# (De)Constructing Bias on Skin Lesion Datasets

arXiv: https://arxiv.org/abs/1904.08818 · HTML: https://mlsanity.com/paper/1904.08818/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 18 Apr 2019.

## Built on by (most cited first, 30 of 30)

- [Domino: Discovering Systematic Errors with Cross-Modal Embeddings](https://mlsanity.com/paper/2203.14960/) (2022)
- [Towards Robust Classification Model by Counterfactual and Invariant Data Generation](https://mlsanity.com/paper/2106.01127/) (2021)
- [Addressing Fairness Issues in Deep Learning-Based Medical Image Analysis: A Systematic Review](https://mlsanity.com/paper/2209.13177/) (2022)
- [Towards Trustable Skin Cancer Diagnosis via Rewriting Model's Decision](https://mlsanity.com/paper/2303.00885/) (2023)
- [GAN-Based Data Augmentation and Anonymization for Skin-Lesion Analysis: A Critical Review](https://mlsanity.com/paper/2104.10603/) (2021)
- [A Survey on Deep Learning for Skin Lesion Segmentation](https://mlsanity.com/paper/2206.00356/) (2022)
- [Skin Deep Unlearning: Artefact and Instrument Debiasing in the Context of Melanoma Classification](https://mlsanity.com/paper/2109.09818/) (2021)
- [Epistemic Uncertainty-Weighted Loss for Visual Bias Mitigation](https://mlsanity.com/paper/2204.09389/) (2022)
- [Artifact-Based Domain Generalization of Skin Lesion Models](https://mlsanity.com/paper/2208.09756/) (2022)
- [An Evaluation of Self-Supervised Pre-Training for Skin-Lesion Analysis](https://mlsanity.com/paper/2106.09229/) (2021)
- [The (de)biasing effect of GAN-based augmentation methods on skin lesion images](https://mlsanity.com/paper/2206.15182/) (2022)
- [EPVT: Environment-aware Prompt Vision Transformer for Domain Generalization in Skin Lesion Recognition](https://mlsanity.com/paper/2304.01508/) (2023)
- [Even Small Correlation and Diversity Shifts Pose Dataset-Bias Issues](https://mlsanity.com/paper/2305.05807/) (2023)
- [Prompt-driven Latent Domain Generalization for Medical Image Classification](https://mlsanity.com/paper/2401.03002/) (2024)
- [Multi-group Learning for Hierarchical Groups](https://mlsanity.com/paper/2402.00258/) (2024)
- [Mask of truth: model sensitivity to unexpected regions of medical images](https://mlsanity.com/paper/2412.04030/) (2024)
- [Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models](https://mlsanity.com/paper/2304.03218/) (2023)
- [A survey on bias in machine learning research](https://mlsanity.com/paper/2308.11254/) (2023)
- [Deep Learning Based Decision Support for Medicine -- A Case Study on Skin Cancer Diagnosis](https://mlsanity.com/paper/2103.05112/) (2021)
- [Test-Time Selection for Robust Skin Lesion Analysis](https://mlsanity.com/paper/2308.05595/) (2023)
- [Targeted Data Augmentation for bias mitigation](https://mlsanity.com/paper/2308.11386/) (2023)
- [Let Samples Speak: Mitigating Spurious Correlation by Exploiting the Clusterness of Samples](https://mlsanity.com/paper/2512.22874/) (2025)
- [Leveraging Contextual Data Augmentation for Generalizable Melanoma Detection](https://mlsanity.com/paper/2212.05116/) (2022)
- [MaskSearch: Querying Image Masks at Scale](https://mlsanity.com/paper/2305.02375/) (2023)
- [Are Explanations Helpful? A Comparative Analysis of Explainability Methods in Skin Lesion Classifiers](https://mlsanity.com/paper/2412.03166/) (2024)
- [Synthetic Melanoma Image Generation and Evaluation Using Generative Adversarial Networks](https://mlsanity.com/paper/2603.13497/) (2026)
- [A Causal Framework for Mitigating Data Shifts in Healthcare](https://mlsanity.com/paper/2603.13595/) (2026)
- [Evaluating the Impact of Medical Image Reconstruction on Downstream AI Fairness and Performance](https://mlsanity.com/paper/2604.10904/) (2026)
- [Searching for Robust Augmentations to Improve Out-of-Domain Generalization in Dermoscopic Skin Cancer Classification](https://mlsanity.com/paper/2607.26765/) (2026)
- [CAMEO: A Class-Activation-Mapped Equitable Overlay Framework for Fair and Robust Deep Learning-based Skin Condition Diagnosis](https://mlsanity.com/paper/2609.36400/) (2026)

## Abstract (arXiv)

Melanoma is the deadliest form of skin cancer. Automated skin lesion analysis plays an important role for early detection. Nowadays, the ISIC Archive and the Atlas of Dermoscopy dataset are the most employed skin lesion sources to benchmark deep-learning based tools. However, all datasets contain biases, often unintentional, due to how they were acquired and annotated. Those biases distort the performance of machine-learning models, creating spurious correlations that the models can unfairly exploit, or, contrarily destroying cogent correlations that the models could learn. In this paper, we propose a set of experiments that reveal both types of biases, positive and negative, in existing skin lesion datasets. Our results show that models can correctly classify skin lesion images without clinically-meaningful information: disturbingly, the machine-learning model learned over images where no information about the lesion remains, presents an accuracy above the AI benchmark curated with dermatologists' performances. That strongly suggests spurious correlations guiding the models. We fed models with additional clinically meaningful information, which failed to improve the results even slightly, suggesting the destruction of cogent correlations. Our main findings raise awareness of the limitations of models trained and evaluated in small datasets such as the ones we evaluated, and may suggest future guidelines for models intended for real-world deployment.