Infectious Disease Modeling & Forecasting
We develop mechanistic, statistical, and AI-based models to understand how infectious diseases spread and evolve across populations. Our work spans epidemic reconstruction, forecasting, contact and mobility modeling, and adaptive models in which disease dynamics and human behavior interact.
Featured Projects
-
Seasonal--Temporal--Spatial Residual Learning framework
We identify systematic errors shared across epidemic forecasting models and develop a learning-based approach to correct them using historical forecasts. The framework improves real-time influenza forecasts, reducing forecasting errors compared with official ensemble forecasts.
-
Feedback-Informed Epidemiological Mode (FIEM)
FIEM integrates individual decision-making with infectious disease dynamics, allowing human behavior and epidemic spread to continuously influence one another. The framework enables more realistic evaluation of public health policies by capturing their interacting health, behavioral, and economic consequences.