Enhancing Data Diversity for Self-training Based Unsupervised Cross-modality Vestibular Schwannoma and Cochlea Segmentation

TL;DR

This paper introduces an unsupervised domain adaptation approach for segmenting vestibular schwannoma and cochlea in MRI images, utilizing data augmentation and self-training to improve segmentation accuracy across different sites and modalities.

Contribution

It proposes a novel cross-site, cross-modality unpaired image translation and rule-based augmentation to enhance unsupervised segmentation performance.

Findings

Achieved mean Dice scores of 0.8178 for VS and 0.8433 for cochlea.

Demonstrated competitive results on the CrossMoDA 2022 leaderboard.

Enhanced data diversity improves unsupervised segmentation accuracy.

Abstract

Automatic segmentation of vestibular schwannoma (VS) and cochlea from magnetic resonance imaging can facilitate VS treatment planning. Unsupervised segmentation methods have shown promising results without requiring the time-consuming and laborious manual labeling process. In this paper, we present an approach for VS and cochlea segmentation in an unsupervised domain adaptation setting. Specifically, we first develop a cross-site cross-modality unpaired image translation strategy to enrich the diversity of the synthesized data. Then, we devise a rule-based offline augmentation technique to further minimize the domain gap. Lastly, we adopt a self-configuring segmentation framework empowered by self-training to obtain the final results. On the CrossMoDA 2022 validation leaderboard, our method has achieved competitive VS and cochlea segmentation performance with mean Dice scores of 0.8178 $\pm$ 0.0803 and 0.8433 $\pm$ 0.0293, respectively.