Algorithms Trained on Normal Chest X-rays Can Predict Health Insurance Types
Published in Proceedings of the 9th International Conference on Medical Imaging with Deep Learning (MIDL), PMLR 315, 2026
This research demonstrates that machine learning algorithms trained solely on normal chest X-rays can inadvertently learn to predict a patient’s health insurance type. The findings highlight critical fairness concerns and potential hidden biases within medical artificial intelligence systems.
An earlier version is available as a 2025 arXiv preprint. The citation and author order above follow the published MIDL 2026 proceedings.
Recommended citation: Chen, C.-Y., Abulibdeh, R., Asgari, A., Ordóñez, S. A. C., Celi, L. A., Goode, D., Hamidi, H., McCague, N., Seyyed-Kalantari, L., Sounack, T., & Kuo, P.-C. (2026). Algorithms Trained on Normal Chest X-rays Can Predict Health Insurance Types. Proceedings of the 9th International Conference on Medical Imaging with Deep Learning, PMLR 315, 4166–4181.
Download Paper