Background

Real world health data contain important knowledge that enables clinical research to assess treatment effect in real world settings. However, there are significant limitations: real-world data are typically imbalanced across different population, diseases and interventions; they contain bias, noise and missing measurements; the process of removing patient identifiable information may take significant time and effort, which also faces the risk of deleting valuable information from the original data. More importantly, observational studies with real-world data have inherent limitations in their ability to identify treatment effects due to their observational nature.

This project is designed to investigate an alternative approach to support clinical research through the use of synthetic data. We study the feasibility of virtual clinical trial emulations through creating synthetic patient populations with the help of the latest AI technology in generative AI and causality learning. The clincail trial emulations aim to answer clinical questions about treatment effect size in a wide range of clinical trial settings.

We have studied the feasibility of this new approach through a specific use case in the context of Type 2 diabetes mellites (T2DM). Through training the AI model with the SCI Diabetes data on the Safe Haven platform, we have carried out virtual trial emulations to assess the effect size of a target drug and compared the outcomes with the real ones. .