Mitigating Individual Skin Tone Bias in Skin Lesion Classification through Distribution-Aware Reweighting
Not on the map. Why?
Abstract (arXiv)
Skin color has historically been a focal point of discrimination, yet fairness research in machine learning for medical imaging often relies on coarse subgroup categories, overlooking individual-level variations. Such group-based approaches risk obscuring biases faced by outliers within subgroups. This study introduces a distribution-based framework for evaluating and mitigating individual fairness in skin lesion classification. We treat skin tone as a continuous attribute rather than a categorical label, and employ kernel density estimation (KDE) to model its distribution. We further compare twelve statistical distance metrics to quantify disparities between skin tone distributions and propose a distance-based reweighting (DRW) loss function to correct underrepresentation in minority tones. Experiments across CNN and Transformer models demonstrate: (i) the limitations of categorical reweighting in capturing individual-level disparities, and (ii) the superior performance of distribution-based reweighting, particularly with Fidelity Similarity (FS), Wasserstein Distance (WD), Hellinger Metric (HM), and Harmonic Mean Similarity (HS). These findings establish a robust methodology for advancing fairness at individual level in dermatological AI systems, and highlight broader implications for sensitive continuous attributes in medical image analysis.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 29 of its 77 references that are in the map. First posted on arXiv 9 Dec 2025.
Builds on
29 of 77 references.
References · 29
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
- 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
- 2016Equality of Opportunity in Supervised Learningcited by 2,007
- 2017StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translationcited by 1,039
- 2019SinGAN: Learning a Generative Model from a Single Natural Imagecited by 348
- 2018Learning Not to Learn: Training Deep Neural Networks with Biased Datacited by 255
- 2021Disparities in Dermatology AI: Assessments Using Diverse Clinical Imagescited by 134
- 2018Turning a Blind Eye: Explicit Removal of Biases and Variation from Deep Neural Network Embeddingscited by 114
- 2024Preserving Fairness Generalization in Deepfake Detectioncited by 70
- 2022FairPrune: Achieving Fairness Through Pruning for Dermatological Disease Diagnosiscited by 59
17 more
- 2020Evaluating and Mitigating Bias in Image Classifiers: A Causal Perspective Using Counterfactualscited by 56
- 2022Third Time's the Charm? Image and Video Editing with StyleGAN3cited by 54
- 2022Towards Transparency in Dermatology Image Datasets with Skin Tone Annotations by Experts, Crowds, and an Algorithmcited by 35
- 2019Estimating Skin Tone and Effects on Classification Performance in Dermatology Datasetscited by 29
- 2021Estimating and Improving Fairness with Adversarial Learningcited by 26
- 2022CIRCLe: Color Invariant Representation Learning for Unbiased Classification of Skin Lesionscited by 17
- 2023A Web-based Mpox Skin Lesion Detection System Using State-of-the-art Deep Learning Models Considering Racial Diversitycited by 17
- 2021Analysis of Manual and Automated Skin Tone Assignments for Face Recognition Applicationscited by 16
- 2023Toward Fairness Through Fair Multi-Exit Framework for Dermatological Disease Diagnosiscited by 14
- 2023Skin Deep: Investigating Subjectivity in Skin Tone Annotations for Computer Vision Benchmark Datasetscited by 11
- 2024Achieving Fairness Through Channel Pruning for Dermatological Disease Diagnosiscited by 9
- 2024BMFT: Achieving Fairness via Bias-based Weight Masking Fine-tuningcited by 6
- 2022Achieving Fairness in Dermatological Disease Diagnosis through Automatic Weight Adjusting Federated Learning and Personalizationcited by 4
- 2022A Fair Loss Function for Network Pruningcited by 4
- 2025DermDiff: Generative Diffusion Model for Mitigating Racial Biases in Dermatology Diagnosiscited by 4
- 2024Biasing & Debiasing based Approach Towards Fair Knowledge Transfer for Equitable Skin Analysiscited by 2
- 2025Evaluating Fairness and Mitigating Bias in Machine Learning: A Novel Technique using Tensor Data and Bayesian Regressioncited by 1
Built on by
Not cited yet.