Progressive Class-Wise Attention (PCA) Approach for Diagnosing Skin Lesions

TL;DR

This paper presents a novel progressive class-wise attention method for skin lesion diagnosis, significantly improving classification accuracy by effectively capturing discriminative features across multiple scales.

Contribution

The paper introduces a new class-wise attention mechanism that enhances skin lesion classification by focusing on class-specific details and integrating multi-scale features.

Findings

Achieved 97.40% accuracy on HAM10000 dataset.

Achieved 94.9% accuracy on ISIC 2019 dataset.

Outperformed over 15 state-of-the-art methods.

Abstract

Skin cancer holds the highest incidence rate among all cancers globally. The importance of early detection cannot be overstated, as late-stage cases can be lethal. Classifying skin lesions, however, presents several challenges due to the many variations they can exhibit, such as differences in colour, shape, and size, significant variation within the same class, and notable similarities between different classes. This paper introduces a novel class-wise attention technique that equally regards each class while unearthing more specific details about skin lesions. This attention mechanism is progressively used to amalgamate discriminative feature details from multiple scales. The introduced technique demonstrated impressive performance, surpassing more than 15 cutting-edge methods including the winners of HAM1000 and ISIC 2019 leaderboards. It achieved an impressive accuracy rate of 97.40% on the HAM10000 dataset and 94.9% on the ISIC 2019 dataset.