Towards Novel Class Discovery: A Study in Novel Skin Lesions Clustering

TL;DR

This paper introduces a novel framework for automatic discovery of new skin lesion categories from dermoscopic images, leveraging contrastive learning and pseudo-label refinement to improve clustering of unknown classes.

Contribution

It proposes a new deep learning approach combining contrastive learning and uncertainty-aware pseudo-labeling for novel class discovery in dermatology images.

Findings

Effective discovery of new skin lesion categories from dermoscopy data.

Improved clustering performance through neighborhood-based pseudo-label refinement.

Validated on ISIC 2019 dataset with extensive ablation studies.

Abstract

Existing deep learning models have achieved promising performance in recognizing skin diseases from dermoscopic images. However, these models can only recognize samples from predefined categories, when they are deployed in the clinic, data from new unknown categories are constantly emerging. Therefore, it is crucial to automatically discover and identify new semantic categories from new data. In this paper, we propose a new novel class discovery framework for automatically discovering new semantic classes from dermoscopy image datasets based on the knowledge of known classes. Specifically, we first use contrastive learning to learn a robust and unbiased feature representation based on all data from known and unknown categories. We then propose an uncertainty-aware multi-view cross pseudo-supervision strategy, which is trained jointly on all categories of data using pseudo labels generated by a self-labeling strategy. Finally, we further refine the pseudo label by aggregating neighborhood information through local sample similarity to improve the clustering performance of the model for unknown categories. We conducted extensive experiments on the dermatology dataset ISIC 2019, and the experimental results show that our approach can effectively leverage knowledge from known categories to discover new semantic categories. We also further validated the effectiveness of the different modules through extensive ablation experiments. Our code will be released soon.