GeNeDis 2018: Computational Biology and Bioinformatics [1st ed.] 9783030326210, 9783030326227

Citation preview

Advances in Experimental Medicine and Biology 1194

Panayiotis Vlamos   Editor

GeNeDis 2018 Computational Biology and Bioinformatics

Advances in Experimental Medicine and Biology Volume 1194

Series Editors Wim E. Crusio, CNRS and University of Bordeaux UMR 5287, Institut de Neurosciences Cognitives et Intégratives d’Aquitaine, Pessac Cedex, France John D. Lambris, University of Pennsylvania, Philadelphia, PA, USA Heinfried H. Radeke, Clinic of the Goethe University Frankfurt Main, Institute of Pharmacology & Toxicology, Frankfurt am Main, Germany Nima Rezaei, Research Center for Immunodeficiencies, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran

More information about this series at http://www.springer.com/series/5584

Panayiotis Vlamos Editor

GeNeDis 2018 Computational Biology and Bioinformatics

Editor Panayiotis Vlamos Department of Informatics Ionian University Corfu, Greece

ISSN 0065-2598     ISSN 2214-8019 (electronic) Advances in Experimental Medicine and Biology ISBN 978-3-030-32621-0    ISBN 978-3-030-32622-7 (eBook) https://doi.org/10.1007/978-3-030-32622-7 © Springer Nature Switzerland AG 2020 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

To my father… who is gone.

Acknowledgment

I would like to thank Konstantina Skolariki for the help she provided during the editorial process.

vii

Contents

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques for the Evaluation of the Strategic Performance of an Emergency Department��������������������������������������������������    1 Panagiotis Manolitzas, Evangelos Grigoroudis, Jason Christodoulou, and Nikolaos Matsatsinis An Architecture for Cooperative Mobile Health Applications��������������������   23 Georgios Drakopoulos, Phivos Mylonas, and Spyros Sioutas An Educational Scenario for the Learning of the Conic Section: Studying the Ellipse with the Use of Digital Tools and Elements of Differentiated Instruction and Cognitive Neurosciences ������������������������   31 Afroditi Pantazi and Spyridon Doukakis Antibody Clustering Using a Machine Learning Pipeline that Fuses Genetic, Structural, and Physicochemical Properties����������������   41 Louis Papageorgiou, Dimitris Maroulis, George P. Chrousos, Elias Eliopoulos, and Dimitrios Vlachakis Neuroeducation and Computer Programming: A Review ��������������������������   59 Panagiota Giannopoulou, Mary-Angela Papalaskari, and Spyridon Doukakis WOL Ecosystem: Secure Remote Power – State Control of Computer(s) Over the Internet for Telemedicine Purposes and Dementia Patients������������������������������������������������������������������������������������   67 George Mantellos Epidemics Fuzzy Decision-Making Applications and Fuzzy Genetic Algorithms Efficiency Enhancement��������������������������������������������������������������   73 Elena Vlamou, Basil Papadopoulos, and Antonia Plerou

ix

x

Contents

Multivariate Data Analysis and Machine Learning for Prediction of MCI-to-AD Conversion������������������������������������������������������   81 Konstantina Skolariki, Graciella Muniz Terrera, and Samuel Danso A Deep Learning Approach for Human Action Recognition Using Skeletal Information ����������������������������������������������������������������������������  105 Eirini Mathe, Apostolos Maniatis, Evaggelos Spyrou, and Phivos Mylonas Artificial Neural Networks in Computer-­Aided Drug Design: An Overview of Recent Advances������������������������������������������������������������������  115 Dionysios G. Cheirdaris Computation of Dolphins’ Sound ASPL While Foraging����������������������������  127 Vasileios Petropoulos, Vasileios Podiadis, and George Verriopoulos Novel Cost Function Definition for Minimum-Cost Tractography in MR Diffusion Tensor Imaging��������������������������������������������������������������������  135 Konstantinos Delibasis, Christos Aronis, Michael Fanariotis, and Ilias Maglogiannis Platform for Managing Medical Data by Using a Passive Electronic Device ����������������������������������������������������������������������������  151 George Makropoulos, Ioannis Dontas, and Nikos Bogonikolos Digital Biomarkers Based Individualized Prognosis for People at Risk of Dementia: the AltoidaML Multi-site External Validation Study����������������������������������������������������������������������������������������������  157 Laura Rai, Rory Boyle, Laura Brosnan, Hannah Rice, Francesca Farina, Ioannis Tarnanas, and Robert Whelan Interconnections and Modeling Schemes of Kinesia Paradoxa ������������������  173 Eirini Banou Knowledge Discovery on IoT-Enabled mHealth Applications��������������������  181 Andreas Menychtas, Panayiotis Tsanakas, and Ilias Maglogiannis Computational Approaches Applied in the Field of Neuroscience��������������  193 Konstantina Skolariki and Themis Exarchos Drugena: A Fully Automated Immunoinformatics Platform for the Design of Antibody-Drug Conjugates Against Neurodegenerative Diseases��������������������������������������������������������������  203 Louis Papageorgiou, Eleni Papakonstantinou, Constantinos Salis, Eleytheria Polychronidou, Marianna Hagidimitriou, Dimitris Maroulis, Elias Eliopoulos, and Dimitrios Vlachakis Brain-Computer Interface Design and Neurofeedback Training in the Case of ADHD Rehabilitation��������������������������������������������������������������  217 Maria Sagiadinou and Antonia Plerou

Contents

xi

Particular Biomolecular Processes as Computing Paradigms��������������������  225 Konstantinos Giannakis, Georgia Theocharopoulou, Christos Papalitsas, Sofia Fanarioti, and Theodore Andronikos Decision Trees and Applications ��������������������������������������������������������������������  239 Georgios Karalis Utilization of New Technologies in the Production of Pharmaceutical Olive Oil����������������������������������������������������������������������������  243 Ioannis Vlachos, Romanos Kalamatianos, Ioannis Karydis, Antonia Spiridonidou, and Markos Avlonitis A Survey of Evolutionary Games in Biology ������������������������������������������������  253 Kalliopi Kastampolidou and Theodore Andronikos DTCo: An Ensemble SSL Algorithm for X-ray Classification��������������������  263 Ioannis Livieris, Theodore Kotsilieris, Ioannis Anagnostopoulos, and Vassilis Tampakas Cognitive Enhancement and Brain-­Computer Interfaces: Potential Boundaries and Risks����������������������������������������������������������������������  275 Polyxeni Kaimara, Antonia Plerou, and Ioannis Deliyannis Informatics and Cognitive Assessment: A RUDAS Scale Paradigm ����������  285 Athanasios Riganis Effective Stochastic Algorithm in Disease Prediction������������������������������    293 Romanos Kalamatianos, Stelios Gavras, Christos Boubouras, Dimitris Kotinas, and Markos Avlonitis A Systems Biology Approach for the Identification of Active Molecular Pathways During the Progression of Alzheimer’s Disease ��������������������������  303 Aristidis G. Vrahatis, Ilias S. Kotsireas, and Panayiotis Vlamos A Brief Survey of the Prisoners’ Dilemma Game and Its Potential Use in Biology����������������������������������������������������������������������  315 Kalliopi Kastampolidou, Maria Nefeli Nikiforos, and Theodore Andronikos Validation of Modeling and Simulation Methods in Computational Biology�������������������������������������������������������������������������������  323 Antigoni Avramouli An Improved Self-Labeled Algorithm for Cancer Prediction ��������������������  331 Ioannis Livieris, Emmanuel Pintelas, Andreas Kanavos, and Panagiotis Pintelas E-Health Application for E-Blood Analysis, E-Diagnosis, and Digital Diet Guidance ������������������������������������������������������������������������������  343 Pavandeep Singh

xii

Contents

Modeling the Critical Activation of Chaperone Machinery in Protein Folding��������������������������������������������������������������������������������������������  351 Georgia Theocharopoulou and Panayiotis Vlamos ANTISOMA: A Computational Pipeline for the Reduction of the Aggregation Propensity of Monoclonal Antibodies����������������������������  359 Katerina C. Nastou, Eleftheria G. Karataraki, Nikos C. Papandreou, Anna-­Isavella G. Rerra, Vassiliki P. Grimanelli, Ilias Maglogiannis, Stavros J. Hamodrakas, and Vassiliki A. Iconomidou A Microscale Mathematical Blood Flow Model for Understanding Cardiovascular Diseases ��������������������������������������������    373 Maria Hadjinicolaou and Eleftherios Protopapas Blockchain for Mobile Health Applications Acceleration with GPU Computing��������������������������������������������������������������������������������������  389 Georgios Drakopoulos, Michail Marountas, Xenophon Liapakis, Giannis Tzimas, Phivos Mylonas, and Spyros Sioutas Finite Size Effects in Networks of Coupled Neurons������������������������������������  397 Nefeli-Dimitra Tsigkri-DeSmedt, Panagiotis Vlamos, and Astero Provata Detecting Common Pathways and Key Molecules of Neurodegenerative Diseases from the Topology of Molecular Networks������������������������������������  409 Aristidis G. Vrahatis, Ilias S. Kotsireas, and Panayiotis Vlamos Robotic Systems Involved in the Diagnosis of Neurodegenerative Diseases������������������������������������������������������������������������  423 Anastasios Xiarchos Undergraduate Students’ Brain Activity in Visual and Textual Programming������������������������������������������������������������������������������  425 Spyridon Doukakis, Mary-Angela Papalaskari, Panayiotis Vlamos, Antonia Plerou, and Panagiota Giannopoulou Neuro-Fuzzy Networks and Their Applications in Medical Fields�������������  437 Elena Vlamou and Basil Papadopoulos Apache Spark Implementations for String Patterns in DNA Sequences��������������������������������������������������������������������������������������������  439 Andreas Kanavos, Ioannis Livieris, Phivos Mylonas, Spyros Sioutas, and Gerasimos Vonitsanos Big Data Analysis and Genetic Liability to Neuropsychiatric Disease��������  455 Panagiotis Roussos

Contents

xiii

Mathematical Models and Computer Simulation of Chemical Carcinogenesis Process and its Inhibition by Anticarcinogenic Substances����������������������������������������������������������������������  457 Paraskeui Koutsi, Spyros Ch. Karkabounas, and George Manis Transitioning Knowledge Levels Through Problem Solving Methods��������  459 Νikolaos Sampanis Cellular Phone User’s Age or the Duration of Calls Moderate Autonomic Nervous System? A Meta-Analysis��������������  475 Styliani A. Geronikolou, Ӧlle Johansson, George Chrousos, Christina Kanaka-­Gantenbein, and Dennis Cokkinos Logical Analysis and Validation of Publications in Bioinformatics������������  489 Konstantinos G. Papageorgiou Index������������������������������������������������������������������������������������������������������������������  493

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques for the Evaluation of the Strategic Performance of an Emergency Department Panagiotis Manolitzas, Evangelos Grigoroudis, Jason Christodoulou, and Nikolaos Matsatsinis

1  Introduction Generally, organizations are facing many obstacles in developing performance measurement systems that correspond to the appropriate measures. The need for such systems has led to the creation of the balanced scorecard (BSC). The BSC method is a strategic planning and management system which is widely used in business and industry fields, as well as in governmental and nonprofit organizations worldwide. The BSC, however, not only provides the ability to create a framework with performance indicators but also helps decision-makers to identify exactly what needs to be measured, the way the various goals are linked together, and what improvements can be made in order to achieve the planned strategic performance of the organization (Kaplan and Norton 1992, 1993, 1996a, b). While BSC was originally designed for business organizations Griffith (1994) reported the applicability of the BSC in healthcare organizations. More in-depth analysis of the applicability of BSC in healthcare organizations can distinguish at least three different philosophies in the development and the implementation of the BSC. These are (i) cases in which the classic model is used (Inamdar et al. 2002; Huang et al. 2004), (ii) cases with a few changes of the architectural logic of the BSC model (Urrutia and Eriksen 2005; Aidemark 2001; Radnor and Lovell 2003a, b; Giorgio Lovaglio and Vittadini 2012), and (iii) cases that the classic model has been transformed in the area of the number of dimensions and KPIs (Josey and Kim 2008; Curtright et al. 2000; Pink et al. 2003; Patel et al. 2008). P. Manolitzas (*) Department of Tourism Management, Ionian University, Corfu, Greece E. Grigoroudis · J. Christodoulou · N. Matsatsinis Technical University of Crete, School of Production Engineering and Management University Campus, Chania, Greece, Greece © Springer Nature Switzerland AG 2020 P. Vlamos (ed.), GeNeDis 2018, Advances in Experimental Medicine and Biology 1194, https://doi.org/10.1007/978-3-030-32622-7_1

1

2

P. Manolitzas et al.

Besides the aforementioned papers, in the last years, many researchers combined the BSC with other tools of operations management. More analytically, Grigoroudis et al. (2012) combined BSC with MCDA techniques in a Greek hospital. The proposed model was able to take into account the preferences of the management organization regarding the achievement of the defined strategic objectives. The results were able to help the organization evaluate and revise its strategy. Another approach that has been proposed was the combination of the BSC with the AHP. Karra and Papadopoulos (2005) through this combination developed the KPIs for a Greek hospital and then applied the AHP in order to select the most important metrics and then to evaluate their weights. Based on this combination, Huang (2009) proposed the BSC knowledge-based system. The scope of this system was to propose the usage of the AHP in order to prioritize all of the measures and strategies in the BSC, and moreover it had the ability to determine the weights of specific strategies. Zhijun et al. (2014), through a literature review, presented the applicability of the BSC at the Chinese hospitals. Moreover, via the data analysis, they found that a positive impact of BSC application on hospital performance is affected by the factors of operational scale, technological quality, and comprehensiveness of medical resources. Fletcher and Smith (2004) focused on the development of a complementary system of managerial metrics linking the EVA system to the BSC using the AHP. Using the AHP methodology, they proved that a comprehensive measurement system can be developed for assessing the performance of an organization. Reisinger et al. (2003) proposed the AHP process as a mechanism to prioritize the measures of the balanced scorecard for an organization. Stewart and Mohamed (2001) proposed a framework which applies the analytic hierarchy process (AHP) and multi-­ attribute utility theory (MAUT) to facilitate the aggregation of the obtained diverse performance measurements. The scope of their approach was to look at potential applications and benefits of using the BSC as a framework to evaluate the performance improvement resulting from IT/IS implementation by a construction organization. Chan (2006) in order to assess the scorecards that have been developed in hospitals, used the AHP process for the performance assessment. Lee et al. (2008) developed an approach based on the fuzzy analytic hierarchy process and balanced scorecard for the evaluation of an IT department. Based on  the aforementioned combination, they proved that said methodology can provide guidance to IT departments regarding strategies for improving department performance. Except AHP, several other multiple criteria methods like TOPSIS, DEMATEL, ANP, and fuzzy AHP have been combined with BSC (Valiris et al. 2005; Bezama et al. 2007; Wu et  al. 2009; Singh et  al. 2018; Hsu et  al. 2011; Ertuğrul and Karakaşoğlu 2009; Yüksel and Dağdeviren 2010). A recent literature review from Asgari and Darestani (2017) revealed that the most used methods that have been combined with BSC are AHP, ANP, and TOPSIS. Gonzalez-Sanchez et al. (2018) studied the applicability of the BSC in the healthcare sector via a literature review in Italy, Spain, and Portugal. In their published studies, they underlined that BSC is not a well-adopted tool for the hospital managers. Besides the combination of BSC with multiple criteria approaches, during the last 5 years, some researchers combined the BSC with the simulation in the health

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

3

sector. For instance, Ismail et  al. (2010) combined simulation with the balanced scorecard in order to improve the performance of an Irish hospital. Abo-Hamad and Arisha (2012) proposed a framework which integrates simulation modeling, balanced scorecard, and multicriteria decision analysis, aiming to provide a decision support system to emergency department managers. For the purpose of this paper, we combine discrete event simulation, balanced scorecard, and UTASTAR algorithm for evaluating the effects of reorganization scenarios in several KPIs that have been selected from the stakeholders. In order to aid the decision-maker (DM) during the decision-making process, we enable the UTASTAR algorithm. The main aim of the UTASTAR method is to enhance the decision-making process of the DM by analyzing the global preferences (judgment policy) in order to identify the criteria aggregation model that underlines the preferred results. The advantage of the UTASTAR method is that it gives the ability to the analyst to model the decision-­ making problem, providing the DM (director of the ED) analytical results that are able to elucidate his/her behavior. This paper is divided into four sections. In the first section (Sect. 1), we demonstrate the combination of the BSC with other approaches. Section 2 analytically presents the proposed integrated methodology. In order to test the applicability of the S-MEDUTA in Sect. 3, we evaluate the strategy of an emergency department in Greece during the economic crisis. For the selection of the KPIs, a literature review has been conducted. In order to examine the effect of each alternative in the selected KPIs, a simulation model has been developed. Finally, in Sect. 4, we summarize some concluding remarks and propose possible extensions for the proposed integrated methodology.

2  S-MEDUTA Methodology The proposed S-MEDUTA (Fig. 1) combines three tools of the operations management science, the balanced scorecard, simulation, and MCDA techniques. The key concept of the methodology is to identify the optimal strategy for the healthcare organization through the combination of BSC and simulation and moreover to enhance the decision-making process. In order to apply the methodology, a three-­ stage process has been developed. The main scope of the first stage is to represent the current situation in the ED. More specifically, through the observations of the processes, we reveal the existent problems and the weak points. In addition, through staff interviews, the research team collects information like personnel swifts, number of personnel, and other data vital for the development of the simulation model. It should be noted that the interviews from different stakeholders aid the operational researcher in more accurately developing the workflow of the emergency department. At the second stage, we analyze the data that have been collected. Based on the interviews with the stakeholders, a workflow diagram will be created in order to use it for the development of the simulation model. In addition, we analyze the data that represent the operation of the ED department, like waiting times, patient triage

4

P. Manolitzas et al.

Fig. 1  S-MEDUTA methodology

scale, type of incidence, etc. The aforementioned details will be used during the development of the simulation model. Based on the information that has been described in the previous paragraphs, a simulation model will be developed. Through the development of the simulation, the research team will have the ability to represent the operation of the simulation department in the software. At the third stage, the director of the emergency department chooses the KPIs that they want to examine using the simulation model. At this stage he/she can evaluate the effects of the hypothetical decisions (alternatives) on several KPIs. Through this process, the director has the chance to examine the effects of their decisions on the ED’s performance. Based on these results, a multicriteria table (i.e., performance matrix) may be developed, taking into account the alternative scenarios of the DM and the results of the SIMUL8 software. Based on the data of the multicriteria table and the preferences of DM, the team uses the UTASTAR algorithm in order to reveal the behavior of the director of the emergency department.

2.1  UTASTAR Method The UTASTAR method proposed by Siskos and Yannacopoulos (1985) is a variation of the UTA method, which aims at inferring a set of additive value functions from a given ranking on a reference set of functions. In the context of this method, the additive value function is assumed to have the following form:

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

5

n



u ( g ) = ∑ ui ( gi ) − σ + + σ − i =1

(1)



Under the following normalization constraints: n ∗ ∑ ui gi = 1 ∀i = 1, 2,…, n  i =1  u (g ) = 0  i i∗

( )



(2)

where g = {g1,g2, ….., gn} is the set of criteria; [gi∗, gi∗] is the criterion evaluation scale with gi∗ and gi∗, the worst and the best level of the i-th criterion; ui (i = 1, 2, ….., n) are the marginal value functions normalized between 0 and 1; and σ+ and σ− are the overestimation and the underestimation error, respectively, with n being the number of criteria. The UTASTAR method infers an unweighted form of the additive value function, equivalent to the form defined by relations (1) and (2), as follows: n

u′ ( g ) = ∑ ui ( gi )



i =1

(3)



under the normalization constraints: n ∗ ∑ ui gi = 1 ∀i = 1, 2,…, n  i =1  u (g ) = 0  i i∗

( )



(4)

where ui(gi∗) has the role of pi (weight of the i-th criterion). On the basis of the additive model (3), the value of each alternative a∈AR may be written as: n



u g ( a )  = ∑ ui  gi ( a )  − σ + ( a ) + σ − ( a ) i =1



(5)

where σ+ and σ− are the overestimation and the underestimation error, respectively, relative to u[g(a)]. Moreover, linear interpolation is used, in order to estimate the corresponding marginal value functions in a piecewise linear form. For each criterion, the interval [gi∗, gi∗] is cut into (αi − 1) equal intervals, and thus the end points gij are given by the formula: gij = gi ∗ +

j −1 ∗ gi − gi ∗ ∀j = 1, 2,…, α i αi − 1

(

)

(6)

6

P. Manolitzas et al.

The marginal value of an action a is approximated by a linear interpolation, and thus, for gi(α)∈[ gij , gij +1 ]:



( )

ui  gi ( a )  = ui gij +

gi ( a ) − gij

(

)

( )

ui gij +1 − ui gij   gij +1 − gij 

(7)

An important modification of the UTASTAR method concerns the monotonicity constraints of the criteria, which are taken into account through the transformations of the variables:

(

)

( )

wij = ui gij +1 − ui gij ≥ 0 ∀i = 1, 2,…, n and j = 1, 2,…,α i − 1



(8)

and thus, the monotonicity conditions for ui can be replaced by the non-negative constraints for the variables wij. Also, the set of reference actions AR = {a1, a2, …, am} is also “rearranged” in such a way that a1 is the head of the ranking (best action) and am its tail (worst action). Since the ranking has the form of a weak order R, for each pair of consecutive actions (ak, ak + 1), it holds either ak > ak + 1 (preference) or ak□ak + 1 (indifference). Thus, if ∆ ( ak , ak +1 ) = u[(g ( ak ) ]−u[ ( g ( ak +1 ) 





(9)

then one of the following holds: ∆ ( ak , ak +1 ) ≥ δ iff   ∆ ( ak , ak +1 ) = 0 iff



ak  ak +1 ak  ak +1

(10)

where δ is a small positive number so as to significantly discriminate two successive equivalence classes of R. Taking into account the previous conditions and assumptions, the UTASTAR algorithm may be summarized in the following steps: Step 1. Express the global value of reference actions u[g(ak)], k = 1, 2, …, m, first in terms of marginal values ui(gi) and then in terms of variable wij according to the formula (8), by means of the following expressions:

( )



 ui gi1 = 0 ∀i = 1, 2,…, n  1 j −  j ui gi = ∑ wit ∀i = 1, 2,…, n and j = 2, 3,…,α i − 1 t =1 

( )

(11)

Step 2. Introduce two error functions σ+ and σ− on AR by writing for each pair of consecutive actions in the ranking the analytics expressions:

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

∆ ( ak , ak +1 ) = u g ( ak )  − σ + ( ak ) + σ − ( ak ) −u g ( ak +1 )  + σ + ( ak +1 ) − σ − ( ak +1 )



7

(12)

Step 3. Solve the LP:



m  + − [ min ] z = ∑ σ ( ak ) + σ ( ak )  k =1  subject to ∆ ( ak , ak +1 ) ≥ δ if ak  ak +1   ∀k  ∆ ( ak , ak +1 ) = 0 if ak  ak +1    n ai −1  ∑ wij = 1 ∑ i =1 j =1  + −  wij ≥ 0, σ ( ak ) ≥ 0, σ ( ak ) ≥ 0 ∀i, j and k 

(13)



Step 4. Test the existence of multiple or near-optimal solutions of the LP (stability analysis); in case of non-uniqueness, find the mean additive value function of those (near) optimal solutions which maximize the objective functions: ai −1

( ) ∑w

ui gi∗ =

j =1

it

∀i = 1, 2,…, n

(14)



On the polyhedron of the constraints of the LP bounded by the new constraint: m



∑ σ + ( ak ) + σ − ( ak ) ≤ z∗ + ε k =1



(15)

where z∗ is the optimal value of the LP in step 3 and ε is a very small positive number.

3  Case Study 3.1  ED Description In order to demonstrate the practical applicability of the S-MEDUTA methodology, a collaboration with the ED of the General Hospital of Chania had been established. General Hospital of Chania is the unique hospital of the town of Chania, Greece, serving 67.000 citizens. It is a 400-bed hospital, providing all the types of health services. The scope of the management committee of the hospital is to provide valuable services to the citizens and moreover to reduce the length of stay, in order to minimize the costs on one hand and to maximize the level of satisfaction on the other hand. Several meetings were arranged with the manager of the hospital and

8

P. Manolitzas et al.

the director of the emergency department. The scope of these meetings was to gather statistical data like the number of patients that visited the ED, type of incidence, level of triage, and demographic data. Moreover, the research team during the phase of observation and through the discussion with the management committee of the hospital developed the workflow of the emergency department (Fig. 2). The General Hospital of Chania has two emergency departments. The first one runs 16 hours per day and the second one 24 hours per day. Generally, patients that arrive between 08:00 and 23:00 have to pass through registration. Depending on the triage (red case – extremely important), patients can skip registration and examination at ED1 and are sent directly to ED2. When a patient arrives at the emergency department prior to 23:00, they have to register at the registration office, provide data like name and age, and pay a flat rate for the examination. Afterward, they have to wait at the waiting room. A nurse will ask the patient about the problem that they face and will characterize the level of the triage. The patients that arrive by ambulance may skip this process and are sent directly to the ED2. It should be noted that the urgent patients, having the worst health problems or injuries, receive the highest priority. The scope of the ED2 is twofold. Patients that enter the emergency department of the hospital after 23:00 will be served from the ED2 because ED1 will be closed. The second scope is that ED2, usually in a 24-hour base, treats patients that face serious problems with their health, in other words those who belong to the yellow and red scale of triage. When a patient enters the room for diagnosis, the nurse will check their temperature, blood pressure, and heartbeat. Then the physician will provide initial examination. Depending on the level of triage, a patient waits for the lab results at the waiting room or on the bed. When the physician delivers the results of the examination, he/ she has three choices. The first choice, if the case is serious, is to send the patient to the appropriate department of the hospital. The second choice is to write a prescription and send the patient back home. The third choice is to decide that the patient will stay at the wards of the emergency department, in order to run more lab tests. Figure 2 depicts how the emergency department is assumed to work.

3.2  Key Performance Indicators Selection One of the most crucial parts of the BSC is the selection of the KPIs for the assessment of the strategy and the mission of an ED by taking into consideration the four perspectives of the BSC method. In order to examine the KPIs that have been globally used for the assessment of a healthcare organization, a literature review has been conducted. The first attempt was to identify as many relevant publications as possible. Based on the literature review, 33 cases were examined. An analytical presentation of the literature is demonstrated in Table 1. The scope of the table is to reveal the number of KPIs and dimensions that have been used for the evaluation of the strategy of healthcare organizations. Analyzing the

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

9

ARRIVAL AMBULANCE NO

YES

REGISTRATION NO

MIDNIGHT

ASSORTMENT ASSORTMENT

NO

YES YES DIAGNOSIS YES

ED II

EXAMINATIONS?

DIAGNOSIS

YES

NO

YES

EXAMINATIONS?

LABORATORY

NO

YES ED II

NO

CLINIC

NO

PRESCRIPTION DECISION YES

YES

PRESCRIPTION

EXIT

Fig. 2  ED workflow

results from the table, we can observe that the number of perspectives, for healthcare organizations, ranges from three to eight. For the development of the balanced scorecard, the operational researchers conducted a meeting with the stakeholders of the emergency department (ED director, doctors, nurses), in order to choose the KPIs that will be examined via the simulation model. The research team also revealed the KPIs that have been selected during the literature review process (Table 2).

10

P. Manolitzas et al.

Table 1  The applicability of the BSC in the healthcare sector via a literature review approach No. Organization 1 Hudson River Psychiatric Center 2 Duke Children’s Hospital 3 University of Colorado Health Sciences Center (Burn Center) 4 Mayo Clinic 5 6 7

8

Cambridge Health Alliance Community Memorial Hospital (CMH) Hospice unit of St Elsewhere Hospital

Location USA

Number of Date perspectives 1998 4

Number of indicators 15

USA

1999 4

22

USA

1999 4

13

USA

2000 8

13

USA

2000 4

44

USA

2000 –

13

USA

2001 4

11

Sweden

2001 4

14

Italy

2001 4

22

10

Long-term planning at Jönköping County Council Montefiore Medical Center Bridgeport Hospital

USA

2002 5

18

11

Bradford PCT

UK

2003 4

30

12

Bradford HIMP

UK

2003 4

29

13

Falls Memorial Hospital, International Falls Dutch Health System

USA

2004 4

37

Netherlands 2004 4

20

Sweden

2004 4

20

Taiwan

2004 4

9

UK

2004 3

35

USA

2005 4

27

9

14 15 16 17

18

A department of Swedish Hospital Emergency department in a hospital Mental Health Trusts and Providers of Services, Healthcare Commission Silver Cross Hospital

Source Wolfersteig and Dunham (1998) Meliones et al. (2001) Wachtel et al. (1999) Curtright et al. (2000) Hermann et al. (2000) Stewart and Bestor (2000) Kershaw and Kershaw (2001) Aidemark (2001) Trotta et al. (2013) Gumbus et al. (2002) Radnor and Lovell (2003a, b) Radnor and Lovell (2003a, b) Mohan (2004)

Ten Asbroek et al. (2004) Kollberg and Elg (2006) Huang et al. (2004) Healthcare Commission (2005) Pieper (2005) (continued)

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

11

Table 1 (continued) No. Organization 19 European Institute of Oncology (IEO) 20 Ospedale Pediatrico Bambino Gesù (OPBG) 21 Benito Menni Hospital 22 23 24

25 26 27 28 29

30 31 32

33

Mental Health NHS Trust (SWYT) National Health Service (NHS) National Hospital of Beijing/National Hospital of Nagoya National Health Service (NHS) Merseyside Health Authority Barbeton Citizens Hospital (BCH) Mackay Memorial Hospital (MMH) Medical Clinic at Högland Hospital (MHH) Private and Public Hospitals Academic medical center in Taipei Presidiums of Health Agency

Location Italy

Number of Date perspectives 2005 5

Italy

2005 6

Spain

2005 6

30+

UK

2006 4

23

UK

2006 3

36

China/ Japan

2006 4

19

UK

2006 3

37

Bevan (2006)

UK

2007 4

31

Chang (2007)

USA

2008 5

27

Taiwan

2008 5

12

Sweden

2009 4

22

Josey and Kim (2008) Chang et al. (2008) Aidemark and Funck (2009)

Lebanon

2011 3

21

Taiwan

2012 4

9

Italy

2012 4

25

2012 4

25

Governmental Hospital Greece of Didimoticho GHD

Number of indicators 19

Source Trotta et al. (2013) Trotta et al. (2013) Urrutia and Eriksen (2005) Schmidt et al. (2006) Patel et al. (2008) Chen et al. (2006)

Gumbus et al. (2002) Chen et al. (2012) Giorgio Lovaglio and Vittadini (2012 Grigoroudis et al. (2012)

The stakeholders, by taking into account the workflow, the weak points of the ED, the proposed KPIs (Table 3) of the research team, and the strategy of the organization, concluded in the evaluation of the following KPIs: personnel costs, revenue per admission for the financial perspective, waiting time, treatment time and length of stay for the customer perspective, bed occupancy rate, no. of admissions in ED2 coming from ED1 (ED{2from1}), and patients with examinations for the ED performance perspective as well as the workload of the personnel for the learning and growth perspective.

12

P. Manolitzas et al.

Table 2  Proposed KPIs Financial perspective Personnel costs Revenue per admission Customer perspective Waiting time

Aidemark (2001), Giorgio Lovaglio and Vittadini (2012), Peters et al. (2007), Chen et al. (2006) Aidemark (2001), Josey and Kim (2008), Chen et al. (2006) Inamdar et al. (2002), Patel et al. (2008), El-Jardali et al. (2011), Chen et al. (2006) Peters et al. (2007) Grigoroudis et al. (2012), Chen et al. (2006)

Treatment time LoS ED performance perspective Bed occupancy rate Grigoroudis et al. (2012), Inamdar et al. (2002), Schmidt et al. (percentage of occupancy) (2006), Aidemark (2001), Chen et al. (2006), Chen et al. (2012) No. of admissions Inamdar et al. (2002), Giorgio Lovaglio and Vittadini (2012), (ED{2froml}) El-Jardali et al. (2011), Chen et al. (2012), Chang et al. (2008) Patients with examinations Gagliardi et al. (2008) Learning and growth perspective Workload Giorgio Lovaglio and Vittadini (2012). Chen et al. (2006), Qingwei (2012) Table 3  KPIs selection from the stakeholders Perspectives Financial Personnel costs (€) Revenue per admission (€)

Customers Waiting time (minutes) Treatment time (minutes) LoS (minutes)

ED performance Bed cccupancy rate (percentage of occupancy) No. of admissions (ED{2from1}) Patients with examinations

Learning and growth Doctors’ workload (percentage) Nurses’ workload (percentage) Administrative employees’ workload (percentage)

3.3  Simulation Model Analysis Based on the workflow that has been designed with the collaboration of the stakeholders, the research team developed the simulation model of the ED. The major problem that they faced during the collaboration with the stakeholders was the lack of the data. The public hospitals in Greece don’t use any smart technologies in order to track the patients during their visit at the emergency department. In order to overcome this limitation, we established a team of six doctors in order to follow each patient. The personnel of the hospital filled one specific document during each patient’s arrival. These documents were, actually, recorded observations. Each recorded observation consisted of the following parameters: entry time at the hospital, registration time, entrance time at the examination room, diagnosis time, exit time, and date of entrance and departure. In addition to these parameters, the number of treatment facility (in our case, two in total), the number of the available doc-

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

13

Fig. 3  Simulation model

tors at the treatment facility, the triage, and the category of the event were recorded. It should be noted that this study is the first study that takes place in Greece in order to collect this kind of data and moreover to simulate an emergency department. Using the workflow of the emergency department and via the discussions that the research team had with the stakeholders, a simulation model using the SIMUL8 software was developed. Figure 3 depicts the simulator that has been developed for the emergency department.

3.4  Scenarios Selection The most crucial phase for the success of a simulation model is the development of alternative scenarios. Based on the literature, the majority of operational researchers develop scenarios using their experiences in similar projects without taking into account the views of the stakeholders. In order to overcome this obstacle, the research team decided to involve all the stakeholders for the development of the alternative scenarios. As we can see from Fig.  4, the research team organized a meeting at the hospital. Through the meeting the stakeholders expressed their views about the healthcare services and the hypothetical scenarios that can be used for the improvement of the healthcare services. Based on the discussions, the manager of

14

P. Manolitzas et al.

Fig. 4 Stakeholders Table 4  Alternatives analysis Alternative 1 Alternative 2

Baseline scenario Adding one doctor at ED II (+1Doc)

Alternative 3

Adding two nurses, one in each ED (+2Nurs)

Alternative 4

Subtracting one doctor and adding one nurse in ED II (−1Doc +2Nurs)

Alternative 5

Fast track design (Fast Track)

Alternative 6

Subtracting one doctor from ED II at the 5th alternative (Fast Track – 1Doc)

Alternative 7

Subtracting one doctor and one nurse from ED II at the 5th alternative (Fast Track – 1Doc-1Nurs)

the hospital and the ED director decided to implement scenarios like personnel adjustment and the implementation of a fast track area. The major reason that the committee decided the implementation of a fast track unit was the economic crisis which forced the patients to use the public hospital because the provided healthcare services are free of charge. Based on the discussion, the operational researchers developed the following alternative scenarios in order to measure the performance of the KPIs (Table 4). Another important factor for the evaluation of the simulation model is the ability of the simulator to mimic the current operation (baseline scenario) of the emergency department. One of the most important stages in order to check the ability of the simulation model to mimic the current operation of the hospital is to compare the values that have been gathered with the results of the simulation model. As we can

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques… Table 5 Simulation model evaluation

15

Model’s deviations from data (%) Queue for ED I 12.00 Queue for ED II 11.66 Queue for ED III 2.70 LoS ED I 11.09 LoS ED II 6.63 LoS ED III 0.62

see in Table 5, the deviations are low, assuming that the simulation model effectively represents the operation of the emergency department.

3.5  Simulation Results As noted in the previous sections, the added value of the simulation is that it can mimic the operations of an organization, and moreover it has the ability to predict the effect of each alternative scenario by taking into account several criteria. Thus, organizations that can implement a simulation project are more competitive among others because they have the advantage to examine the effects of decisions regarding several criteria. In our case, Table 6 demonstrates the applicability of the simulation at the emergency department and moreover the combination of simulation with the BSC approach. Studying the results, it is obvious that with the new directions that had been designed, the ED performs better than in its current situation. Analyzing the results from the learning and growth perspective, we observe that solutions like the implementation of a fast track unit improve the productivity of the healthcare resources like doctors’ and nurses’ workload. The alternatives that examine the adjustment of the personnel are not so effective, considering the four dimensions of the proposed BSC.

3.6  KPI Weights Based on the data that have been derived from the simulation (Table 6), the operational researchers asked the DM, in our case the ED manager, to rank the alternatives by taking into account the KPIs of each dimension of the balanced scorecard (Table 7). These rankings may be considered as a weak-order preference structure, while the process is repeated for each perspective of the BSC (Grigoroudis et al. 2012). Using the DM’s ranking, the UTASTAR algorithm will elucidate the preference system of the DM. Figure 5 depicts the criteria weights for each dimension of the BSC. Financial Perspective  Concerning the financial perspective, the most important KPI is the personnel cost with a relative importance of 59.18. The revenue per

16

P. Manolitzas et al.

Table 6  Simulation results (−1Doc BSC Baseline (+1Doc) (+2Nurs) +2Nurs) Financial perspective Personnel costs (€) 3200 3605 3610 3205 Revenue per 2.22 2.22 2.22 2.22 admission (€) Customer perspective Waiting time 11.08 11.08 7.24 7.63 (minutes) Treatment time 83.34 83.34 79.38 78.66 (minutes) Length of stay 95.92 95.92 88.12 87.79 (minutes) ED performance perspective Bed occupancy rate 32.95 32.95 30.46 30.06 (%) 5.00 5.00 5.00 5.00 No. of admissions (Tep{2from1}) Patients with 37.00 37.00 41.00 41.00 examinations (#) Learning and growth perspective Doctors’ workload 53.20 44.80 50.99 60.37 (%) Nurses’ workload 51.87 51.87 40.30 41.41 (%) 3.87 3.87 3.87 3.87 Administrative employees’ workload (%)

Fast Track

Fast Track Fast Track (−1Doc-­ (−1Doc) 1Nurs)

3200 2795 4.24 4.24

2545 4.24

8.69

8.69

28.60

82.37

82.37

83.21

94.15

94.15

114.90

29.07

29.07

31.33

0.00

0.00

0.00

38.00

38.00

35.00

55.69

68.54

72.15

59.11

59.11

72.45

11.89

11.89

11.89

admission seems to be not very important (40.82) for the ED director because the hospital belongs to the public sector. The revenue per admission had been established by the Ministry of Health in order to minimize visits from patients that did not face critical problems with their health status. Customer Perspective  Analyzing the results from the customer perspective, it is obvious that the most important KPI for the decision-maker is the treatment time. These results demonstrate the willingness of the manager to optimize the healthcare services and moreover to improve stages like exam results, etc. Based on this analysis, another important criterion for the DM is the LOS which has been affected by the treatment time. Contrary, the least important criterion for the DM is the waiting time (20.3). ED Performance Perspective  The most important criterion for the DM is the patients with examinations. The ED director gives special attention at this KPI because the numbers of examinations are affecting the economics of the emergency

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

17

Table 7  Alternatives ranking by the ED director (−1Doc BSC Baseline (+1Doc) (+2Nurs) +2Nurs) Financial perspective Personnel costs (€) 3200 3605 3610 3205 Revenue per 2.22 2.22 2.22 2.22 admission (€) Ranking 4 6 7 5 Customer perspective Waiting time (min) 11.08 11.08 7.24 7.63 Treatment time 83.34 83.34 79.38 78.66 (min) Length of stay 95.92 95.92 88.12 87.79 (min) Ranking 4 4 2 1 ED performance perspective Bed occupancy rate 32.95 32.95 30.46 30.06 (%) 5.00 5.00 5.00 No. of admissions 5.00 (Tep{2ffom1}) 37.00 37.00 41.00 41.00 Patients with examinations (#) Ranking 3 3 4 4 Learning and growth perspective Doctors’ workload 53.20 44.80 50.99 60.37 (%) Nurses’ workload 51.87 51.87 40.30 41.41 (%) 3.87 3.87 3.87 3.87 Administrative employees’ workload (%) Ranking 5 5 6 3

Fast Track

Fast Track Fast Track (−1Doc-­ (−1Doc) 1Nurs)

3200 4.24

2795 4.24

2545 4.24

3

2

1

8.69 82.37

8.69 82.37

28.60 83.21

94.15

94.15

114.90

3

3

5

29.07

29.07

31.33

0.00

0.00

0.00

38.00

38.00

35.00

2

2

1

55.69

68.54

72.15

59.11

59.11

72.45

11.89

11.89

11.89

4

2

1

department and moreover the LOS and the treatment time. Another crucial KPI for the decision-maker is the patients that moved from ED1 to ED2, because this KPI reveals medical errors that occurred during the triage evaluation process. The least important KPI at the performance perspective is the bed occupancy rate (17.22). This result reveals that the ED director does not give emphasis on the effectiveness of resource allocation and the operating performance of the emergency department. Learning and Growth Perspective  The least important KPI for the ED manager is the administrative employees’ workload because this stage of process does not have a direct impact on the performance of the ED. The most important KPI for the DM is the doctors’ and nurses’ workload, because these KPIs have a direct impact

18

P. Manolitzas et al.

Fig. 5  KPI weights for each perspective of the BSC

on the effectiveness of the emergency department and moreover can impact other KPIs like LOS and treatment time.

3.7  Performance Evaluation Another important result that has been derived from the UTASTAR algorithm is the evaluation of the ED performance using the hypothetical solutions that have been produced via the simulation model. Figure 6 presents the values and the results of the UTASTAR method (marginal utility functions) that are used to estimate the performance scores of the ED.  It is obvious that the ED director can effectively evaluate and revise the strategy of the ED, by using this type of graphs. By analyzing the ED’s performance, it is obvious that this strategic goal can been achieved if the organization develops a fast track unit. On the other hand, actions like the adjustment of the personnel are not so effective for the implementation of the strategy. The customer dimension actions like adding personnel or implementing a fast track unit are performing better among the rest of the actions. Observing the financial perspective, the best alternatives are the implementation of a fast track unit by adjusting the personnel. The least effective solutions are actions like the adjustment of the personnel. Analyzing the performance of the ED by taking into account the four dimen-

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

19

Fig. 6  Performance indicators

sions of the BSC, it is clear that the most effective solution is the implementation of a fast track unit (−1) doc. The worst alternative is the baseline scenario that represents the current situation at the ED. In addition, the adjustment of the personnel does not have a great impact on the strategy of the ED like the alternatives that implement the development of a fast track unit.

4  Conclusions The scope of this paper was to propose a new integrated methodology for the evaluation of the strategy of a healthcare organization. In order to find the optimal strategy, we propose the combination of BSC, simulation, and UTASTAR algorithm. Through the simulation model, the stakeholders have the ability to evaluate the effects of their decisions on a number of KPIs that are important for the successful implementation of the strategy of the ED. In addition, one of the most important results of the methodology is the weights of each KPI that have been produced by the UTASTAR algorithm. The weights represent which KPI is important for the decision-maker. It should be noted that the advantage of the UTASTAR method is its ability to represent the preferred system of the organization’s management with the minimum necessary information. Moreover, the method is able to provide a set of completed results (e.g., scores, weights, value functions, etc.), which may help the organization evaluate and revise its strategy (Grigoroudis et al. 2012).

20

P. Manolitzas et al.

Based on the recent advances on the area of operations research, the provided methodology could be improved in the near future with process mining techniques. Nowadays, these techniques are playing a vital role because they have the ability to quickly discover the real operations of an ED, aiding the research team to develop the simulation model more precisely. In addition, future research efforts must be focused in areas like group decision-making by enabling more decision-makers for the design and the execution of ED strategy.

References Abo-Hamad W, Arisha A (2012) Multi-criteria framework for emergency department in Irish hospital. In: Proceedings of the winter simulation conference, Winter Simulation Conference, p 88 Aidemark LG (2001) The meaning of balanced scorecards in the health care organisation. Financ Account Manage 17(1):23–40 Aidemark LG, Funck EK (2009) Measurement and health care management. Financ Account Manage 25(2):253–276 Asgari N, Darestani SA (2017) Application of multi-criteria decision making methods for balanced scorecard: a literature review investigation. Int J Serv Oper Manage 27(2):262–283 Bevan G (2006) Setting targets for health care performance: lessons from a case study of the English NHS. Natl Inst Econ Rev 197(1):67–79 Bezama A, Szarka N, Wolfbauer J, Lorber KE (2007) Application of a balanced scorecard system for supporting decision-making in contaminated sites remediation. Water Air Soil Pollut 181(1–4):3–16 Chan YCL (2006) An analytic hierarchy framework for evaluating balanced scorecards of healthcare organizations. Can J Admin Sci/Revue Canadienne des Sciences de l’Administration 23(2):85–104 Chang LC (2007) The NHS performance assessment framework as a balanced scorecard approach: limitations and implications. Int J Public Sect Manag 20(2):101–117 Chang WC, Tung YC, Huang CH, Yang MC (2008) Performance improvement after implementing the Balanced Scorecard: a large hospital’s experience in Taiwan. Total Qual Manag 19(11):1143–1154 Chen XY, Yamauchi K, Kato K, Nishimura A, Ito K (2006) Using the balanced scorecard to measure Chinese and Japanese hospital performance. Int J Health Care Qual Assur 19(4):339–350 Chen HF, Hou YH, Chang RE (2012) Application of the balanced scorecard to an academic medical center in Taiwan: the effect of warning systems on improvement of hospital performance. J Chin Med Assoc 75(10):530–535 Curtright JW, Stolp-Smith SC, Edell ES (2000) Strategic performance management: development of a performance measurement system at the Mayo Clinic. J Healthc Manag 45(1):58–68 El-Jardali F, Saleh S, Ataya N, Jamal D (2011) Design, implementation and scaling up of the balanced scorecard for hospitals in Lebanon: policy coherence and application lessons for low and middle income countries. Health Policy 103(2):305–314 Ertuğrul İ, Karakaşoğlu N (2009) Performance evaluation of Turkish cement firms with fuzzy analytic hierarchy process and TOPSIS methods. Expert Syst Appl 36(1):702–715 Fletcher HD, Smith DB (2004) Managing for value: developing a performance measurement system integrating economic value added and the balanced scorecard in strategic planning. J Bus Strateg 21(1):1 Gagliardi A, Lemieux-Charles L, Brown A, Sullivan T, Goel V (2008) Stakeholder preferences for cancer care performance indicators. Int J Health Care Qual Assur 21(2):175–189

S-MEDUTA: Combining Balanced Scorecard with Simulation and MCDA Techniques…

21

Gonzalez-Sanchez MB, Broccardo L, Martins Pires AM (2018) The use and design of the BSC in the health care sector: A systematic literature review for Italy, Spain, and Portugal. The International journal of health planning and management, 33(1): 6–30 Giorgio Lovaglio P, Vittadini G (2012) The balanced scorecard in health care: a multilevel latent variable approach. J Model Manag 7(1):38–58 Griffith JR (1994) Reengineering health care: management systems for survivors. J Healthc Manag 39(4):451 Grigoroudis E, Orfanoudaki E, Zopounidis C (2012) Strategic performance measurement in a healthcare organisation: a multiple criteria approach based on balanced scorecard. Omega 40(1):104–119 Gumbus A, Lyons B, Bellhouse DE (2002) Journey to destination 2005. Strategic Finance 84(2):46 Healthcare Commission (2005) NHS performance ratings 2004/2005. Commission for Healthcare Audit and Inspection, London, pp 82–83 Hermann RC, Regner JL, Erickson P, Yang D (2000) Developing a quality management system for behavioral health care: the Cambridge Health Alliance Experience. Harv Rev Psychiatry 8(5):251–260 Hsu CW, Hu AH, Chiou CY, Chen TC (2011) Using the FDM and ANP to construct a sustainability balanced scorecard for the semiconductor industry. Expert Syst Appl 38(10):12891–12899 Huang HC (2009) Designing a knowledge-based system for strategic planning: a balanced scorecard perspective. Expert Syst Appl 36(1):209–218 Huang SH, Wen-Yin C, Ping-Ling C, Lee HJ, Ming-Chin Y (2004) Using a balanced scorecard to improve the performance of an emergency department. Nurs Econ 22(3):140 Inamdar N, Kaplan RS, Reynolds K (2002) Applying the balanced scorecard in healthcare provider organizations/practitioner’s response. J Healthc Manag 47(3):179 Ismail K, Abo-Hamad W, Arisha A (2010) Integrating balanced scorecard and simulation modeling to improve emergency department performance in Irish hospitals. In: Simulation conference (WSC), Proceedings of the 2010 winter, December 2010, IEEE, pp 2340–2351 Josey C, Kim IW (2008) Implementation of the balanced scorecard at Barberton citizens hospital. J Corporate Account Finance 19(3):57–63 Kaplan RS, Norton DP (1992) The balanced scorecard - measures that drive performance. Harv Bus Rev 70:71–79 Kaplan RS, Norton DP (1993) Putting the balanced scorecard to work. Harv Bus Rev 71:134–147 Kaplan RS, Norton DP (1996a) Using the balanced scorecard as a strategic management system. Harv Bus Rev 4126:1–11 Kaplan RS, Norton DP (1996b) The balanced scorecard: translating strategy into action. Harvard Business Press, USA Karra ED, Papadopoulos DL (2005) Measuring performance of Theagenion Hospital of Thessaloniki, Greece through a balanced scorecard. Oper Res 5(2):289 Kershaw, R., & Kershaw, S. (2001). Developing a balanced scorecard to implement strategy at St. Elsewhere Hospital Kollberg B, Elg M (2006) Exploring the use of balanced scorecards in Swedish health care organizations. Asian J Qual 7(2):1–18 Lee AH, Chen WC, Chang CJ (2008) A fuzzy AHP and BSC approach for evaluating performance of IT department in the manufacturing industry in Taiwan. Expert Syst Appl 34(1):96–107 Meliones JN, Ballard R, Liekweg R, Burton W (2001) No mission no margin: it’s that simple. J Health Care Finance 27(3):21–29 Mohan J (2004) Monitoring quality through a balanced scorecard. Wiley, Hoboken Patel B, Chaussalet T, Millard P (2008) Balancing the NHS balanced scorecard! Eur J Oper Res 185(3):905–914 Peters DH, Noor AA, Singh LP, Kakar FK, Hansen PM, Burnham G (2007) A balanced scorecard for health services in Afghanistan. Bull World Health Organ 85(2):146–151 Pieper SK (2005) Reading the right signals: how to strategically manage with scorecards. Healthc Exec 20(3):8

22

P. Manolitzas et al.

Pink GH, Murray MA, McKillop I (2003) Hospital efficiency and patient satisfaction. Health Serv Manag Res 16(1):24–38 Qingwei F (2012) Research on evaluation index system of management effectiveness on hospital human resource based on balanced scorecard. Procedia Environ Sci 12:1040–1044 Radnor Z, Lovell B (2003a) Defining, justifying and implementing the Balanced Scorecard in the National Health Service. J Med Mark 3(3):174–188 Radnor Z, Lovell B (2003b) Success factors for implementation of the balanced scorecard in a NHS multi-agency setting. Int J Health Care Qual Assur 16(2):99–108 Reisinger H, Cravens KS, Tell N (2003) Prioritizing performance measures within the balanced scorecard framework. MIR: Manage Int Rev 43:429–437 Schmidt S, Bateman I, Breinlinger-O’Reilly J, Smith P (2006) A management approach that drives actions strategically: balanced scorecard in a mental health trust case study. Int J Health Care Qual Assur 19(2):119–135 Siskos Y, Yannacopoulos D (1985) UTASTAR: An ordinal regression method for building additive value functions. Investigaçao Operacional, 5(1):39–53 Singh S, Olugu EU, Musa SN, Mahat AB (2018) Fuzzy-based sustainability evaluation method for manufacturing SMEs using balanced scorecard framework. J Intell Manuf 29(1):1–18 Stewart LJ, Bestor WE (2000) Applying a balanced scorecard to health care organizations. J Corporate Account Finance 11(3):75–82 Stewart RA, Mohamed S (2001) Utilizing the balanced scorecard for IT/IS performance evaluation in construction. Constr Innov 1(3):147–163 Ten Asbroek AHA, Arah OA, Geelhoed J, Custers T, Delnoij DM, Klazinga NS (2004) Developing a national performance indicator framework for the Dutch health system. Int J Qual Health Care 16(suppl_1):i65–i71 Trotta A, Cardamone E, Cavallaro G, Mauro M (2013) Applying the Balanced Scorecard approach in teaching hospitals: a literature review and conceptual framework. Int J Health Plann Manag 28(2):181–201 Urrutia I, Eriksen SD (2005) Application of the balanced scorecard in Spanish private health-care management. Meas Bus Excell 9(4):16–26 Valiris G, Chytas P, Glykas M (2005) Making decisions using the balanced scorecard and the simple multi-attribute rating technique. Perform Meas Metrics 6(3):159–171 Wachtel TL, Hartford CE, Hughes JA (1999) Building a balanced scorecard for a burn center. Burns 25(5):431–437 Wolfersteig J, Dunham S (1998) Project report. Performance improvement a multidimensional model. Int J Qual Health Care 10(4):351–354 Wu HY, Tzeng GH, Chen YH (2009) A fuzzy MCDM approach for evaluating banking performance based on Balanced Scorecard. Expert Syst Appl 36(6):10135–10147 Yüksel İ, Dağdeviren M (2010) Using the fuzzy analytic network process (ANP) for Balanced Scorecard (BSC): a case study for a manufacturing firm. Expert Syst Appl 37(2):1270–1278 Zhijun LIN, Zengbiao YU, Zhang L (2014) Performance outcomes of balanced scorecard application in hospital administration in China. China Econ Rev 30:1–15

An Architecture for Cooperative Mobile Health Applications Georgios Drakopoulos, Phivos Mylonas, and Spyros Sioutas

Mathematics Subject Classification (2010)  05C12 05C20 05C80 05C85

1  Introduction Mobile smart applications for monitoring human health or processing health-related data are increasing at an almost geometric rate. This can be attributed to a combination of social and technological factors. The accumulated recent multidisciplinary research on biosignals and the quest for improved biomarkers create the need for advanced biosignal processing algorithms. Smartphone applications are progressively becoming popular in all age groups, albeit with a different rate for each such group. Moreover, mobile subscribers tend to be more willing to provide sensitive health data such as heartbeat rate, blood pressure, or eye sight status to applications for processing. Thus, not only technological but also financial factors favor the development of digital health applications. The primary contribution of this chapter is a set of guidelines toward a cross-­ layer cooperative architecture for mobile health applications. The principal motivation behind them are increased parallelism and consequently lower turnaround or wall clock time; additional redundancy, which translates to higher reliability; and lower energy consumption. All these factors are critical for mobile health applications.

G. Drakopoulos (*) · P. Mylonas (*) Department of Informatics, Ionian University, Corfu, Greece e-mail: [email protected]; [email protected] S. Sioutas (*) Computer Engineering and Informatics Department, University of Patras, Patras, Greece e-mail: [email protected] © Springer Nature Switzerland AG 2020 P. Vlamos (ed.), GeNeDis 2018, Advances in Experimental Medicine and Biology 1194, https://doi.org/10.1007/978-3-030-32622-7_2

23

24 Table 1  Notation of this chapter

G. Drakopoulos et al. Symbol ≜ {s1, …, sn} |S| or |{s1, …, sn}| E[X] Var [X] γ1

Meaning Definition or equality by definition Set comprising of elements s1, …, sn Cardinality of set S Mean value of random variable X Variance of random variable X Skewness coefficient

The remaining of this chapter is structured as follows. Section 2 briefly summarizes the recent scientific literature in the fields of edge computing, mobile applications, mobile services, and digital health applications. Section 3 presents the proposed architecture. Finally, Sect. 4 recapitulates the main points of this chapter. The notation of this chapter is shown at Table 1.

2  Previous Work Mobile health applications cover a broad spectrum of cases as listed, for instance, in Sunyaev et al. (2014) or in Fox and Duggan (2010). These include pregnancy as described in Banerjee et al. (2013), heartbeat as mentioned in Steinhubl et al. (2013), and blood pressure as stated in Logan et al. (2007). Using mobile health applications results in increased awareness of the digital health potential (Rich and Miah 2014). A major driver for the latter is the formation of thematically related communities in online social media as stated in Ba and Wang (2013). Another factor accounting for the popularity as well as for the ease of health applications is gamification (Lupton 2013; Pagoto and Bennett 2013), i.e., the business methodologies relying on gaming elements as their names suggest (Deterding et  al. 2011a, b; Huotari and Hamari 2012). Gamification can already be found at the very core of such applications as described in Cugelman (2013). The processing path of digital health may take several forms as shown in Serbanati et al. (2011). For an overview of recent security practices of mobile health applications, see Papageorgiou et  al. (2018). Path analysis as in Kanavos et  al. (2017) plays a central role in graph mining in various contexts, for instance, in social networks as in Drakopoulos et  al. (2017). Finally, the advent of advanced GPU technologies can lead to more efficient graph algorithms as in Drakopoulos et al. (2018). Finally, although it has been only very recently enforced (May 2018), GDPR, the EU directive governing the collection, processing, and sharing of sensitive personal information, seems to be already shaping more transparent conditions that the smartphone applications ecosystem is adapting to. In fact, despite the original protests that GDPR may be excessively constraining under certain circumstances as described in Charitou et  al. (2018), consumers seem to trust mobile applications

An Architecture for Cooperative Mobile Health Applications

25

which clearly outlines their intentions concerning any collected piece of personal information as Bachiri et al. (2018) found out.

3  Architecture This section presents and analyzes the proposed cooperative architecture for mobile digital health applications. Figure 1 visualizes an instance of a mobile health application running on a smartphone and a number of peers which can be reached either by Wi-Fi or by regular mobile services. As with the majority of mobile architectures, the proposed architecture is conceptually best described with graphs, as concepts such as connectivity and community structure can be naturally expressed. To this end, the cell phones, the base stations, and the Wi-Fi access points are represented as vertices, each device category being represented as a different type. Moreover, connections between these are represented as edges, where each edge is also of different type depending on the connection. These can be easily programmed in a graph database like Neo4j.

Fig. 1  Instance of a mobile application surrounded by peers

26

G. Drakopoulos et al.

The general constraints that will be the basis for the subsequent analysis are as follows: –– Assume that a mobile health application monitoring a biomarker or a biosignal must deliver results every T0 time units, usually seconds. Additionally, assuming that the required computation can be split into n + 1 parts to be distributed to the available n neighbors, then:

Ta + Tp + Ts + 2Tc ≤ T0

(1)



where Ta, Tp, Ts, and Tc denote, respectively, (i) the time required for analysis, namely, breaking down the computation and assigning each neighbor a task; (ii) processing, namely, the time of the slowest task; (iii) synthesis, namely, assembling the solution of each task to create the general solution; and (iv) communication. The latter term counts twice as the data and the task need to be communicated and then the results need to be collected. –– In mobile, communication for the minimization of the energy dedicated to a single task is of paramount importance. In general, the relationship between a given task and the energy spent for its accomplishment is unknown. However, given that tasks have a short duration, it is fairly reasonable to assume that the same function f(•) links the task and the energy at each neighbor. Then the following inequality should also be satisfied:



( n + 1) f (Tp ) + f (Ta ) + f (Ts ) + 2 ( n + 1) f (Tc ) ≤ f (T0 ) ⇔ f ( T0 ) − f ( Ta ) − f ( Ts ) −1 ≥ n f ( Tp ) + 2 ( Tc )

(2)

Given the fundamental constraints (1) and (2), let us estimate the key parameter Tc, since Ta, Ts, and Tp depend on the problem and T0 is a constraint. Let ei,j denote the communication link between vertices vi and vj and it has a given capacity Ci,j as well as a propagation delay τi,j. Then, the number of bits bi,j which can be transmitted over edge ei,j in a time slot of length t0 is, assuming the variables are expressed in the proper units:

bi , j = Ci , j (τ 0 − τ i , j )



(3)

If the link delay τi,j is expressed as a percentage 0