融合多模型与不确定性检测，提升皮肤病变分类的准确与可信度。
Uncertainty-Aware and Explainable Ensemble Deep Learning Framework for Multi-Class Skin Lesion Classification
- 用视觉变换器与卷积网络集成，结合蒙特卡洛丢弃估计预测不确定性。
- 在HAM10000数据集上达96%准确率，99%ROC-AUC，三类指标均超94%。
- 通过Grad-CAM++提供可视化解释，适合临床辅助诊断场景。
皮肤癌的皮肤镜图像诊断仍面临类内差异大、类间相似、类别不平衡及深度学习模型可解释性差等挑战。本文提出一种不确定性和可解释性兼备的深度集成学习框架，用于多类皮肤病变分类。该框架融合视觉变换器（MaxViT-Tiny）与基于CNN的模型（ConvNeXt-Tiny和EfficientNetV2-B0），采用深度集成学习策略。通过蒙特卡洛丢弃（MC Dropout）估计预测不确定性，识别不可靠预测；利用梯度加权类激活映射++（Grad-CAM++）技术，生成可视化解释，揭示影响决策的病灶区域。在HAM10000数据集上评估，经不确定性过滤（熵 < 1.0，置信度 ≥ 0.7）后，准确率达到96%，ROC-AUC为99%，宏平均精确率、召回率和F1分数分别为94%、95%和95%，加权平均三项指标均为96%。结果表明，该框架实现高精度、可解释且具备不确定性感知的皮肤病变分类，支持可信的计算机辅助诊断。
原文摘要 · Abstract (English)
Skin cancer diagnosis from dermoscopic images remains challenging due to high intra-class variability, inter-class similarity, class imbalance, and the limited interpretability of deep learning models. This paper proposes an uncertainty-aware and explainable deep learning framework for multi-class skin lesion classification. The framework combines a vision transformer model (MaxViT-Tiny) with CNN-based models (ConvNeXt-Tiny and EfficientNetV2-B0) through deep ensemble learning. Monte Carlo (MC) Dropout estimates predictive uncertainty and identifies unreliable predictions, while Grad-CAM++, an explainable AI (XAI) technique, provides visual explanations by highlighting lesion regions that influence model decisions. Evaluated on the HAM10000 dataset, the framework achieves 96% accuracy and 99% ROC-AUC under uncertainty-aware filtering (entropy < 1.0, confidence >= 0.7), with macro-average precision, recall, and F1-score of 94%, 95%, and 95%, respectively, and 96% weighted-average scores across all three metrics. The results demonstrate accurate, interpretable, and uncertainty-aware skin lesion classification for trustworthy computer-aided diagnosis.
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。