MelanomaNet让皮肤病变分类模型可解释，提升临床可信度。
MelanomaNet: Explainable Deep Learning for Skin Lesion Classification
- 融合GradCAM++与临床准则提取，可视化模型关注点。
- 在ISIC 2019上达85.61%准确率，加权F1为0.8564。
- 支持不确定性量化，适合医生审阅高风险预测。
基于深度学习的自动皮肤病变分类已展现卓越准确率，但因模型“黑箱”特性，临床应用受限。本文提出MelanomaNet，一种用于多类别皮肤病变分类的可解释深度学习系统，通过四种互补的可解释性机制解决该问题。方法结合EfficientNet V2主干网络，采用GradCAM++进行注意力可视化，自动提取ABCDE临床评估准则，使用FastCAV进行概念解释，并引入蒙特卡洛丢弃法量化不确定性。在包含25,331张皮肤镜图像、涵盖9个诊断类别的ISIC 2019数据集上，模型取得85.61%准确率和0.8564的加权F1分数。模型注意力与既定皮肤病学评估标准高度一致，不确定性模块可分解预测置信度为认知与随机不确定性成分，实现对不可靠预测的自动标记以供临床复核。结果表明，高性能与全面可解释性可共存，有望推动其在临床皮肤科工作流中的信任与采纳。源码已公开于https://github.com/suxrobgm/explainable-melanoma。
原文摘要 · Abstract (English)
Automated skin lesion classification using deep learning has shown remarkable accuracy, yet clinical adoption remains limited due to the "black box" nature of these models. We present MelanomaNet, an explainable deep learning system for multi-class skin lesion classification that addresses this gap through four complementary interpretability mechanisms. Our approach combines an EfficientNet V2 backbone with GradCAM++ attention visualization, automated ABCDE clinical criterion extraction, Fast Concept Activation Vectors (FastCAV) for concept-based explanations, and Monte Carlo Dropout uncertainty quantification. We evaluate our system on the ISIC 2019 dataset containing 25,331 dermoscopic images across 9 diagnostic categories. Our model achieves 85.61% accuracy with a weighted F1 score of 0.8564, while providing clinically meaningful explanations that align model attention with established dermatological assessment criteria. The uncertainty quantification module decomposes prediction confidence into epistemic and aleatoric components, enabling automatic flagging of unreliable predictions for clinical review. Our results demonstrate that high classification performance can be achieved alongside comprehensive interpretability, potentially facilitating greater trust and adoption in clinical dermatology workflows. The source code is available at https://github.com/suxrobgm/explainable-melanoma
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。