Organized by GMDS AG MoCoMed
28-05-2025 16:00
Artificial Intelligence (AI) is poised to revolutionize biomedicine by enhancing diagnosis, treatment, and prognostic insights across complex biomedical datasets, including clinical records, imaging, and omics data. However, applying AI to biomedicine presents unique challenges: specialized data types such as sequences, limited sample sizes, severe class imbalances, the critical need for model interpretability, and stringent data protection regulations. This talk outlines these challenges and explores cutting-edge solutions, including privacy-preserving strategies like Federated and Swarm Learning. Special focus is given to the pressing issue of antimicrobial resistance (AMR), demonstrating how AI models can predict resistance patterns and cross-resistance using innovative machine learning approaches. Case studies on AMR in environmental samples and decentralized differential abundance analysis illustrate the potential of AI-driven collaboration in medicine without compromising patient data security.
University of Münster, Institute of Medical Informatics
Dominik Heider is the director of the Institute of Medical Informatics at the University of Münster since September 2024 and a visiting professor at the T.H. Chan School of Public Health, Harvard University since 2023. Before he held full professorships at the University of Düsseldorf and the University of Marburg, was associate professor at the TUM Campus Straubing, and Associate Director at Qiagen. He studied Computer Science and did his PhD in 2008 at the University of Münster.
Organized by GMDS AG MoCoMed / Impressum / Privacy