发表机构
Victoria University; Bangaldesh University of Business and Technology; Daffodil Institute of Information Technology; American International University-Bangladesh (AIUB); Multimedia University（维多利亚大学; 孟加拉国商业与技术大学; 达芙妮信息技术学院; 孟加拉国美国国际大学; 多媒体大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对无掩码皮肤病变分类问题，提出特权病变-上下文关系蒸馏框架。训练时利用病变掩码，通过多种机制训练仅图像学生。该框架能将特权病变注释转化为知识，在相关数据集上取得较好分类结果，提供实用可解释的分类方法。
AI 中文摘要
准确的皮肤病变分类可受益于病变分割掩码，但推理时需要掩码或辅助分割模型会降低临床实用性并增加计算复杂性。本文介绍了特权病变-上下文关系蒸馏（PLCRD），这是一种师生框架，在训练期间仅利用病变掩码，同时保留仅图像推理。特权教师联合分析原始皮肤镜图像及其掩码引导的病变区域，以学习病变特定和上下文诊断表示。然后通过互补知识转移机制训练仅图像的学生，该机制传达教师的诊断分布、病变聚焦注意力、病变间关系几何和病变-上下文结构。PLCRD将深度表示分解为病变和上下文嵌入，并通过病变间相似性对齐、病变-上下文亲和力匹配、分离正则化和类感知关系学习来转移它们的关系组织。该框架在HAM10000上使用病变不相交数据分区进行评估，并在ISIC 2018上进行外部验证而无需重新训练。PLCRD在HAM10000上实现了病变级宏观F1为0.773±0.018、平衡准确率为0.764±0.023和宏观AUROC为0.976±0.002，在ISIC 2018上宏观F1为0.732±0.008。结果表明，特权病变注释可以转化为可转移的关系知识，产生一种实用且可解释的无掩码皮肤病变分类方法。
英文摘要
Accurate skin lesion classification can benefit from lesion segmentation masks, but requiring masks or an auxiliary segmentation model during inference reduces clinical practicality and increases computational complexity. This work introduces Privileged Lesion-Context Relational Distillation (PLCRD), a teacher-student framework that exploits lesion masks exclusively during training while preserving image-only inference. The privileged teacher jointly analyzes the original dermoscopic image and its mask-guided lesion region to learn lesion-specific and contextual diagnostic representations. An image-only student is then trained through complementary knowledge-transfer mechanisms that convey the teacher's diagnostic distribution, lesion-focused attention, inter-lesion relational geometry, and lesion-context structure. PLCRD decomposes deep representations into lesion and contextual embeddings and transfers their relational organization through inter-lesion similarity alignment, lesion-context affinity matching, separation regularization, and class-aware relational learning. This formulation avoids direct feature matching between heterogeneous teacher and student architectures and enables the student to internalize mask-informed diagnostic structure without accessing masks at deployment. The framework was evaluated on HAM10000 using lesion-disjoint data partitioning and externally validated on ISIC 2018 without retraining. PLCRD achieved a lesion-level macro-F1 of 0.773 +/- 0.018, balanced accuracy of 0.764 +/- 0.023, and macro-AUROC of 0.976 +/- 0.002 on HAM10000, together with a macro-F1 of 0.732 +/- 0.008 on ISIC 2018. The results indicate that privileged lesion annotations can be transformed into transferable relational knowledge, yielding a practical and interpretable approach to mask-free skin lesion classification.
CommentsManuscript submitted for consideration. The paper presents a mask-free skin lesion classification framework evaluated on HAM10000 and externally validated on ISIC 2018
