Federated Learning in Healthcare: from roadmap to real-world impact
We recently came across a valuable new publication that strongly resonates with the mission of the BETTER Project:
“A roadmap for federated learning projects using health data to guide sustainable artificial intelligence development in the European Union”, published in the International Journal of Medical Informatics by Janne Kommusaar together with Silja Elunurm, Taridzo Chomutare, Mari Kangasniemi, Sanna Salanterä and Laura-Maria Peltonen
The paper proposes an end-to-end roadmap for federated learning in healthcare, integrating technical, ethical, legal, and administrative dimensions — a perspective that is increasingly essential as AI systems move from pilots to scalable, cross-institutional deployments.
Why this matters
Healthcare AI cannot scale without trust. Federated learning offers a concrete answer to the privacy-vs-innovation dilemma, enabling collaborative model training while keeping sensitive health data local and compliant with EU regulations.
How this connects to Better AI Health
Within BETTER, coordinated by Datrix | Embrace the AI Challenge , we are working precisely at this intersection:
privacy-preserving AI for health data
federated and distributed learning architectures
governance-by-design aligned with EU values
sustainable, trustworthy AI beyond the lab
This publication reinforces a key message we strongly believe in: federated learning is not only a technical choice — it is an organizational, ethical, and policy-driven one.
Congratulations to the authors for contributing a much-needed, practical reference for the European healthcare AI ecosystem.