BIH Lecture | Andrea Ganna "Disease Prediction Using Nation-Wide Health Data and Genetics"

“Disease Prediction Using Nation-Wide Health Data and Genetics”

The presentation will outline recent advances in disease prediction through the integration of national health registries, genomic data, and artificial intelligence. Using data from the Finnish FinRegistry (over 7 million individuals and 6.5 billion records), large-scale machine learning models achieve high predictive accuracy for disease outcomes but also reveal disparities across regions and socioeconomic groups, emphasizing fairness and generalizability challenges. The presentation will further demonstrate how polygenic scores (PGS) capture lifelong disease risk and complement electronic health record–derived phenotype risk scores (PheRS), with each excelling for different disease categories. Combining genomic and EHR data enhances trial emulation, strengthens causal inference, and supports the design of more representative clinical studies. The talk will underscore that equitable, ethically guided AI and genetic integration are key to realizing precision prevention at a population scale.

BIO:  Andrea Ganna is an Associate Professor at FIMM and HiLIFE and a research associate at Harvard Medical School and Massachusetts General Hospital. He is also Associate Faculty at the ELLIS Institute Finland and an ELLIS member. 

Andrea's research interests lie at the intersection between epidemiology, genetics, and statistics. He is co-leading the INTERVENE consortium, which aims to integrate AI and human genetics tools for disease prevention and diagnosis across biobanks in Europe. He has been honored with the Leena Peltonen Prize for Excellence in Human Genetics and an ERC Starting Grant. Under this grant, he launched the FinRegistry project, one of the most comprehensive registry-based health studies in the world. Within FinRegistry, Andrea’s team, utilizes AI and machine learning to improve early disease detection and improve public health interventions.