MS-MT: Multi-Scale Mean Teacher with Contrastive Unpaired Translation for Cross-Modality Vestibular Schwannoma and Cochlea Segmentation

TL;DR

This paper introduces a multi-scale mean teacher framework with contrastive unpaired translation for improved cross-modality segmentation of brain structures, addressing domain shift in medical imaging without requiring labels in the target domain.

Contribution

It proposes a novel multi-scale self-ensembling UDA framework combining contrastive unpaired translation and self-training to enhance segmentation across different imaging modalities.

Findings

Achieved mean Dice scores of 83.8% for VS and 81.4% for Cochlea.

Reduced domain gap through contrastive unpaired image translation.

Demonstrated effectiveness in the crossMoDA 2022 challenge.

Abstract

Domain shift has been a long-standing issue for medical image segmentation. Recently, unsupervised domain adaptation (UDA) methods have achieved promising cross-modality segmentation performance by distilling knowledge from a label-rich source domain to a target domain without labels. In this work, we propose a multi-scale self-ensembling based UDA framework for automatic segmentation of two key brain structures i.e., Vestibular Schwannoma (VS) and Cochlea on high-resolution T2 images. First, a segmentation-enhanced contrastive unpaired image translation module is designed for image-level domain adaptation from source T1 to target T2. Next, multi-scale deep supervision and consistency regularization are introduced to a mean teacher network for self-ensemble learning to further close the domain gap. Furthermore, self-training and intensity augmentation techniques are utilized to mitigate label scarcity and boost cross-modality segmentation performance. Our method demonstrates promising segmentation performance with a mean Dice score of 83.8% and 81.4% and an average asymmetric surface distance (ASSD) of 0.55 mm and 0.26 mm for the VS and Cochlea, respectively in the validation phase of the crossMoDA 2022 challenge.