Foundation models in healthcare are large AI systems trained on vast amounts of medical and biological data. They learn general patterns about health and disease and can then be adapted to a wide range of healthcare, research, and policy applications.
With ageing populations, the global demand for governed, real-time solutions that can boost the efficiency of healthcare is growing rapidly. Although many international initiatives are emerging, they often rely on fragmented or biased clinical datasets that are neither representative nor fully compliant with regulatory frameworks.
Now
The FINe-Health Foundry aims to build the world’s first nationwide healthcare foundation model, capable of supporting doctors in making clinical decisions in real time and generating population-level scenarios based on different interventions. For example, the AI models could predict how different treatments, policies, or health trends may affect future outcomes.
The project serves as the spearhead initiative of the first strategic focus area in health at the newly established ELLIS Institute Finland and brings together researchers from FIMM and partner organisations. This joint effort builds on the research team's world-class expertise in AI and biomedicine.
“The new foundation model will be able to predict disease risk for over 200 conditions for every living person in Finland,” says Associate Professor Andrea Ganna, the project’s lead principal investigator at the University of Helsinki.
The FINe-Health Foundry has received nearly five million euros in Business Finland funding for an initial three-year period. The Rise to Challenge programme supports radical and impactful research ideas that create the expertise and capabilities needed for future growth sectors.
“Our research in the Foundry will move machine learning from using correlations to cause-and-effect reasoning, while integrating expert knowledge from healthcare practitioners. The resulting foundation model will accelerate and improve healthcare decision-making,” says Professor Samuel Kaski, director of ELLIS Institute Finland.
The strategic ELLIS Institute Finland–FIMM/HiLIFE partnership spans Aalto University, the Department of Computer Science at the University of Helsinki, and the University of Turku.
“For real impact, we need to integrate the foundation model and the decision-making tools into existing research and clinical workflows. It is not just about making them available but also thinking about how they change and improve the daily work,” says Professor Arto Klami from the Department of Computer Science at the University of Helsinki.