Artificial Intelligence for In Silico Clinical Trials: A Review

Another closely related research area is synthetic patient data generation. This research area seeks to simulate patient records, usually for specific disease conditions. For example, one can predict future patient EHR data using current patient baseline metrics or historical EHR data. Early works on patient record generation methods used rule-based methods [lombardo2008ta, buczak2010data, mclachlan2016using]. However, the rule-based methods cannot provide realistic and complex patient data to support general machine learning (ML) tasks.