发表机构
University of Basel; Lucerne University of Applied Sciences and Arts; University Hospital of Basel（巴塞尔大学; 卢塞恩应用科学与艺术大学; 巴塞尔大学医院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出SkinLex数据集，整合48个临床视觉概念于四个皮肤病数据集，通过九分类实验发现概念冗余，表明粗略诊断需少量多样概念，可提升模型效率和可解释性。
AI 中文摘要
AI系统在数字皮肤病学中的临床整合在很大程度上依赖于人类的信任。临床上可解释的视觉概念可以作为中间表示，增强信任和可靠性。然而，该领域的研究目前受到分散、异构数据集注释的限制。在这项工作中，我们引入了SkinLex，这是一个协调的数据集，包含四个公共数据集（SkinCon、DermaCon-IN、MM-Skin和PASSION）中的48个临床形态学属性，共计20,411条记录。对皮肤状况进行有监督的九分类表明，将特征限制在特定的视觉组（如仅形状或仅颜色）会降低诊断准确性。自助式向后消除法显示，在所检查的数据集上，48个视觉概念集对于算法九分类诊断存在一定程度的冗余。这表明，在所选数据集上进行粗略诊断需要相对较小但多样化的临床概念组合，并激励进一步研究以改进概念分类法。结果可以通过减少基于概念模型的输入、提高注释和建模效率以及进一步增强可解释性，转化为临床益处。代码和提示模板可在以下https URL获取。
英文摘要
The clinical integration of AI systems in digital dermatology relies heavily on human trust. Clinically interpretable visual concepts can act as intermediate representations enhancing trust and reliability. However, research in this domain is currently limited by scattered, heterogeneous dataset annotations. In this work, we introduce SkinLex, a harmonized dataset of 48 clinical morphological attributes across four public datasets (SkinCon, DermaCon-IN, MM-Skin, and PASSION) for a total of 20,411 records. Supervised nine-partition classification of skin conditions shows that limiting features to specific visual groups, like shapes or colors alone, reduces diagnostic accuracy. Bootstrapped backward elimination reveals that the set of 48 visual concepts has some degree of redundancy for algorithmic nine-partition diagnosis on the examined dataset. This demonstrates that coarse diagnosis on the selected dataset requires a relatively small but varied combination of clinical concepts, and motivates further research to improve concept taxonomy. Results can be translated into clinical benefits by reducing inputs for concept-based models, improving efficiency for annotation and modeling, and further enhancing interpretability. Code and prompt templates are available at https://github.com/Digital-Dermatology/SkinLex.
CommentsAccepted at the MICCAI ISIC Workshop 2026. 11 pages, 3 figures, 3 tables. Code and dataset: https://github.com/Digital-Dermatology/SkinLex
