# Revisiting the Shape-Bias of Deep Learning for Dermoscopic Skin Lesion Classification

arXiv: https://arxiv.org/abs/2206.06466 · HTML: https://mlsanity.com/paper/2206.06466/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 34 references that are in the map.
- First posted on arXiv 13 Jun 2022.

## Builds on (most cited first, 10 of 10)

- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [Sequence to Sequence Learning with Neural Networks](https://mlsanity.com/paper/1409.3215/) (2014)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [Adversarial Examples Are Not Bugs, They Are Features](https://mlsanity.com/paper/1905.02175/) (2019)
- [Last Layer Re-Training is Sufficient for Robustness to Spurious Correlations](https://mlsanity.com/paper/2204.02937/) (2022)
- [A Fourier-based Framework for Domain Generalization](https://mlsanity.com/paper/2105.11120/) (2021)
- [A Fourier Perspective on Model Robustness in Computer Vision](https://mlsanity.com/paper/1906.08988/) (2019)
- [Amplitude-Phase Recombination: Rethinking Robustness of Convolutional Neural Networks in Frequency Domain](https://mlsanity.com/paper/2108.08487/) (2021)
- [Boundary-aware Transformers for Skin Lesion Segmentation](https://mlsanity.com/paper/2110.03864/) (2021)
- [Towards Frequency-Based Explanation for Robust CNN](https://mlsanity.com/paper/2005.03141/) (2020)

## Built on by (most cited first, 4 of 4)

- [ImageNet-trained CNNs are not biased towards texture: Revisiting feature reliance through controlled suppression](https://mlsanity.com/paper/2509.20234/) (2025)
- [A Systematic Review of Generalization Research in Medical Image Classification](https://mlsanity.com/paper/2403.12167/) (2024)
- [DocXplain: A Novel Model-Agnostic Explainability Method for Document Image Classification](https://mlsanity.com/paper/2407.03830/) (2024)
- [DocVCE: Diffusion-based Visual Counterfactual Explanations for Document Image Classification](https://mlsanity.com/paper/2508.04233/) (2025)

## Abstract (arXiv)

It is generally believed that the human visual system is biased towards the recognition of shapes rather than textures. This assumption has led to a growing body of work aiming to align deep models' decision-making processes with the fundamental properties of human vision. The reliance on shape features is primarily expected to improve the robustness of these models under covariate shift. In this paper, we revisit the significance of shape-biases for the classification of skin lesion images. Our analysis shows that different skin lesion datasets exhibit varying biases towards individual image features. Interestingly, despite deep feature extractors being inclined towards learning entangled features for skin lesion classification, individual features can still be decoded from this entangled representation. This indicates that these features are still represented in the learnt embedding spaces of the models, but not used for classification. In addition, the spectral analysis of different datasets shows that in contrast to common visual recognition, dermoscopic skin lesion classification, by nature, is reliant on complex feature combinations beyond shape-bias. As a natural consequence, shifting away from the prevalent desire of shape-biasing models can even improve skin lesion classifiers in some cases.