Evaluating the quality of tabular synthetic data in health care
PLOS Digital Health · 07.07.2026 · 5 (7) · e0001522
DOI: 10.1371/journal.pdig.0001522 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Digital Health, online erschienen am 07.07.2026. Verfasst von Ivana Nanevski und 7 weiteren Autor:innen. Beteiligte Einrichtung: Berliner Hochschule für Technik, Berlin, Germany. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Bundesministerium für Bildung und Forschung, Deutsche Forschungsgemeinschaft.
Abstract (Originalsprache)
Machine Learning (ML) research in healthcare remains challenging as large, privacy-preserving open datasets are lacking. Synthetic data could offer a solution, but the value of synthetic data depends on diverse and conflicting criteria such as utility, fidelity, and privacy, which are rarely evaluated comprehensively. To close this gap, we explore the trade-off between these metrics in an empirical evaluation across a broad spectrum of generative models, datasets and metrics. In order to include as many metrics and models as possible and to ensure both applicability and comparability with other studies, we focus on the most widely available data modality and task setting: tabular data associated with a classification task. Extending prior work our results demonstrate that no single generative model excels across all metrics and datasets. Across 9 datasets and 11 generative models, the first principal variance direction of all metrics captures the dominant trade-off between fidelity and utility metrics on one side and the privacy metrics on the other side. Sensitivity analyses indicate that the privacy–fidelity/utility trade-off captured by the first principal variance direction remains consistent across several datasets and may support model selection. These insights highlight the potential of synthetic data for responsible data sharing in health care as well as the need for better tooling in synthetic data generation with a higher degree of automation when optimizing for metrics capturing fidelity, utility and privacy.
Autor:innen und Institutionen
- Berliner Hochschule für Technik, Berlin, Germany.
- Berlin Institute of Health at Charité - Universitätsmedizin Berlin, Berlin, Germany.
- Freie Universität Berlin, Berlin, Germany.
- Charité - Universitätsmedizin Berlin, Berlin, Germany.
- Einstein Center Digital Future, Berlin, Germany.
ORCID-Profile: Ivana Nanevski
Förderung
- Bundesministerium für Bildung und Forschung
- Deutsche Forschungsgemeinschaft
Zitieren
APA
Nanevski, I., Mohebi, M., Jäger, S., Otte, K., Prasser, F., Schulte-Althoff, M., Fürstenau, D., & Biessmann, F. (2026). Evaluating the quality of tabular synthetic data in health care. PLOS Digital Health, 5(7), e0001522. https://doi.org/10.1371/journal.pdig.0001522
Vancouver
Nanevski I, Mohebi M, Jäger S, Otte K, Prasser F, Schulte-Althoff M, et al. Evaluating the quality of tabular synthetic data in health care. PLOS Digital Health. 2026;5(7):e0001522. doi:10.1371/journal.pdig.0001522
BibTeX
@article{Nanevski2026,
author = {Nanevski, Ivana and Mohebi, Maryam and Jäger, Sebastian and Otte, Karen and Prasser, Fabian and Schulte-Althoff, Matthias and Fürstenau, Daniel and Biessmann, Felix},
title = {Evaluating the quality of tabular synthetic data in health care},
journal = {PLOS Digital Health},
year = {2026},
volume = {5},
number = {7},
pages = {e0001522},
doi = {10.1371/journal.pdig.0001522}
}
EUVIMED