https://deisygysi.github.io/categories/ai-for-healthcare/Recent content in AI for Healthcare on Wonderland of Deisy GysiHugo -- gohugo.ioenTue, 01 Sep 2026 00:00:00 +0000https://deisygysi.github.io/projects/ai-healthcare/Tue, 01 Sep 2026 00:00:00 +0000https://deisygysi.github.io/projects/ai-healthcare/Summary Artificial intelligence can help connect complex biomedical data to useful predictions, but high benchmark performance does not guarantee that a model will generalize to new molecules, populations or clinical settings. Our research examines both the opportunities and the limits of machine learning in healthcare. We combine AI with network-based representations to improve prediction, interpretability and evaluation. The emphasis is on methods that reveal the evidence behind a prediction, avoid shortcuts hidden in biomedical datasets and remain useful beyond well-annotated examples.https://deisygysi.github.io/projects/drugrepurpusing/Tue, 01 Sep 2026 00:00:00 +0000https://deisygysi.github.io/projects/drugrepurpusing/Summary Discovering a therapy requires connecting evidence across molecular profiles, chemical structures, protein targets, pathways, clinical data and experimental models. We develop and evaluate computational strategies that bring these sources together to prioritize drug-target relationships and therapeutic candidates. Our work combines network medicine, machine learning, molecular docking and experimental validation. A central goal is to understand when predictive models generalize to new proteins, compounds and diseases, rather than performing well only on well-annotated examples.