A federated paradigm of real-world data sources for diagnosis, prognosis and risk assessment of cardiovascular conditions

It aims to implement a privacy-by-design European-wide federated platform-as-a-service (PaaS) for the delivery of effective data-driven interventions and the advancement of research and management in the cardiovascular domain. The high level objectives are:
-Analysis of gaps and pathways in management of cardiovascular diseases and current data landscape challenges. 4 use cases: The risk of long-term sudden cardiac death, The early recognition of heart failure with preserved ejection fraction (HFpEF) and identification of specific phenotypes, Non-invasive detection of dyselectrolytemia for risk stratification of cardiac arrhythmias, The risk for post-operative atrial fibrillation.
-Integration of real-world data sources in a FAIR, open and federated repository 
-Implementation of data and AI-driven federated tools for enhancing the management of cardiovascular conditions, setting a paradigm on the exploration of available data sources. 
-Performance of validation studies and proof-of-concept demonstrations. 
-Pave the way for the sustainable and wide adoption of data and AI-driven federated strategies and in the management of cardiovascular and other chronic conditions
-Active engagement of stakeholders and wide steering of CVDLINK results towards their future upscale

Vicomtech aims to lead the way in the development of AI model for medical images and participate in the development of AI models from signal and omics data with a special focus on  robustness, trustworthiness and generalizability of our models. We will translate models to pipelines and user services and develop federate learning and inference mechanisms.

Partners:  Romania (SIMAVI-Coordinator, UMFCD)
Spain (VICOM)
Finland (TAU, TAUH, VER)
Greece (AUTH, AMC, CERTH, WCS)
Ireland (TRI)
Deutschland (TUD, UHAM)
Israel (HYMC, KIHDL)
Moldova (IMSP)
Lithuania (KTU)
Estonia (UTARTU)
Austria (SYNYO)