MDSkin-Net：由模式分析先验和空间对齐正则化驱动的多任务皮肤病变分析
MDSkin-Net: Multi-Task Skin Lesion Analysis Driven by Pattern Analysis Priors and Spatial Alignment Regularization
浏览论文内容
中文总结 AI 辅助
MDSkin-Net通过混合CNN-Transformer架构融合模式分析先验与空间对齐正则化，实现多任务皮肤病变分割与分类，在零样本评估中取得高DSC和AUC，展现跨数据集稳健泛化。
中文摘要 AI 辅助
可靠的皮肤病变分割和分类是皮肤镜计算机辅助诊断的核心。现有的多任务框架在架构上将这两个任务耦合在一起，但缺乏临床知识，而知识注入方法依赖于宏观的ABCD规则，该规则并非为皮肤镜检查而设计。皮肤镜诊断基于模式分析，这是一个围绕皮肤镜特征构建的微观框架。我们提出了MDSkin-Net，它将线索级模式分析先验融入混合CNN-Transformer架构中。其核心是一个模式分析引导的注意力模块（PAGAM），包含由不同皮肤镜线索驱动的三个先验：改进的高效通道注意力（iECA）、多尺度空间注意力（MSSA）和偏置不对称注意力（BAA）。我们进一步引入了一种多尺度空间对齐正则化（MSAR），它使用分割地面真值掩码作为分层软监督，将分类头限制在病变局部证据上，并通过共享的空间先验将两个任务路径耦合在一起。仅在ISIC 2017训练集上训练，无需外部皮肤镜数据，MDSkin-Net集成在零样本评估下稳健迁移，在PH2上达到92.38%的Dice相似系数（DSC）和97.84%的黑色素瘤AUC，在ISIC 2018任务1测试集上达到88.92%的DSC。在域内ISIC 2017基准上，集成在两个分类任务（黑色素瘤和脂溢性角化病与其余类别）中达到91.60%的平均曲线下面积（AUC），分割的DSC为84.72%。分类与基线保持竞争力；域内分割落后于单任务专家，但所提出的先验和对齐正则化产生了在不同规模队列中一致泛化的表示。
英文摘要
Reliable skin lesion segmentation and classification are central to dermoscopic computer-aided diagnosis. Existing multi-task frameworks couple the two tasks architecturally without clinical knowledge, while knowledge-injecting approaches rely on the macroscopic ABCD rule, which was not designed for dermoscopy. Dermoscopic diagnosis is grounded in Pattern Analysis, a microscopic framework structured around dermoscopic features. We propose MDSkin-Net, which incorporates cue-level Pattern Analysis priors into a hybrid CNN-Transformer architecture. At its core is a Pattern Analysis-Guided Attention Module (PAGAM) comprising three priors motivated by distinct dermoscopic cues: an improved Efficient Channel Attention (iECA), a Multi-Scale Spatial Attention (MSSA), and a Biased Asymmetry Attention (BAA). We further introduce a multi-scale spatial alignment regularization (MSAR) that uses the segmentation ground-truth mask as hierarchical soft supervision, confining the classification head to lesion-localized evidence and coupling both task pathways through a shared spatial prior. Trained exclusively on the ISIC 2017 training split without external dermoscopy data, the MDSkin-Net ensemble transfers robustly under zero-shot evaluation, reaching a Dice Similarity Coefficient (DSC) of 92.38% and a melanoma AUC of 97.84%on PH2, and a DSC of 88.92% on the ISIC 2018 Task 1 test set. On the in-domain ISIC 2017 benchmark, the ensemble attains a mean Area Under the Curve (AUC) of 91.60% across the two classification tasks (melanoma and seborrheic keratosis vs. rest), and a DSC of 84.72% for segmentation. Classification remains competitive with baselines; in-domain segmentation trails single-task specialists, yet the proposed priors and alignment regularization yield representations that generalize consistently across cohorts of different scales.
发表机构
- University of Minnesota（明尼苏达大学）
机构由 AI 辅助整理，请以论文原文为准。
