Generative AI for synthetic data across multiple medical modalities: A systematic review of recent developments and challenges

Mahmoud K. Ibrahim, Yasmina Al Khalil, Sina Amirrajab, Chang Sun, Marcel Breeuwer, Josien P. W. Pluim, Bart Elen, Gökhan Ertaylan, Michel Dumontier

Abstract: Highlights•Broad review of generative models for medical imaging, EHR, text, and signal synthesis.•Focus on overlooked aspects: conditional generation and robust evaluation methods.•Uncovers gaps in leveraging clinical knowledge, patient-specific context, and clinical validation.•Highlights opportunities for cross-modality innovation and standard benchmarks.•Proposes future directions to advance synthetic data use beyond augmentation.