Chen Xingran, Wu Zhenke, Shi Xu, Cho Hyunghoon, Mukherjee Bhramar
Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109, United States.
Department of Biomedical Informatics and Data Science, Yale University, New Haven, CT 06520, United States.
J Am Med Inform Assoc. 2025 Jul 1;32(7):1227-1240. doi: 10.1093/jamia/ocaf082.
To conduct a scoping review (ScR) of existing approaches for synthetic Electronic Health Records (EHR) data generation, to benchmark major methods, and to provide an open-source software and offer recommendations for practitioners.
We search three academic databases for our scoping review. Methods are benchmarked on open-source EHR datasets, Medical Information Mart for Intensive Care III and IV (MIMIC-III/IV). Seven existing methods covering major categories and two baseline methods are implemented and compared. Evaluation metrics concern data fidelity, downstream utility, privacy protection, and computational cost.
Forty-eight studies are identified and classified into five categories. Seven open-source methods covering all categories are selected, trained on MIMIC-III, and evaluated on MIMIC-III or MIMIC-IV for transportability considerations. Among them, Generative Adversarial Network (GAN)-based methods demonstrate competitive performance in fidelity and utility on MIMIC-III, rule-based methods excel in privacy protection. Similar findings are observed on MIMIC-IV, except that GAN-based methods further outperform the baseline methods in preserving fidelity.
Method choice is governed by the relative importance of the evaluation metrics in downstream use cases. We provide a decision tree to guide the choice among the benchmarked methods. An extensible Python package, "SynthEHRella", is provided to facilitate streamlined evaluations.
GAN-based methods excel when distributional shifts exist between the training and testing populations. Otherwise, CorGAN and MedGAN are most suitable for association modeling and predictive modeling, respectively. Future research should prioritize enhancing fidelity of the synthetic data while controlling privacy exposure, and comprehensive benchmarking of longitudinal or conditional generation methods.
对现有的合成电子健康记录（EHR）数据生成方法进行范围审查（ScR），对主要方法进行基准测试，并提供开源软件并为从业者提供建议。
我们在三个学术数据库中进行范围审查。在开源EHR数据集“重症监护医学信息库III和IV（MIMIC-III/IV）”上对方法进行基准测试。实施并比较了涵盖主要类别的七种现有方法和两种基线方法。评估指标涉及数据保真度、下游效用、隐私保护和计算成本。
识别出48项研究并将其分为五类。选择了涵盖所有类别的七种开源方法，在MIMIC-III上进行训练，并出于可移植性考虑在MIMIC-III或MIMIC-IV上进行评估。其中，基于生成对抗网络（GAN）的方法在MIMIC-III上的保真度和效用方面表现出竞争力，基于规则的方法在隐私保护方面表现出色。在MIMIC-IV上也观察到了类似的结果，只是基于GAN的方法在保持保真度方面进一步优于基线方法。
方法的选择取决于下游用例中评估指标的相对重要性。我们提供了一个决策树来指导在基准方法之间进行选择。提供了一个可扩展的Python包“SynthEHRella”，以促进简化评估。
当训练和测试人群之间存在分布变化时，基于GAN的方法表现出色。否则，CorGAN和MedGAN分别最适合关联建模和预测建模。未来的研究应优先提高合成数据的保真度，同时控制隐私暴露，并对纵向或条件生成方法进行全面基准测试。
