Scorecards for Synthetic Medical Data Evaluation and Reporting

• DOI: 10.48550/arxiv.2406.11143
• Publication Date: 2024-01-01
• Authors: Zamzmi, Ghada; Subbaswamy, Adarsh; Sizikova, Elena; Margerrison, Edward; Delfino, Jana; Badano, Aldo
• Keywords: FOS: Computer and information sciences; Computer Science - Computers and Society; Artificial Intelligence (cs.AI); Computer Science - Databases; Computer Science - Artificial Intelligence; Computers and Society (cs.CY); Databases (cs.DB)

7 pages, 2 figures<br/>Although interest in synthetic medical data (SMD) for training and testing AI methods is growing, the absence of a standardized framework to evaluate its quality and applicability hinders its wider adoption. Here, we outline an evaluation framework designed to meet the unique requirements of medical applications, and introduce SMD Card, which can serve as comprehensive reports that accompany artificially generated datasets. This card provides a transparent and standardized framework for evaluating and reporting the quality of synthetic data, which can benefit SMD developers, users, and regulators, particularly for AI models using SMD in regulatory submissions.<br/>