[DSC MENA 24] Amal_Elgammal_-_QUALITOP_presentation.pptx

Background: The digital twin paradigm holds great promise for healthcare, most importantly efficiently integrating many disparate healthcare data sources and servicing complex tasks like personalizing care, predicting health outcomes, and planning patient care, even though many technical and scientific challenges remain to be overcome. Objective: As part of the QUALITOP project, we conducted a comprehensive analysis of diverse healthcare data, encompassing both prospective and retrospective datasets, along with an in-depth examination of the advanced analytical needs of medical institutions across five European Union countries. Through these endeavors, we have systematically developed and refined a formal Personal Medical Digital Twin (PMDT) model subjected to iterative validation by medical institutions to ensure its applicability, efficacy, and utility. Findings: The PMDT is based on an interconnected set of expressive knowledge structures that are calibrated to capture an individual patient’s psychosomatic, cognitive, biometrical and genetic information in one personal digital footprint in a manner that allows medical professionals to run various models to predict an individual’s health issues over time and intervene early with personalized preventive care.Conclusion: At the forefront of digital transformation, the PMDT emerges as a pivotal entity, positioned at the convergence of Big Data and Artificial Intelligence. This paper introduces a PMDT environment that lays the foundation for the application of comprehensive big data analytics, continuous monitoring, cognitive simulations, and AI techniques. By integrating stakeholders across the care continuum, including patients, this system enables the derivation of insights and facilitates informed decision-making for personalized preventive care.

1. Medical Digital Twins for Personalized Chronic Care Assoc. Prof. Amal Elgammal Head of Software Dept., Egypt University of Informatics, Egypt Adjunct Senior Researcher, Scientific Academy for Service Technology ( ServTech ), Germany

2. MONITORING MULTIDIMENSIONAL ASPECTS OF QUALITY OF LIFE AFTER CANCER IMMUNOTHERAPY: AN OPEN SMART DIGITAL PLATFORM FOR PERSONALIZED PREVENTION AND PATIENT MANAGEMENT WP4 Big Health Data Smart Digital Platform and Shared Data Lake

3. Agenda QUALITOP Project Motivation Problem Definition General Objective QUALITOP consortium WP4: Big Health Data Digital Platform and shared Datalake

4. QUALITOP: Motivation Cancer immunotherapy research revolutionized medicine. Qualified as the fifth pillar of cancer therapy after surgery , radiation , chemotherapy , and precision medicine , cancer immunotherapy brought about significant progress in cancer treatment Compared to surgery, radiotherapy or chemotherapy, immunotherapy is considered one of the most complicated treatment options because it activates the “ body’s natural anti-cancer immune response to attack and destroy cancer By enabling the immune system, immunotherapy causes toxicities or side effects that are challenging to predict because they are not caused by mechanisms involved in the other cancer treatment types. These are referred to as immune-related adverse events (IR-AEs).

5. QUALITOP: Problem Definition There are currently three main challenges that impede improving cancer patients’ QoL after immunotherapy: First, there is a crucial need to determine “ predictive markers ” or “ patients' sub-populations ” associated with the development of IR-AEs that help prevent , predict , and manage IR-AEs and improve patient’s health status. A second barrier is the lack of knowledge regarding patients in the “ real-life " after the start of immunotherapy (life-style, polymedication to treat symptoms associated with IR-AEs). To overcome the two previous challenges about the determinants of the “medical condition” (Health status) and QoL , significantly more diversified sources of data and higher numbers of patients should be included in research studies The recourse to a vast patients’ data heterogeneity (e.g., different cancer types, patients’ characteristics, and various sources of medical and patients life data) is a great opportunity , but it would trigger major data complexities (e.g., high volume and velocity of supply) and make it tremendously difficult to collect , aggregate , and analyse these data, especially within the context of the General Data Protection Regulation (GDPR).

6. QUALITOP: GENERAL OBJECTIVE “ Develop and implement an IT-based European immunotherapy platform and use big data analysis , artificial intelligence , and simulation modelling approaches to collect and aggregate efficiently and effectively real-world QoL data , monitor patients' health status, conduct causal inference analyses, create harm-reduction recommendations for patients and other stakeholders, and disseminate the findings”

7. QUALITOP Consortium

8. Implementation of QUALITOP through WPs 8 This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 875171

9. implementation Architecture of Big Health Data Smart Digital Platform 9

10. Phase#1: Normalized Patient-Centric Data Capture 10 This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 875171 * Map of existing databases and cohorts [D2.1]

11. France ( at Hospices Civils de Lyon - HCL) : With more than 500 patients (adults only) treated by immunotherapy each year at HCL, more than 1000 patients are expected to be included in the prospective data collection for QUALITOP in France. Additionally, about 200 lymphoma’s patients are expected to be enrolled in the QUALITOP study in France. Spain (Hospital Clinic de Barcelona - IDIBAPS) : Up to 100 patients treated with immunotherapy/CAR-T are expected to be included in the cohort of QUALITOP. Portugal (Portuguese Institute of Oncology Lisbon- IPOL) : Up to 30 Patients will be involved in the cohort of QUALITOP, at the beginning of treatment or within the first 6 months. Netherlands (Amsterdam UMC - AMC) : Up to 100 patients treated with CPIs are expected to be included in the cohort of QUALITOP Netherlands (University Medical Center Groningen- UMCG) : 150 patients are expected to be enrolled. Patients Recruitment

12. Data Interoperability: Standardization Data structure/technical interchange standard Data value Standards Schemas, data element sets expressed in machine-readable form Data content Standards Medical Digital Twins Controlled vocabularies, thesauri, terms coding schemes Cataloging rules and codes. Guidelines for the format and syntax of the data values that are used to populate data elements

13. https://www.altexsoft.com/blog/data-standards-healthcare/#:~:text=Data%20standards%20are%20created%20to,of%20domain%20or%20software%20provider.

14. Medical Digital Twins 14 This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement N° 875171 CREATE A SMART HEALTH COMMUNITY/PLATFORM: an entity that operates largely outside of the traditional health care system centred around the after cancer ImmunoTherapy patient and encourages disease prevention, improved QoL and overall well-being in a virtual community setting in which stakeholders respond more and more to mutual, shared challenges. MEDICAL DIGITAL TWIN = Capturing and Tracking Patient Data construct a digital model of the patient ( personal digital replica ) that can be used as a virtual test-bed for improved QoL & future treatment ; enable doctors and other healthcare providers to capture & track patient data in order to tailor treatment to each patient; incorporate a variety of care data, including vital medical information from medical records, current medication, imaging studies, lifestyle, genetic, & patient-provided health data from exercise or health monitoring applications & medical pathways; improve post operative planning , reduce medical risks , and generate more accurate therapy & improved QoL for patients.

15. Medical Digital Twins Models

16. Detailed conceptual ontology model of the PMDT

17. Data Homogenization & Federated Query Processing Tool Security & Privacy-Preserving Tool Hybrid Federated/Data-Mesh Management Approach Hospital-1 Database Exported schema information & meta-data Hospital-2 Database Exported schema information & meta-data Hospital-3 Database Exported schema information & meta-data Local (edge) data sources Results Aggregation Query Translation Queries to local (edge) sources Partial answers from local (edge) sources Storage of DB structure and relationships meta-data, unifying knowledge model entities & correspondence to databases. Database Meta-data & Knowledge Model Repository Query submission Response/ answer Tools & users Access control, Anonymisation Local Data Model Data Product APIs Access control, Anonymisation Local Data Model Data Product APIs Access control, Anonymisation Local Data Model Data Product APIs Retrospective Data Routing Logic

18. PMDT designed and Implemented as an Ontology An ontology is a formal explicit description of concepts in a domain of discourse (classes), properties of each concept describing various features and attributes of the concepts (roles or properties), and restrictions on slots (facets or role restriction) An ontology together with a set of individual instances of classes forms a knowledge base The PMDTs conceptual models of the blueprints models are designed/modelled and implemented as an ontology using the Ontology Web Language (OWL) standard as part of W3C semantic web stack

19. W3C Sematic Web Stack

20. PMDT designed and Implemented as an Ontology An ontology is a formal explicit description of concepts in a domain of discourse (classes), properties of each concept describing various features and attributes of the concepts (roles or properties), and restrictions on slots (facets or role restriction) An ontology together with a set of individual instances of classes forms a knowledge base The PMDTs conceptual models of the blueprints models are designed/modelled and implemented as an ontology using the Ontology Web Language (OWL) standard as part of W3C semantic web stack

21. Adding and revising information contained in blueprints (e.g., adding stamped periodic PGHD) A Domain Specific Language is a programming language with a higher level of abstraction optimized for a specific class of problems. A DSL uses the concepts and rules from the field or domain A DSL Is usually less complex than a general-purpose language that must be developed in close coordination with the expert in that field Viewing and querying blueprints (by linking and cross-correlating information from diverse PMDT blueprints) Extending blueprint definitions by composing multiple blueprints of the same type, typically organized into large, composite blueprint