Digital twins of biological systems

Our research focuses on developing computational approaches to personalized medicine, with the goal of understanding how individual biological differences influence disease progression and treatment response.

We use a combination of mechanistic modeling, probabilistic modeling, and predictive analytics to represent complex biological systems and generate patient-specific predictions. Our work with ordinary differential equation (ODE) systems examines the dynamics of tumor-immune interactions and treatment response, including how therapeutic timing may be optimized for individual disease states. I also develop probabilistic graph models to represent biological and anatomical connectivity and explore how uncertainty can be incorporated into individualized predictions. Complementing these mechanistic approaches, I work with data-driven prediction models to identify patterns that can improve risk prediction and treatment decisions.

Across these projects, my broader goal is to develop computational models that connect biological data with individualized predictions, ultimately supporting more precise and adaptive approaches to healthcare.