Multi-Scale Cross Contrastive Learning for Semi-Supervised Medical Image Segmentation

TL;DR

This paper introduces a novel multi-scale cross supervised contrastive learning framework that enhances semi-supervised medical image segmentation by capturing high-level semantic relations and addressing class imbalance, outperforming existing methods.

Contribution

The paper proposes a new MCSC framework that jointly trains CNN and Transformer models with multi-scale contrastive regularization for improved semi-supervised segmentation.

Findings

Outperforms state-of-the-art semi-supervised methods by over 3% in Dice score.

Effectively captures intra- and inter-slice relationships across datasets.

Reduces the performance gap between semi-supervised and fully supervised methods.

Abstract

Semi-supervised learning has demonstrated great potential in medical image segmentation by utilizing knowledge from unlabeled data. However, most existing approaches do not explicitly capture high-level semantic relations between distant regions, which limits their performance. In this paper, we focus on representation learning for semi-supervised learning, by developing a novel Multi-Scale Cross Supervised Contrastive Learning (MCSC) framework, to segment structures in medical images. We jointly train CNN and Transformer models, regularising their features to be semantically consistent across different scales. Our approach contrasts multi-scale features based on ground-truth and cross-predicted labels, in order to extract robust feature representations that reflect intra- and inter-slice relationships across the whole dataset. To tackle class imbalance, we take into account the prevalence of each class to guide contrastive learning and ensure that features adequately capture infrequent classes. Extensive experiments on two multi-structure medical segmentation datasets demonstrate the effectiveness of MCSC. It not only outperforms state-of-the-art semi-supervised methods by more than 3.0% in Dice, but also greatly reduces the performance gap with fully supervised methods.