Enhancing Fairness in Skin Lesion Classification for Medical Diagnosis Using Prune Learning

2025-08-31

AI Summary

This study addresses diagnostic bias in skin lesion classification models arising from skin tone variation. We propose a feature-map skewness-driven channel pruning method that operates without external statistical priors or annotations of sensitive attributes. The method is universally applicable to both convolutional networks (e.g., VGG) and vision transformers, automatically identifying and removing redundant channels strongly correlated with skin tone—thereby enhancing focus on lesion regions and reducing model sensitivity to skin tone. Crucially, it preserves classification accuracy while significantly improving cross-skin-tone prediction consistency (i.e., fairness) and reducing parameter count and computational cost. Experiments on multiple benchmark datasets demonstrate a 42% reduction in fairness disparity metrics (e.g., Equal Opportunity Difference) and a 23% speedup in inference latency, underscoring its potential for clinical deployment.

Technology Category

• Computer Vision: Bias, Fairness & Privacy
• Machine Learning: Ethics, Bias, and Fairness
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Recent advances in deep learning have significantly improved the accuracy of skin lesion classification models, supporting medical diagnoses and promoting equitable healthcare. However, concerns remain about potential biases related to skin color, which can impact diagnostic outcomes. Ensuring fairness is challenging due to difficulties in classifying skin tones, high computational demands, and the complexity of objectively verifying fairness. To address these challenges, we propose a fairness algorithm for skin lesion classification that overcomes the challenges associated with achieving diagnostic fairness across varying skin tones. By calculating the skewness of the feature map in the convolution layer of the VGG (Visual Geometry Group) network and the patches and the heads of the Vision Transformer, our method reduces unnecessary channels related to skin tone, focusing instead on the lesion area. This approach lowers computational costs and mitigates bias without relying on conventional statistical methods. It potentially reduces model size while maintaining fairness, making it more practical for real-world applications.

Problem

• Addressing skin color bias in lesion classification models
• Reducing computational costs while ensuring diagnostic fairness
• Focusing on lesion areas instead of skin tone features

Innovation

• Prune learning reduces skin tone bias
• Analyzes VGG and Vision Transformer features
• Lowers computational cost while maintaining fairness