ECL: Class-Enhancement Contrastive Learning for Long-tailed Skin Lesion Classification
This addresses the problem of imbalanced data distribution in skin disease diagnosis for medical imaging, representing an incremental improvement over existing supervised contrastive learning methods.
The paper tackles long-tailed skin lesion classification by proposing class-Enhancement Contrastive Learning (ECL) to enrich minority class information and treat classes equally, achieving superior and effective results in experiments on imbalanced skin lesion data.
Skin image datasets often suffer from imbalanced data distribution, exacerbating the difficulty of computer-aided skin disease diagnosis. Some recent works exploit supervised contrastive learning (SCL) for this long-tailed challenge. Despite achieving significant performance, these SCL-based methods focus more on head classes, yet ignoring the utilization of information in tail classes. In this paper, we propose class-Enhancement Contrastive Learning (ECL), which enriches the information of minority classes and treats different classes equally. For information enhancement, we design a hybrid-proxy model to generate class-dependent proxies and propose a cycle update strategy for parameters optimization. A balanced-hybrid-proxy loss is designed to exploit relations between samples and proxies with different classes treated equally. Taking both "imbalanced data" and "imbalanced diagnosis difficulty" into account, we further present a balanced-weighted cross-entropy loss following curriculum learning schedule. Experimental results on the classification of imbalanced skin lesion data have demonstrated the superiority and effectiveness of our method.