Multimodal Machine Learning for Patient Stratification and Outcome Prediction

15.08.2026 bis 14.08.2029

This PhD project explores machine learning methods for integrating multimodal biomedical data to better understand disease heterogeneity and predict clinical outcomes. The focus is on combining diverse data sources such as clinical variables, biomarkers, genomics, and large-scale electronic health records (EHRs) to learn unified patient representations. Methodological work will address challenges such as missing data, irregular sampling, and differences across datasets, with the aim of developing scalable and interpretable models for longitudinal data. Overall, the goal is to contribute generalizable methods for data-driven patient stratification and outcome prediction in real-world clinical environments.