Synthetic data generation: a privacy-preserving approach to accelerate rare disease research
Jorge M. Mendes, Aziz Barbar, Marwa Refaie
Abstract
Rare disease research faces significant challenges due to limited patient data, strict privacy regulations, and the need for diverse datasets to develop accurate AI-driven diagnostics and treatments. Synthetic data-artificially generated datasets that mimic patient data while preserving privacy-offer a promising solution to these issues. This article explores how synthetic data can bridge data gaps, enabling the training of AI models, simulating clinical trials, and facilitating cross-border collaborations in rare disease research. We examine case studies where synthetic data successfully replicated patient characteristics, and supported predictive modelling and ensured compliance with regulations like GDPR and HIPAA. While acknowledging current limitations, we discuss synthetic data's potential to revolutionise rare disease research by enhancing data availability and privacy file enabling more efficient and effective research efforts in diagnosing, treating, and managing rare diseases globally.