Nordic Biobank x NVIDIA Federated Learning Hackathon for Biomedical Applications

Organised jointly by NSHG-PM, NVIDIA, and Rigshospitalet, Copenhagen University Hospital in conjunction with the NCFH 2026 conference, the hackathon showcased the potential of increased Nordic collaboration in precision medicine, health data, and AI.

The hackathon brought together 75+ selected researchers, clinicians, data scientists, bioinformaticians, and AI specialists to work onsite and online in interdisciplinary teams, exploring federated learning approaches for biomedical research.

The interdisciplinary teams collaborated to develop secure cross-biobank analytical workflows, leveraging NVIDIA FLARE, explore strategies for data harmonization across Nordic biobanks and apply federated, multimodal analyses to health-related outcomes. All projects followed open science principles, with outcomes being shared publicly (e.g., via GitHub and potential publications).
We wish to warmly thank Ben Busby for his expertise, engagement, and enthusiasm in planning and running the hackathon and Malene Fischer for offering Rigshospitalet, Copenhagen University Hospital as host and venue of the hackathon and its participants. We also extend our sincere thanks to the Danish Centre for AI Innovation (DCAI) for providing the hackathon with access to working on the Gefion AI supercomputer as well as to Accenture for supporting the hackathon financially.

For NSHG-PM this has been an exciting experiment and an inspiring demonstration of what is possible through closer Nordic collaboration in health data, artificial intelligence, and precision medicine. We look forward to building on the momentum created during these three days and continue exploring similar initiatives in the future.

One of the teams gathered for review and planning of their coding.

Final presentation by the working groups on Rigshospitalet – Copenhagen University Hospital at the end of hackathon. Work is now ongoing to write one or more publications.