Search papers, labs, and topics across Lattice.
This paper introduces Privileged Lesion-Context Relational Distillation (PLCRD), a novel teacher-student framework that leverages lesion segmentation masks solely during training to enable mask-free inference for skin lesion classification. By employing a privileged teacher that analyzes both the original dermoscopic image and its corresponding mask, the method effectively learns lesion-specific and contextual representations, which are then transferred to an image-only student through various knowledge-transfer mechanisms. The approach was validated on the HAM10000 and ISIC 2018 datasets, achieving competitive classification metrics, thereby demonstrating the potential of transforming privileged annotations into practical, interpretable knowledge for clinical applications.
Transforming lesion segmentation masks into actionable insights allows for accurate skin lesion classification without the need for masks during inference.
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.
