I lead a research group on AI for clinical decision support in intensive care at the Medical University of Innsbruck.

My research focuses on building machine learning models that work reliably in real clinical environments — across different hospitals, patient populations, and over time. I’m particularly interested in how we can leverage large-scale ICU data to develop foundation models that generalise beyond the settings they were trained on.

Starting February 2026, I’m the PI of GPT-MEDIC, a €1.5M ERC Starting Grant to develop and validate multicentre foundation models for ICU time-series data. I also coordinate MUI’s participation in INDICATE, a consortium building federated data infrastructure for intensive care across Europe.

Before joining Innsbruck, I was a Humboldt Research Fellow at Charité – Universitätsmedizin Berlin, where I worked on EHR-based early warning scores for ICU settings. I also worked on trustworthy AI methods at Fraunhofer IKS in Munich and as a medical statistician at Sensyne Health in Oxford. I hold a PhD in Health Data Science from University College London.