An Explainable Epidemiology System for Vaccine Safety and Proactive Care

This project aims to develop an explainable epidemiology system that combines generative AI, biostatistics, and epidemiological reasoning to support healthcare analysis and decision-making. The system will take structured health data as input, analyse it multimodally, and build a knowledge base that can be queried by epidemiologists and health researchers. The goal is to enable users to ask practical questions such as whether a vaccine or exposure is associated with increased risk, how strong the signal is, which methods support the finding, and what assumptions or parameter settings were used.

The project is designed to support both vaccine safety surveillance and proactive care in general practice. In vaccine safety, it can help identify potential adverse event signals and explain why they may warrant attention. In general practice, it can help identify patients at rising risk of deterioration, support earlier intervention, and reduce administrative burden by surfacing clinically relevant patterns in a transparent and auditable way. The emphasis is on interpretability, trust, and practical usefulness rather than black-box prediction.