Biomedical research generates enormous amounts of data, ranging from molecular measurements to brain images and clinical records; however, it remains challenging to leverage this data to better understand pathological mechanisms. This chair is developing explainable AI that links heterogeneous biomedical data to validated biomedical knowledge, enabling researchers and clinicians to explore hypotheses while ensuring that every result is traceable back to its source. Her work combines data representation (knowledge graphs, ontologies, and community data standards), statistical and machine learning models that account for data structure, and collaborative design with the scientists who use them. Together, these elements form an end-to-end methodology in which the needs and gaps in the field define the computational problems to be solved.
MIAI Cluster is co-funding two doctoral projects at the chair :
MEDiQ applies explainable reinforcement learning to biomedical knowledge graphs to help researchers explore the mechanisms underlying facioscapulohumeral muscular dystrophy (FSHD1), a rare muscle disorder, in collaboration with biologists from the University of Savoie Mont Blanc and University College London.
PRIME-AI (Precision Reporting of Individual Mapping for Enhanced AI) develops AI methods for brain mapping at the individual level and makes neuroimaging data interoperable with biomedical knowledge standards, in collaboration with the LREN laboratory (CHUV – University of Lausanne).
The AI4Health group holds monthly meetings open to the public, during which biomedical researchers and computer scientists translate complex real-world questions into computational solutions, and then reinterpret the computational results as biological and clinical questions. Anyone interested is welcome to attend.
Share the linkCopyCopiedClose the modal windowShare the URL of this pageI recommend this page:Available at this address:The page will then be accessible from your "My Favorites" menu.Stop the videoPlay the videoMutePlay audioChat: Have a question?Robo Fabrica ChatbotMatomo Traffic StatisticsX (formerly Twitter)
