Positional Contrastive Learning for Volumetric Medical Image Segmentation

TL;DR

This paper introduces a positional contrastive learning framework that leverages spatial information in volumetric medical images to improve segmentation performance, especially when labeled data is scarce.

Contribution

The novel PCL framework effectively generates contrastive pairs using position information, reducing false negatives and enhancing segmentation accuracy in medical imaging.

Findings

Significant improvement in segmentation accuracy on CT and MRI datasets.

Effective in semi-supervised and transfer learning scenarios.

Reduces false-negative pairs in contrastive learning for medical images.

Abstract

The success of deep learning heavily depends on the availability of large labeled training sets. However, it is hard to get large labeled datasets in medical image domain because of the strict privacy concern and costly labeling efforts. Contrastive learning, an unsupervised learning technique, has been proved powerful in learning image-level representations from unlabeled data. The learned encoder can then be transferred or fine-tuned to improve the performance of downstream tasks with limited labels. A critical step in contrastive learning is the generation of contrastive data pairs, which is relatively simple for natural image classification but quite challenging for medical image segmentation due to the existence of the same tissue or organ across the dataset. As a result, when applied to medical image segmentation, most state-of-the-art contrastive learning frameworks inevitably introduce a lot of false-negative pairs and result in degraded segmentation quality. To address this issue, we propose a novel positional contrastive learning (PCL) framework to generate contrastive data pairs by leveraging the position information in volumetric medical images. Experimental results on CT and MRI datasets demonstrate that the proposed PCL method can substantially improve the segmentation performance compared to existing methods in both semi-supervised setting and transfer learning setting.