Lesion-Aware Contrastive Representation Learning for Histopathology Whole Slide Images Analysis

TL;DR

This paper introduces Lesion-Aware Contrastive Learning (LACL), a novel self-supervised framework for histopathology WSI analysis that improves representation learning by using a lesion queue and refinement strategy to address class collision issues.

Contribution

LACL is the first to incorporate a lesion-aware memory bank and queue refinement for contrastive learning in histopathology WSIs, enhancing class discrimination.

Findings

LACL outperforms existing methods on multiple datasets.

LACL achieves superior WSI classification accuracy.

The lesion queue improves negative pair selection in contrastive learning.

Abstract

Local representation learning has been a key challenge to promote the performance of the histopathological whole slide images analysis. The previous representation learning methods followed the supervised learning paradigm. However, manual annotation for large-scale WSIs is time-consuming and labor-intensive. Hence, the self-supervised contrastive learning has recently attracted intensive attention. The present contrastive learning methods treat each sample as a single class, which suffers from class collision problems, especially in the domain of histopathology image analysis. In this paper, we proposed a novel contrastive representation learning framework named Lesion-Aware Contrastive Learning (LACL) for histopathology whole slide image analysis. We built a lesion queue based on the memory bank structure to store the representations of different classes of WSIs, which allowed the contrastive model to selectively define the negative pairs during the training. Moreover, We designed a queue refinement strategy to purify the representations stored in the lesion queue. The experimental results demonstrate that LACL achieves the best performance in histopathology image representation learning on different datasets, and outperforms state-of-the-art methods under different WSI classification benchmarks. The code is available at https://github.com/junl21/lacl.