Innovative interventions for Parkinson's disease patients using iPrognosis games

Innovative interventions for Parkinson's disease patients using iPrognosis Games: An evaluation analysis by medical experts Sofia Dias* Ioannis Ioakeimidis Kosmas Dimitropoulos CIPER, Faculdade de Motricidade Humana Lisbon Portugal sbalula@fmh.ulisboa.pt Karolinska Institutet Sweden ioannis.ioakimidis@ki.se Information Technologies Institute, Centre for Research and Technology Hellas Thessaloniki Greece dimitrop@iti.gr Athina Grammatikopoulou Information Technologies Institute, Centre for Research and Technology Hellas Thessaloniki, Greece ngramm@iti.gr Nikos Grammalidis Information Technologies Institute, Centre for Research and Technology Hellas Thessaloniki Greece agramm@iti.gr José A. Diniz CIPER, Faculdade de Motricidade Humana Lisbon Portugal jadiniz@fmh.ulisboa.pt Theodore Savvidis Vicky Zilidou Lab of Medical Physics, Aristotle University of Thessaloniki Thessaloniki Greece vickyzilidou@gmail.com Lab of Medical Physics, Aristotle University of Thessaloniki Thessaloniki Greece teo.savvidi@gmail.com Evdokimos Konstantinidis Lab of Medical Physics, Aristotle University of Thessaloniki Thessaloniki Greece evdokimos@gmail.com Michael Stadtschnitzer Panagiotis D. Bamidis Lab of Medical Physics, Aristotle University of Thessaloniki Thessaloniki Greece pdbamidis@gmail.com Fraunhofer Institute IAIS Sankt Augustin Germany Michael.Stadtschnitzer@iais.fraunh ofer.de Dhaval Trivedi King’s College Hospital NHS Foundation Trust London United Kingdom dhaval.trivedi1@nhs.net Sevasti Bostantzopoulou Lisa Klingelhoefer Department of Neurology, Technical University Dresden Dresden Germany lisa.klingelhoefer@uniklinikumdresden.de Stelios Hadjidimitriou Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki Thessaloniki Greece stellios22@gmail.com Medical school, Aristotle University of Thessaloniki Thessaloniki Greece bostkamb@otenet.gr Zoe Katsarou Medical school, Aristotle University of Thessaloniki Thessaloniki Greece katsarouzoe@gmail.com Vasileios Charisis Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki Thessaloniki Greece vcharisis@ee.auth.gr Leontios J. Hadjileontiadis Dept. of Electrical Engineering and Computer Science, Khalifa University, Abu Dhabi, UAE/Dept. of Electrical and Computer Engineering, Aristotle University of Thessaloniki, Thessaloniki Greece leontios.hadjileontiadis@ku.ac.ae; leontios@auth.gr * All authors contributed equally to this work. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Permissions@acm.org. PETRA’20, June 30-July 3, 2020, Corfu, Greece © 2020 Association for Computing Machinery. ACM ISBN 978-1-4503-77737/20/06...$15.00 https://doi.org/10.1145/3389189.3397974 ABSTRACT Parkinson’s disease (PD) is a chronic and progressive neurodegenerative disease that affects ~7 million people worldwide, without any cure to date; however, it can be 288 PETRA’20, June, 2020, Corfu Island, Greece S. Dias et al. Motivated from the aforementioned perspectives, the principal objective of the ecosystem of the EC H2020 funded project iPROGNOSIS (www.i-prognosis.eu) is the development of: (i) an ICT-based behavioral analysis approach for identifying, as early as possible, the PD symptoms appearance, and (ii) the application of ICT-based interventions countering identified risks. To those identified and clinically validated as early stage PD patients, ICT-based interventions are provided via the iPROGNOSIS Intervention Platform, including: (i) a Personalized Game Suite (ExerGames, DietaryGames, EmoGames, Handwriting/VoiceGames) for physical/emotional support, and (ii) assistive interventions for voice enhancement and gait rhythm guidance. Overall, i-PROGNOSIS leverages and extends the stateof-the-art in a number of different areas, such as behavioral, physiological and lifestyle monitoring, motion capture, physical activity evaluation, personalized gaming, home-based humancomputer interfaces, multi-parametric data modelling and decision support systems, ensuring valuable intellectual property. In this vein, for the present study, part of the i-PROGNOSIS ecosystem is presented, namely the iPrognosis Games, along with the opinion/feedback of health and medical experts from the three collaborating medical centres of the i-PROGNOSIS consortium, in Greece (Aristotle University of Thessaloniki-Medical School, AUTH-MED), in Germany (Technischen Universität Dresden, TUD) and in the United Kingdom (King's College London, KCL). symptomatically treated. In this vein, innovative technologies can be used for the objective assessment of clinical symptoms and to provide supportive therapies at home. The present work explores the processes and the outcomes of the i-PROGNOSIS (www.iprognosis.eu) intervention deployment in three PD clinical centres (Greece, UK, and Germany). For that purpose, 36 PD patients were recruited to voluntarily participate in the i-PROGNOSIS feasibility study, spread across the three different countries. The PD patients interacted with the i-PROGNOSIS system for up-to-three months, mainly within the clinical environment, using the provided iPrognosis Games in dedicated gaming stations that were setup in the corresponding clinical centres. Overall, the results show that the iPrognosis Games were positively evaluated by medical experts. Moreover, based on the collected feedback, the iPrognosis Games have achieved their main goals of providing an innovative, objective and usable system for the monitoring of early PD (motor and non-motor) symptomatology, by providing tools for complementing existing clinical interventions for the improvement of PD patients’ quality of life. CCS CONCEPTS • Human-centered computing • PD management • Serious Games • PD interventions • PD monitoring tools • Healthy and Active Ageing KEYWORDS Parkinson’s disease (PD), i-PROGNOSIS, iPrognosis Games, iPROGNOSIS Motor Assessment Tool (iMAT), Medical Experts’ Evaluation 2 THE iPROGNOSIS GAMES Overall, the iPrognosis Games aim to monitor and support older adult’s physical and emotional status enhancement, towards the decrease of the PD-related risks and increase of their QALYs. The iPrognosis Games application, an Android mobile application, already available on Google Play Store1, consists of 14 different games, taking into account the PD symptoms (i.e., Exergames, Dietarygames, Emogames, Handwriting and Voicegames), constituting the Personalized Game Suite (PGS), along with the Warming up game and motor Assessment Tests. Moreover, the iPrognosis Games differ in the required equipment for playing them, that also defines different target uses; more specifically: i) Games requiring the MentorAge®2 sensor device (four games) are targeting to be used inside specialised PD clinical centres or at home if the user purchases the sensor device, and ii) Games requiring a smartwatch or tablet (with microphone and frontal camera) (ten games) can be played either at home environment or be played under supervision in clinical settings. ACM Reference format: Sofia Balula Dias, Ioannis Ioakeimidis, Kosmas Dimitropoulos, Athina Grammatikopoulou, Nikos Grammalidis, José Alves Diniz, Vicky Zilidou, Theodore Savvidis, Evdokimos Konstantinidis, Panagiotis D. Bamidis, Michael Stadtschnitzer, Dhaval Trivedi, Lisa Klingelhoefer, Sevasti Bostantzopoulou, Zoe Katsarou, Stelios Hadjidimitriou, Vasileios Charisis, and Leontios J. Hadjileontiadis. 2020. Innovative interventions for Parkinson's disease patients using the iPrognosis Games: An evaluation analysis by medical experts. In Proceedings of ACM PETRA conference (PETRA’20). ACM, Corfu Island, Greece, 8 pages. https://doi.org/10.1145/3389189.3397974 1 INTRODUCTION Parkinson’s disease (PD) is progressive and one of the most common neurodegenerative disorder, affecting approximately 1% of individuals over the age of 60, causing progressive disability that results in a burden of ~2.2 million disability-adjusted life years (DALYs), exhibiting the greatest loss of quality-adjusted life years (QALYs) among 29 major chronic conditions [1], [2]. In the area of neurological disease, research is being conducted into novel detection, assessment and treatment technologies based on motion analysis [3], robotics [4], and rehabilitation fields [5], among others. 1 2.1 Warm-up Game The i-PROGNOSIS Games application includes a Warm-up Game (see Figure 1) that was developed with a dual purpose. Initially, this 2 https://play.google.com/store/apps/details?id=com.TeoS.iPGS&hl=en 289 http://www.nively.com/en/ Innovative interventions for Parkinson's disease patients using iPrognosis Games: An evaluation analysis by medical experts PETRA’20, June, 2020, Corfu Island, Greece Figure 1: Indicative example of the user interface of the i-PROGNOSIS Warm-up Game. (Top panel) Gaming session with one fixed virtual object (apple) for the patient to collect using hand movements. (Bottom panel) Game’s static sequence of apple positions (green numbers: objects in front of the patient in the 3D space; red numbers: objects in parallel to the patient’s). Figure 2: (Top panel) PD patient performing the Leg Agility Test of the i-PROGNOSIS Motor Assessment Tool (iMAT). (Bottom) User interface of the Assessment Test 2 (Leg Agility). The user’s and the expert’s performance are compared in real-time side-by-side, and quantified through the level of agility agreement depicted in the score in the middle of the screen. game was designed in order to familiarize the user with the virtual environment and the gamification element in the intervention platform. Next, this game served as a real warming-up session designed to increase blood flow and oxygen to the main muscles, reducing the chance of soft tissue injuries (i.e., ligament, tendon and muscle), facilitating muscles and joints to move through a greater range of motions. The Warming-up game scenario consisted of an avatar controlled by the players’ body movements that were detected using a depth sensor. The patients could move their hands around freely to collect the apples that appear on their screen. For each appearing apple, players had a time window of 5 seconds to successfully catch it. Each game session consisted of a static sequence of nineteen positions where apples appear one after the other. For each successful try, players’ score was boosted and positive auditory feedback was provided. After a player’s performance was completed, an overall evaluation score was communicated, along with an appropriate feedback message. In order to estimate the player’s game performance, the time required to catch an apple for each position, the overall game duration, the failures and their score are recorded. In addition, during each gameplay, the RGB stream of the depth sensor was optionally captured and saved at 16fps, in order to be used for further motion data analysis. More detailed information can be found in [6]. 2.2 Motor Assessment Tool Apart from the Warming up Exergame, in order to evaluate the effect of the iPrognosis Games on the PD patient’s status, the iPROGNOSIS Motor Assessment Tool (iMAT) was constructed to complement the i-PROGNOSIS PGS (see section 2.3), and also creating the opportunity to deploy this component as a stand-alone assessment module. It is important to note that the motor aspects within iMAT are also reflected within the general design of the Exergames included in the i-PROGNOSIS Games. Thus, iMAT can be seen as a complement to PD patient’s conventional motor status monitoring, but also it can be used as a tool for measuring the intervention’s progress [7]. 2.3 Personalized Game Suite Finally, the i-PROGNOSIS Personalized Game Suite (PGS) includes a variation of Serious Games (SGs) in a unified and personalized platform for PD symptoms management. The PGS consists of 14 different games, namely: three Exergames targeting whole-body motor symptomatology (i.e., Fishing, Picking Citrus Fruit, and Kinematic Orchestra), three Dietarygames targeting improvements in dietary characteristics (i.e., Eatwell Plate, Sudoku and Retrain Eating Behaviour), two Emotional games targeting 290 PETRA’20, June, 2020, Corfu Island, Greece S. Dias et al. Table 1. Basic characteristics of the PD patients included in the intervention data sample across the three iPROGNOSIS medical centres Clinical Centre Male Total Age (years); mean (st.dev) AUTH 2 11 13 65.5 (8.9) 7 5 12 58.5 (8.8) 5 6 11 65.7 (10.3) 22 36 63.2 (9.7) KCL TUD Female 14 More detailed information about the design and development of the games can be found in [8]-[13]. 2.1 Figure 3: Indicative example of the user interface of the iPrognosis Game selection. (Top panel) Sub-categories of the Personalised Game Suite, and (Middle and bottom panels): Available games in each sub-category. emotional symptomatology aspects (i.e., Face Rhapsody and Imitation game), three Handwriting games for training finger finemotor skills (i.e., Driving, Find me, and Writing letters) and three Voice games for the improvement of speech characteristics (i.e., Catching Voicimons, Pop the balloons and Good night story) (see Figure 3). Moreover, the developed battery of 14 interventional and qualityof-life improving Games are based on the following PD targeted symptomatology: 1. Motor-improvement (six games): posture, gait, postural instability, fine motor skills impairment and tremor; 2. Non-motor Improvement (five games): facial expression/depression and constipation; and 3. Speech-improvement (three games): voice and speech difficulties. Tackling the most often appearing PD symptoms via the gamification concept, iPrognosis Games can provide a userfriendly environment with bilateral functionality, i.e., both engaging the user into long-lasting supportive activities, which could retard the evolution of the symptoms, and providing to the physician, the necessary information for monitoring the PD patient’s status in a quantitative and detailed way. The iPrognosis Games differ in the required equipment for playing them, that also defines different target uses; more specifically: i) Games requiring the MentorAge®3 sensor device (four games) are targeting to be used inside specialised PD clinical centres or at home if the user purchases the sensor device (~€700), and ii) Games requiring a smartwatch or tablet (with microphone and frontal camera) (ten games) can be played either at home environment or be played under supervision in clinical settings. 3 Intervention Protocol The i-PROGNOSIS interventional data feasibility study included 36 PD patients spread across the three clinical sites of the iPROGNOSIS project, i.e., AUTH-MED), in Germany (TUD) and in the United Kingdom (KCL). The PD patients were recruited and participated in the study from September 2019 to January 2020. Table 1 presents the basic characteristics of the study sample and their distribution among the three clinical centres. The protocol included periodic sessions (one per month) of the iPrognosis Assessment Tests under the controlled environment at the three medical centres. The PD patients could perform each Assessment Test up to four times per session. At the beginning of each session of the Assessment Tests, an exercise game (Warm-up Exergame), was first performed by the PD patients, mainly to provide the PD patient with an initial contact with the sensor device and related environment and to gradually increase their heart rate and body temperature, including general coordination activities to prepare the PD patients for the preceding Assessment Tests. At the same time, all PD patients were motivated to engage with every category of games, according to their availability and participant’s willingness, but they were motivated to focus on the specific groups of games and follow the medical protocol based on the specific set of symptoms of the PD patient. In addition, each medical center involved in the present study was equipped with the depth sensor tracking system Mentorage® and assisted with technical support. Overall, the PD patients were able to perform the iPrognosis Warm-up game, the motor Assessment Tests and the PGS by moving in front of the MentorAge® sensor, placed at least 1.5 meters in front of it, as it can be seen in Figure 2. The acquired data were de-identified and securely stored in the backend database of the i-PROGNOSIS. 3 METHODOLOGY A largely influential report by Wyatt and colleagues [14] sets the aims of medical evaluation of novel medical informational systems based on three main questions: http://www.nively.com/en/ 291 Innovative interventions for Parkinson's disease patients using iPrognosis Games: An evaluation analysis by medical experts PETRA’20, June, 2020, Corfu Island, Greece Are they effective? Are they safe? Do they change the use of health care resources? Indeed, this basic framework has since resulted in formative efforts of evaluating the real-world medical value of novel informational medical platforms, including both systems for optimal use of existing resources (e.g., Electronic Health Records [15]) and novel diagnostic [16], monitoring [17] and interventional [18] systems for a variety of diseases and medical conditions. In many of the above examples, the medical expert evaluation processes place the medical and clinical experts’ opinions for a system’s performance as the central point of reference. Understandably, these user groups are the eventual mediators for the widespread adoption of such systems within the pre-medical and medical practices, having the potential of reducing the, typically long [19], adoption periods for translational and methodological innovations. A more systematic assessment of medical expert evaluations [20] sets three potential purposes for the evaluation process, namely focusing on: i) comparing the results with the goals and the expected effects of a novel system, ii) directing further development work towards the expected result, and iii) using the findings and outcomes of the evaluation process as an experience base for the next round of development. However, past experiences point towards various methodological difficulties when innovative informational medical systems are being evaluated. Such issues mainly arise due to [20]: a) the innate complexity of modern informational platforms, making it difficult for the medical expert to evaluate the methodological nuances of the innovative approach, and b) unfeasibility of proper randomised and blinded medical system evaluation protocol, due to the, often, lack of comparable traditional medical evaluations and processes. Nevertheless, the need to evaluate medical systems, despite the difficulties, is often emphasized [21], stating the central role of collecting valid expert feedback through careful and proper analysis and presentation of system components and outcomes to the target user and expert groups [22]. Figure 4: Medical Expert evaluation Form regarding the i-PROGNOSIS interventions. custom, need-focused expert evaluation form (with some inspiration from contemporary similar efforts [24]). The final format of the Expert Evaluation Form regarding the i-PROGNOSIS interventions can be found in Figure 4. 3.2 Data analysis The analysis of the collected expert feedback was performed using traditional qualitative analysis techniques, focusing on the descriptive analysis of the results, in order to evaluate the overall attitudes of the experts towards the i-PROGNOSIS’ achievements. Specifically, all the ordinal evaluation questions were descriptively analysed, based on the ranges of the collected responses [25]. Thus, for the provided seven-step evaluation inquiries, answers were coded from 1 to 7 and the responses within the 0-2 were ranked as “Strong disagreement”, answers coded as 3-5 were rated as “Neutral opinion” and 6-7 answers were coded as “Strong agreement”. Similarly, thematic analysis was used for the analysis of the freetext answer, where content analysis of each provided answer was performed [26]. To achieve that, a matrix of open codes per item was created, in order to support code repetition analysis [27]. In summary, each provided answer was contextually analysed and repeated common concepts were quantified. Finally, the frequency rates for contextually independent concepts, within the overall pool of quantified concept-mentions, were calculated. The results are 3.1 Data collection Methodologically, for the formulation of the “Expert Evaluation Form” regarding the i-PROGNOSIS interventions (see Figure 4), the existing knowledge and literature-based framework presented in previous section was considered. Initial efforts towards this end were focused on identifying pre-existing questionnaires and evaluation frameworks for evaluations of novel medical informational and decision-support systems that would be appropriate for the i-PROGNOSIS framework. However, due to the system-specific characteristics and the innovative nature of the iPROGNOSIS effort, the above task has not been possible. Indeed, the difficulties for the creation of proper, all-encompassing evaluation frameworks for existing decision-support systems have been identifies many times (e.g., [23]) this was a challenging task. Thus, the i-PROGNOSIS project has opted for the creation of a 292 PETRA’20, June, 2020, Corfu Island, Greece S. Dias et al. Table 3. Frequency analysis of the expert opinions with regards to the i-PROGNOSIS Intervention system Table 2. Descriptive characteristics of the responders that provided feedback through the Expert Evaluation Forms Clinical Centre Experts (n) TUD KCL AUTH 3 5 10 Strong disagreement Average Clinical Experience (yrs) in PD 7.0 7.0 6.4 Strong Agreement Question 1: The i-PROGNOSIS system has the potential to add value to existing clinical interventions provided by specialised PD centres (n=18) 0.0% presented in the following subsection in the form of ranked concepts mentioned, sorted from the most frequent to the least common ones. 33.3% 66.7% Question 2: The i-PROGNOSIS system has the potential to add value to existing clinical interventions provided by primary care centres and/or outpatient care providers (n=18) 4 RESULTS AND DISCUSSION 0.0% After reaching out to the clinical network of the i-PROGNOSIS clinical partners (AUTH-MED, TUD, and KCL), 18 Expert Evaluation Forms were collected. The sample included clinicians with hands-on experience with the i-PROGNOSIS platform (n=15), as well as a limited number of “innocent” responders (n=3) without prior i-PROGNOSIS experience. All the respondents had significant clinical experience in providing health care for PD populations (mean years of experience: 6.2), while a significant proportion of the respondents (n=11), also had experience from clinical and medical research in the field of PD development, intervention and management. The descriptive characteristics of the respondent sample can be seen in Table 2. The Expert Evaluation Form was focused on collecting feedback from the i-PROGNOSIS expert network concerning the potential of the intervention data system module that concerns the development and the deployment of the i-PROGNOSIS Interventions through 14 interventional and quality-of-life improving SGs. These were grouped into three main categories based on the types of targeted symptomatology: 22.2% 77.8% Question 3: The i-PROGNOSIS system has the potential to add value to existing clinical interventions, through patient self-monitoring of disease progression (n=18) 5.6% 33.3% 61.1% Question 4: The i-PROGNOSIS system could be intergraded with current patient monitoring and support systems (n=17) 5.9% 11.8% 82.4% In general, the experts’ opinions were also fairly positive for the self-monitoring and integration potential of the system to current monitoring and support systems, despite a small segment of the sample that was negative. The results can be seen in Table 3. In addition, overall percentages of strong agreement among the respondents are presented in Figure 5. Overall, the results are positive, despite the comparatively immature state of deployment and evaluation process. These results create optimism about the clinical value of the developed system once the evaluation efforts are expanded and properly disseminated among the medical community (beyond the lifecycle of the iPROGNOSIS project). The experts also evaluated the three (3) main advantages of the developed Interventions data module with regards to its value as a quality-of-life improving tool throughout the PD development (see Table 4). Their feedback mainly focused on the accessibility of the system for patients and clinicians and the attractive aspect of the game design that has the potential to promote longer-term use. Another frequently mentioned point was the fact that the platform facilitated self-monitoring of the disease, often conceptually connected with the increase of patient self-awareness about the progress of the monitored symptoms. With regards to the main challenges for the use of the Interventions data module as a tool for improving quality-of-life, PD monitoring and interventions throughout the PD development, the experts emphasized the need for further optimisation in the design, the functionality, the feedback to the patients and the technical 1. Motor-improvements (6 games), 2. Non-motor improvements (5 games), and 3. Speech-improvement (3 games). Specifically, during the distribution of the evaluation form (January 2020) the data collection was ongoing, with the i-PROGNOSIS project having decided that the Intervention deployment would have the nature of a feasibility trial, rather than a long-term monitoring trial. Thus, this prohibited the evaluation of this system module based on quantified quality-of-life improvements. Similarly, the clinical effects of the games could not be properly evaluated and communicated to the experts. Instead, the experts were provided with a preliminary analysis of user evaluations (based on the data included in the collected CFIII forms in the OpenClinica platform4 ). Based on the provided user evaluations and the experts’ own experience with the intervention data system, the collected respondent opinions were fairly positive for the potential of the intervention data system to add value to existing clinical interventions in Specialised and Primary/Outpatient Care. 4 Neutral opinion www.openclinica.com 293 Innovative interventions for Parkinson's disease patients using iPrognosis Games: An evaluation analysis by medical experts PETRA’20, June, 2020, Corfu Island, Greece Table 5. The main challenges for the adoption of the iPROGNOSIS module as an intervention and patient monitoring system, as identified by 18 medical experts Identified i-PROGNOSIS challenges Fine tuning of the interfaces and the functionality is required. The innovative nature of the platform requires initial effort to convince patients to use the system. A long-term feasibility trial for the evaluation of the compliance of use is needed. Technical improvements are required. The patient directed feedback needs to be optimised. The system effects on PD symptomatology during long term use should be validated. More flexibility when patients select the initial game difficulty is required. Specialised equipment (partly) is needed for clinical deployment. The emotional state of patients might be a hinder for longterm use. More games might be required for additional symptoms and added variety might be needed. Users’ data privacy concerns. Integration with current Health Care Systems and Records will be challenging. iOS support is required for the mobile/tablet-based games. Figure 5: Percentage of strong expert agreement for the potential of the i-PROGNOSIS system characteristics. Table 4. The main advantages of the i-PROGNOSIS module, as identified by 17 medical experts Identified i-PROGNOSIS system advantages Attractive design. Accessible to patients despite age group. Facilitates self-monitoring of the disease progress. Has the potential for long-term disease monitoring It targets the full range of clinically significant PD symptomatology. It can be (partly) deployed in the home environment. It targets at improving quality of life for PD patients. It promotes self-awareness in PD patients regarding symptomatology progression. It has high potential for use in other relevant medical fields. The games have a desirable social element. The platform is innovative. The integrated progressive reward system improved compliance. It promotes exercise and increased physical activity for patients. Mentions % of Experts* 12 71% 11 65% 7 41% 6 35% 6 35% Mentions % of Experts* 44% 8 28% 5 22% 4 3 3 17% 17% 17% 3 11% 2 11% 2 11% 2 5 29% 5 29% 5 29% 3 18% 2 12% 2 12% 2 12% project, technical refinement and increased patient participation were achieved in four-month feasibility study. 2 12% 5 CONCLUSIONS 11% 2 1 6% 6% 1 1 6% *at least one mention per subject The i-PROGNOSIS system is well on its way to provide interventional tools, such as the iPrognosis Games and the iMAT, for complementing existing clinical practices for the management of PD symptomatology, always targeting the improvement of patients’ quality of life. However, beyond the end of the iPROGNOSIS project, key parameters should be taken in consideration, in order to ensure the wider adoption of the iPROGNOSIS by the medical community. *at least one mention per subject integration of the platform, in order to support longer-term use (see Table 5). Also, the respondents pointed the need for long term feasibility trials to evaluate the compliance of use and the effects of the games on the disease state of the patients. Towards such recommendations, upon the completion of the i-PROGNOSIS 294 PETRA’20, June, 2020, Corfu Island, Greece S. Dias et al. [11] Dias, S.B., Diniz, J.A., Konstantinidis, E., Savvidis, T., Bamidis, P., Jaeger, H., Stadtschnitzer, M., Klingelhoefer, L., Trivedi, D., Bostantzopoulou, S., Charisis, V., Hadjidimitriou, S., and Hadjileontiadis, J.L. (2018). On Exploring Design Elements in Assistive Serious Games for Parkinson's Disease Patients: The iPROGNOSIS Exergames Paradigm. In Proceedings of the 2nd International Conference on Technology and Innovation in Sports, Health and Wellbeing (TISHW 2018), IEEE Conference Location: El Paso, Texas USA. [12] Savvidis, T., Konstantinidis, E., Dias, S.B., Diniz, J.A., Hadjileontiadis, L.J., and Bamidis, P. (2018). Exergames for Parkinson’s Disease patients: How participatory design led to technology adaptation. Studies in Health Technology and Informatics, Data, Informatics and Technology: An Inspiration for Improved Healthcare (vol. 251, pp. 78-81) doi: 10.3233/978-1-61499-880-8-78. [13] Savvidis, T., Konstantinidis, E., Dias, S.B., Zilidou, V., Romanopoulou, E., Hadjileontiadis, L.J., and Bamidis, P., (2019). Co-creating Exergames with Parkinson’s Disease Patients. In Proceedings of the 8th Pan-Hellenic Conference on Biomedical Technology (ELEVIT 2019), 9-10 May, Athens, Greece. [14] Wyatt, J., & Spiegelhalter, D. (1991). Evaluating medical expert systems: what to test, and how? In Knowledge Based Systems in Medicine: Methods, Applications and Evaluation (pp. 274-290). Springer, Berlin, Heidelberg. [15] Ben-Assuli, O., Sagi, D., Leshno, M., Ironi, A., & Ziv, A. (2015). Improving diagnostic accuracy using EHR in emergency departments: A simulation-based study. Journal of Biomedical Informatics, 55, 31-40. [16] Gamberger, D., Krstačić, G., & Šmuc, T. (2000, September). Medical expert evaluation of machine learning results for a coronary heart disease database. In International Symposium on Medical Data Analysis (pp. 159-168). Springer, Berlin, Heidelberg. [17] Crossley, G. H., Boyle, A., Vitense, H., Chang, Y., Mead, R. H., & Connect Investigators. (2011). The CONNECT (Clinical Evaluation of Remote Notification to Reduce Time to Clinical Decision) trial: the value of wireless remote monitoring with automatic clinician alerts. Journal of the American College of Cardiology, 57(10), 1181-1189. [18] Smithburger, P. L., Buckley, M. S., Bejian, S., Burenheide, K., & Kane-Gill, S. L. (2011). A critical evaluation of clinical decision support for the detection of drug–drug interactions. Expert opinion on drug safety, 10(6), 871-882. [19] Morris, Z. S., Wooding, S., & Grant, J. (2011). The answer is 17 years, what is the question: understanding time lags in translational research. Journal of the Royal Society of Medicine, 104(12), 510-520. [20] Bürkle, T., Ammenwerth, E., Prokosch, H. U., & Dudeck, J. (2001). Evaluation of clinical information systems. What can be evaluated and what cannot? Journal of evaluation in clinical practice, 7(4), 373-385. [21] Fritz, F., Balhorn, S., Riek, M., Breil, B., & Dugas, M. (2012). Qualitative and quantitative evaluation of EHR-integrated mobile patient questionnaires regarding usability and cost-efficiency. International Journal of Medical Informatics, 81(5), 303-313. [22] Boon, H., MacPherson, H., Fleishman, S., Grimsgaard, S., Koithan, M., Norheim, A. J., & Walach, H. (2007). Evaluating complex healthcare systems: a critique of four approaches. Evidence-Based Complementary and Alternative Medicine, 4(3), 279-285. [23] Kilsdonk, E., Peute, L. W., & Jaspers, M. W. (2017). Factors influencing implementation success of guideline-based clinical decision support systems: A systematic review and gaps analysis. International journal of medical informatics, 98, 56-64. [24] Simon, A. (2020). Usability of electronic patient record systems: Instrument validation study conducted for hospitals in Germany. Health Informatics Journal, 1460458219895910. [25] Smith, A. B., Rush, R., Fallowfield, L. J., Velikova, G., & Sharpe, M. (2008). Rasch fit statistics and sample size considerations for polytomous data. BMC Medical Research Methodology, 8(1), 33. [26] Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77-101. [27] Bengtsson, M. (2016). How to plan and perform a qualitative study using content analysis. NursingPlus Open, 2, 8-14. In an effort to probe further, the evaluation of the level of the PD patient engagement with i-PROGNOSIS system will be taken into consideration, through the analysis of the monitored useriPrognosis Game interaction. ACKNOWLEDGMENTS The authors would like to thank all clinical partners of the iPROGNOSIS consortium (i.e., AUTH-MED, Greece; TUD, Germany; and KCL, United Kingdom), who participate in this study for their time and contribution. The research leading to these results has received funding from the European Union's Horizon 2020 Research and Innovation Programme under grant agreement No 690494—i-PROGNOSIS: Intelligent Parkinson early detection guiding novel supportive interventions. REFERENCES [1] Dorsey, E. R., Constantinescu, R., Thompson, J. P., Biglan, K. M., Holloway, R. G., Kieburtz, K., ... and Tanner, C.M. (2007). Projected number of people with Parkinson disease in the most populous nations, 2005 through 2030. Neurology 68(5), 384-386. [2] Tanner, C.M., Brandabur, M. and Dorsey, E R. (2008). Parkinson Disease: A Global View. Parkinson Report, 9-11. [3] Procházka, A., Vyšata, O., Vališ, M., Ťupa, O., Schätz, M., and Mařík, V. (2015). Bayesian classification and analysis of gait disorders using image and depth sensors of Microsoft Kinect. Digital Signal Processing 47, 169-177. [4] Dixit, S., & Tedla, J. S. (2019). Effectiveness of robotics in improving upper extremity functions among people with neurological dysfunction: a systematic review. International Journal of Neuroscience, 129(4), 369-383. [5] Knippenberg, E., Verbrugghe, J., Lamers, I., Palmaers, S., Timmermans, A., and Spooren, A. (2017). Markerless motion capture systems as training device in neurological rehabilitation: A systematic review of their use, application, target population and efficacy. J. Neuroeng. Rehabil 14, 61. [6] Grammatikopoulou, A., Dimitropoulos, K., Bostantjopoulou, S., Katsarou, Z., and Grammalidis, N. (2019). Motion analysis of Parkinson diseased patients using a video game approach. In Proceedings of the 12th ACM International Conference on Pervasive Technologies Related to Assistive Environments, pp. 523-527. [7] Dias et al. (2020). Innovative Parkinson´s Disease Patients´ Motor Skills Assessment: The i-PROGNOSIS Paradigm, Frontiers in Computer Science, section Human-Media Interaction (under review). [8] Dias, S.B., Diniz, J.A., Hadjidimitriou, S., Charisis, V., Konstantinidis, E., Bamidis, P.D., and Hadjileontiadis, L.J. (2016). Personalized Game Suite: A unified platform to sustain and improve the Quality of Life of Parkinson’s Disease patients. Frontiers in Human Neuroscience. doi: 10.3389/conf.fnhum.2016.220.00023. [9] Dias, S.B., Diniz, J.A., Konstantinidis, E., Bamidis, P., Charisis, V., Hadjidimitriou, S., Stadtschnitzer, M., Fagerberg, P., Ioakeimidis, I., Dimitropoulos, K., Grammalidis, N., and Hadjileontiadis, L.J. (2017). On supporting Parkinson's Disease patients: The i-PROGNOSIS Personalized Game Suite design approach. In Proceedings of the IEEE International Symposium on Computer-Based Medical Systems - CBMS 2017, June 22-24, 2017, Thessaloniki, Greece. [10] Dias, S.B., Konstantinidis, E., Diniz, J.A., Bamidis, P., Charisis, V., Hadjidimitriou, S., Stadtschnitzer, M., Fagerberg, P., Ioakeimidis, I., Dimitropoulos, K., Grammalidis, N., and Hadjileontiadis, L.J. (2017). Serious Games as a means for holistically supporting Parkinson’s Disease patients: The i-PROGNOSIS Personalized Game Suite framework. In Proceedings of the 9th International Conference on Virtual Worlds and Games for Serious Applications (VS-Games 2017), Sept 6-8, 2017, Athens, Greece. 295