, and the feedback given was used to implement changes into the application for further improvement. The most significant changes made concerned UI elements of the application, for example, the placements of the UI elements, specifically the home buttons, next buttons, and information buttons, were adjusted so that they would appear more clearly on smaller mobile phone screens; this was not found to be an issue on larger tablet screens used during development. There were comments regarding the animations, claiming that adding arrows or highlighting the parts of the bone being talked about as each voiceover played in the modules would help students follow along with the modules. Given more time, this would have been another change implemented into the application.

6.6.4

Future Development

The results from the study are promising and can therefore provide reason to develop the research project further in the future. As discussed in the beginning of the chapter, previous studies regarding the impact of understanding etymology on learning anatomy have resulted in similar findings to this study. New cohorts of medical students entering the field of medicine could benefit from more research into this topic, as it could provide insight into new methods of learning anatomy that are simple and enjoyable. According to previous studies, preliminary feedback from anatomy instructors, and the participant feedback from this study, there seems to be a scope for an application such as the one created in this research project as a teaching and study tool for medical students. Furthermore, mobile devices are extremely common among today’s population, including medical students, making a mobile application like this one extremely accessible and easy to use. It is therefore important that further studies are conducted on this topic to further understand how teaching anatomy to medical

students can be improved through the use of etymology and digital learning tools.

6.7

Conclusion

The creation and evaluation of this mobile application with medical students have demonstrated the potential that digital learning tools can have in improving learning and retention of knowledge of complex anatomical concepts, like that of the carpal bones. The mobile application in this research project was developed with a clear concept in mind: to aid medical students in learning carpal bone anatomy by visualising the links between the bones and their etymologies. The use of animations, illustrations, and 3D models in addition to anatomical and etymological information allowed for medical students to improve their learning and retention of knowledge of the carpal bones, which was demonstrated through the pre-test and post-test results. The positive results in this study can therefore justify the scope of etymology visualisation through digital learning tools in the field of medical teaching. Mobile applications for teaching and learning anatomy can benefit medical students by providing anatomy content that is simplified, entertaining, and accessible. The research field would benefit from further study into the feasibility of mobile device integration into medical education and testing sessions with larger participant pools. The results presented in this chapter support the use of a mobile application that visualises the link between carpal bones and their etymologies to aid medical students in their learning and retention of knowledge of carpal bone anatomy.

References Bice MR et al (2016) The use of Mobile application to enhance learning of the skeletal system. Intro Anatomy Physiol Stud 27(1):16–22 Brewer DN et al (2012) Evaluation of neuroanatomical training using a 3D visual reality model. Stud Health Technol Inform 173:85–91. https://doi.org/10.3233/ 978-1-61499-022-2-85

6

Visualising the Link Between Carpal Bones and Their Etymologies

Brown AO (2014) Lexical access, knowledge transfer and meaningful learning of scientific terminology via an etymological approach. Int J Biol Edu 3(2). https://doi. org/10.20876/ijobed.92616 Drury NE, Powell-Smith E, McKeever JA (2002) Medical practitioners’ knowledge of Latin. Med Educ 36 (12):1175. https://doi.org/10.1046/j.1365-2923.2002. 01369.x Ebenezer DA, Mohanraj S (2020) Understanding etymology : awareness among doctors and a tool in successful medical education. Int J Med Sci Edu 7(1):8–11 Gilbert JK (2004) Models and modelling: routes to more authentic science education. Int J Sci Math Educ. https://doi.org/10.1007/s10763-004-3186-4 Haghshenas MSM, Hashemian M (2016) A comparative study of the effectiveness of two strategies of etymological elaboration and pictorial elucidation on idiom learning: a case of young EFL Iranian learners. Engl Lang Teach 9(8):140. https://doi.org/10.5539/elt. v9n8p140 Lerner EB et al (2000) Medical communication: do our patients understand? Am J Emerg Med 18(7):764–766. https://doi.org/10.1053/ajem.2000.18040 Lewis TL et al (2014) Complementing anatomy education using three-dimensional anatomy mobile software applications on tablet computers. Clin Anat 27 (3):313–320. https://doi.org/10.1002/ca.22256 Lund AM (2001) Measuring usability with the USE questionnaire. Usability Interface 8(2):3–6 Mayer RE (2002) Rote versus meaningful learning. Theory Pract 41(4):219–225. https://doi.org/10.1207/ s15430421tip4104 Murgitroyd E et al (2015) 3D digital anatomy modelling– practical or pretty? Surgeon Elsevier Ltd 13 (3):177–180. https://doi.org/10.1016/j.surge.2014.10. 007 Nicholson DT et al (2006) Can virtual reality improve anatomy education? A randomised controlled study of a computer-generated three-dimensional anatomical ear model. Med Educ 40(11):1081–1087. https://doi. org/10.1111/j.1365-2929.2006.02611.x

135

Pampush JD, Petto AJ (2011) Familiarity with Latin and Greek anatomical terms and course performance in undergraduates. Anat Sci Educ 4(1):9–15. https://doi. org/10.1002/ase.189 Pierson HD (1989) Using etymology in the classroom. ELT J 43(1):57–63. https://doi.org/10.1093/elt/43.1.57 Pujol S et al (2016) Using 3D modeling techniques to enhance teaching of difficult anatomical concepts. Acad Radiol 23(4):507–516. https://doi.org/10.1016/j. acra.2015.12.012 Ross A (1969) Etymology. Andre Deutsch, London Saidin NF, Halim NDA, Yahaya N (2015) A review of research on augmented reality in education: advantages and applications. Int Educ Stud. Canadian Center of Science and Education 13:1–8. https://doi.org/10. 5539/ies.v8n13p1 Sevinc A, Buyukberber S, Camci C (2005) Medical jargon: obstacle to effective communication between physicians and patients. Med Princ Pract 14(4):292. https://doi.org/10.1159/000085754 Smith SB et al (2007) Latin and Greek in gross anatomy. Clin Anat 20(3):332–337. https://doi.org/10.1002/ca. 20342 Spielmann PM, Oliver CW (2005) The carpal bones: a basic test of medical students’ and junior doctors’ knowledge of anatomy. Surgeon Royal College of Surgeons of Edinburgh and Royal College of Surgeons in Ireland 3(4):257–259. https://doi.org/10.1016/ S1479-666X(05)80087-3 Teoh B and Neo TK (2007) Interactive multimedia learning: students’ attitudes and learning impact in an animation course. Online Submission Valenza MC et al (2012) Comparison of third-year medical and physical therapy students’ knowledge of anatomy using the carpal bone test. Journal of manipulative and physiological therapeutics. National University of Health Sciences 35(2):121–126. https://doi.org/10. 1016/j.jmpt.2011.12.005 White TD, Folkens PA (2005) The human bone manual. Elsevier Academic, Amsterdam

7

Augmented Reality Application of Schizocosa ocreata: A Tool for Reducing Fear of Arachnids Through Public Outreach Rohne Nyberg, Matthieu Poyade, Paul M. Rea, and Jeremy Gibson

Abstract

This study aims to create a mobile application for public interaction around the subject of wolf spiders, specifically the brush-legged wolf spider. The hope is that the public will have a reduced level of fear towards arachnids when given a chance to view arachnids in a digital setting. To assist this, the application employs augmented reality animation, which has been shown to have a positive impact on the viewer’s interest and learning. With the opportunity to view a digital spider interacting with surfaces in a repeatable and informative R. Nyberg (*) Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK School of Simulation and Visualisation, The Glasgow School of Art, Glasgow, UK M. Poyade School of Simulation and Visualisation, The Glasgow School of Art, Glasgow, UK The Hub, Pacific Quay, Glasgow, UK e-mail: [email protected] P. M. Rea (*) Anatomy Facility, School of Life Sciences, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, UK e-mail: [email protected] J. Gibson Zoology, Kentucky Wesleyan College, Owensboro, USA Kentucky Wesleyan College, Owensboro, KY, USA e-mail: [email protected]

manner, viewers may find that their fears are more understood and can be controlled for both their benefit and that of the local arachnids. In order to accommodate multiple audiences, the application was set up to have a more cartoonish and simpler text as well as a realistic and advanced text. The area of information covered the brush-legged wolf spider’s anatomy, mating behaviour, and general safety practices for handling and avoiding wolf spiders. The modelling and animation were created using Zbrush and Blender, respectively. The programming and application creations used were Unity with Android and AR plugins. Keywords

Augmented reality · Public interaction · Brushlegged wolf spider · Animation · Schizocosa ocreata · 3D model · Unity · Zbrush · Blender

7.1 7.1.1

Introduction Background Review

Augmented reality (AR) has shown promises for learning in many fields ranging from museum exhibits to new ways to learn in the classroom. As a digital technology, AR has all the advantages of computerized education, being adaptable for both mobile and computer

# The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. M. Rea (ed.), Biomedical Visualisation, Advances in Experimental Medicine and Biology 1334, https://doi.org/10.1007/978-3-030-76951-2_7

137

138

programs. AR expands upon this by visualizing real-world interaction with digital objects. Such interaction has been shown to increase a student’s ability to learn difficult concepts (MartínGutiérrez et al. 2017). This is especially true when compared to textbooks, which students often have difficulty understanding and visualizing, particularly with more complicated subjects (Saidin et al. 2015). Even when those who find digital reading distracting or irritating and prefer the use of print are taken into account, there is a large majority who learn better through digital means (Sun et al. 2006). The ability for objects to be viewed and manipulated can better portray subjects that are often unable to be wholly visualized in a 2D circumstance; this is magnified when paired with the interest garnered by AR (Krause and Santos 2017). When objects or features are too small to see with the naked eye, having a digital object can show these features in greater visual detail than a viewer might normally be able to see (Saidin et al. 2015). This same effect is demonstrated in public interaction at museums where AR has been used to reduce the necessary text on exhibitions and provide more adaptable options and imagery (Ding 2017). These examples of AR have been found to have increased effect when structured in a game or game-like setting, stimulating the viewer and assisting with retention of information when tested later (Selviany et al. 2018). As measured at the Heinz Nixdorf Museums Forum, 74% of individuals chose AR-enabled devices over the older alternative method when viewing an exhibit (Huang et al. 2011). The ability of AR to positively affect student learning extends into the public outreach of the exhibition. It was found that students had an easier time learning when AR was used in conjunction with exhibits (Wu et al. 2013). Furthermore, when applied to zoological information, the use of AR allows for the viewing of a digital surrogate in place of going into active habitats to find specimens, potentially disrupting them and causing damage to the local ecology (Kozak and Uetz 2019). Multiple studies have shown that zoological applications that employ animation and interactive methods have the potential to create a high level of interest in

R. Nyberg et al.

students and maintain retention of the information learned (Perry et al. 2004; Selviany et al. 2018; Munoz-Cristobal et al. 2018). This same potential is shown to work in lab studies, where AR and interactive applications have been proven to increase immersion and give a feeling of control and focus when working (Florea et al. 2019). While the price of AR and VR technologies is often expensive for many, this barrier is lessening as technologies became cheaper over time (Saidin et al. 2015; Krause and Santos 2017). Furthermore, AR and VR applications can be created for use on smartphones, significantly dropping the cost of the technology, as most users already own a device capable of running such programs. When AR is used with mobile devices, the smartphone’s camera is used to capture the environment for the program. This makes the application both adaptable and portable, with the user being able to change location and environments, potentially changing the experience. Often users are more comfortable using an application when it is on their device; they already know how to use the hardware and know its capabilities. AR can be programmed to work for both Android and IOS systems extending its reach to most smartphones (Kesim and Ozarslan 2012). When MunozCristobal et al. (2018) created a study around the effectiveness of a mobile game that was intended to make students invested by bringing them to the location they were learning about, an AR application was their choice. The game used the phone’s GPS location, and the use of discreet QR codes to make a scavenger hunt of various sites, when the code was scanned, displayed information about the site. The goal was to have students become more engaged in on-site learning and eased the input of information at these sites. Many subjects such as history can benefit from on-site learning to build student interest and learning. Mobile AR proves to be a boon to such tasks. The one complaint found when using this technology was the need for technical support of the many devices used. Many students did not know how to use the devices that were supplied, and this taxed the instructors. Such a problem is often less extreme when used on the user’s personal devices. This method was tested in a study to find if AR could

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

support the need of museums to create adaptable exhibitions that do not require a change to the supporting infrastructure such as displays. Through the use of QR code-based AR, visitors could use their devices to scan a code and view the display information, preventing the need of museums to replace text displays when an item or exhibit piece was changed out (Ding 2017). Such examples show the potential of AR applications to affect and improve the learning experience of users in a positive way. Using such technology could create an opportunity to help change the perception of arachnids.

7.1.2

Rationale

Arachnids often incite fear and disgust in many individuals. This inherent bias can spread and affect actions beyond simple opinion. Studies have shown that even in education the bias can have consequences, resulting in lessons featuring arachnids being shortened or entirely skipped (Wagler and Wagler 2017). This only serves to further enforce the mistrust of arachnids and a general lack of public knowledge about them (Hebets et al. 2018). Spiders rank in the top five most feared animals in the world, garnering distinct aggression towards them (Gerdes et al. 2009). For the application, Schizocosa ocreata, the brush-legged wolf spider, was chosen to be the species featured. Because of the application’s nature as a tool to reach the public, the brushlegged wolf spider is ideal for acting as a digital ambassador due to its habitat range overlaying with the app’s target area. The brush-legged wolf spider has a broad range and may be found in temperate forests of the Eastern United States, wandering and hunting in the leaf litter. It is so named for the male’s characteristic bristles that adorn its front-most pair of legs. These adornments on the legs of males are used in courtship displays (Scheffer et al. 1996; Hebets and Uetz 1999). During bouts of courtship, males wave their legs that are adorned with these brushes (Uetz and Stoffer 2015). In addition, females of closely related species appear nearly

139

identical and often differentiate males of different species (with overlapping habitat ranges) by male courtship behaviour (McClintock and Uetz 1996; Scheffer et al. 1996; Hebets and Uetz 1999; Stratton and Uetz 2014). Males tend to have darker leg colouration on their front pair of legs, while females tend to have light tan legs (Fig. 7.1). There is little sexual dimorphism in size between males and females, although females are just slightly bigger in size on average (Uetz et al. 2002). Courtship behaviour is usually initiated when a male encounters silk draglines left behind by an adult female. The silk contains pheromones, which stimulate mating responses in the male. The courtship display of male brushlegged wolf spiders is collectively called “jerky tapping” because it involves four different behaviours, arching, cheliceral striking, stridulation, and double-tapping. Arching describes the motion where the male lifts one of his front legs upwards in a curve, creating an arched curve (potentially showing off the bristles of the foreleg). Usually while engaging in arching, the male is also simultaneously generating stridulations. Stridulation results in the generation of a substrate-borne vibrational signal. Stridulation is generated by males flexing the tibiotarsal joint of the pedipalps, which contains a file and scrapper (Rovner 1975; Elias et al. 2006). Cheliceral striking is when the wolf spider’s fangs are struck against the ground by dropping its cephalothorax to the substrate, this

Fig. 7.1 An image of a Schizocosa ocreata male with focus on bristled limbs (Uetz and Stoffer 2015)

140

R. Nyberg et al.

action also imparts vibrations into the substrate. Double-tapping is that the wolf spider male uses its front pair of legs to tap the ground while moving forward, but this visual component is not known to generate any additional substrate vibrations (Uetz et al. 2002; Hebets 2005). Females choose males based on a combination of their multi-modal signal and the information contained within those signals about male quality (Hebets and Uetz 1999; Gibson and Uetz 2012). Female receptivity is closely gauged by males as females will often attempt to eat males that fail to notice their aggression and try to enter into courtship. The level of aggression is often measured by the amount of pheromones left in the female’s silk and visual cues. It was shown that in the dark, males had trouble assessing the level of females’ aggression (Taylor et al. 2005). Males are known to be highly aggressive during copulation, grasping with their fangs during mating, which can puncture the female’s carapace and leave other wounds in efforts to assure successful fertilization (Johns 2007). Despite this aggression being common, research shows that females actually maintain a higher level of receptiveness after mating with less aggressive males, meaning that they are more likely to allow for further mating in the future (Aisenberg and Costa 2004). When taken together, all of these incredibly complex communication behaviours of the brush-legged wolf spider can be displayed through AR, bringing this world to a new audience. When added to the information on how to properly avoid and handle any encounters with arachnids, it may further the application’s effect to lessen fear. Such information can be avoiding an area of heavy leaf litter in a forested area (stay on well-travelled trails), the main habitat in which the brush-legged wolf spider lives and hunts. Regardless of where you find yourself, if a spider does appear on your skin, it is best to brush the spider off instead of attempting to crush them (Diaz 2004).

7.1.3

Objectives

This chapter presents research that aims to build a body of knowledge to accurately depict the

Schizocosa ocreata in an AR environment as well as to build the supporting content that will be used to engage and educate the public with enough depth. This will involve detailed images to have a comprehensive view of the wolf spider’s anatomy, along with a review of the literature describing specific behaviours, with a focus on courtship, which has a series of complex signals that may be unique and unknown to the public, such as its multiple methods of substrate-borne vibrations. Then, this chapter looks at a methodological and technological framework to support the creation of a mobile application. This framework includes a detailed storyboard of the application’s flow from the menu to exit with all possible viewer paths explored; and the design of a complete outline of the information and details needed for the application to be accurate to life. This chapter shows how 3D models to be incorporated for a public outreach and AR application. The models will be made at different ranges of detail, from cartoonish to realistic, allowing for a broader potential audience to be reached. Finally, this chapter works to function as a mobile application. The application will use Unity Engine to construct the UI and program the necessary functions. The application will need functionality to play an animation on command, the ability to select different versions of the application from a starting menu, and a UI capable of showing all the required information.

7.2 7.2.1

Methods Application Purpose and Goal

This application intends to create a way of teaching the public about their local arachnids and increase an understanding of them. It is the hope that in doing so, the prevalent fear around arachnids, and in the case of the application of the brush-legged wolf spider, will not be as pronounced. In order to be effective in reaching a public audience and thereby delivering the information and visuals, the application needs to be adaptable and suitable for a wide range of audiences. This

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

translates to having the application be built to work on devices that have a large outreach, such as mobile Android, which can be used on multiple different smartphones as well as tablet devices. This makes any outreach programs able to use the application on easily portable devices, as well as the ability to be downloaded and used by any of the potential public wanting to take a look themselves. In addition, to be viewable by as many people as possible, the application needs to have the potential to switch between more accurate information and visuals and an alternative for those that are younger or more afraid of arachnid

141

imagery. This means creating a secondary set-up with the application that allows for a selection of visual and textual preference and building secondary models and animations. Though this factors to an increased amount of work in the application’s creation, it makes for greater benefits in its use for public outreach, generating a greater opportunity for the purpose of the application to be met.

7.2.2

Materials (Table 7.1)

Table 7.1 Software and hardware used Software

Use High poly sculpting of S. ocreata models

Publisher Pixilogic www.pixilogic.com

Texturing and animation of high poly models

Blender www.blender.org

Application development

Unity www.unity.com

AR development add-on for Unity allows for non-targeted surface AR

Unity www.unity.com/unity/ features/arfoundation

Android Studio Software Development Kit provides the ability to create and build Android executable applications

Android Developer.android.com

A vector-based 2D digital application used for the creation of sprites within the application

Inkscape www.inkscape.org

AR Unity development add-on that adds additional support to Unity

Google developer.google.com/ar

Smartphone for testing the application during its development. Android v10

Google Local Mobile Phone Provider

ZbrushCore

Blender

Unity

ARFoundation

AndroidSDK

Inkscape

ARCore Hardware

Google Pixel 2

142

7.2.3

R. Nyberg et al.

Design and Development

7.2.3.1 Unity Basic Set-up In order to begin the creation of the application, a frame must be created, proceeded by digital assets to fill it (Fig. 7.2). To build a frame, the application begins in Unity where a basic scene UI is set up. First, a skeleton of buttons is created so that the application’s flow can be followed for the final UI. This translates to designing where text and images will be placed and how the buttons trigger them to be hidden or revealed as is necessary for the information and visuals. After the skeleton is created, the UI imagery can be built using vectored png images that can be converted into sprites usable for the application. The UI uses a forested theme to match with the brush-legged wolf spiders’ habitat. This theme was intentionally kept neutral and basic to accommodate both younger and older audiences. This is set up in a branching path where the user begins at a title screen, allowing them to continue into the application or to close it. To create a more impactful experience for the user, it is better to have a more tailored experience. To do this, the application has an age and content selection page where a preference for more cartoonish and simplified content can be selected or a more advanced and realistic content page. This is to ensure that an individual can use the application even if they do not want to be exposed to the realistic imagery of spiders. Following this screen, there is a warning allowing the user to make an informed decision about proceeding with the possible fearinducing or otherwise uncomfortable content within the application. This second confirmation is to make sure that an individual is sure about using the application, even if they have already chosen a preferred set of content. Once all content and consent have been selected, the user is brought to the main screen, where they may place the first of the AR models and move to the three content pages of safety, anatomy, and courtship behaviour. This is where the user can move through the application’s information and visuals.

Fig. 7.2 A diagram of the workflow for the creation of the application and its design

7.2.3.2 3D Modelling In order to create 3D digital assets, they must be built in a 3D modelling program. Zbrush was used as the main modelling program. It was used with image reference provided by Dr Jeremy Gibson to act as a comparison with which to build accurate anatomy. The imagery featured a Schizocosa sp. male and female as the males are similar in anatomy and appearance, lacking the iconic bristles of Schizocosa ocreata and possessing a slight difference in colouration. The females within the genus appear with nearidentical colouration and anatomy. Pictures of a male of S. ocreata that was preserved in alcohol was also provided. Modelling began with constructing the basic shape of the brush-legged

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

wolf spider out of basic shapes using the sub-tool feature of ZbrushCore. This tool enables 3D models to exist on separate layers, meaning that they can be visualized together without interacting (Fig. 7.3). Using this, the basic shape of the arachnid could take form. This resulted in a merged sub-tool of the cephalothorax and abdomen, the left pedipalp, and left four legs to be mirrored after details are completed. This set-up of sub-tools allowed for control of size and editing without affecting other important parts of the anatomy. This made it easier to build up proper proportions and create larger detail such as the addition of the brush-legged wolf spiders’ multiple eyes and the creation of the left chelicerae and fang. From this state, each sub-tool was brought to a higher level of polygons through the use of subdivisions. Using this method instead of using dynamism meant that the overall number of polygons for the final model could be better controlled from the beginning. At this higher number of polygons, smaller details could take shapes such as defining the leg joints and the chelicerae. It was at this point that the models began to diverge, first into male and female models. The female model is bigger and lacks the swollen pedipalps and areas of bristles that are on the male, which meant that only a slight adjustment to the proportions of the abdomen was needed. The male required editing of the pedipalps to swell and grooves for stridulating. The area of the front-most legs that will have added bristles was changed in order to match the slight swelling seen in its real-life version. After the male and female were fully detailed, the models diverged again, this time into a high poly realistic model pair and a lower poly cartoonish pair. This was to match the UI frame built into Unity and provide assets for the two versions of the application. The cartoonish versions feature inflated proportions, such as more ballooned legs and overly large abdomens, in an effort to make them less threatening. The eyes were also intentionally enlarged, but only to a slight degree to prevent any visual confusion with jumping spiders, which ordinarily have much larger eyes than wolf spiders. The realistic versions utilized higher polygon counts to allow for greater exoskeleton detail, such as the use of geometry to

143

Fig. 7.3 An image of the sub-tool feature within the ZbrushCore program. Basic shape constructs can be on different layers

create the small hairs across the abdomen. Once all four models were completed, they were each imported to Blender as obj file types for texturing and animation.

7.2.3.3 Texturing Once in Blender, the first step towards building a model’s texture is to unwrap, also known as UV unwrapping, the polygons into a flattened state. This allows for an image to be projected onto the polygons, effectively “skinning” the model in colours that act as texturing and shading. To unwrap in Blender, there are multiple methods and tools that offer varieties of control and speed. For the models, automatic UV unwrapping was chosen as it was very quick to do and only required a slight adjustment of the flattened

144

texture to create an acceptable flattened polygon map. This map was placed onto a 2048
2048 texture for the models to allow for a high level of texture detail, as the increase over the standard 1024
1024 doesn’t increase the load on processing significantly, especially with a maximum of two to three models on screen at any given time. This process was repeated for each model. After unwrapping and mapping were completed, Blender’s texture painter tool was used for texture generation. This tool translates the vertex coordinates of the model to flattened polygons for the texture map. This means that through the use of brush tools, the 3D model can be painted and have the colours placed onto the corresponding image texture created earlier. This allowed for controlled painting without needing to constantly check between painting on a 2D image and its skinned model. The realistic models were textured to have colours and shading to match as close as possible to their real-life counterparts. For the female model, this meant having a reddish-brown abdomen and cephalothorax, with ruddy yellow legs. The male has a black abdomen, cephalothorax, and front pair of legs, with the last three pairs having increasing ratios of sand yellow colouration as they move towards the back most legs, as well as a white stripe along the top of the cephalothorax and abdomen. The cartoon models followed the same colouration patterns with less detail and variation to attempt to bring it further from their realistic counterparts. After texturing was finished, a grouping of long triangular pyramid geometries was aligned in cylindrical patterns and pointed outwards to form the bristles for the Fig. 7.4 The realistic male model of the brush-legged wolf spider. Textured without bristles

R. Nyberg et al.

male models. Once finished with texture painting, the resulting image textures were added to the material tool, and additional parameters were set up for the realistic and cartoon versions. The realistic versions used a more gloss finish, mainly with clearcoat and metallic options. This created a reflective quality that appears similar to the shiny gloss of the actual chitin of their exoskeletons. A slight transparency filter was also added to copy the transparency of the chitin (Fig. 7.4). The cartoon models use a much more matte finish mainly derived by lowering the metallic values. This allowed for the models to appear more toy-like in quality; no transparency was added to build this effect (Fig. 7.5). Once texturing was fully completed, bone armatures and rigging had to be built and weighted before animation could begin.

7.2.3.4 Rigging and Animation Rigging is the process of placing “bones” on a model as a tool to deform the topology in a controlled way, allowing for animation. Bones are empty objects that do not render or show in a program’s camera. They are effectively two points, an originating point and an ending point, and can be treated as a sort of hybrid between an anatomical bone and muscle. For simpler armatures such as the one for the wolf spider models, the bones only act as a set-up for mimicking the joints of the body in question. Bones can be linked together, with each bone’s ending point acting as the parent of the next bone’s origin. To begin an armature, a root bone is placed acting as the foremost parent, each other bone in the armature branches off this bone. For

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

145

Fig. 7.5 The cartoon male model of the brush-legged wolf spider. Textured without bristles

the wolf spiders, the armatures of the realistic versions had 70 bones and the cartoon version had 67. The extra three bones in the realistic versions were placed along the cephalothorax and abdomen in case greater control was needed in these areas. The armatures were set up so that each segment of the arachnids had a bone that allowed it to move separately. The fangs and pedipalps received the same treatment and were parented directly to the root bone, which was located at the front of the cephalothorax. Once armatures were completed, they were weighted via Blender’s automatic weighting tool, which is very effective at creating accurate weight maps for armatures. A weight map works similarly to a texture map, except that there is no actual image; instead, the colour is used to visualize the values assigned to the model’s vertexes. This value corresponds to how much that vertex moves with the bone; a high value means that that vertex moves directly, whereas a lower value lags behind. When these values move in a gradient across the model, it creates smooth deformation of the topology, imitating realistic movements. With weighting, finished animation can begin; these animations were set up using cycling animations. In animation, cycles are a series of keyframes that can be repeated to create a continuous motion. The first cycling animation was a simple walking animation; this allowed for the spider’s path to be followed while showing the spider to appear as though it was walking. This walking cycle acts as the base from which all other animation cycles will be layered onto, making sure that motions will lead to one another. After the base cycle was completed, the various

mating behaviours could begin to be animated. There were four animated motions: arching, cheliceral striking, stridulation, and double-tapping. These were adding to the animation at semirandom points to imitate the varying actions and frequency that the male performs. Each behaviour was set up in its own keyframe cycle and placed onto the dope sheet as needed. The animation itself is meant to pair with text explaining each action and its purpose during the male’s mating behaviour. Once the animation was completed for the male models, all of the Blender assets were exported as fbx file types and brought into Unity for use in the application.

7.2.3.5

Augmented Reality Development In order to make the Unity engine capable of creating AR applications, plugins must be added that allow for the generation of tools that can interface with a device’s camera and contain pre-built programs that allow for the recognizing of planes or targets that are used for digital object generation. For this application, the plugins of ARCore and ARFoundation were used, which are free plugins that come with their own custom scene managers to assist in the use of AR. Non-targeting AR was used as the form of AR within the application, meaning that instead of a specific image tracking digital actions such as the spawning of an object, the camera tracks flat surfaces forming digital representations of them that can have digital interactions. This allows digital objects to be able to appear as though they are resting upon physical surfaces. This is done through the use of point clouds, which uses

146

the camera to recognize similar points stretching a plane across them, making a digital surface that mimics the surfaces that the camera sees. It is this method that allows for the wolf spider models to be placed on the screen and for the animations to move along surfaces. Once all of the AR scene managers were properly set up and the Unity project had the correct setting for Android builds, the physical programming could begin. First, the script for the user’s ability to place the anatomy models (unanimated and rotatable) and the full animations needed to be created. This script allowed for the user to place an object onto a surface plane by touching the screen, thereby choosing its location as well. The script also prevents objects from being placed multiple times, which often could otherwise occur with long-duration touches or mistakenly doublelayered planes. From this base, a nested structure of script controls was built, which means that If-Else statements were placed within one another to create a branching descent of controls, much like how parent and child controls work in animation. First, the script searched if certain empty objects were active in the scene. These empty objects were turned to active only if specific buttons were pressed in the UI. This allowed for control of whether the cartoon models and animation were placed or the realistic versions. After this, the prefabs of the models and animation were given specific tags. If the second set of empty objects were active or not, it allowed for prefabs with the corresponding tags to be spawned (instantiated). If the second tier of empty objects were inactive, the anatomy models could be spawned; if active, the animation was allowed to be spawned. By using this layered If-Else statement set-up, a series of gates are created with the script, gates that can only be unlocked by the user going down a specific path. This prevents a user from choosing the simplified and cartoonish version from accidentally spawning realistic models and vice versa. After the different gates were created within the script, an integer counter was added to measure if a model was placed. This was used to prevent a user from infinitely spawning models and potentially crashing the program. As the integer was

R. Nyberg et al.

less than 1, a model could be placed; if a model was placed, it set the integer to 1, thereby preventing further spawning. If the previously mentioned empty object that allowed the animation to be spawned was active, the integer reset and allowed the animation to be spawned (Fig. 7.6). An additional script was added to the prefabs of the anatomy models. This measured the first touch on screen to the location of that touch as the finger moves, effectively measuring the location of a finger’s dragging motion. With this information, it causes a relative change to the rotation value of the anatomy models, specifically on the Y-axis, allowing the models to be spun as though on a turn-table (Fig. 7.7).

7.2.3.6

Implementation of Textual Information The text within the application is how the user actually obtains the information about the brushlegged wolf spider as well as how they are directed through the application. Naturally, this makes it a very important part of the development process. Other than directional text, which describes the consent pages, the majority of the text in the application is built up in three different menus: the safety page, the anatomy page, and the courtship page. This text is what is changed between the simpler and more complicated versions. Although the content was kept the same between versions, it was the level of language used that changed and how detailed the information was. This was most distinguishable in the anatomy and courtship pages, where any terms that could not be generalized were explained in simpler vocabulary. This was to assist any younger audiences that chose the cartoon version. The anatomy page acts to explain the various parts of the external anatomy of the brush-legged wolf spider; it is accompanied by rotating models of the male and female versions so that the text matches a visual. The courtship page gives general knowledge of the mating of the brush-legged wolf spider, as well as the specific actions the male takes during courtship. This text is paired with the animation, which takes the various courtship actions in a loop around whatever surface the user has chosen. With everything

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

Fig. 7.6 The script dictating the control of AR object spawing

147

148

R. Nyberg et al.

Fig. 7.7 The script dictating the rotation of AR anatomy model

finished, the application is created by Unity as an apk file that can be used on any android device.

7.3

Results

The application follows a branching structure that the user can navigate to find all the information and feature of the application as well as allowing for control of content (Fig. 7.8). A regular play through the application begins at the title page. This page serves as an introduction and a way of crediting those involved in the project. If a user intended to enter the application, they can view the page and progress through the application from there; if not, the title page has an exit button that allows for the application to be closed without progressing any further. The title page shows an image of the brush-legged wolf spider with the three logos of the Glasgow School of Art, the University of Glasgow, and Kentucky Wesleyan College, which helped to support the creation of this application. Other than an exit button, the title page, also known as a splash page, has two other buttons: a play button to move to the next page or a credit button. The credit button contains acknowledgements and credits for the individuals responsible for creating and assisting in the creation of the application as well as an additional form of recognition for the schools involved. After the user proceeds to the next page, they

Fig. 7.8 A diagram of the paths through the application available to the user

will find the version selection page. This is one of two consent pages that the user must read and go through before being allowed to see the content within the application. In the version selection page, there are explanations of each version, how the realistic version may be too much for those who find arachnids frightening or are of a young age, and clarifies the cartoon and simplified version for those who require it. This page does ask that if a user is under 12 that they

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

choose the cartoon version as it was decided that any younger audiences should be restricted to less frightening imagery. Each selection leads to a consent form that once again warns the user of potentially frightening imagery and content and asks if they would like to continue forward, giving them an exit button if they would like to close the application. After continuing through all of the consent pages, the user is brought to the main menu page of their respective versions. These pages contain information on how the user can make use of the AR models with instructions on the right side of the screen. These instructions detail the finding of a surface, making sure that it is large enough in size to be able to fit the models and to make sure the user sees a series of dots forming signalling that the camera is able to register the surface. It is here that the user is allowed to place their first model for AR, the anatomy model. This is so that they can get a feel for the use of AR, and as this model does not move except for rotation, it is less distracting as a first model. To the left of the page are the three directory buttons that lead to each version’s respective safety, anatomy, and courtship pages. Once entered, there is a menu button that will

149

allow for re-entry on this page (Fig. 7.9). At the top left is an exit button that leads the user out of the application. The safety page is the only informational page that lacks AR representation. This was because the varied situations and circumstances described in the safety page would have made it very difficult to build 3D or animation visuals. Instead, the visuals were 2D vectored images to represent the circumstances. The information within the safety page is split into two columns, one regarding how to handle spiders when encountered, the other with how to avoid encounters entirely. When handling an encounter with a wolf spider, the recommended action is not to smack or crush the spider but to brush it away with a sideways motion. This not only prevents the unnecessary killing of the spider, but it also prevents any bites as most wolf spiders have fangs that face inwards and are unable to bite human skin unless pressing into contact (Diaz 2004). To avoid encounters, it is recommended that, first and foremost, try not to enter a wolf spider’s habitat, the detritus, and leaf litter of forest floors. Second, frequently cleaning darken areas of homes with dust and debris also helps deter the presence of wolf spiders in your

Fig. 7.9 The main page with AR instructions and directory buttons

150

R. Nyberg et al.

Fig. 7.10 The anatomy page with AR models and anatomical information buttons

home (Diaz 2004). The anatomy page corresponds with the two rotatable models of the male and female of the brush-legged wolf spider. On this page, the various pieces of external anatomy are broken down. There are seven sections of anatomy whose information is viewable when its button, at the left side of the screen, is pressed. The sections are as follows: the chelicera, the pedipalps, the legs, the tagmata, the spinnerets, the eyes, and the lyriforms, which make up much of the major anatomy that can be seen on the exoskeleton of the wolf spider. The information for each section is shown in a panel to the right, beginning with an introductory message explaining the page and its use. The button related to each segment closes the introductory message and brings up the relevant information as well as a new button that can be used to close the text and bring back the introductory text (Fig. 7.10). This text is one of the more significant variations between versions with the cartoon, below 12 versions being simpler in content. The courtship page is the second page of major content within the application as well as the corresponding page to the animation. Within the courtship page, there is information in a panel to

the right, the same as the anatomy page. This is also the page that triggers the mentioned empty object that allows for the animation to be placed onto a surface. To the left, there is introductory text on what the courtship page is and instructions on good surfaces for the AR animation to be placed on. The animation shows the various parts of “jerky tapping” performed by the male brush-legged wolf spider as it progresses around the room. The panel on the right can be navigated to view the information surrounding the courtship behaviour, going in-depth on the parts of “jerky tapping” such as arching or double-tapping. Though there are only two pages that involve AR visuals, its impact and relevance make it incredibly important to the overall design and effect of the application. The ability of AR to create interaction and imitate real objects is important for an application that seeks to help an audience understand an animal that could likely be encountered in their local environment. A 3D model that is only seen in a single viewpoint and perspective can only provide a fraction of the visual information that can be gained from being able to see such a model in a live environment with imitated movement, one that can be

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

viewed from multiple perspectives. When dealing with anatomy and behaviour, the ability to view detail from multiple angles and in a realistic setting can have an immense impact (Krause and Santos 2017). The hope is that when these impactful visuals are paired with understandable text, the information will be retained and help build better relationships between the public and their arachnid neighbours.

7.4

Discussions

The completed application met all of the necessary requirements that it was set out to fulfil in its conception. It works only on Android mobile devices by being buildable as an apk file. The application can inform the public as it covers the general anatomy, courtship, and safety. It was also important for the visuals to involve AR to assist in the impact and retention of the information the application provides; this was done by the use of AR with the anatomy models and the animation of courtship behaviour. In order to be helpful and useful as a public outreach tool, the application needed to be able to adapt to audiences of different ages and mentalities; this is especially integral when the subject matter is something so commonly feared, such as arachnids. This adaptability was covered by the creation of two versions, one that was realistic and held a higher level of information, and a more cartoonish version, one that can be more suited for those who are more easily frightened or younger such as those under the age of 12. This paired with the safety information and the ability of AR to act as a safe surrogate to interacting with real wolf spiders creates the potential for an application that can reduce fear and anxiety that arachnids cause in many people.

7.4.1

Limitations

Multiple points of the application lacked refinement or could be improved to increase the performance of the application. The biggest quality that needs refinement is the smoothness of the

151

animation; underperforming equipment prevented proper viewing of the animation during its creation and significantly slowed rendering, creating a span of loss time. This meant that the animation had to move forward in a state of polish that was less than desired for the level of quality in the application. Due to the preview of the animation being faulty and slow, it meant that in order to properly view the animation in a way reflective of its actual end result, the animation would have to be rendered and compiled into a video for review. But due to the same hardware constraints that caused the issues with the animation preview, the renders would subtract too much time from the building of other aspects of the application. This did not affect the accuracy of the content of the animation, but it does cause a stuttering motion when viewed, which can be distracting to the viewer. The differences between models could also be given further work after review in their use in the final application. Once used in the scale of the application AR, the differences between the realistic model and the cartoon models are lost. This made room for the cartoon models to become less realistic and increased cartoonish proportions to try and heighten the differences and curb any fear they may still cause. The realistic models functioned well in their role, acting as reliable digital double to their real-life counterparts. The UI buttons could use further work; although no mistakes were made in their creation and they function for their intended purpose, some pathways between pages and content are inefficient, and with more time, they could be worked into a smoother interface for the user. As it stands, unless the user makes unusual choices through the applications, these inefficiencies will not be visible, but at certain times, a button will not close or remove anything despite being present, as another button already did so. In these cases, a reworking of the UI could assist in the application’s overall usability, such as exchanging multiple buttons that close down different blocks of text, thereby opening room errors when multiple buttons are pressed quickly or in a strange order; instead, a singular button could be exchanged and used to open all relevant text

152

R. Nyberg et al.

eliminating such potential errors. The sprites used for the UI were satisfactory and needed no further editing to create a significant improvement to the application, but if the other aspects were refined and improved, the sprite could have altered vectored images for a more compelling forested theme. This would enhance the overall appearance of the application and hopefully build better immersion and interest when a user is moving throughout the application. In order to properly ascertain the true effectiveness of the application, it must undergo testing by individuals un-associated with the application’s design in order to create feedback that is unbiased and diverse. Unfortunately, circumstances meant that volunteers were not able to undertake testing activities.

7.4.2

Future Development

In the future, a series of changes could be made to build up the application to a better state. The biggest improvement would be to find willing participants to find unbiased data about the application. This would provide valuable input on what is wrong with the application and what needs to be changed in order for it functions as a public outreach application. Other notable changes that could be made in the future would be the possible improvements to the structure and visuals of the application mentioned earlier. Fixing the animation would not only bring a smoothness to the visuals but allow for further action to be taken to highlight the different parts of “jerky tapping” that the male brush-legged wolf spider takes during mating. This could be done purely through animation, perhaps through more exaggerated movement, but this risks inaccuracy. Instead, shaders could be used to control the colouration and emission of the model during the time that actions are taken, outlining or bringing attention to important parts and movements, such as the pedipalps during stridulation. Potentially, the animation could be edited and programmed to imitate the random nature of the male’s jerky tapping as well. This would require extensive effects as each action would have to be

animated separately, and programmed or altered in such a way that they blend together, no matter the combination. They would then have to be programmed so that each animation is randomly placed into the pattern fluidly and without the loss of the overall motion and path of the model. Though difficult, this would build not only a more accurate animation but a more visually stimulating one. It could greatly improve the audience’s interest in the information involved and perhaps their retention as well. The visual shader effects could greatly enhance the anatomy page, enabling specific anatomy to emit light or change colouration when its related information was selected. This could be taken even further with additional “zoomed in” models of details that are difficult to see on a scale model due to their size in relation to other anatomies, such as lyriforms. These smaller details could have secondary models with increased size and more room for details, which are only shown when their information was selected. A further effect for future upgrades to the models would be to use fur shaders for the male’s bristles and the exoskeleton of the models. This was originally intended to be used in this application, but the chosen fur shader was found to be incompatible with the model when it was attempted to be used. Despite its inability to be used currently, finding a compatible fur shader or altering the models would allow for highly realistic visuals and additional opportunities to create a more friendly appearance in the cartoon models without significantly altering important anatomy. The cartoonish model is meant to be an ambassador for audiences who are more sensitive to arachnid imagery; therefore, having a friendlier model is better as long as the anatomy is maintained and is accurate. This can be a difficult task to make more changes than the proportional and textural changes that have been made, as any major deviances break from the characteristic that is important to keep a visual representation of the brush-legged wolf spider. Fur shader is a good way of resolving this issue as it can alter the texture and overall appearance of the silhouette of the wolf spider without changing the underlying anatomy. The visuals of the safety page could

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

also be taken further in future projects. As it stands, the safety page fills in the audience with important safety information, but the 2D imagery is lacklustre compared with the AR of the other pages. The information in these safety pages could be animated and used in AR as well. Hopefully, this would cause this information to become more memorable by those who use the application and assist with the public’s interaction with arachnids from day to day. The UI currently uses simplistic vectored imagery for its forested theme. As mentioned before, this could be improved even though it does not negatively affect the usability of the application. This would be because using higher quality and higher detailed images for the UI’s visuals could have positive effects on user experience, making the application more effective in its main goal of public outreach. Repeating sprites of leaf litter or other nods to the brush-legged wolf spider’s habitat would make for good additions to the UI’s visuals as it would further tie the application to its subject. Adding other additions such as visual indicators of button clicks and toggles could improve the usability of the application as well. Further streamlining of button pathways could be performed to assist this as well, such as unifying panels and buttons into simpler parenting controls to allow for easier editing in the future. Other more intensive improvements that are beyond the scope of this study could be made in the future as well. These would be theoretical as they would require further research and work to understand the potential effects that they could have. Using the AR visuals as a method of generating surrogate mating partners for the brush-legged wolf spider would be one such change as the alteration of the AR animation would change how the application operated, but in doing so would expand the application’s use beyond public outreach into a tool for further research into Schizocosa ocreata.

7.5

Conclusion

The ultimate goal of this study was to create an application that could build interest and understanding through public outreach. In doing so, the

153

application would use newer technology, specifically AR, to help with the viewer’s interest and retention of the information within the application. Schizocosa ocreata, the brush-legged wolf spider, was chosen to act as the arachnid subject of the application. It was chosen due to its nature as a model species creating a large body of information from which to draw and also due to its habitat being in the area of intended testing for the application. The brush-legged wolf spider is native to temperate forests of the Eastern United States and hunts in the leaf litter. It was also chosen for its complicated form of mating behaviour that will hopefully build interest in the application and its subject. The hope is that through this application’s use, the general public will have less fear and more understanding of their local arachnids. To create an application to fill these goals, it was required to be adaptable and capable of enabling AR in a way that complemented the information and visuals being brought forward. To do this, the application employed the use of Android compatibility for the use of mobile devices, allowing the application to be portable and installable on the audience’s devices. The AR was employed as anatomy models and as an animation of the brush-legged wolf spider male’s mating behaviour, known as “jerky tapping.” These are the main points of interest in the subject of the application and provided interesting visuals that complimented the information. The safety information was not done in AR due to its much greater time requirement. It did, however, cover important information on how to avoid interactions with spiders as well as how to deal with a wolf spider if it has been encountered and is on your skin. The application contained two different versions of information and visuals, one which was more realistic and was intended for those less afraid or over 12, the second was more cartoonish and contained simpler information for those more fearful or under the age of 12. This was done to make sure that the application could be used for a majority of users, even those who have a high level of fear around arachnids. The application was built in Unity using additional plugins for AR and the tool of mobile android

154

apk builds. The models were created using ZbrushCore and were built on image reference of the subject. Animation and texturing took place in Blender. The animation had some complications due to the equipment being unable to fully handle the graphics load, causing a drop in quality and smoothness of motion. Despite this setback, the application achieved its main goals for its creation. The application functions with AR visuals that are accurate and pair with information regarding the brush-legged wolf spider that informs users about its external anatomy and the process of its complex mating behaviour. The application’s ability to work on mobile devices lends itself to public situations and opens routes to widespread use by being downloadable on personal devices. By splitting the application into two versions, it can reach audiences without cutting off possible users by using visuals that could frighten them or with intimidating information levels. This means that in its current state it has tools to function through the rigours of public outreach to attempt to reduce the fear of arachnids that is in the local public.

References Aisenberg A, Costa FG (2004) Research papers. Ethology 110:81–95 Diaz JH (2004) The global epidemiology, syndromic classification, management, and prevention of spider bites. Am J Trop Med Hyg 71(2):239–250. https://doi.org/ 10.4269/ajtmh.2004.71.2.0700239 Ding M (2017) ‘Augmented reality in museums’. Available from https://static1.squarespace.com/static/ 51d98be2e4b05a25fc200cbc/t/ 5908d019f5e2314ab790c269/1493749785593/Aug mented+Reality+in+Museums.pdf Elias DO, Lee N, Hebets E, Mason AC (2006) Seismic signal production in a wolf spider: parallel versus serial multi-component signals. J Exp Biol 209:1074–1084 Florea C et al (2019) ‘Extending a user involvement tool with virtual and augmented reality’, 26th IEEE Conference on Virtual Reality and 3D User Interfaces, VR 2019–Proceedings, pp. 925–926. https://doi.org/10. 1109/VR.2019.8798299 Gerdes, A. B. M., Uhl, G. and Alpers, G. W. (2009) ‘Spiders are special: fear and disgust evoked by pictures of arthropods’, Evolution and human behavior. Elsevier Inc, 30(1), pp. 66–73. doi: https://doi.org/ 10.1016/j.evolhumbehav.2008.08.005

R. Nyberg et al. Gibson JS, Uetz GW (2012) Effect of rearing environment and food availability on seismic signalling in male wolf spiders (Araneae: Lycosidae). Anim Behav 84 (1):85–92. Available from http://www.scopus.com/ inward/record.url?eid¼2-s2.0-84862649021& p a r t n e r I D ¼4 0 & md5¼0f8cc3f9b56e255f9f97baeb31eadd12 Hebets EA (2005) Attention-altering signal interactions in the multimodal courtship display of the wolf spider Schizocosa uetzi. Behav Ecol 16(1):75–82. https:// doi.org/10.1093/beheco/arh133 Hebets EA, Uetz GW (1999) Female responses to isolated signals from multimodal male courtship displays in the wolf spider genus Schizocosa (Araneae: Lycosidae). Anim Behav 57(4):865–872. Available from http:// www.scopus.com/inward/record.url?eid¼2-s2.00033118165&partnerID¼40 Hebets EA et al (2018) Eight-legged encounters– arachnids, volunteers, and art help to bridge the gap between informal and formal science learning. Insects 9(1). https://doi.org/10.3390/insects9010027 Huang Y et al (2011) Handbook of augmented reality. In: Handbook of augmented reality, pp 707–720. https:// doi.org/10.1007/978-1-4614-0064-6 Johns JL (2007) ‘Coercive Male Mating Behavior in the Brush-Legged Wolf Spider Schizocosa ocreata (Hentz). In: Dizertačná práca. University of Cincinnati, vol 81 Kesim M, Ozarslan Y (2012) ‘Augmented reality in education: current technologies and the potential for education’, Procedia Soc Behav Sci. Elsevier B.V., 47 (222), pp. 297–302. doi: https://doi.org/10.1016/j. sbspro.2012.06.654 Kozak EC, Uetz GW (2019) Male courtship signal modality and female mate preference in the wolf spider Schizocosa ocreata: results of digital multimodal playback studies. Current Zoology 65(6):705–711. https:// doi.org/10.1093/cz/zoz025 Krause FC, Santos GL (2017) Augmented reality (AR) in biology and environmental sciences education: the state of the art’nuevas ideas en informatica educativa (13) pp. 272–280 Martín-Gutiérrez J et al (2017) Virtual technologies trends in education. Eurasia J Math Sci Tech Edu 13 (2):469–486. https://doi.org/10.12973/eurasia.2017. 00626a McClintock WJ, Uetz GW (1996) Female choice and pre-existing bias: visual cues during courtship in two Schizocosa wolf spiders (Araneae: Lycosidae). Anim Behav 52(1):167–181. Available from http://www. s c o p u s . c o m / i n w a r d / r e c o r d . u r l ? e i d ¼2 - s 2 . 0 0 0 3 0 1 7 6 7 7 5 & p a r t n e r I D ¼4 0 & md5¼6177a6f9aad58d231f047351e0068ea5 Munoz-Cristobal JA et al (2018) Game of blazons: helping teachers conduct learning situations that integrate web tools and multiple types of augmented reality. IEEE Trans Learn Technol 11(4):506–519. https://doi.org/ 10.1109/TLT.2018.2808491

7

Augmented Reality Application of Schizocosa ocreata: A Tool for. . .

Perry J et al (2004) ‘AR gone wild : two approaches to using augmented reality learning games in zoos overview : augmenting learning through augmented reality zoo scene investigators : challenges of designing a mystery themed AR game for students ages 10–14 in a U.S. Zoo’, Management, pp. 322–329. Available from http://dl.acm.org/citation.cfm?id¼1600034 Rovner JS (1975) Sound production by nearctic wolf spiders: a substratum-coupled stridulating mechanism. Science 190:1309–1310 Saidin NF, Halim NDA, Yahaya N (2015) A review of research on augmented reality in education: advantages and applications. Int Educ Stud 8(13):1–8. https://doi. org/10.5539/ies.v8n13p1 Scheffer SJ, Uetz GW, Stratton GE (1996) Sexual selection, male morphology, and the efficacy of courtship signalling in two wolf spiders (Araneae: Lycosidae). Behav Ecol Sociobiol 38(1):17–23. Available from http://www.scopus.com/inward/record.url?eid¼2-s2. 0 - 0 0 2 9 6 6 7 5 9 4 & p a r t n e r I D ¼4 0 & md5¼282028b4d53ccb74db78be86e8f1c42a Selviany A, Kaburuan ER, Junaedi D (2018) ‘User interface model for Indonesian animal apps to kid using augmented reality’, Proceedings of the 2017 International Conference on Orange Technologies, ICOT 2017, 2018-Janua, pp. 134–138. https://doi.org/10. 1109/ICOT.2017.8336106 Stratton GE, Uetz GW (2014) ‘The inheritance of courtship behavior and its role as a reproductive isolating

155

mechanism in two species of schizocosa wolf spiders (Araneae ; Lycosidae Author (s): The inheritance of courtship behavior and its role’, 40(1), pp. 129–141 Sun SY, Sheig C, Huang Y (2006) ‘A summary of research on the’, 16(16), pp. 87–101 Taylor PW, Roberts JA, Uetz GW (2005) Flexibility in the multi-modal courtship of a wolf spider, Schizocosa ocreata. J Ethol 23(1):71–75. https://doi.org/10.1007/ s10164-004-0129-z Uetz GW, Papke R, Kilinc B (2002) Influence of feeding regime on body size, body condition and a male secondary sexual character in Schizocosa Ocreata wolf spiders (Araneae, Lycosidae): condition-dependence in a visual signaling trait. J Arachnol 30(3):461–469. https://doi.org/10.1636/0161-8202(2002)030[0461: iofrob]2.0.co;2 Uetz GW, Stoffer B (2015) Social experience affects female mate preferences for a visual trait in a wolf spider. Behav Ecol 27(1). https://doi.org/10.1093/ beheco/arv143 Wagler R, Wagler A (2017) Understanding how preservice teachers’ fear, perceived danger and disgust affects the incorporation of arachnid information into the elementary science classroom. Int J Environ Sci Edu 12(2):213–231 Wu HK et al (2013) ‘Current status, opportunities and challenges of augmented reality in education’, Computers and Education. Elsevier Ltd 62:41–49. https://doi.org/10.1016/j.compedu.2012.10.024

The Surgical Art Face©: Developing a Bespoke Multimodal Face Model for Reconstructive Surgical Education

8

Partha Vaiude, Mark Roughley, Amy Redman, and Aenone Harper-Machin

Abstract

Surgical and educational challenges exist in relation to the teaching of facial reconstructive surgery due to the complexities of the facial landscape and training models available. This chapter will describe the development and implementation of alternative modes of teaching facial reconstructive techniques in a multidisciplinary setting, pioneered by Surgical Art (www.surgical-art.com), through the use of a bespoke multimodal training model—the Surgical Art Face#. Keywords

Facial reconstruction · Plastic surgery · Surgical simulation · Facial topography · P. Vaiude Department of Plastic Surgery, Noble’s Hospital, Douglas, Isle of Man, UK Liverpool School of Art and Design, Liverpool John Moores University, Liverpool, UK Surgical Art, Douglas, Isle of Man, UK M. Roughley (*) Liverpool School of Art and Design, Liverpool John Moores University, Liverpool, UK e-mail: [email protected] A. Redman Surgical Art, Douglas, Isle of Man, UK http://www.surgical-art.com A. Harper-Machin Department of Plastic Surgery, Whiston Hospital, Prescot, UK

Facial anisotropy · Facial sub-units · Surgical training

8.1

Introduction

The face is a complex hybrid structure with non-homogeneous layers that differ between facial sub-units. Anatomical variation and the specific characteristics of each layer are unique to an individual and present challenges to the surgeon. Reconstructive surgeons are often considered ‘sculptors of the flesh’; however, they do not work with rigid materials like a sculptor of marble or stone. Instead, the surgeon works with living, moving, ageing and evolving ‘clay’. In facial reconstructive surgery, the surgeon needs to take an approach whereby they adapt to variations in anatomical form, movement and ageing and consider their effect on the overall topography of the facial landscape. Obtaining an in-depth understanding of facial lines, tether points and facial contours aids a surgeon in guiding the reconstructive process and repairing defects using tissue recruited during reconstructions more creatively. Surgical reconstructive procedures climb a ladder of complexity. The lowest rungs of complexity include leaving a wound to heal by secondary intention and direct closure of skin, and the procedures advance in difficulty as the rungs of the ladder are climbed to include skin grafts and the movement of vascularised tissue in the

# The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. M. Rea (ed.), Biomedical Visualisation, Advances in Experimental Medicine and Biology 1334, https://doi.org/10.1007/978-3-030-76951-2_8

157

158

P. Vaiude et al.

form of local flaps or distant flaps. Regardless of the surgical procedure, it is critical to bear in mind how the skin creases and folds, the surface topography, the muscle layers of the sub-unit operated on, and the symmetry of the body, in order to achieve a functional and refined aesthetic outcome. Facial flap operations are surgical and aesthetic challenges due to the cosmetically sensitive nature of the face. Flaps involve transferring tissue from one area of the face to an (often) adjacent defect. These flaps, composed of skin and other component structures, are designed using geometric principles that are customised per individual and are one of the most elegant examples of the marriage of art, geometry and surgery. In performing these operations, there exists the potential to cause asymmetry and distortion to the facial topography, however, if done sympathetically, the functional and aesthetic results can be very satisfactory. The transformation from a trainee surgeon, who relies on geometry and pre-prescribed standardised measurements, to an experienced surgeon who is able to artistically use the patient’s facial topography to create a bespoke operation can be a long process of training and development. Whilst the cadaver remains the Holy Grail of teaching models, it is a costly and resource limited teaching tool. To overcome these technical, aesthetic and financial challenges, whilst maintaining the key elements of the face, Isle of Man based Surgical Art has designed a training model known as the Surgical Art Face# to aid and enhance this learning process.

8.1.1

Reconstructive Surgery

Plastic surgery and reconstructive surgery are interchangeable names for the specialty of medicine that attempts to return form and function to a part of the body that is affected by trauma, tumour and congenital conditions or for aesthetic purposes. The term ‘plastic’ surgery is derived from the ancient Greek adjective “plastikos”, meaning ‘to mould’ or ‘to give form’ (Macionis 2018). The first mention of plastic repair dates

back to 3000 BC in the translation of an Ancient Egyptian medical text—the Edwin Smith Papyrus (Shiffman 2012), referring to the reconstruction of a broken nose. Whilst the current spectrum of plastic surgery is a relatively new phenomena, its roots can be firmly traced back to circa two millennia in India, where arts and craft techniques and materials were used to recreate missing parts of the face (Bennett 1983). Sushruta was an Indian Physician considered by many as the Father of Plastic Surgery and his writing ‘The Sushruta Samhita’ is considered one of the most important texts on surgery from the ancient world (Champaneria et al. 2014). It was here that rhinoplastic methods (as we understand them today) began. However, there was little understanding of these surgical advances in the western world until the 1400s. It was then that the first European Surgeon, Sicilian Branca de’Branca, restored a lost nose by taking adjacent tissue from the cheek (Keil 1978). His son, Antonia Branca, furthered this reparative surgery by taking a flap of skin from the upper arm and attaching it elsewhere on the body. This concept of transferring tissue from one part of the body to the other, by connecting the two with local flaps, was named the ‘Italian Method’ (Keil 1978). It needed the patient to be positioned or strapped into a position to allow the tissue to revascularise at the new site before being disconnected from its origin. Branca also reconstructed lips and ears, all of which were documented in Gasparo Tagliacozzi’s ‘De Curtorum Chirurgia per Insitione’ in 1597 (Keil 1978). Tagliacozzi further improved on the ‘Italian Method’ that the Branca family had developed and was known as one of the first Plastic Surgeons, writing “we restore, rebuild, and make whole those parts which nature hath given, but which fortune has taken away. Not so much that it may delight the eye, but that it might buoy up the spirit, and help the mind of the afflicted” (Tagliacozzi 1597). Soon after Tagliacozzi’s death reconstructive methods such as these were abandoned until the end of the 1700s. It was in 1794 that the Indian rhinoplasty methods came to the attention of European surgeons. British surgeons travelled to India to

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

see these being performed, with some like Joseph Constantine Carpue staying to study local plastic surgery methods (Lock et al. 2001). In 1815, Carpue performed the first major surgery of the western world, using instruments first described by Sushruta and modified for modern use (Lock et al. 2001). The most significant advances in modern plastic surgery have been achieved in the last century, instigated by the advances in warfare that occurred over the two world wars. Sir Harold Gilles, generally considered the father of modern day plastic surgery, was a New Zealand born otolaryngologist who worked in London caring for soldiers injured in World War I by reconstructing large facial defects caused by gunshot wounds (Chambers and Ray 2009). Through World War I, Gilles and his team performed over 11,000 operations on over 5000 victims of artillery fire (Chambers and Ray 2009) using both historic and modern surgical principles that connected tubed vascularised tissues to large defects. Gilles was often accompanied by a General Surgeon and later a Slade Professor of Fine Art at University College London, Henry Tonks. Tonks became a Lieutenant in the Royal Army Medical Corps and worked for Gilles illustrating the injuries of soldiers prior to surgery and after their subsequent reconstructions (Biernoff 2010). These portraits, created in chalk pastels, were recognised as the first images to reflect the brutality of modern warfare and also the aesthetic and psychological aspect of reconstructive surgery. They are often used as an example to highlight the merging of art and science. Suzannah Biernoff, a Reader in Visual Culture in the Department of History of Art at Birkbeck College, University of London writes that “Like the ‘strange new art’ of facial reconstruction, Tonks’ drawings blur the line between art and medicine, and, by disturbing the conventional categories of medical illustration and portraiture, they highlight the ambiguities that lie at the heart of those representational practices” (Biernoff 2010). Medical illustrations, both two-dimensional (2D) and three-dimensional (3D), have the

159

power to educate and transfer knowledge of anatomy and surgery through passive observation. By physically drawing the patients throughout their surgical journey, Tonks was not only documenting the injuries and procedures but he was also learning more about the threedimensionality of the face and the person it belonged to. Medical illustrations also have the ability to only show what is required to be seen. Dorothy Davison, one of the founding members of the Medical Artist Association of Great Britain (MAA) was once quoted saying that the “medical artist can be particularly useful in elucidating obscure and difficult points, for she never draws the obvious: that is photographed” (John Rylands Library 2017). Through artistic practices we are not only able to ‘see’ the face in new ways but we can also selectively highlight information and bring this to the attention of the viewer. These are crucial factors to consider when creating a surgical simulator such as the Surgical Art Face#.

8.1.2

Reconstructive Ladder

Reconstructive surgery is a wide-ranging discipline, working with most body systems, from head to toe, from skin to bone, and everything in between. The plastic surgeon uses a spectrum of options to help heal, repair, rejuvenate or replace parts of the body that are lost to disease, trauma or time. Below is a list of options, graded by complexity, called the ‘reconstructive ladder’ (Simman 2009): • Healing by secondary intention with dressings only. • Direct closure involving edge to edge wound closure where there is adequate skin to allow this. • A skin graft where skin is harvested from another area, disconnected and secured to the wound to help it ‘take’ or reintegrate with the body. This can take the form of a superficial removal of the layers of skin called a ‘split skin graft’ or a ‘full thickness skin graft’, utilising the full depth of skin.

160

• Tissue expansion where skin and other tissues are “stretched” with a balloon or expander placed under them. This stimulates tissue growth, eventually providing enough skin for closure of the wound. • Local flap(s) where local available tissue is partly disconnected, retaining a blood supply, is inserted into the defect using geometric and artistic principles. • Free tissue transfer or auto-transplantation where one part of the body is carefully dissected, disconnected and reconnected to another part of the same individual by separating it from its source vessels and then reconnecting the blood vessels at the new site using microsurgery. The part that is moved can contain skin, fascia, fat, muscle, bone, other organs or a combination of these. It can provide both form and/or function. More recently, the rungs of this reconstructive ladder (Fig. 8.1) have been extended, with

Fig. 8.1 The reconstructive ladder is a stepped progression of reconstructive options that are considered when addressing a tissue defect that may be congenital or acquired from trauma, tumour or for aesthetic reasons

P. Vaiude et al.

advances in medicine, surgical techniques and technology, allowing for • Allotransplantation, which is free tissue transfer but from one individual to another. • Robotics where a robotic device is used to enhance the body either externally or by integrating with the body. The description of these available options as a ladder could imply that the surgeon utilises the options of the ladder from the most basic to the most complex in an incremental manner until the desired result is reached. This is not the case. The ladder is simply a guide of options, and when assessing the wound and planning for reconstruction, the surgeon will work through the choices available from the simplest to most complex (Simman 2009). The ‘reconstructive elevator’ (Fig. 8.2) described by Gottlieb and Krieger (1994) describes the process by which the surgeon bypasses one or multiple rungs of the ladder

Fig. 8.2 The ‘reconstructive elevator’ described by Gottlieb and Krieger (1994) is an adaptation of the reconstructive ladder highlighting the philosophy of selective choice based on form and function without necessarily exploring every ‘rung’ of the ladder

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

to reach the ideal option for the reconstructive challenge faced (Glat and Davenport 2017). The aim of modern surgery is not to simply get the wound healed but rather to replace the missing tissue with the closest matching tissue in an attempt to return both form and function.

8.2

Facial Reconstructive Surgery

In countries with a significant Caucasian population, the majority of facial surgeries are performed for skin cancer excision and reconstruction, followed by surgery for trauma, aesthetic reasons, congenital conditions such as cleft lips/palates and conditions requiring craniofacial surgery. In most skin cancer reconstructions, the malignant lesion or tumour is excised and the area reconstructed. The choice of rung from the reconstructive ladder is based on a variety of factors, but at a minimum, the wound is either closed directly, with a skin graft or by using a local/regional flap. It is less common for free tissue transfer to be needed after skin cancer excision. This surgical choice—the assessment the surgeon makes of the reconstructive ‘challenge’ at hand and the intended path to wound healing—is one of skilled decision making that considers risk, disease, anatomy, procedural complexity and patient expectation (Sandberg 2016).

8.2.1

What Knowledge and Skills Do Facial Surgeons Need?

To make an effective surgical choice and to execute it successfully, the surgeon needs a breadth of knowledge and skills. When considering facial surgery, the need for this understanding is even greater as the face is arguably the most visible and cosmetically sensitive part of the body. For any form of facial reconstruction, the surgeon must possess a number of skills to deliver an effective and aesthetic outcome, from analysis of the patient and the condition all the way through to the surgical execution of the reconstruction and the post-operative scar modulation or, in some cases, scar revision. Here we describe the

161

required skills that an effective facial surgeon should possess; some of which overlap with other surgical disciplines. The list is not exhaustive and does not take into account the assumed generic skills of critical analysis, communication and leadership: Clinical Anatomy Skills The surgeon needs a detailed understanding of clinical anatomy, specifically the anatomical forms that will be engaged with directly and in the surrounding area. Skin cancers and other lesions “affect both function and aesthetics, which are based on complex anatomical features” (Marur et al. 2014). Every potential risk and danger area must be anticipated, and the potential impact the structures in the surrounding facial sub-units needs to remain at the forefront of the surgeon’s mind. This is ‘selective knowledge’ that has been distilled from practical experience of operating in the specific anatomical area. Facial Topography Knowledge ‘Facial topography’ is a term that has been described by a number of surgeons and researchers including Gruber et al. (2013); however, Surgical Art uses the term ‘facial topography’ to describe the contours and undulations of the face when teaching the aesthetic approach to facial surgery. An understanding of the rise and fall of the facial landscape is important in defining the chiaroscuro of the face—areas of highlights and shadows. It is a critical exercise in understanding sites of volume and how they change over time, and how to use shadows to hide scars and enhance other areas where highlights are desired. As demonstrated in Fig. 8.3, sculpting the human face using clay or wax helps to give the trainee a haptic and visual reinforcement of facial topography. This understanding provides a foundation to reconstruct the functional face and still retain the character of the face. Understanding of Facial Lines, Units and Sub-units, and Incision Lines Lines of skin are a product of inherent skin anisotropy, muscle movement and facial tethering to deeper layers. In 1861, Karl Langer used a

162

P. Vaiude et al.

Fig. 8.3 Facial topography: an understanding of the contours of the face. Examples of training with clay at Surgical Art

round tipped awl to create defects in skin, but these stretched and created axial patterns based on skin tension (Langer 1978). The lines that resulted from mapping these axial defects are termed Langer’s lines. Later, Kraissl (1951) based his lines perpendicular to the underlying muscles and using photographs to help map these lines or wrinkles. Borges (1984) described resting skin tension lines (RSTL), which are mapped by the straight lines formed when skin is pinched in one direction, and S-shaped when pinched in other directions. This description is useful when exploring lines in a youthful face where there are no wrinkles. When these lines are used as incision guides, the wound experiences minimal stretch and scars heal optimally. Also, because these lines are parallel or within natural crease lines, scars are hidden better. Whilst these lines are unique to every individual, they do follow a general pattern and can be considered as the “facial grid” or “roadmap” (Fig. 8.4). In addition, the face is also divided into anatomic and aesthetic components known as ‘subunits’ (Fig. 8.4), which are intended to help the surgeon respect the functional and aesthetic boundaries within the face, to avoid either a functional complication or aesthetic failure. Within the sub-units there are some accepted facial

landmarks that should avoid being distorted, including the hairline, eyebrow, eyelid, nasal tip, nasal ala, earlobe, philtrum, vermillion and oral commissure. In contrast to these landmarks, there are facial units that provide an ideal source of tissue recruitment for local flap reconstruction whilst still respecting the facial unit boundaries, including cheek, neck, forehead, chin and submenton (Patel and Sykes 2011). The plastic surgeon must understand and use these dual “facial grids” in synchronicity with one another to maintain the functional and aesthetic harmony of the face. Knowledge of Anisotropy, Skin Laxity and Tethering The skin of the face, not unlike other parts of the body, displays elasticity in varying degrees that is related to age and skin quality but is also ‘direction dependent’. This property of directional dependence is termed ‘anisotropy’. Isotropic materials display the same properties in all directions while anisotropic materials display different properties in different directions. Anisotropy has been documented in other biological tissues like the transversalis fascia of the abdominal wall (Kureshi et al. 2008). In skin, anisotropy is based on the skin’s attachments to deeper structures, like fascia or bone via retaining ligaments and other fibrous connections that

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

163

Fig. 8.4 On the mask, the left side of the face has been divided into facial sub-units and the right is patterned with resting skin tension lines

tether the skin (Alghoul and Codner 2013). These connections to the underlying tissues influence the capacity of the skin to be ‘moved’ in different directions. The quality and relationship of the different planes is also not static and changes with age and in response to disease and surgery. The face therefore displays anisotropy that varies in every individual and in changing circumstances. The analysis of an individual’s facial anisotropy along with skin laxity is critical in selecting the most appropriate type of reconstruction and in designing local flaps. Understanding of Local Flap Design Local flaps are an elegant reconstructive option that have seldom been taught outside of the operating theatre. It is, however, an important skill that bridges the disciplines of surgery, geometry and art. The ability to design the appropriate geometrical pattern (the local flap) to fill a ‘hole’ (the tissue defect) and understanding the pattern of movement without the added ‘complication’ of anatomy is itself a steep learning curve. The process of local flap reconstruction applies this concept of geometrically designed patterns and uses tissue adjacent or near to a defect, which can be moved into the area previously occupied by a defect. The application of this process on the

human form requires a combination of surgical and anatomical understanding, and artistic flair to create a harmonious result (Patel and Sykes 2011). As part of the decision-making process into the choice and design of local flaps, the following key components are considered:

1. Vascularity—traditionally local flaps are based on random pattern blood supply with no named specific blood vessel feeding the flap. Instead, an appropriately wide base is selected that allows a rich network of blood vessels to enter the flap at different levels within and below the skin, and even through other tissues that can be incorporated within the local flap, such as muscle. However, there are some local flaps that are based on a named vessel or that can be designed to incorporate a named vessel. 2. Uni/bi/multi-directional laxity—the surgeon assesses the tissue around the defect to gauge laxity and availability of tissue that can be recruited into the defect. 3. Shape and long axis relating to resting skin tension lines—the shape and long axis of the defect plays a crucial role in the choice and

164

design of the local flap. We refer to certain axes as favourable or unfavourable, which has a profound effect on the choice of flap design and orientation. 4. Adjacent functional or aesthetic landmarks— the surgeon has to take into account the proximity of the defect to a landmark organ or aesthetic line when designing a reconstruction. Although a number of reconstructive choices may work for the same defect, a skilful surgeon will choose a design that will least interfere with a functional or aesthetic landmark, for example, the lower eyelid. 5. Thickness of the skin—the quality and thickness of skin is important but does not preclude the use of a local flap with a different thickness of skin and subcutaneous tissue, to that of the defect site. 6. Hair pattern—hair pattern is an important matter of consideration, especially in areas where loss or distortion of hair can be obvious and disabling, for example, the eyebrows. It is also important for the surgeon to consider the aesthetics of facial hair and to attempt to replace like for like, not just in terms of hair bearing skin but also factoring in direction and quality of hair, where possible. Other areas where hair is an important consideration in reconstruction is the hairline, moustache/ beard and sideburns. The three primary flaps are transposition, advancement and rotation flaps. Based on the key factors above, transposition flaps are used primarily when there is minimal tissue available from any direction. The other critical component in deciding to use a transposition flap is the vascularity. A flap design that incorporates a known vessel into it is likely to survive all the way to its tip but also can be raised on a much narrower pedicle. Typical sites for transposition flaps are the scalp, where hair pattern is also an important consideration that needs to be factored in. Advancement flaps are best applied when there is either a uni- or bi-directional laxity available or affordable, and the axis of the defect is at its widest perpendicular to the RSTLs. Both

P. Vaiude et al.

components are important factors in making reconstruction by advancement flap achievable. We have observed that the reach of this type of flap is heavily dependent on the laxity of tissue available, with a huge spectrum of variation amongst individuals, and therefore the trainee surgeon will need to learn to gauge defects in individuals, preferably pre-operatively but sometimes intra-operatively. This flap allows the surgeon to avoid distortion perpendicular to its direction of movement, which is a critical feature that can be used in cosmetically sensitive areas like the forehead, where it is important to avoid lifting one eyebrow, thus changing the quality of the patients face and expression completely. Thickness of the skin and vascularity are less important in these situations as these are considered random pattern flaps, fed by blood vessels at multiple levels and by the network of vessels over the wide flap base. In the third primary flap type—rotation flaps— there is limited uni- or bi-directional laxity but small amounts of multi-directional laxity that can be recruited from a large surface area. A typical site for this flap is the scalp. This flap must be designed significantly larger than the defect as it works on the recruitment of small quantities of tissue from a large arc of rotation, to feed the much smaller defect. Vascularity and hair pattern are important considerations in this flap. We have characterised other flaps as secondary flaps with tertiary variants. These are beyond the scope of this chapter but broadly fall within the design parameters defined by the three primary flaps described above. Figure 8.5 demonstrates the Abbe–Estlander flap that uses a combination of rotation and transposition (Ebrahimi et al. 2011). Facial Reconstruction Experience Transferring the broad clinical understanding of reconstruction and applying this to the facial topography, skin lines, facial units, laxity and tethering of the human face, combined with the design and application of local flaps, is a critical step in the surgeon’s developmental journey. The

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

165

Fig. 8.5 The Abbe–Estlander flap—using part of the upper lip to create a tissue defect of the lower lip. Image courtesy of Jessica Irwin#. Produced as part of a

collaborative project with Surgical Art during her MA Art in Science study at Liverpool John Moores University

ability to merge these individual facets of training is the goal of this facial reconstructive education process.

adapt it to their personal flair, based on their own experience—an accumulation of hundreds if not thousands of operations—to become an ‘artistic’ surgeon. We consider the ‘artistic surgeon’ to be one who has evolved, not purely with surgical knowledge and experience but also with a broader maturity and understanding of the body, the disease and the surgical solutions. However, a surgeon’s transformation from engineer to artist, through this process, is inconsistent and unpredictable both in terms of time and quality.

The surgeon must learn to apply direct closure principles in respect of facial ‘gridlines’, to understand skin/tissue qualities in different parts of the face/head and thus make informed judgments on the appropriate reconstructive option, type and thickness of graft and finally to design and apply local flaps that adapt to the convexities and concavities of the face, and the varying functional and aesthetic sub-units. A final, but crucial, aspect of this learning process is the ability to apply this holistic approach to an individual’s unique facial characteristics and form, the individuals use of their face and their expectations of surgery. The ‘plastic surgery compass’ described by Sandberg (2016) must be balanced and weighed up with the surgical options, risk and the defect/disease/ enhancement in question. Creative Skill The trainee surgeon is taught to apply techniques with prescribed detail for surgical design and incision all the way to closure and dressing. This standardisation provides an important framework, especially for the less mature surgeon. We refer to the surgeon in this phase of training as the ‘engineer’ surgeon. As surgeons progress, they use the principle of this analytical, prescribed process and

We believe that this evolution from engineer to artist is critical but more significantly the rate and quality of this transformation is down to the type of training received. A multi-disciplinary approach to training is vital and for the most part unrecognised in traditional surgical education, which typically includes an apprenticeship with senior surgeons in the specialty or sub-specialty.

8.2.2

What Is the Ideal Simulation Tool to Train Surgeons to Perform Facial Surgery?

There is no question that the living human face is the ideal model for planning facial reconstruction, and that the living patient’s dynamic tissues are the perfect canvas to design a surgical plan. In addition to this, the patient not only provides a

166

body but also a mind with personality, character and an expectation of what the surgeon should achieve. However, access to the patient for surgical planning as a learning tool is not always possible, and when such opportunities arise, they can be stressful learning processes, usually with an awake patient under local anaesthetic, and where the inevitable mistakes of learning cannot be afforded. This otherwise ideal model is followed closely by the human cadaveric face; however, this is a limited resource primarily due to cost implications both for the trainee and training provider. Whilst it would be ideal to have one easily available, cost effective and simple to manufacture model to teach the skills described above, this unfortunately does not exist. However, there are a number of different training models that are used at different stages of the learning process: Biological Models: Porcine Skin Porcine skin has been extensively used in skin research and surgical training (Hassan et al. 2014). Although there are many biological and healing similarities between porcine skin and human skin, porcine skin has limitations for surgical training. It is a reasonable model for basic skin suturing and harvesting split skin graft training as it has a thick dermis and can tolerate trauma at the hands of a trainee surgeon (Fig. 8.6). The Fig. 8.6 A Surgical Art porcine skin training model being used during a ‘Plastic Surgery Skills’ training course

P. Vaiude et al.

face however has a much thinner and more pliable skin that is delicate and allows a significant amount of movement and displays the property of anisotropy. Porcine skin does not simulate these properties and so the elegance of facial local flap techniques are not well simulated on this model alone. Biological Models: Chicken Skin Chicken skin on the other hand is very elastic and pliable. However, the skin is very thin with minimal dermis, which does not allow dermal suturing and can sometimes tear. The chicken model works very well when used as a full carcass to train surgeons in the principles of local flap design (Fig. 8.7). This model has been developed by Surgical Art and can be used to demonstrate and practice the full spectrum of local flap types. Biological Model: Excess Human Skin in Theatre Ibrahim et al. (2016) describes the use of excess human skin that is usually discarded during surgery to be used by the trainee after surgery, away from the operating field. This approach, which intends to be economical is an excellent albeit limited resource for the trainee. More often than not, the tissue that is made available is not facial skin and will not have the same properties; however, it is excellent for skin suturing techniques

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

167

Fig. 8.7 A Surgical Art chicken skin training model being used during a ‘Local Flaps and Skin Graft’ training course

but limited for facial local flap techniques beyond the understanding of flap concepts.

medical prosthetists, and thus will have higher cost implications.

Synthetic Models: Sponge Sponge is the most popular material for suturing demonstration and practice. There are a wide variety of synthetic sponges available for varied applications. A number of these sponges have been used as available or specifically designed for surgical training. They are effective models for teaching basic concepts of flaps and patient education (Villafane et al. 1999) but do not have the refinement to be considered a close simulation of local flap reconstruction.

Synthetic Models: Double Layer Silicone Face The silicone face created by Powell et al. (2019) has been developed using a two layer silicone model of different grades to represent the skin and subcutaneous fat and placed on a 3D printed base as a support. The model, although life-like and multi-layered, again does not have the property of anisotropy built into it.

Synthetic Models: Single Layer Silicone Face Silicone face models often have excellent facial topography and can be coloured or textured to provide a close simulation of the human form. Using prosthetic technology, this model can be developed even further to provide a very aesthetic simulation (Liew et al. 2004). These models are excellent devices for patient education and teaching facial contours and surface qualities but lack anisotropy and only allow uniform movement, which is not a feature of the living face. A further drawback is that this model can only be manufactured by skilled professionals, such as

We have not directly used all of the above models but base our opinions on the qualities tested and presented by the manufacturing teams and researchers. There is no doubt that the appeal of biological tissue models is the feel of handling real tissue with all its complexity, and whilst synthetic models can simulate the shape of the face well, they lack the quality of facial anisotropy.

8.3

Development of the Surgical Art Face©

Surgical Art, originally a UK-based organisation and now located on the Isle of Man, was conceived as a vehicle for change in surgical

168

P. Vaiude et al.

Fig. 8.8 Mark Roughley and Partha Vaiude using a 3D facial scanner and digital manipulation software during a course in the Surgical Art studio-laboratory

education and a way to deliver a novel philosophy: • To provide a multi-disciplinary approach to teaching anatomy, disease processes, and reconstructive concepts and surgery with the intention to provide the surgeon with a wider perspective and therefore be equipped to deliver comprehensive solutions. • To employ the creative training models used in these varied, often non-medical disciplines to develop its own relevant training models. This adaptation injects creativity into an otherwise rigid training system. • The immersion of both the principles and practice of other disciplines. This has been trialled with success in the Surgical Art studiolaboratory (Fig. 8.8), which is a non-clinical environment to promote receptivity and innovation. It is within the bounds of this philosophy that the concept of the Surgical Art Face# was conceived.

The face has been depicted differently by a variety of professionals namely the sculptor, the mask maker, the medical artist, the forensic facial anthropologist, the beautician and the surgeon. The sculptor understands artistic anatomy and sees the face as a combination of positive and negative spaces, contours and textures. They work to recreate the relationships between the different biological tissues that produce an overall appearance and imprint or express personality through often hard and rigid but malleable materials (Fig. 8.9). The mask maker takes an impression of the face and creates a mask. They enhance facial contours in their designs to introduce a character, create an emotional affect, or to strengthen or soften the face. They use facial shaping, colour, pattern and texture to create an artistic, historical or theatrical depiction of a person (Fig. 8.10). The medical artist might use their understanding of human anatomy to build models of faces for teaching and training purposes. As anatomists themselves or alongside anatomists, medical

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

169

Fig. 8.9 The sculptor—working in three dimensions has the freedom to explore each facial sub-unit with creativity as is seen with Athar Jabber’s stone carvings (https:// www.atharjaber.com/). The surgeon can be taught the

effect of subtle changes in each sub-unit, thus rendering a major change in the overall impression. Images courtesy of Athar Jabber

artists study the anatomical structures and rich network of vessels and nerves beneath the skin. They understand the interplay of the different structures and layers and how their complex relationships work in a functional face.

Historically, wax moulages or ceroplastics were created by medical artists who demonstrated extraordinary craft skills to produce hyperrealistic models for educational purposes (Ballestriero 2010). They possessed a clear

Fig. 8.10 The mask-maker’s perception of the face. Note the similarities in this Venetian mask to facial sub-units and surgical designs. The mask shown here is created by Angela and Victor of Schegge Art & Craft, Venice

170

P. Vaiude et al.

Fig. 8.11 A forensic facial anthropologist reconstructing the face of an ancient Egyptian mummy using the 3D software ‘Geomagic Freeform’ and a haptic interface device. Image courtesy of Face Lab at Liverpool John Moores University

understanding of the human anatomy often learned by observing dissections first-hand, and the models could be tailored to create bespoke models that showed the relationships between muscles and vessels, demonstrate the appearance of diseases on hard and soft tissues and highlight the subtle differences in anatomical variation between individuals, for example. The forensic facial anthropologist interprets the face from inside out and builds the face of an individual from interpretations of skeletal human remains. They construct the face using wax, clay or 3D software from replicas or 3D scans of the skull (Fig. 8.11). Following an assessment of the skull morphology, the muscles, fat, skin and facial features are modelled using established scientific methods. The forensic facial anthropologist applies knowledge of anatomy and morphology determination to accurately reconstruct faces of the dead for presentation to public audiences (Wilkinson 2010). These depictions are primarily used for forensic identification purposes or to bring us closer to historical figures from our past. The surgeon negotiates the different layers of the face, respecting form and protecting function, retaining vascularity and nerve supply. They move parts of the face, often to replace missing or diseased segments, with healthy and carefully chosen local or distant components. The surgeon

sees the face from outside in, and in most operations the surgical approach is from skin to bone. The beautician, more precisely the contouring expert, does not physically change shape but uses colour, highlights, shading and even texture to achieve similar albeit impermanent results. They use their understanding of the face to camouflage scars, enhance the quality of skin, increase or decrease volume and lift or move certain features (Fig. 8.12). These professionals approach the face from different perspectives; however, multiple components of their practice overlap. This crossdisciplinary overlap can foster novel multidisciplinary collaborations where professionals can learn from each other and innovation flourishes. Multi-disciplinary collaborations often arise in reflexive spaces, such as the Surgical Art studio-laboratory, where there are opportunities to “understand, critique and evolve the dominant discourses of the parent disciplines” by bringing together different but potentially similar disciplinary practices to tackle challenges (Marshall and Bleecker 2010). In doing this, there are also un-disciplinary opportunities to understand or experience new or different ways of working with ideas, processes or materials. This might require re-learning ways of thinking and doing or the creation of new vocabularies, but

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

171

Fig. 8.12 Beautician teaching contouring during a Surgical Art course— using highlights and shadows to adjust the appearance of facial contours and volume

the outcomes often benefit all involved. We have adopted this multi-disciplinary, cross-pollination approach to enhance the training programmes at Surgical Art. In comparing and evaluating the practices of these varying disciplines and their approaches to the face, we recognise the incredible complexity of the living human face, which is impossible to recreate fully as an affordable and accessible simulation model. In designing the Surgical Art Face#, it was decided that it was critical to first tease out the individual components of reconstructive surgical training requirements and then to identify the most effective available models to address these training needs. The Surgical Art Face# needed to • Simulate contours of the face • Simulate anisotropic properties of the face • Be able to be drawn on and wiped off for multiple uses • Be cut, reshaped and sutured • Be cost and time efficient to produce • Be made without technical or artistic skills • Be mounted for ease of use The Surgical Art Face# model has moved through a number of different iterations with input at each stage from sculptors, medical artists, forensic facial anthropologists and surgeons.

Surgical Art Face# Version 1 Initially a silicone-only face was produced using different grades of widely available silicone. These cost-effective silicones were cast using a face mould. Different thicknesses of silicone were used to create varying levels of pliability and anisotropy in the face. The silicone thickness did not mirror that of the skin in different sub-units of the face but was based on facial pliability. The resulting face had excellent contours but lacked adequate anisotropy and the capacity to be sutured effectively, with sutures often tearing through the silicone layers. This model was considered inadequate for comprehensive training. Surgical Art Face# Version 2 In version 2, the addition of fabric to the silicone to enhance the anisotropic properties of the model was explored. Several fabrics were identified that displayed different degrees of anisotropy, with different capacities of adhering to silicone. A second experimental model was developed using small pieces of anisotropic fabrics embedded on to one side of the silicone face, which worked with varying levels of success. The fabric undoubtedly enhanced the anisotropy of the silicone ‘skin’, but the major drawback of this approach was the tendency of the fabric to peel away from the silicone when handled and sutured.

172

Surgical Art Face# Version 3 The issue of fabric extrusion was resolved with the application of silicone from both sides of the silicone skin in the third iteration of the model. Silicone was used as in version 1 but with varied thickness to further enhance anisotropy. The face moved better, but the different fabric components felt ‘detached’. Surgical Art Face# Version 4 Taking this into account, version 4 incorporated a complete fabric sheet that was the same size as the face with silicone on both sides, with silicone thickness selected from the development of versions 1 and 3. This was effective and was further enhanced by pre-tensioning the fabric before it was embedded. This concept was adapted from the Surgical Art Z-Plasty Simulation Model#, which uses pre-tensioned anisotropic fabric embedded in silicone on an embroidery ring to simulate a tight scar. This is then released using a technique called Z-Plasty. Version 4 of the Surgical Art Face# (Fig. 8.13) is the model currently used for training reconstructive surgeons.

P. Vaiude et al.

8.4

Facial Surgery Simulation Using the Surgical Art Face© in Multi-disciplinary Settings

Training surgeons to assimilate generic concepts and techniques is possible with simulation models. However, this does not allow for individuality both in terms of the patient’s anatomy or condition but also the surgeon’s preferences. Every patient presents with individual facial characteristics and clinical problems that are unique to them. These need to be analysed carefully and a bespoke surgical design devised for that person. For the most part, this can be achieved by studying the face at rest and with directed or palpated movement. Every surgeon has their own interpretation of a patient’s face and knowledge of how they will tackle the reconstructive challenge and therefore will approach it differently. In training, these two components can only be explored with a real patient and not with a model. We acknowledge the limitations of the Surgical Art Face# but also its strengths and have developed a system to use the Surgical Art Face# in combination with other activities such as anatomical wax modelling of facial musculature and drawing on the faces of

Fig. 8.13 Inspecting a completed Surgical Art Face# before its use in training (left). The Surgical Art Face#, after been used for different drawing and surgical exercises (right)

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

human participants during Surgical Art training sessions. In these multi-disciplinary sessions, participants might 1. Build a wax anatomical model using a cast of a human skull, guided by a forensic facial anthropologist and/or medical artist. 2. Simulate sculpting of facial topography and understand facial lines and sub-units in conjunction with the wax models, living faces, colour mapping exercises and the Surgical Art Face#. 3. Simulate clinical scenarios by overlaying the Surgical Art Face# on top of the wax models. 4. Simulate skin laxity, affordability and tether points through interactions with living faces and editing of the Surgical Art Face#. 5. Engage in contouring exercises with living faces through the application of makeup guided by a beautician. 6. Simulate local flap design and surgery using Surgical Art’s biological chicken models. 7. Simulate facial local flaps using the Surgical Art Face#. 8. Explore creativity in facial reconstruction design and surgery using the Surgical Art Face#. Surgical art training sessions take place in non-clinical, studio-laboratory settings, and in the following sections, we detail a number of activities that occur during a number of different training sessions. Using the Surgical Art Face# to Understand Facial Anatomy and Topography: ‘Understanding the Terrain’ This training activity involves the analysis of skeletal morphology and the sculpting of wax muscles onto a plaster cast of a skull, with the aim that the muscles morphologically fit the skull. Established methods used by anatomical and forensic facial reconstruction practitioners are followed. For Surgical Art courses, we follow the Manchester method of forensic facial reconstruction devised by Richard Neave and Caroline Wilkinson (Wilkinson 2010) that builds the face from skull to skin, and the superficial musculoaponeurotic system (SMAS) that focuses on the

173

watershed layer of anatomical structures that defines tissue planes and generates facial expression. The fat, vessels, nerves and facial ligaments are also sculpted; acting as a revision exercise before the Surgical Art Face# is draped on top (Fig. 8.14). This system uses the Surgical Art Face# in a collaborative manner for surface anatomy and enables discursive understanding points of subcutaneous tethering between skin and muscle with retaining ligaments. The objective of the exercise is to learn the clinical anatomy and contour of the face through the ‘assembly’ of the face from skull to skin. Using the Surgical Art Face# to Learn the Lines of the Face: ‘Understanding Road Maps’ A living human volunteer is used to demonstrate facial lines (RSTLs) through drawing exercises on their face. RSTLs are not visible at rest in a youthful face but can be seen as the face ages, and these are referred to as static rhytids. These facial lines can be enhanced in all faces with active movement, for example, by asking the volunteer to smile, frown etc. and are referred to as dynamic rhytids. They can also be produced by passive movement; for example, the surgeon manipulates the skin to elicit facial skin creasing as described by Borges (1984). Hair patterns and previous scars are also important factors to consider in this analytical process. The lines define the facial grid, which we describe as the ‘roadmap for surgery’. Every face is slightly different, and this variation is best appreciated on a human subject. However, once these lines are drawn on the subject, they are then transposed onto the Surgical Art Face# to practice the grid pattern. The objective of the exercise is to learn the lines of the face with individual variations. Using the Surgical Art Face# to Understand Facial Units and Sub-units: ‘Understanding Town Planning’ Once the Surgical Art Face# has RSTLs depicted (as seen in Fig. 8.15), it has the overall road map of the face. However, the face is not a uniform structure, both in form or function. The

174

P. Vaiude et al.

Fig. 8.14 Learning anatomy by building a face on a plaster cast of a human skull; layer by layer using modelling wax (left). The Surgical Art Face# mounted on the wax face to perform exercises (right)

face is divided into units and sub-units from both an artistic and a surgical perspective (Fig. 8.16). The understanding of the different units and sub-units is important to help plan reconstructions that are contained within sub-units, so as not to cause functional or aesthetic asymmetry. The sub-units often have a specific component either as major facial sensory organs or other anatomically discreet functions. The objective of this exercise is to learn the different facial units and sub-units and to understand the boundaries or borders that should be respected when reconstructing. Using the Surgical Art Face# to Learn Surgical Facial Reconstruction: ‘Mapping the Journey’ Once the trainee has grasped the concepts of direct closure, skin grafts and local flaps by using basic models like sponge, pig skin and chicken skin models, they can apply learned

concepts to the Surgical Art Face# (Fig. 8.17). The trainee will explore each facial unit/sub-unit and the suitable reconstructive options such as local flaps that can be used without aesthetically or functionally distorting other facial units. This session starts with approaching each sub-unit individually, working first within the sub-unit and then exploring the challenge of bridging over into another sub-unit. It is important for the trainee to understand the dynamics of the site being reconstructed and how this may differ greatly a few millimetres away in one direction, generating a different reconstructive design, which may be entirely different in the opposite direction. The three primary local flap movements or their secondary variations can then be used creatively based on the type of movement desired and the effects, be them major or minor, on the surrounding tissue. This process starts by drawing

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

175

Fig. 8.15 Facial lines drawn on a human subject (left) to identify and map individual variation before this understanding is applied to the Surgical Art Face# (right)

defects on the Surgical Art Face# at sites of typical reconstructive solutions. At this point, the trainee assesses the distribution of tension that the flap will exert on the surrounding tissue, especially on functional or aesthetic landmark

areas, to assess the potential distortion and the functional or aesthetic issues that arise from this distortion. An element of initial distortion is accepted in certain areas to allow tissue stretch and scar modulation, which completely or

Fig. 8.16 Facial lines and sub-units explored using colour exercises and beads on the wax face (left) before this knowledge is applied to the Surgical Art Face# (right)

176

P. Vaiude et al.

Fig. 8.17 Using the Surgical Art Face# to learn facial reconstruction after foundation exercises are completed

partially neutralises the effect. This is an important part of the training programme where faculty can use both the mask and volunteers to demonstrate the effects of different tissue movements in different sub-units, producing varied results. The trainees then cut out the disease (defect) on the Surgical Art Face# and apply the concepts taught to design the reconstruction, which is then performed by raising the flap segment and moving it into the area of disease/defect. Once positioned at its final site, the tension of the

Fig. 8.18 Using the Surgical Art Face# to learn facial reconstruction after foundation exercises are completed

flap is assessed and the order of the flap inset is then practiced. This order usually starts with closure of the donor site or, in certain flaps, anchoring the point of maximum tension to a less mobile point to help hold the flap in place. The flap is then fully inset (Fig. 8.18). The next step of the training is creative design in bespoke terrain, where human volunteers are used and tissue defects are drawn on their faces using skin pencils in different sub-units or bridging two sub-units. First, they are drawn at the site

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

of typical reconstructions, followed by more complex sub-unit junctions or functional and cosmetic junctions. The trainee starts by mapping the face as described in sect. 1.4 before analysing each tissue defect and its site in a bespoke manner. Different flap options are explored, their pros and cons are discussed, and a final reconstructive design is drawn on the volunteer’s face. As these tissue defect sites are randomly chosen for each volunteer, a large number of potential variations in tissue defects both in terms of size, shape and location are explored and every trainee presents their reconstructive solutions to the group.

8.5

Conclusion

The innovative Surgical Art Face# version 4 is the product of collaborative engagements and teaching approaches, using inputs from surgeons, sculptors, medical artists and facial anthropologists, plus experts from a number of other disciplines. It is a cost-effective training model that does not need specialist expertise to produce. The model allows trainees to understand the face and its relevant components both technically and artistically. It recreates in 3D the topography of the face, both for demonstrations and practice, and is highly effective in the training of facial anatomy, facial topography, RSTLs and facial units/sub-units to build an understanding of facial roadmaps for surgical planning. Finally, it can be ‘operated’ upon to deliver haptic training for a surgeon. The Surgical Art Face# plays a critical role in almost every step of the Surgical Art training program and is highly regarded by trainees and faculty alike. More than 500 Surgical Art Faces# have been made and used in surgical training to date of this publication with overwhelmingly positive feedback. It is a userfriendly and multi-functional device that is envisaged to play a significant role in medical education. It can also be used for patient education to describe scars from simple to complex reconstructions and has been used successfully in public engagement events.

177

Whilst the Surgical Art Face# has been used effectively in training scenarios, it has its limitations. These are aesthetic, functional and cost related. Currently, the training models are time-consuming to produce, and there is a lack of refinement and difficulty in maintaining standardisation, which effects more sensitive facial sub-units like the peri-orbital and nasal regions. Our vision is to use the principles set out in this chapter to further enhance the Surgical Art Face# through collaboration and technological transformation. We will continue to engage with mask makers, sculptors, beauticians and forensic anthropologists to develop a 3D digital construct that can be 3D printed. Through a research exercise and innovation process, we will evaluate the current Surgical Art Face# through self-evaluation and evaluation of feedback obtained from trainees to further develop a more functional and aesthetically appropriate training model. Version 5 of the Surgical Art Face# will be constructed using a combination of 3D face scanning, complex 3D modelling to embed facial anisotropy capabilities directly between layers that make up the face model, and 3D printing of the final model as one or more piece(s). The overall aim is for 3D printed face models to be printed en-masse but also more easily customised to present anatomical variations between models. Whilst this research and development process is undertaken, the Surgical Art Face# version 4 will continue to work harmoniously in tandem with other training models, and the greatest model of them all—the living human face.

References Alghoul M, Codner MA (2013) Retaining ligaments of the face. Review of anatomy and clinical applications. Aesthet Surg J 33(6):769–782. Available at: https:// academic.oup.com/asj/article/33/6/769/198068 Ballestriero R (2010) Anatomical models and wax Venuses: art masterpieces or scientific craft works? J Anat 216(2):223–234. Available at: https:// onlinelibrary.wiley.com/doi/full/10.1111/j.1469-7580. 2009.01169.x

178 Bennett JP (1983) Aspects of the history of plastic surgery since the 16th century. J R Soc Med 76(2):152–156. Available at: https://journals.sagepub.com/doi/pdf/10. 1177/014107688307600214 Biernoff S (2010) Flesh poems: Henry Tonks and the art of surgery. Vis Cult Br 11(1):25–47. Available at: https:// www.tandfonline.com/doi/full/10.1080/ 14714780903509979 Borges AF (1984) Relaxed skin tension lines (RSTL) versus other skin lines. Plast Reconstr Surg 73 (1):144–150. Available at: https://europepmc.org/arti cle/med/6691065 Chambers JA, Ray PD (2009) Achieving growth and excellence in medicine. Ann Plast Surg 63 (5):473–478. Available at: https://journals.lww.com/ annalsplasticsurgery/Abstract/2009/11000/Achieving_ Growth_and_Excellence_in_Medicine__The.2.aspx Champaneria MC, Workman AD, Gupta SC (2014) Sushruta: father of plastic surgery. Ann Plast Surg 73 (1):2–7. Available at: https://journals.lww.com/ annalsplasticsurgery/Abstract/2014/07000/Sushruta__ Father_of_Plastic_Surgery.2.aspx Ebrahimi A, Maghsoudnia GR, Arshadi AA (2011) Prospective comparative study of lower lip defects reconstruction with different local flaps. Journal of Craniofacial Surgery 22(6):2255–2259. Available at: https://journals.lww.com/jcraniofacialsurgery/ Abstract/2011/11000/Prospective_Comparative_ Study_of_Lower_Lip_Defects.61.aspx Glat PM, Davenport T (2017) Current techniques for burn reconstruction: using dehydrated human amnion/chorion membrane allografts as an adjunctive treatment along the reconstructive ladder. Ann Plast Surg 78(2): S14–S18. Available at: https://journals.lww.com/ annalsplasticsurgery/FullText/2017/02001/Current_ Techniques_for_Burn_Reconstruction__Using.5.aspx Gottlieb LJ, Krieger LM (1994) From the reconstructive ladder to the reconstructive elevator. Plast Reconstr Surg 93:1503–1504 Gruber RP, Levine SM, Levine JP (2013) Facial topography: clinical anatomy of the face. Plast Reconstr Surg 132(1):249. Available at: https://journals.lww.com/ plasreconsurg/Fulltext/2013/07000/Facial_Topogra phy___Clinical_Anatomy_of_the_Face.49.aspx Hassan Z, Hogg F, Graham K (2014) A 3-dimensional model for teaching local flaps using porcine skin. Ann Plast Surg 73(4):362–363. Available at: https:// journals.lww.com/annalsplasticsurgery/Abstract/2014/ 10000/A_3_Dimensional_Model_for_Teaching_ Local_Flaps.2.aspx Ibrahim N, Chauhan I, Varma S (2016) A novel method for practising local skin flaps. The Annals of The Royal College of Surgeons of England 100(4):338–338. Available at: https://publishing.rcseng.ac.uk/doi/full/ 10.1308/rcsann.2016.0252 John Rylands Library (2017) Visual medical collections, The John Rylands library special collections blog (April 26, 2017). Available at https://

P. Vaiude et al. rylandscollections.com/2017/04/25/visual-medicalcollections/ Keil G (1978) The history of plastic surgery. Laryngol Rhinol Otol (Stuttg) 57(7):581–591. Available at: https://pubmed.ncbi.nlm.nih.gov/355747/ Kraissl CJ (1951) The selection of appropriate lines for elective surgical incisions. Plast Reconstr Surg 8 (1):1–28. Available at: https://journals.lww.com/ plasreconsurg/Citation/1951/07000/THE_SELEC TION_OF_APPROPRIATE_LINES_FOR_ELEC TIVE.1.aspx Kureshi A, Vaiude P, Nazhat SN, Petrie A, Brown RA (2008) Matrix mechanical properties of transversalis fascia in inguinal herniation as a model for tissue expansion. J Biomech 41(16):3462–3468. Available at: https://www.sciencedirect.com/science/article/abs/ pii/S0021929008004454 Langer K (1978) On the anatomy and physiology of the skin: I. the cleavability of the cutis. Br J Plast Surg 31 (1):3–8. Available at: https://www.jprasurg.com/arti cle/0007-1226(78)90003-6/pdf Liew SH, McPhail J, Morton JD, Darcy CM, Alvi R, Graham KE (2004) A non-animal facial model for teaching local flaps to trainees. Br J Plast Surg 57 (4):374–375 Lock S, Last JM, Dunea G (2001) The Oxford illustrated companion to medicine. Oxford University Press, USA Macionis V (2018) History of plastic surgery: art, philosophy, and rhinoplasty. J Plast Reconstr Aesthet Surg 71 (7):1086–1092. Available at: https://www. sciencedirect.com/science/article/pii/ S1748681518300913 Marshall J, Bleecker J (2010) Undisciplinarity. In: Rodgers P, Smyth M (eds) Digital blur: creative practice at the boundaries of architecture, design and art. Libri Publishers, pp 216–223 Marur T, Tuna Y, Demirci S (2014) Facial anatomy. Clin Dermatol 32(1):14–23. Available at: https://www. sciencedirect.com/science/article/abs/pii/ S0738081X13000898 Patel KG, Sykes JM (2011) Concepts in local flap design and classification. Oper Tech Otolaryngol Head Neck Surg 22(1):13–23. Available at: https://www. sciencedirect.com/science/article/abs/pii/ S1043181010000710 Powell AR, Srinivasan S, Green G, Kim J, Zopf DA (2019) Computer-aided design, 3-D–printed manufacturing, and expert validation of a high-fidelity facial flap surgical simulator. JAMA Facial Plastic Surgery 21(4):327–331. Available at: https://www. liebertpub.com/abs/doi/10.1001/jamafacial.2019.0050 Sandberg LJM (2016) The plastic surgery compass: navigating the reconstructive ladder in the personalized health care era. Plast Reconstr Surg Glob Open 4 (9) Available at: https://www.ncbi.nlm.nih.gov/pmc/ articles/PMC5055014/ Shiffman M A, Di Giuseppe A (2012) Cosmetic surgery: art and techniques. Springer Science & Business Media

8

The Surgical Art Face©: Developing a Bespoke Multimodal Face. . .

Simman R (2009) Wound closure and the reconstructive ladder in plastic surgery. The Journal of the American College of Certified Wound Specialists 1(1):6–11. Available at: https://www.sciencedirect.com/science/ article/abs/pii/S1876498308000088 Tagliacozzi, G (1597) De Curtorum Chirurgia per Insitionem. Venezia

179

Villafane O, Southern SJ, Foo IT (1999) Simulated interactive local flaps: operating room models for surgeon and patient alike. Br J Plast Surg 52(3):241–241 Wilkinson C (2010) Facial reconstruction–anatomical art or artistic anatomy? J Anat 216(2):235–250. Available at: https://onlinelibrary.wiley.com/doi/full/10.1111/j. 1469-7580.2009.01182.x

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens Kristen E. Pearlstein, Terrie Simmons-Ehrhardt, Brian F. Spatola, Bernard K. Means, and Mary R. Mani

Abstract

The anatomical collections at the National Museum of Health and Medicine (NMHM) contain skeletal specimens that highlight the history of military and civilian medicine dating from the American Civil War and the founding of the museum as the Army Medical Museum in 1862. Today, NMHM curates over 6400 gross skeletal specimens consisting primarily of pathological or anomalous single bone elements that display a variety of pathological conditions, including congenital anomalies, neoplasms, healed and unhealed trauma and infectious diseases, and surgical interventions such as amputations and excisions. In an effort to increase accessibility to these pathological specimens, NMHM is collaborating with Virginia Commonwealth University (VCU) and the Laboratory Division of the Federal Bureau of Investigation (FBI) to digitize and disseminate high-quality 3D models via online

K. E. Pearlstein (*) · B. F. Spatola National Museum of Health and Medicine, Defense Health Agency, Silver Spring, MD, USA e-mail: [email protected] T. Simmons-Ehrhardt · B. K. Means Virginia Commonwealth University, Richmond, VA, USA e-mail: [email protected] M. R. Mani Federal Bureau of Investigation, Laboratory Division, Richmond, VA, USA

portals, enabling scholars and educators to manipulate, analyze, and 3D print the models from anywhere in the world. Many institutions with courses in paleopathology and forensic anthropology do not have reference collections or access to museum collections for hands-on teaching. Therefore a digital repository of osteological specimens can provide an unprecedented and unique resource of exemplars for scholars and educators. The sharing of these military medical assets improves historical knowledge and diagnostic capabilities in the fields of medicine and anthropology. This chapter outlines the digitization processes that are being utilized to increase access to these pathological skeletal specimens through multimodal 3D capture. Keywords

Medical museum · Digitization · Pathology · 3D modeling · Micro-computed tomography · 3D scanning

9.1

Background

The anatomical collections at the National Museum of Health and Medicine (NMHM) contain skeletal specimens that highlight the history of military and civilian medicine dating from the American Civil War and the founding of the museum as the Army Medical Museum in 1862.

# The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 P. M. Rea (ed.), Biomedical Visualisation, Advances in Experimental Medicine and Biology 1334, https://doi.org/10.1007/978-3-030-76951-2_9

181

182

The original purpose of the museum was to collect medical reports, documents, artifacts, and anatomical specimens pertaining to injuries and diseases of war in order to analyze health outcomes for soldiers undergoing medical treatment. In the years after the Civil War, the collecting agenda for skeletal specimens included examples of general medical interest, such as transportation accidents, noncombat gunshot wounds, healed fractures, and “diseases of civil life” (Woodward 1871:236). The purpose and scope of the museum expanded to embrace “all forms of injuries and diseases, so that eventually it shall become a general pathological museum, accessible for study to all medical men who are prosecuting original inquiries” (Woodward 1876:3). Today, NMHM curates over 6400 gross skeletal specimens consisting primarily of pathological or anomalous single bone elements that display a variety of pathological conditions, including congenital anomalies, neoplasms, healed and unhealed trauma and infectious diseases, and surgical interventions such as amputations and excisions. The term “specimen” is used throughout this chapter to refer to osteological elements because it describes the specific medical context under which these partial human remains were collected: as singular examples of types of pathological conditions. In support of the early design and purpose of the museum, NMHM is making a selection of these pathological specimens available for study by modernizing access through three-dimensional (3D) digitization. To facilitate the goal of accessibility, NMHM is collaborating with Virginia Commonwealth University (VCU) and the Laboratory Division of the Federal Bureau of Investigation (FBI) to digitize and disseminate high-quality 3D models via online portals, enabling scholars and educators to manipulate, analyze, and 3D print the models from anywhere in the world. Many institutions with courses in paleopathology and forensic anthropology do not have reference collections or access to museum collections for hands-on teaching. Therefore, a digital repository of osteological specimens can provide an unprecedented and unique resource of exemplars for

K. E. Pearlstein et al.

scholars and educators. The sharing of these military medical assets improves historical knowledge and diagnostic capabilities in the fields of medicine and anthropology. This chapter outlines the effort to increase access to these skeletal specimens through multimodal 3D capture.

9.2 9.2.1

Digitization and Processing Specimen Selection and Digitization Methods

The osteological specimens selected for this 3D digitization project represent both unique and typical exemplars of pathological conditions. Some specimens are considered one-of-a-kind examples, while others represent typical bony responses to force, inflammation, or fracture repair. A variety of pathological conditions were included in the scanning project to provide access to outliers and realistic examples of conditions that students and researchers might find when analyzing human remains in the field. Generally, the specimens can be grouped into three broad categories: American Civil War, paleopathology, and comparative nonhuman. Over half of the specimens scanned at the time of this writing have been of injuries or pathological conditions sustained during the American Civil War (n ¼ 68) between 1862 and 1865. These specimens primarily represent perimortem gunshot wounds, advanced osteomyelitis, healed fractures and callus formation, and amputation. Since the American Civil War took place before the acceptance and understanding of bacteriology and antisepsis, many of these specimens exhibit various stages of bony response to infection. Associated medical case reports provide additional biomedical and biohistorical information. Specimens categorized under paleopathology (n ¼ 53) represent time periods ranging from the 1870s to the modern era. While the American Civil War specimens primarily represent battlefield trauma and associated infectious processes, the paleopathology specimens include a broader range of pathological conditions such as

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens

neoplasms, congenital abnormalities, fractures, age-related changes, ankylosis, dislocation, and infectious disease. These examples are consistent with the types of pathological conditions that are typically taught in paleopathology courses or explored in paleopathology literature. No Native American human remains were scanned or included in the project. Nonhuman specimens (n ¼ 3) were selected as comparative material. Two specimens exhibit conditions that affect both humans and nonhuman animals, that is, osteomyelitis and osteofluorosis, and can be used as teaching materials for veterinary pathology, or in instances where the use of human remains is not considered appropriate. A third nonhuman specimen represents the comparative skeletal anatomy of a nonhuman primate. Of the 124 specimens scanned at the time of writing, 63 have been surface scanned by VCU’s Virtual Curation Laboratory, 58 have been scanned with micro-computed tomography (micro-CT) by the Laboratory Division of the FBI, and 3 have been both surface and microCT scanned to compare digitization methods. Specimens were selected for surface scanning versus micro-CT scanning based on factors such as the complexity of the pathological condition, the stability of the specimen, the size of the specimen, and the presence or absence of embedded metal or wiring. Specimens with complex pathological conditions that benefit from visualization of internal structures were selected for micro-CT scanning. These included specimens with advanced osteomyelitis, neoplasms, endocranial lesions, trauma, infectious disease, and amputation. Specimens that did not require the visualization of internal structures were selected for surface scanning. These specimens included simple fractures, gunshot wounds, DISH, periostitis, ankylosis, and other pathological conditions. Many of these latter specimens may provide better visualization of the condition with micro-CT scanning, but limitations on time, digital storage space, and modeling capabilities meant that surface scanning could produce more models at a faster and more regular pace than micro-CT.

183

Specimen size was also a factor in selection. Very few complete femora have been scanned for this project due to the length of the long bone and space considerations for the internal chamber of the micro-CT scanner and scanning platform of the surface scanner. Other selection factors included moving parts, such as the bone fragments of comminuted fractures that have a tendency to shift during surface or CT scanning, thus diminishing the quality of the scan. Additionally, these fracture pieces are often wired together with pieces of metal that creates extraneous noise and scatter on the scans. Finally, the fragility and stability of each pathological condition were considered during specimen selection. One specimen, a mandible of a 21-year-old female with a “sunburst” periosteal reaction in a case of osteosarcoma, is featured in several textbooks on paleopathology (e.g., Aufderheide and Rodriguez-Martin 1998) and is often requested for hands-on teaching demonstrations. Over the years, physical handling of the specimen caused permanent damage to the fragile bone spicules and degraded the appearance of the pathological condition. MicroCT scanning allows institutions like NMHM to minimize handling of fragile and irreplaceable specimens through the use of digital surrogates. NMHM can now disseminate a 3D model of the mandible for teaching and research purposes instead of using the physical specimen, which will preserve the integrity and longevity of the specimen and the manifested pathological condition (Fig. 9.1). Museums with fragile skeletal or pathological material are increasingly turning to 3D scanning for preservation and research, so specimens are handled less frequently, while public access to the material is simultaneously improved (Kuzminsky and Gardiner 2012; Rea et al. 2017; Jedrzejewski et al. 2020). Virtual curation of fragile material is particularly important in cases where funding is limited or not currently available for conservation (Means et al. 2013a, b).

184

K. E. Pearlstein et al.

Fig. 9.1 Mandible, 21-year-old female, osteosarcoma (AFIP 0384642). Micro-CT model (89μm) with ambient occlusion shading

9.2.2

External Surface Capture

External surface capture of pathological specimens was implemented with the NextEngine (nextengine.com) laser scanner (HD and UltraHD) and the Go!Scan structured light scanner (for larger specimens such as complete femora). Specimens scanned with the NextEngine laser scanner were edited in ScanStudio™ software to remove extraneous noise and align two or more scans and then were fused in RapidWorks into a solid mesh, which was exported as Wavefront objects (OBJ) and stereolithography (STL) models. Specimens scanned with the Go! Scan structured light scanner were edited and aligned using VXElements, exported as raw

OBJ files, and converted into solid OBJ and STL files using MeshMixer.

9.2.2.1 NextEngine Scanner Specimen scanning and model production with the NextEngine 3D scanner followed operation protocols outlined by Means et al. (2013b) and McCuistion et al. (2013). The scanner uses lasers to record topological (surface) attributes of a specimen. In normal operation, specimens are placed on a small platform that attaches to the NextEngine 3D scanner. This platform turns a specimen a full 360 degrees, and the lasers emanating from the NextEngine 3D scanner record the specimen as it rotates (Fig. 9.2). The small platform does not rotate continuously but rather at fixed intervals or “divisions.” At each

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens

185

Fig. 9.2 Anatomical specimen being surface scanned by the NextEngine 3D scanner at the National Museum of Health and Medicine, Silver Spring, Maryland, in May 2019. Photograph courtesy of NMHM

rotation, the NextEngine 3D scanner uses its lasers to record a “panel.” The small platform upon which specimens are placed needs 4 to 16 intervals to complete a 360-degree arc designed to capture all morphological data in a broad horizontal zone from the base to the tip of a specimen. As the number of panels increases, the amount of data generated to create a digital topological model—and the time needed to gather that data—increases as well. The opening setup screen for the ScanStudio™ HD Pro software allows the user to select from a number of options, including positioning (Fig. 9.3). The user has three positioning options from which to select: 360, bracket, and single. In order to create a full model of a specimen, the 360-degree positioning option is normally chosen. The divisions option allows the user to select the number of panels that will be recorded. Although the VCU Virtual Curation Laboratory initially recorded specimens at the highest number of divisions, this took considerable time for the complete rotation of a specimen—approximately 46 minutes—and resulted in large data files that were difficult to manipulate. A review of the literature and extensive experimentation by personnel associated with the Virtual Curation Laboratory indicates that the optimal scanning setup for most osteological specimens is an eight-panel interval. Fewer

panels are generally not advisable, and some specimens may require as many as 12 panels, but usually not more. To obtain a complete digital model of an entire specimen, all scanning times must be at least doubled, as detailed below. The user has the option to select the number of points per square inch recorded for each panel. These include quick, standard definition (SD), and high definition (HD) options. The HD option was ideal for most specimens. Quick and SD options create digital models that are relatively crude—especially if one intends for the digital model to be measurable with any degree of accuracy or serve as a form of virtual curation. The user next selects a target option: dark, neutral, or light. Target basically refers to the reflective qualities of a specimen. For most specimens, the neutral option should be selected. The light option should be reserved for bright objects, usually very white objects, such as historic ball clay pipes. The dark setting rarely works, as the lasers used by the NextEngine 3D scanner are simply absorbed by dark objects and do not reflect adequately back to the scanner. The final option the user must select is the range, which refers to the distance between the NextEngine’s 3D desktop scanner and the object being scanned. The option screen for ScanStudio™ HD Pro indicates the minimum, maximum, and ideal distances for each range

186

K. E. Pearlstein et al.

Fig. 9.3 Software interface of ScanStudio™ showing options for scan settings with the NextEngine laser scanner

option. It is very important to ensure that a specimen does not cross outside of the minimum or maximum distances of each range option—otherwise, a distorted digital model is generated. Time and memory are not options that the user directly selects but rather change depending on what options the user chooses for position, divisions, points per square inch, and target. The range option does not directly influence the time of the scan; the time indicated is in minutes and should be seen as a rough estimate. In some cases, the estimate was very accurate and in other cases was off by as many as 10 minutes. The ScanStudio™ HD Pro also does initial processing of an image after the lasers complete documenting a specimen—this is referred to as global aligning. The time needed for global

aligning is also extremely variable. Global aligning shows no obvious direct correlation to a specimen’s size, shape, or reflective characteristics. To the right of these options on the computer screen is a small viewing window created by the camera built into the NextEngine 3D scanner. It is important that a specimen fits completely within this window to ensure the entire item is scanned. Just above the viewing window are two arrows that allow the user to rotate the platform that attaches to the scanner. The user should do a full rotation of the specimen to ensure that the specimen is completely within the window—and also stable on the rotating platform. Each specimen needs to be held immobile as the small platform rotates. A problematic scan of a specimen will be

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens

generated if it moves or vibrates during the scanning process. Specimens were placed directly on a small adjustable steel mount covered with a removable rubber disk; the steel mount extends from a steel pole fixed into the rotating platform. The adjustable rubber-covered mount has sufficient flexibility to grip the base of the specimen, without being so rigid that damage to the specimen is a concern. An adjustable gripper with a rubber tip that can be moved and tightened against the top of a specimen extends from the steel rod. The gripper provides sufficient flexibility and limited rigidity to hold a specimen into place securely without damaging it. While each specimen is held secure on the rotating platform, the very base and very tip of the specimen is obscured from the NextEngine 3D scanner’s lasers. To capture these missing data, each specimen has to be physically rotated approximately 90 degrees and scanned again. More complicated shapes, such as human crania, may need more than two scans to fully capture all morphological data. When successful, the digital models generated from the NextEngine 3D scanner are generally accurate in how they record surface topology and can produce precise measurements but cannot always accurately capture important attributes such as color and fine texture details (Fig. 9.4). Digital topological models are dynamic Fig. 9.4 Vertebral column, 82-year-old male, DISH (NMHM 2012.0014). Photograph of actual specimen (left); NextEngine laser scan model (middle); photograph of the 3D printed and painted replica (right). Photograph courtesy of NMHM

187

representations that go beyond static images but still can lose critical details. The NextEngine 3D scanner is equipped with a basic camera that records color attributes but does not always record these attributes well. This is a significant limitation of the scanner. The NextEngine 3D scanner does a particularly poor job in recording the color of light-colored specimens under artificial lighting conditions—a greenish tint is added to the normal color of the specimen. This situation can be mitigated by 3D scanning with all external light sources turned off, but this is not always feasible in shared workspaces. Another option is to take high-quality color photographs of each scanned specimen and pair those with the digital models. It is even possible to use sophisticated photographic-editing software to “map” color photographs onto the digital models (Rea et al. 2017).

9.2.2.2 Go!Scan 50 3D Scanner The Go!Scan 50 uses structured white light rather than lasers to create a digital topological model and is best suited for medium to large specimens (Creaform 2016). The Go!Scan 50 compliments the NextEngine 3D scanner because the latter is essentially restricted to small and medium-sized specimens. With the Go!Scan 50, specimens do not need to be positioned in a particular way, unlike with the NextEngine 3D and its attached

188

K. E. Pearlstein et al.

rotating platform. As with the NextEngine, most specimens will need at least two scans. For the Go!Scan 50, a scanning episode needs to capture sufficient surface characteristics to ensure that it can be aligned with subsequent scanning episodes. If a particular specimen requires external support, this can be edited out through the VXelements software program, which is also used to align different scanning episodes and to create a final digital model. The Go!Scan 50 has a resolution of 0.5 mm and, as a consequence, is challenged with 3D scanning thin surfaces or edges. The scanning process is dynamic, and a user can watch the creation of a digital model as they scan. The Go!Scan 50 has a digital indicator to show if the scanner is held too far or too close to a specimen, but for practical applications, the operator does need to watch the digital model during the scanning process to ensure that a specimen is being properly captured in 3D. Operation of the Go!Scan 50 is computer-intensive. The VCU Virtual Curation Laboratory operates the Go!Scan 50 with a laptop that has 64 GB RAM; however, this is barely sufficient for large objects with complex shapes.

9.2.3

Internal Surface Capture

Internal surface capture of pathological specimens was implemented with a North Star Imaging micro-CT scanner. Specimens scanned with the North Star Imaging micro-CT were modeled and edited in Mimics or 3D Slicer and then exported as polygon file format (PLY) or STL files.

9.2.3.1

North Star Imaging Micro-CT Scanner The internal details of pathological skeletal specimens were captured using a North Star Imaging micro-CT machine X View CT model X5000, similar to a cone-beam scanner. This machine uses a rotating table rather than a sliding table like a medical CT scanner. During the scanning process, the programs utilized include the control program, which alters the beam parameter settings, and a visualization window that

interprets the effects of the settings on the data collected. The beam’s voltage and ampere parameters were modified in the controls program to generate the clearest vision of the specimen projected in the visual window. In the viewing window, a histogram box restricts what raw values of gray are visible. Values between 0 and 1 appeared as variations of gray, raw values closer to 0 were darker, and those closer to 1 were lighter. When determining the frequency and amplification to conduct a scan, the automatically populated histogram values were not cropped. The desired raw values range is 0.80–0.90 for the background, 0.10–0.20 for the darkest area of bone, and 0.45–0.60 for the majority of the bone. Filter overlay functions indicated if the parameters were distorting any pixels. For example, the Wallace filter identified defective pixels, which appeared smooth against a background that resembled static white noise. If there were too many of these smooth pixels, the amplification was lowered to lessen the image’s exposure. Once the optimal parameters for scanning each specimen were determined, gains were added to the scan calibration settings using the parameters created using gray-level mapping. With the specimen moved out of view in the visual window, kV and μA were adjusted to set the background to white (0.85–0.95), then black (0.10–0.20), and then a median (0.50–0.60). The last gain used the optimal parameters found for each specific specimen. When each gain was added, the histogram was cropped to focus on the peak’s raw values. Focusing on the pixels in the peak’s range showed defective pixels, which were added to a Defect Map. This allowed these pixels to be filled during capture to prevent empty pixels in the completed reconstruction. For scanning, specimens were wrapped in radiolucent egg crate foam and packaging tape and secured in a Styrofoam box. The packaging ensured that each specimen would not move or shift positions as the table turned during the scan, as this would create a blurry or inaccurate reconstruction. At points of contact with the foam packing and tape, the specimen was first wrapped

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens

189

Fig. 9.5 The image on the left represents a radiograph collected as a single projection during a micro-CT scan. Radiographs are reconstructed into a 3D volume exported

as Y-slices in DICOM format that can be viewed as orthogonal slices and/or 3D renderings or surfaces with CT viewing software such as 3D Slicer (right)

with laboratory paper to ensure that fragile osteological features were not altered. Several scanned specimens had features that differed from the average density of bone. These included bullets or metal fixtures that were much denser than the surrounding bone, meaning that the x-ray photons did not easily pass through the feature and subsequently altered the image produced from the scan. For this reason, higher kV and μA beam values were used with these specimens to enhance x-ray penetration of the more radio-dense areas. Some specimens included areas of low bone density or damage that were penetrated by too many photons, resulting in excluded features. These specimens used lower kV and μA values to accommodate the lower density areas. Using the specified parameters, each specimen was then scanned with a total of 500–1000 individual radiographs as the specimen was rotated a full 360 degrees. Following the specimen scan, a geometry scan was conducted using the large calibration tool. This scan provides a calibration for the reconstruction software based on the known shape of the geometry tool. One long bone specimen that did not fit entirely in the

scanning area was scanned using a Vortex scan. This scan requires the same steps as the other micro-CT scans but also includes a vertical change during the table rotation, which allows capture of the whole specimen. Three-dimensional reconstructions of each specimen were produced using the NorthStar reconstruction software, which incorporates individual radiographs and the calibration information from the geometry scan. Y-slices, similar to the axial slices (z-axis) of medical CT scan data, were extracted in both Digital Imaging and Communications in Medicine (DICOM) and TIFF file formats using optimal resolutions based on scan parameters, object size, and magnification with isotropic voxels (du Plessis et al. 2017) (Fig. 9.5). Final voxel sizes for micro-CT scanned specimens ranged from 60 μm to 159 μm.

9.2.3.2 Mimics Workflow Mimics (Materialise Medical Image Control System) is a fee-based 3D engineering software package produced by Materialise (Leuven, Belgium) for use in medical and anatomical applications. This workflow was developed using Mimics

190

version 22.0.0.524. Mimics uses segmentation to model stacked image data (e.g., DICOM and TIFF) from CT, micro-CT, and other medical imaging technologies. Mimics was selected for data modeling due to its user-friendly software interface and editing tools. Mimics Innovation Suite Base was used for this workflow, and models were finalized with Meshlab using a protocol similar to that devised by SimmonsEhrhardt, which is outlined in the following section. Importing DICOM and TIFF Users begin by creating a New Project and importing an image stack from the CT or micro-CT scan of the selected specimen (DICOM, TIFF, etc.). Copy the resolution information in the DICOM metadata to add to the online record (e.g., slice thickness ¼ 0.1213 mm). Once the stack is imported, adjust the radiograph contrast for the most accurate view of the specimen. The best visualization of the range of densities found throughout the specimen can be seen when adjusted to a near-black background and an appreciable range of gray values where only the densest of structures such as enamel and petrous region are near white. At this point, it may be desirable to reorient the 3D model if it is not in an anatomical position. This step is not absolutely necessary but can help with the efficiency and functionality of multi-slice editing, Smart Fill, and other program functions. Examine the model orientation in the 2D windows with the model oriented properly and not as a mirror image of the original. Use thresholding to select the range of grayscale values that best approximate the true appearance of the specimen and minimize artifacts such as scatter, then select Mask 3D Preview. If the specimen exhibits new bone proliferation from infection or a neoplasm, the best representation of the actual appearance should be selected. Instructions for adding back any thin, low-density features that are lost at this stage are outlined below. If the model is not positioned properly in the 2D and 3D windows, it can be resliced or oriented prior to any segmentation work. The visible

K. E. Pearlstein et al.

appearance of any masks should be checked against the specimen or a photo for accuracy. If the model position and orientation are accurate, the process can advance to refining the model. If the position is accurate and orientation is inaccurate, select Image ! Change Orientation and correct any orientations by right-clicking on the appropriate label (R/L, A/P, T/B). To reslice a model, select Image ! Reslice ! Draw Line to mark a known anatomical plane. Adjust the image width and height so that the box contains the entire model, and scroll through each of the views and 3D preview to confirm. When finished reslicing, readjust contrast and reselect layout again as outlined above. Set orientation as described above if necessary before making the first mask that will be used to make the model. Masks and Modeling Select New Mask and adjust threshold values to maximize cortical and trabecular bone, thin bony plates, and teeth and to minimize scatter. For specimens with complex morphology such as exostoses or osteomyelitis, use photos for reference or have the specimen available during modeling to ensure accuracy. To make a second mask without unconnected fragments, select Segment ! Region Grow. Leave the original mask and multilayer buttons checked and select 6-connectivity and then use the cursor to select the mask in one of the 2D views. The new mask will be the color automatically selected as Target, and any disconnected pieces will be eliminated. Eliminating Scatter/Unwanted Structures Eliminating scatter or unwanted structures utilizes the Split Mask function, as this separates bulk material into two masks: one to keep and one to ignore or discard. Select Segment ! Split Mask and label Regions A and B in different colors (e.g., scatter vs. osteological structures). Split Mask can also be used to separate larger bone elements and bone fragments. When selecting regions using Split Mask, it is easier and more precise to use a 2D view at full screen. Any remaining low relief can be smoothed later using Meshlab or another preferred mesh editing software.

9

Modernizing Medical Museums Through the 3D Digitization of Pathological Specimens

Replacing Lost Structures It may be necessary to replace modeled areas of thin bone that were lost during the first round of thresholding, particularly for areas of the mid-face or cranial base. This step can also be used to reduce scatter in a particular region. Use the Crop Mask function to select just the area that needs to be edited. Begin with Segment ! Multi-slice Edit ! Threshold and experiment with adjusting minimum (e.g., lower to add missing segments) and maximum (e.g., higher to remove scatter). Adjust the size of the tool and zoom in on the 2D view as needed. Make sure Auto-interpolate is selected. The program will interpolate data on slices between selected layers with a different color. The outcome may require additional smoothing in another program, but this can fill in thresholding artifact holes in thin areas of bone. Repairing Lost Structures Users may find it necessary to repair modeled areas of bone, particularly where the process of removing metal scatter or noise during thresholding has also removed bony areas that should remain intact. For this 3D modeling project, postmortem drill holes and similar taphonomic artifacts were preserved in as much detail as possible to reflect the appearance of the physical specimen. Select Smart Fill ! Local ! Mark Hole, then select the diameter, and manually fill identified holes. Modeling Under Masks in Project Manager, select Calculate Part ! Optimal. Under Options, check Shell Reduction to remove floating pieces. Set Largest Shells to 1. Select Smoothing if needed. Export the model as an STL file. By using custom parameters, the user can adjust the smoothing factor and triangle reduction method. The exported STL model can then be imported for finalization into Meshlab or another preferred mesh software.

9.2.3.3 3D Slicer Workflow 3D Slicer (www.slicer.org) (Fedorov et al. 2012) is a free, open-source program for visualization and analysis of imaging and volume data and 3D models. Image sequences can be imported in

191

several formats and converted to a single volume file with the Nearly Raw Raster Data (NRRD) extension. Additionally, all work in progress can be saved in a scene file in the Medical Reality Markup Language (MRML), linking the CT volume (NRRD), segmentation volume (SEG. NRRD), volume rendering settings (VP), and any generated 3D surface models as Visualization Toolkit (VTK), STL, OBJ, or PLY. Multiple options for importing micro-CT images are available in 3D Slicer depending on image format and software version. Importing DICOM For DICOM images with the DCM file extension, import into 3D Slicer is straightforward: select the Load DICOM Data button, and the DICOM browser opens as a separate window in v. 4.10.2 and as an embedded window in v. 4.11.20200930. Select the Import option (Import DICOM files button in 4.11.20200930) and navigate to the folder containing the DICOM slices. Once the images are loaded, some metadata appears in the browser, and a Series can be selected; click Examine, the scan will populate the Dicom data box, and then select the Scalar Volume and click Load. Data will be fully loaded when the orthogonal slice views (red, yellow, green) are populated. DICOM images contain image spacing information for all three axes; this can be checked with the Volumes module by opening the Volume Information tab. Image processing can proceed as described below, or the volume can be saved for later editing by clicking Save on the top toolbar and saving the NRRD volume (double-click on the name to change it), select the folder to save in, and click Save. Optionally, the scene MRML can also be saved at this point but is not necessary if no editing has been made. The saved NRRD volume or the scene MRML can be later loaded using the Load Data button or by simply dropping the file into the 3D Slicer workspace. Importing TIFF Sequences of TIFF images can be imported in three ways: (1) by importing them as an Image Sequence into Fiji/ImageJ (Schindelin et al. 2012) and saving as an NRRD volume (Buser et al. 2020), (2) importing via the

192

K. E. Pearlstein et al.

and dragging left and right and up and down on the CT slice views in v. 4.10 to minimize background materials/noise and maximize the visibility of the full range of bone voxels.

Load Data button in 3D Slicer v. 4.10.2 (click Choose File(s) to Add button, navigate to the first TIFF file, click Show Options, and uncheck Single File), or (3) using the ImageStacks module in 3D Slicer v. 4.11.20200930 to select all the TIFF files to be loaded (www.github.com/SlicerMorph/ SlicerMorph) (Rolfe et al. 2020). The first two methods require the image spacing to be manually entered in the Volumes module’s Volume Information tab after importing (a default spacing of 1 mm in all three axes is applied) and re-saving the updated NRRD. Optionally, for TIFF stacks imported into Fiji/ImageJ, the spacing can be adjusted in Fiji/ImageJ under Image ! Properties and saved as NRRD. For the third method, the ImageStacks module appears after the installation of the SlicerMorph extension and allows for the manual entry of image resolution before image loading, as well as an option for downsampling upon import by 50%, and an option for skipping slices upon loading (Rolfe et al. 2020). For this project, downsampling was not applied upon import to allow for evaluation and visualization of the original CT data before further processing.

Reducing Resolution Some volumes needed further resampling to reduce the resolution to facilitate 3D modeling, either in the z-axis alone or in all three axes using the Resample Scalar Volume module, entering the resampling values for each axis (equal to or a multiple of the original voxel resolution) and saving as a new NRRD volume. The resampled NRRD volume was re-imported to generate a new scene for segmentation and 3D modeling. The Volumes module was used to verify that image spacing in all axes matched the resampled resolutions. Note that resampling may result in a loss of some grayscale values if those voxels were “removed” in the resampling, so that the scalar range may show fewer values than 0 to 255 (see discussion at https://discourse.slicer.org/t/downsamplingnrrds/14033/4). For dry bone specimens with no wet soft tissue, the loss of these values is minimal and likely would not affect 3D models.

Converting to 8-Bit For images with few features (no desiccated soft tissue or little porous bone) or with very large file sizes (crania), images were converted from 16-bit (65,536 grayscale values) to 8-bit (256 grayscale values), using the workflow in 3D Slicer outlined on the SlicerMorph project website (https://github.com/ SlicerMorph/W_2020/tree/master/Lab11_ SlicerPlusPlus#rescalecast). The 16-bit NRRD volumes generated from either TIFF or DICOM sequences were imported, rescaled to grayscale values of 0 to 255, and saved as new 8-bit NRRD volumes. Proper scaling of 16-bit images to 8-bit images resulted in significantly reduced (approximately half) and more manageable CT volumes (usually