Skin Lesion Classification using Class Activation Map

TL;DR

This paper introduces a two-stage CNN framework for skin lesion classification that improves accuracy by focusing on the most relevant image regions, achieving higher AUC scores than using original images alone.

Contribution

A novel two-stage CNN approach that enhances skin lesion classification accuracy by region-focused retraining, outperforming baseline methods.

Findings

Achieved a mean AUC of 0.857 on ISIC-2017 validation set.

Improved AUC by 0.04 over using original images.

Demonstrated effectiveness of region cropping in skin lesion analysis.

Abstract

We proposed a two stage framework with only one network to analyze skin lesion images, we firstly trained a convolutional network to classify these images, and cropped the import regions which the network has the maximum activation value. In the second stage, we retrained this CNN with the image regions extracted from stage one and output the final probabilities. The two stage framework achieved a mean AUC of 0.857 in ISIC-2017 skin lesion validation set and is 0.04 higher than that of the original inputs, 0.821.