Unveiling the hidden relations between different diseases with cutting-edge data science and health records
We use large-scale health data and network-based approaches to understand how to categorize diseases, an classify diseases based on physiological measures. Drawing from millions of electronic health records across several countries, we develop machine learning models to predict disease onset, identify risk factors, and personalize treatment.
We are particularly interested in mapping connections between diseases and understanding how individuals transition between health states. Using network science, we model these relationships to uncover common pathways and physiological patterns across people. In collaboration with Harvard and Yale Medical Schools, our research aims to support early diagnosis, causal inference, and intervention strategies.