Learning Hub

< BACK Federated Learning Aggregation based on Weight Distribution Analysis Oct 4, 2023 Enhancing Federated Learning: Leveraging Weight Distribution Analysis for Superior Aggregation Federated learning has recently been proposed as a solution to the problem of using private or sensitive data for training a central deep model, without exchanging the local data. In federated learning, local models are trained on the client side using the available data, while a server is responsible for aggregating the weights of these models into a global model. However, the traditional weight averaging approach does not take into consideration the importance of the different weights for the performance of a model. To this end, this work proposes a novel federated learning weight aggregation method that estimates the statistical distance of each client's parameters from the Gaussianity, and weighs the contribution of each client to the global model accordingly so that the most significant information is retained and enhanced. To create an accurate global model, a complex weighted averaging of the parameters of clients' models at the layer level is performed, considering as low quality the parameters following the Gaussian distribution. The proposed method can be employed to both convolutional and linear layers and it is based on the notion that parameters following a Gaussian distribution do not significantly affect the output of a model. Experiments with different network architectures and a comparison with a plethora of state-of-the-art approaches on three well-known image classification datasets demonstrate the superiority of the proposed method for federated learning weight aggregation. Read the full publication: https://www.researchgate.net/publication/374440973_Federated_Learning_Aggregation_based_on_Weight_Distribution_Analysis 1/1 PREVIOUS NEXT

< BACK Ongoing preparation for the iPROLEPSIS-PDPID study Mar 4, 2024 iPROLEPSIS-PDPID Study: Advancing Psoriatic Arthritis through Innovative Digital Biomarkers Patients with Psoriatic arthritis (PsA) experience difficulties in dealing with unpredictable disease activity, which can affect their quality of life. Developing a digital biomarker using a smartphone and smartwatch would allow for unobtrusive monitoring of the disease activity in these patients. Besides, understanding what factors trigger flare would allow for better disease control. The iPROLEPSIS-PDPID study is a multicentre observational cohort study that aims to develop an unobtrusive and affordable digital biomarker capable of detecting changes in disease activity, including flare, and to identify triggers of flare in patients with PsA. The study will be conducted in four countries: the Netherlands, the UK, Portugal and Greece. Ethical approval for initiating the iPROLEPSIS-PDPID study in the Netherlands and the UK has been granted. The OpenClinica platform will be used in the collection and storage of data for the iPROLEPSIS-PDPID study. Future steps include obtaining ethical approvals for Portugal and Greece as well as local approvals from centres participating in each country, and inclusion of patients from each participating country. The miPROLEPSIS phone app , currently under development, will be used in this study. The app will be installed on the patient smartphone and utilised as a data collector, which will be then used for algorithms development and training. The app will measure physical activity, the length of sleep, self-registered flare, and register daily pain, fatigue, sleep and morning stiffness. The Garmin smartwatch Vivoactive 5 device will be used to collect data on physical activity, heart rate, heart rate variability and sleep quality. The collected data from Garmin will be downloaded from the miPROLEPSIS app by Wellics software (developer in the project), which allows extracting, visualising and exporting collected both company-processed data, as well as raw data. 10.png 13.png 14.png 10.png 1/5 PREVIOUS NEXT

< BACK iPROLEPSIS: updates on user research and co-creation Oct 17, 2023 Co-creation sessions for advancing the development of the mIPROLEPSIS app iPROLEPSIS active engagement with users through ongoing co-creation sessions helps to advance the development of the mIPROLEPSIS app, scheduled for delivery before January 2024. In October 2023, the project partner Erasmus MC (EMC) conducted a co-creation session with patient partners to assess the User Interface (UI) of the PDPID-app. The focus was on improving usability, ensuring users’ understanding of expectations and enhancing users’ engagement to minimise missing data. The project partner FACULDADE DE MOTRICIDADE HUMANA (FMH-ULISBOA), is also developing storyboards for the personalised gaming suite . Three co-creation sessions were conducted: one with healthcare professionals and two with patient partners. Project partners The Centre for Research and Technology Hellas (CERTH) and FACULDADE DE MOTRICIDADE HUMANA (FMH-ULISBOA), in collaboration with the clinical partners, worked on identifying movements for the video analysis that will be included in the iPROLEPSIS-PDPID study. Multiple hand gestures and fullbody movements were identified for further assessment. Main points: mIPROLEPSIS App’s general looks and UI were approved by the patient partners; Despite being a data-collection app for research purposes, patients emphasise the significance of reviewing their own data entries; Participants noted a need for enhanced visual cues regarding specific activities in the app. Main points: Personalised gaming suite Smart games meant for disease management need to be highly tailored to the capabilities of the patients; Patients view gaming as a form of relaxation, emphasising the importance of stress-free design; Patient partners expressed interest in the proposed gaming categories. 1/0 PREVIOUS NEXT

< BACK Innovative Digital Solutions for Psoriatic Arthritis: A Spotlight on the iPROLEPSIS Project by Prof. Leontios Hadjileontiadis Feb 14, 2024 Empowering Patients, Pioneering Solutions, and Bridging Gaps in Psoriatic Arthritis Knowledge Prof. Leontios Hadjileontiadis, the coordinator of the iPROLEPSIS project, shares how the iPROLEPSIS project addresses psoriatic arthritis research . iPROLEPSIS aims to identify triggers behind the transition from psoriasis to psoriatic arthritis, employing digital models and everyday devices like smartphones and smartwatches to understand disease dynamics. Beyond research, iPROLEPSIS introduces practical solutions such as biAURA for pain relief through sound and serious games targeting exercise, emotion, nutrition, and pain management. Collaborating with 15 partners, including patient advocates, iPROLEPSIS adopts a co-creation approach , bridging technological innovation with real-world healthcare practices. Focusing on predictive and preventive healthcare, iPROLEPSIS strives to fill the knowledge gap surrounding psoriatic arthritis, transforming the research landscape. Watch the full interview: leontios_edited.jpg leontios_edited.jpg 1/1 PREVIOUS NEXT

< BACK World Psoriasis Day: iPROLEPSIS featured by HaDEA Oct 29, 2025 On World Psoriasis Day, the European Health and Digital Executive Agency (HaDEA) highlights how the iPROLEPSIS project is advancing early detection and personalised care for psoriatic arthritis Psoriatic arthritis (PsA) is a chronic inflammatory condition that can cause irreversible joint damage if left undiagnosed or untreated. To address this challenge, iPROLEPSIS is developing AI-driven predictive models and digital tools that help identify individuals at risk, enable earlier diagnosis, and support more personalised treatment pathways. In a recent interview with Prof. Leontios Hadjileontiadis , project coordinator, HaDEA explores how iPROLEPSIS brings together clinicians, researchers, and data scientists to transform psoriatic arthritis management across Europe. The discussion covers the project’s use of AI and biological data to build comprehensive patient profiles, its collaborative approach between dermatology and rheumatology, and the importance of EU funding and ethical, cross-border cooperation in enabling innovation. Read the full article on HaDEA’s website: https://hadea.ec.europa.eu/news/world-psoriasis-day-exploration-how-horizon-europe-project-improving-early-detection-psoriatic-2025-10-29_en World Psoriasis Day 2025.png World Psoriasis Day 2025.png 1/1 PREVIOUS NEXT

< BACK iPROLEPSIS digital ecosystem's architecture Oct 5, 2023 iPROLEPSIS ecosystem's architecture: a glimpse into technical progress One of the main objectives of the iPROLEPSIS project is to provide a digital health ecosystem to support healthcare professionals in disease screening, monitoring, and treatment while empowering individuals with/at risk of Psoriatic Arthritis with personalised insights and preventive interventions. In September 2023, the project’s technical partners introduced the first version of the iPROLEPSIS ecosystem architecture , including the initial technical specifications and product backlog. __________________ Amalia Ntemou from Netcompany-Intrasoft , consortium partner leading technical specification, system architecture and product backlog task within iPROLEPSIS, shares insights on the latest updates on the iPROLEPSIS ecosystem architecture. The first version of the iPROLEPSIS ecosystem's architecture and technical specifications has been defined. Can you explain the importance of this achievement in the project's development process? One of the main objectives of the iPROLEPSIS project is to provide a digital health ecosystem to support healthcare professionals in disease screening, monitoring, and treatment, while empowering individuals with/at risk of PsA with personalised insights and preventive interventions. The design of the architecture and the definition of functional and non-functional requirements are essential for the successful implementation of the iPROLEPSIS ecosystem as it constitutes a clear and comprehensive guide for partners’ development efforts. Various systems’ elements, components, connectors, processes and interactions are illustrated thereby enabling common understanding among project’s partners and adherence to common approaches and specifications that promote effective collaboration and delivery. A well-defined architecture with clear functional and non-functional requirements can be the basis for delivering sustainable solutions. What are the main components of the iPROLEPSIS digital ecosystem to be developed? The initial conceptual architecture of the iPROLEPSIS ecosystem consists of the miPROLEPSIS patient mobile app (including the embedded subcomponent/feature biAURA), the iPROLEPSIS healthcare professionals (HCP) web app , the cloud-based Data Management System , the AI-driven Lifestyle Recommendation Engine , the xAI-driven Models Engine and the Serious Games Suite . Finally, the operation of the entire system is supported by the Orchestration and Monitoring component . For whom is the iPROLEPSIS ecosystem specifically intended, and how does it cater to their unique needs? Patients are the primary users of the miPROLEPSIS patient app which is the tool for personalised disease monitoring. The miPROLEPSIS patient app operates as a tool for capturing data from sensors and wearable devices, as well as communicating feedback from the recommendation engine to the user. The miPROLEPSIS patient app incorporates additional applications/features (e.g., biAURA, keyboard etc.) for further monitoring of patients’ conditions, interventions and engagement hooks. Via the miPROLEPSIS HCP app, doctors and health care professionals are able to track their patients’ health status, review projections of the PsA progression, and utilise recommendations of optimised personalised suggestions towards a better management of PsA. Aggregated patients’ data about their health status, and outcomes of the xAI Models Engine are displayed intuitively and informatively to HCPs through dashboards. Doctors can monitor patients’ adherence to medical, dietary, and physical activity plans and issue warnings and notifications or adjust treatment regimens. In this way, accurate interventions are supported, while efficient and optimized management of PsA patients’ needs can be achieved and communicated to fellow colleagues and related stakeholders. What was the main focus while designing the iPROLEPSIS ecosystem architecture? The iPROLEPSIS digital health ecosystem serves as the enabler for personalised preventive care and the empowerment of citizens, patients and Healthcare Professionals (HCP). This emphasis on personalisation and empowerment guided the collection and definition of functional and non-functional requirements, as well as the internal architecture of components and technical specifications. Ultimately, these efforts led to the design of the ecosystem’s architecture. Have any methodologies been used in designing the iPROLEPSIS ecosystem architecture, and why is it important to use them? To organise the design and architecture of the iPROLEPSIS ecosystem, several established frameworks/methodologies were evaluated based on a thorough literature review in order to conclude to a framework that provides a holistic and comprehensive view of a software system’s architecture. Finally, we used a widely known methodology developed in 1995 by Kruchten , due to its ability to offer a structured approach to the system’s design, enabling multidisciplinary teams to represent different aspects of the system effectively and in detail. Utilising already established methodologies and following best practices, enhances efficiency, reduces risks, leads in better design decisions and contributes to overall success of the project. As the project progresses, how will the architecture and product backlog be refined and adapted during the development phases of the iPROLEPSIS? Development and establishment of the iPROLEPSIS digital health ecosystem is an ongoing process that runs approximately until the end of the project. Technical testing and verification of new versions of ecosystem’s components, as well as the monitoring of their performance and stability once released to users, have also been considered. The outcomes of development and monitoring processes will continuously populate the product backlog, representing the changing needs and priorities as the development of the components progresses. All updates and enhancements will lead to a refined version of the architecture and a fully operational iPROLEPSIS ecosystem. Stay tuned as we embark on this transformative journey towards a brighter future in Psoriatic Arthritis management! News & Events iPROLEPSIS - news.png News & Events iPROLEPSIS - news.png 1/1 PREVIOUS NEXT

< BACK Call for Workshop Papers: AI-enabled Digital Health Tools for Non-Communicable Diseases: From Concepts to Impact May 16, 2025 A joint workshop by iPROLEPSIS and AI-PROGNOSIS at IEEE HealthCom 2025 in Abu Dhabi, focusing on AI-enabled Digital Health Tools for Non-Communicable Diseases The iPROLEPSIS and AI-PROGNOSIS projects are co-organising a workshop titled: “AI-enabled Digital Health Tools for Non-Communicable Diseases: From Concepts to Impact” as part of the IEEE HealthCom 2025 conference , taking place on 21–23 October 2025 in Abu Dhabi, United Arab Emirates. More information: https://healthcom2025.ieee-healthcom.org/program/ai-enabled-digital-health-tools-non-communicable-diseases-concepts-impact We invite authors to submit original research, position papers, or case studies with practical insights. Topics of interest include (but are not limited to): Predictive models for early detection, risk assessment, and prognosis of NCDs Digital biomarkers and wearable sensors for NCDs Machine learning for personalized treatment, disease management, and precision medicine AI in remote care, mHealth, and telemedicine for NCDs Regulatory and ethical considerations in AI-enabled software as medical device Data quality, fairness, and explainability in AI health applications User-centered design in AI-enabled health technology for chronic conditions Translational pathways from research prototypes to scalable digital health solutions The workshop aims to feature 10 selected papers. Important dates: Workshop paper submission July 1, 2025 Acceptance notifications August 15, 2025 Camera Ready papers August 30, 2025 For any questions, please contact: Dr. Vasileios Charisis , Aristotle University of Thessaloniki, Greece – vaschar@auth.gr Dr. Stelios Hadjidimitriou , Aristotle University of Thessaloniki, Greece – shadjidim@ece.auth.gr HealthCom 2025_Call for WS papers_Digital Health Tools for NCDs .pdf Download PDF • 181KB iPROLEPSIS_Call for workshop papers.png iPROLEPSIS_Call for workshop papers.png 1/1 PREVIOUS NEXT

< BACK iPROLEPSIS presented at MEDICA 2023 Nov 14, 2023 iPROLEPSIS project presented at the world’s largest event for the medical sector Project partners Amalia Ntemou ( Netcompany-Intrasoft ) and Georgios Apostolidis ( Aristotle University of Thessaloniki (AUTH) ) presented the iPROLEPSIS project at the MEDICA - Leading International Trade Fair in Germany. MEDICA is the world’s largest event for the medical sector. It attracts several thousand exhibitors from more than 50 countries in the halls. Furthermore, each year, leading individuals from business, research, and politics grace this top-class event with their presence. MEDICA 2023 took place on 13-16 November in Düsseldorf, Germany. IMG_20231115_135346.jpg IMG_20231115_135346.jpg 1/1 PREVIOUS NEXT

< BACK iPROLEPSIS presented at PhD DEI 2023 Nov 13, 2023 Highlights from PhD DEI 2023 Organised by the Universidade de Lisboa (the Instituto Superior Técnico), the PhD DEI 2023 event on 10 November 2023 served as a platform for introducing new PhD students to the Department of Engineering and Informatics (DEI). The event fostered an environment of knowledge-sharing and collaboration among students from the Doctoral Program in Computer Science and Engineering and the Doctoral Program in Digital Media. One of the standout presentations at PhD DEI 2023 was the poster presentation on the iPROLEPSIS Exergames. This presentation, delivered by Bárbara Ramalho, a PhD student from the FMH-ULisboa team, highlighted the general concepts and objectives of the iPROLEPSIS project. The iPROLEPSIS Exergames were developed through a co-design process, emphasizing collaborative and innovative approaches to game design aimed at promoting physical activity and health. Read more: We now know the audience's favorite posters from PhD DEI 2023! - Departamento de Engenharia Informática (DEI) ( ulisboa.pt ) 1/1 PREVIOUS NEXT

< BACK Initiation of the iPROLEPSIS-PDPID Study Oct 22, 2024 Initiating Patient Enrollment for Psoriatic Arthritis Research The PsA digital phenotyping and inflammation drivers study (iPROLEPSIS-PDPID) has officially been initiated. The first patient was enrolled in the Netherlands in early September. The iPROLEPSIS-PDPID study focuses on using smart devices and clinical data to develop AI-driven models for personalised monitoring and flare prediction. This approach aims to empower patients by providing tools to better understand their condition. The study seeks to develop unobtrusive and affordable digital biomarkers capable of detecting changes in disease activity, including flares, and identifying their triggers in patients with psoriatic arthritis (PsA). The study is being conducted in four countries: the Netherlands, the UK, Portugal, and Greece. Patient recruitment is planned to start in the UK by the end of October, with recruitment in Greece and Portugal expected later this year. The miPROLEPSIS phone app is used in this study. The app is installed on patients' smartphones and acts as a data collection tool, which will be used to develop and train algorithms. The app measures physical activity, sleep duration, self-registered flares, and daily records of pain, fatigue, sleep quality, and morning stiffness. Additionally, the Garmin Vivoactive 5 smartwatch is used to collect data on physical activity, heart rate, heart rate variability, and sleep quality. 1/1 PREVIOUS NEXT

Learning Hub Explore resources to help you understand and manage psoriatic arthritis. Learning hub Key Facts Handbook News Feed Quizzes Search How will psoriatic arthritis affect me? Work See related Handbook section PREVIOUS NEXT

< BACK iPROLEPSIS presented to Eli Lilly in the Netherlands Nov 8, 2023 Presentation to Eli Lilly introduces iPROLEPSIS as the next step in personalised psoriatic arthritis care On the 7th of November 2023, Ilja Tchetverikov from CICERO Rheumatology presented the results and prospects of the DEPAR cohort and introduced iPROLEPSIS clinical studies to the representatives of Eli Lilly company in the Netherlands. DEPAR, Dutch south-west Psoriatic Arthritis Registry, is a registry set up by Cicero Rheumatology, a partnership of rheumatologists working in 15 different hospitals in the south-west of the Netherlands. DEPAR, with the participation of well over 900 patients, supports rheumatologists to gain insight into the treatment and course of psoriatic arthritis. The presentation highlighted iPROLEPSIS as a logical next step in the further development of DEPAR, emphasising the project’s vision and aim to explain psoriatic arthritis inflammation through multi-source data analysis, guiding to a novel personalised digital care ecosystem. Slide1.JPG Slide2.JPG Slide6.JPG Slide1.JPG 1/6 PREVIOUS NEXT