AI summaryⓘ
The authors address the challenge of accurately diagnosing skin diseases using AI, especially when there is limited data and when similar-looking skin conditions cause confusion. They developed a new model called IViT that uses a special mathematical method (Quadratic Programming) to pick out important features that doctors usually look at. This helps the model explain its decisions better without losing accuracy. Tests on six skin disease datasets showed that their model maintained nearly the same accuracy as standard methods but was more interpretable and less redundant. The authors suggest this approach could support doctors in diagnosing skin diseases more reliably, even with limited training data.
Vision Transformer (ViT)Skin disease diagnosisFew-shot learningQuadratic ProgrammingFeature selectionInterpretabilityTransfer learningMulti-objective lossActivation distributionDeep learning
Authors
Haibiao Li, Di Lin, Xue Jiang, Weiwei Wu, Yanxi Li, Yugang Chi
Abstract
The clinical diagnosis of skin diseases is susceptible to interference from inter-class similarity of skin lesions, and over-reliance on clinicians'experience easily leads to subjective bias. Although existing deep learning aided diagnosis methods achieve competitive accuracy, they suffer from the black-box opacity of Vision Transformer (ViT) and poor adaptability to medical few-shot scenarios. Moreover, mainstream explainable algorithms generally face the bottleneck of significant accuracy degradation when improving interpretability. This paper proposes an interpretable ViT (IViT) constrained by Quadratic Programming (QP). The introduced pre-trained transfer learning adapts to few-shot feature extraction. A discrete QP feature selection framework is constructed to screen generic and discriminative features consistent with clinical diagnostic logic. A multi-objective loss function is designed to reduce feature redundancy and optimize activation distribution while preserving classification performance. Experimental results on six standard skin disease datasets show that IViT achieves an accuracy of 93.80%, only 0.21% lower than the baseline, with feature redundancy reduced by 29.5%. Its core activation regions are consistent with clinically concerned lesion areas. The proposed model balances accuracy and interpretability, providing a reliable solution for the clinical deployment of few-shot intelligent skin disease diagnosis.