Generative AI mitigates representation bias and improves model fairness through synthetic health data
PLOS Computational Biology · 19.05.2025 · 21 (5) · e1013080
DOI: 10.1371/journal.pcbi.1013080 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 19.05.2025. Verfasst von Raffaele Marchesi und 5 weiteren Autor:innen. Beteiligte Einrichtung: Data Science for Health (DSH), Fondazione Bruno Kessler, Trento, Italy. Hauptthemen laut MEDLINE: Artificial Intelligence. Die Arbeit wurde bisher 27-mal zitiert (Crossref) und nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Representation bias in health data can lead to unfair decisions and compromise the generalisability of research findings. As a consequence, underrepresented subpopulations, such as those from specific ethnic backgrounds or genders, do not benefit equally from clinical discoveries. Several approaches have been developed to mitigate representation bias, ranging from simple resampling methods, such as SMOTE, to recent approaches based on generative adversarial networks (GAN). However, generating high-dimensional time-series synthetic health data remains a significant challenge. In response, we devised a novel architecture (CA-GAN) that synthesises authentic, high-dimensional time series data. CA-GAN outperforms state-of-the-art methods in a qualitative and a quantitative evaluation while avoiding mode collapse, a serious GAN failure. We perform evaluation using 7535 patients with hypotension and sepsis from two diverse, real-world clinical datasets. We show that synthetic data generated by our CA-GAN improves model fairness in Black patients as well as female patients when evaluated separately for each subpopulation. Furthermore, CA-GAN generates authentic data of the minority class while faithfully maintaining the original distribution of data, resulting in improved performance in a downstream predictive task.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Data Science for Health (DSH), Fondazione Bruno Kessler, Trento, Italy.
- Department of Mathematics, University of Pavia, Pavia, Italy.
- Department of Computer Science, University of Manchester, Manchester, United Kingdom.
- Centre for Big Data Research in Health, University of New South Wales, Sydney, New South Wales, Australia.
- Queensland Digital Health Centre, University of Queensland, Brisbane, Queensland, Australia.
- Department of Biomedical Sciences, Humanitas University, Milan, Italy.
ORCID-Profile: Raffaele Marchesi, Sebastiano Barbieri, Giuseppe Jurman, Venet Osmani
Zitieren
APA
Marchesi, R., Micheletti, N., I-Hsien Kuo, N., Barbieri, S., Jurman, G., & Osmani, V. (2025). Generative AI mitigates representation bias and improves model fairness through synthetic health data. PLOS Computational Biology, 21(5), e1013080. https://doi.org/10.1371/journal.pcbi.1013080
Vancouver
Marchesi R, Micheletti N, I-Hsien Kuo N, Barbieri S, Jurman G, Osmani V. Generative AI mitigates representation bias and improves model fairness through synthetic health data. PLOS Computational Biology. 2025;21(5):e1013080. doi:10.1371/journal.pcbi.1013080
BibTeX
@article{Marchesi2025,
author = {Marchesi, Raffaele and Micheletti, Nicolo and I-Hsien Kuo, Nicholas and Barbieri, Sebastiano and Jurman, Giuseppe and Osmani, Venet},
title = {Generative AI mitigates representation bias and improves model fairness through synthetic health data},
journal = {PLOS Computational Biology},
year = {2025},
volume = {21},
number = {5},
pages = {e1013080},
doi = {10.1371/journal.pcbi.1013080}
}
EUVIMED