My research develops multiscale computational health frameworks that integrate mechanistic modeling, constrained artificial intelligence, and heterogeneous biomedical data to support precision and population health. Biological and health systems are governed by interacting processes spanning molecular, cellular, tissue, behavioral, and population scales. At the same time, modern health data arise from heterogeneous and often incomplete sources, including multi-omics, physiological, clinical, behavioral, and public health measurements. A major challenge in computational health is to integrate these data while preserving mechanistic interpretability and biological consistency.

To address this challenge, we develop mechanistic and hybrid AI frameworks that combine multiscale mathematical modeling with interpretable and trustworthy machine learning. Our goal is to build computational and digital twin systems that uncover hidden biological dynamics, generate predictive insights, and support decision-making under uncertainty. These frameworks are designed to remain biologically grounded while enabling rapid inference, forecasting, and optimization from sparse and heterogeneous data.

Our research spans infectious diseases, immune dynamics, blood coagulation, tumor growth, developmental biology, behavioral health, and population health systems. Across these applications, we develop computational methods that connect intracellular signaling, cell-cell interactions, tissue dynamics, and population-level processes. We aim to create scalable and clinically relevant computational frameworks that support personalized therapeutics, public health decision-making, and translational biomedical research.