Artificial Intelligence for Bioscientific Research – Prof. Laura Azzimonti

This is an online event, tune in via Webex here.

Physics-informed machine learning and causal artificial intelligence for personalized diabetes prevention

Personalised medicine requires tools capable of integrating heterogeneous expert knowledge, from mechanistic models to causal relationships, with real-world clinical data. Physics-Informed Machine Learning (PIML) and Causal Artificial Intelligence (Causal AI) represent two complementary methodological frameworks that address this challenge: PIML embeds domain knowledge encoded in differential equations directly into the learning process, while Causal AI enables the estimation of intervention effects from real-world data. Together, they offer a principled way to move beyond black-box predictions towards interpretable, mechanistically grounded models that can inform individualised clinical decisions. This talk illustrates both approaches through applications to type 2 diabetes prevention, developed within the Horizon Europe project PRAESIIDIUM. Specifically, we show how PIML and Causal AI can jointly support clinicians by enabling physics-grounded personalised predictions and guiding prevention strategies across heterogeneous patient populations.

About the speaker

Laura Azzimonti is a senior researcher and lecturer at the Dalle Molle Institute for Artificial Intelligence (SUPSI-IDSIA) in Lugano, Switzerland, and a SIB Group Leader. She develops machine learning methods for personalised medicine, focusing on physics-informed machine learning, causal AI, federated learning and Bayesian modelling, with applications in bioinformatics, clinical decision support and biomedical modelling

• This lecture is part of the Causal Analysis of Biomedical Data Lecture Series

The Causal Analysis of Biomedical Data Lecture Series is supported by the Luxembourg National Research Fund (FNR) RESCOM Program.