# The Role of Subgroup Separability in Group-Fair Medical Image Classification

arXiv: https://arxiv.org/abs/2307.02791 · HTML: https://mlsanity.com/paper/2307.02791/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 27 references that are in the map.
- First posted on arXiv 6 Jul 2023.

## Builds on (most cited first, 17 of 17)

- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning](https://mlsanity.com/paper/1711.05225/) (2017)
- [Learning Not to Learn: Training Deep Neural Networks with Biased Data](https://mlsanity.com/paper/1812.10352/) (2018)
- [Towards Fairness in Visual Recognition: Effective Strategies for Bias Mitigation](https://mlsanity.com/paper/1911.11834/) (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)
- [Evaluating Deep Neural Networks Trained on Clinical Images in Dermatology with the Fitzpatrick 17k Dataset](https://mlsanity.com/paper/2104.09957/) (2021)
- [A Fine-Grained Analysis on Distribution Shift](https://mlsanity.com/paper/2110.11328/) (2021)
- [Reading Race: AI Recognises Patient's Racial Identity In Medical Images](https://mlsanity.com/paper/2107.10356/) (2021)
- [Turning a Blind Eye: Explicit Removal of Biases and Variation from Deep Neural Network Embeddings](https://mlsanity.com/paper/1809.02169/) (2018)
- [MEDFAIR: Benchmarking Fairness for Medical Imaging](https://mlsanity.com/paper/2210.01725/) (2022)
- [Leveling Down in Computer Vision: Pareto Inefficiencies in Fair Deep Classifiers](https://mlsanity.com/paper/2203.04913/) (2022)
- [The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by default](https://mlsanity.com/paper/2302.02404/) (2023)
- [Towards Transparency in Dermatology Image Datasets with Skin Tone Annotations by Experts, Crowds, and an Algorithm](https://mlsanity.com/paper/2207.02942/) (2022)
- [Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcuts](https://mlsanity.com/paper/2009.10132/) (2020)
- [Potential sources of dataset bias complicate investigation of underdiagnosis by machine learning algorithms](https://mlsanity.com/paper/2201.07856/) (2022)

## Built on by (most cited first, 11 of 11)

- [FairVision: Equitable Deep Learning for Eye Disease Screening via Fair Identity Scaling](https://mlsanity.com/paper/2310.02492/) (2023)
- [No Fair Lunch: A Causal Perspective on Dataset Bias in Machine Learning for Medical Imaging](https://mlsanity.com/paper/2307.16526/) (2023)
- [Rethinking Fair Representation Learning for Performance-Sensitive Tasks](https://mlsanity.com/paper/2410.04120/) (2024)
- [Understanding challenges to the interpretation of disaggregated evaluations of algorithmic fairness](https://mlsanity.com/paper/2506.04193/) (2025)
- [Towards objective and systematic evaluation of bias in artificial intelligence for medical imaging](https://mlsanity.com/paper/2311.02115/) (2023)
- [On Biases in a UK Biobank-based Retinal Image Classification Model](https://mlsanity.com/paper/2408.02676/) (2024)
- [Exploring the interplay of label bias with subgroup size and separability: A case study in mammographic density classification](https://mlsanity.com/paper/2507.17996/) (2025)
- [Towards Fairness under Label Bias in Image Segmentation: Impact, Measurement and Mitigation](https://mlsanity.com/paper/2605.06891/) (2026)
- [Evaluating the Fairness of Neural Collapse in Medical Image Classification](https://mlsanity.com/paper/2407.05843/) (2024)
- [A Primer on Causal and Statistical Dataset Biases for Fair and Robust Image Analysis](https://mlsanity.com/paper/2509.04295/) (2025)
- [Is invariance all you need for algorithmic fairness? Removing demographic information can create new bias](https://mlsanity.com/paper/2609.32004/) (2026)

## Abstract (arXiv)

We investigate performance disparities in deep classifiers. We find that the ability of classifiers to separate individuals into subgroups varies substantially across medical imaging modalities and protected characteristics; crucially, we show that this property is predictive of algorithmic bias. Through theoretical analysis and extensive empirical evaluation, we find a relationship between subgroup separability, subgroup disparities, and performance degradation when models are trained on data with systematic bias such as underdiagnosis. Our findings shed new light on the question of how models become biased, providing important insights for the development of fair medical imaging AI.