NextGen: next generation tools for genome-centric multimodal data integration in personalized cardiovascular medicine.

Funded by Horizon Europe (2024-2027)

Personalised medicine, comprising tailored approaches for prevention, diagnosis, monitoring and treatment is essential to reduce the burden of disease and improve the quality of life. Integration of multiple data types (multimodal data) into artificial intelligence models is required for the development of accurate and personalized interventions.

This is particularly true for the inclusion of genomic data, which is information-rich and individual-specific, and more routinely available as the cost of sequencing continues to fall. Multimodal data integration is complex due to privacy & governance requirements, the presence of multiple standards, distinct data formats, and underlying data complexity and volume. NextGen tools will remove barriers in data integration for several cardiovascular use cases. NextGen deliverables will include tooling for multimodal data integration and research portability, extension of secure federated analytics to genomic computation, more effective federated learning over distributed infrastructures, more effective and accessible tools for genomic data analysis; improved clinical efficiency of variant prioritization; scalable genomic data curation; and improved data discoverability and data management. Several real-world pilots will demonstrate the effectiveness of NextGen tools and will be integrated in the NextGen Pathfinder network of five collaborating clinical sites as a self-contained data ecosystem and comprehensive proof of concept.

In this context, IDSIA team will contribute to the following objectives:

• Develop tools for the prediction, prevention, diagnosis, monitoring, and treatment of cardiovascular disease using multiomic, multimodal data.
• Improve the effective use of genomic data through advanced integration and workflow tooling. NextGen will develop tools to overcome barriers in health data integration, including algorithms for federated learning, and enhance the effective use and incorporation of genomic data.