# Risk of Bias in Chest Radiography Deep Learning Foundation Models

arXiv: https://arxiv.org/abs/2209.02965 · HTML: https://mlsanity.com/paper/2209.02965/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 35 references that are in the map.
- First posted on arXiv 7 Sep 2022.

## Builds on (most cited first, 13 of 13)

- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [Hidden Stratification Causes Clinically Meaningful Failures in Machine Learning for Medical Imaging](https://mlsanity.com/paper/1909.12475/) (2019)
- [Causality matters in medical imaging](https://mlsanity.com/paper/1912.08142/) (2019)
- [Reading Race: AI Recognises Patient's Racial Identity In Medical Images](https://mlsanity.com/paper/2107.10356/) (2021)
- [CheXclusion: Fairness gaps in deep chest X-ray classifiers](https://mlsanity.com/paper/2003.00827/) (2020)
- [Deep Learning for Chest X-ray Analysis: A Survey](https://mlsanity.com/paper/2103.08700/) (2021)
- [On the limits of cross-domain generalization in automated X-ray prediction](https://mlsanity.com/paper/2002.02497/) (2020)
- [Robust and Efficient Medical Imaging with Self-Supervision](https://mlsanity.com/paper/2205.09723/) (2022)
- [Self-supervised Learning from 100 Million Medical Images](https://mlsanity.com/paper/2201.01283/) (2022)
- [Potential sources of dataset bias complicate investigation of underdiagnosis by machine learning algorithms](https://mlsanity.com/paper/2201.07856/) (2022)
- [Debiasing Deep Chest X-Ray Classifiers using Intra- and Post-processing Methods](https://mlsanity.com/paper/2208.00781/) (2022)

## Built on by (most cited first, 27 of 27)

- [Foundation Models in Radiology: What, How, When, Why and Why Not](https://mlsanity.com/paper/2411.18730/) (2024)
- [EVA-X: A Foundation Model for General Chest X-ray Analysis with Self-supervised Learning](https://mlsanity.com/paper/2405.05237/) (2024)
- [Demographic Bias of Expert-Level Vision-Language Foundation Models in Medical Imaging](https://mlsanity.com/paper/2402.14815/) (2024)
- [Slicing Through Bias: Explaining Performance Gaps in Medical Image Analysis using Slice Discovery Methods](https://mlsanity.com/paper/2406.12142/) (2024)
- [Foundation Models in Medical Imaging: A Review and Outlook](https://mlsanity.com/paper/2506.09095/) (2025)
- [Low-Rank Knowledge Decomposition for Medical Foundation Models](https://mlsanity.com/paper/2404.17184/) (2024)
- [LoRKD: Low-Rank Knowledge Decomposition for Medical Foundation Models](https://mlsanity.com/paper/2409.19540/) (2024)
- [Bias and Generalizability of Foundation Models across Datasets in Breast Mammography](https://mlsanity.com/paper/2505.10579/) (2025)
- [Preventing Shortcut Learning in Medical Image Analysis through Intermediate Layer Knowledge Distillation from Specialist Teachers](https://mlsanity.com/paper/2511.17421/) (2025)
- [Fair Foundation Models for Medical Image Analysis: Challenges and Perspectives](https://mlsanity.com/paper/2502.16841/) (2025)
- [Generalist Models in Medical Image Segmentation: A Survey and Performance Comparison with Task-Specific Approaches](https://mlsanity.com/paper/2506.10825/) (2025)
- [Post-hoc Orthogonalization for Mitigation of Protected Feature Bias in CXR Embeddings](https://mlsanity.com/paper/2311.01349/) (2023)
- [Understanding Dataset Bias in Medical Imaging: A Case Study on Chest X-rays](https://mlsanity.com/paper/2507.07722/) (2025)
- [Multi Anatomy X-Ray Foundation Model](https://mlsanity.com/paper/2509.12146/) (2025)
- [DUCX: Decomposing Unfairness in Tool-Using Chest X-ray Agents](https://mlsanity.com/paper/2603.00777/) (2026)
- [Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift](https://mlsanity.com/paper/2607.10358/) (2026)
- [The Importance of Model Inspection for Better Understanding Performance Characteristics of Graph Neural Networks](https://mlsanity.com/paper/2405.01270/) (2024)
- [Generalizing Orthogonalization for Models with Non-Linearities](https://mlsanity.com/paper/2405.02475/) (2024)
- [Does Data-Efficient Generalization Exacerbate Bias in Foundation Models?](https://mlsanity.com/paper/2408.16154/) (2024)
- [Predict Patient Self-reported Race from Skin Histological Images](https://mlsanity.com/paper/2507.21912/) (2025)
- [The Impact of Preprocessing Methods on Racial Encoding and Model Robustness in CXR Diagnosis](https://mlsanity.com/paper/2603.05157/) (2026)
- [ViX-Ray: A Vietnamese Chest X-Ray Dataset for Vision-Language Models](https://mlsanity.com/paper/2603.15513/) (2026)
- [Intersectional Disentangling of Temporal and Acquisition Bias in Fetal Ultrasound](https://mlsanity.com/paper/2605.02942/) (2026)
- [On-Device Continual Learning with Dual-Stage Buffer and Dynamic Loss for Point-of-Care Pneumonia Diagnosis](https://mlsanity.com/paper/2605.19201/) (2026)
- [Effect of Demographic Bias on Skin Lesion Classification](https://mlsanity.com/paper/2606.03214/) (2026)
- [Subgroup performance analysis of adaptation strategies for chest X-ray foundation models](https://mlsanity.com/paper/2608.19078/) (2026)
- [Representation Disentanglement for Fair Chest X-Ray Diagnosis](https://mlsanity.com/paper/2610.06041/) (2026)

## Abstract (arXiv)

Purpose: To analyze a recently published chest radiography foundation model for the presence of biases that could lead to subgroup performance disparities across biological sex and race. Materials and Methods: This retrospective study used 127,118 chest radiographs from 42,884 patients (mean age, 63 [SD] 17 years; 23,623 male, 19,261 female) from the CheXpert dataset collected between October 2002 and July 2017. To determine the presence of bias in features generated by a chest radiography foundation model and baseline deep learning model, dimensionality reduction methods together with two-sample Kolmogorov-Smirnov tests were used to detect distribution shifts across sex and race. A comprehensive disease detection performance analysis was then performed to associate any biases in the features to specific disparities in classification performance across patient subgroups. Results: Ten out of twelve pairwise comparisons across biological sex and race showed statistically significant differences in the studied foundation model, compared with four significant tests in the baseline model. Significant differences were found between male and female (P < .001) and Asian and Black patients (P < .001) in the feature projections that primarily capture disease. Compared with average model performance across all subgroups, classification performance on the 'no finding' label dropped between 6.8% and 7.8% for female patients, and performance in detecting 'pleural effusion' dropped between 10.7% and 11.6% for Black patients. Conclusion: The studied chest radiography foundation model demonstrated racial and sex-related bias leading to disparate performance across patient subgroups and may be unsafe for clinical applications.