ConCL: Concept Contrastive Learning for Dense Prediction Pre-training in Pathology Images

TL;DR

This paper introduces ConCL, a novel self-supervised learning framework for dense prediction in pathology images, demonstrating superior performance over existing methods and providing a simple, dependency-free concept generation approach.

Contribution

The paper benchmarks SSL methods for pathology dense prediction and proposes ConCL, a new contrastive learning framework with a simple, dependency-free concept generation method.

Findings

ConCL outperforms previous SSL methods in pathology dense prediction tasks.

The proposed concept generation method does not rely on external segmentation or saliency models.

Extensive experiments validate the effectiveness of ConCL across different settings.

Abstract

Detectingandsegmentingobjectswithinwholeslideimagesis essential in computational pathology workflow. Self-supervised learning (SSL) is appealing to such annotation-heavy tasks. Despite the extensive benchmarks in natural images for dense tasks, such studies are, unfortunately, absent in current works for pathology. Our paper intends to narrow this gap. We first benchmark representative SSL methods for dense prediction tasks in pathology images. Then, we propose concept contrastive learning (ConCL), an SSL framework for dense pre-training. We explore how ConCL performs with concepts provided by different sources and end up with proposing a simple dependency-free concept generating method that does not rely on external segmentation algorithms or saliency detection models. Extensive experiments demonstrate the superiority of ConCL over previous state-of-the-art SSL methods across different settings. Along our exploration, we distll several important and intriguing components contributing to the success of dense pre-training for pathology images. We hope this work could provide useful data points and encourage the community to conduct ConCL pre-training for problems of interest. Code is available.