Healthcare systems restrict clinical data access due to patient privacy concerns. However, restrictive data access slows the development of novel artificial intelligence tools in healthcare. Creating synthetic clinical datasets that could be shared while securing patient privacy would facilitate collaborations with research groups and other partners to develop new methodology, validate existing tools, or compare patient populations. This project involves adapting generative adversarial networks to create synthetic structured electronic health record data along with a consolidated software package to train, generate, and validate synthetic data in the context of realism and privacy. This project was led by BIDS I4H Data Science Health Innovation Fellow Haley Hunter-Zinck.