crossMoDA Challenge: Evolution of Cross-Modality Domain Adaptation Techniques for Vestibular Schwannoma and Cochlea Segmentation from 2021 to 2023

2025-06-13

AI Summary

This work addresses the domain adaptation challenge in unsupervised cross-modal medical image segmentation, specifically for vestibular schwannoma (VS) and cochlea segmentation from T1- to T2-weighted MRI. We propose a multi-stage progressive benchmark paradigm—from single- to multi-center settings, from whole-tumor annotation to Koos grading, and further to intra-/extra-tumoral subregion delineation—to systematically evaluate how data heterogeneity enhances model robustness. Our method integrates both paired and unpaired multi-institutional T1/T2 MRI data within an unsupervised domain adaptation framework, enabling generalization across scanning protocols and imaging devices. Evaluated on the 2023 international challenge, our approach achieved first place: it significantly reduced outliers on the cross-year test set and delivered stable VS segmentation performance. Although cochlear Dice scores slightly decreased due to the complexity of subregional annotations, the overall framework advances low-cost, clinically deployable VS management.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Security and Privacy: Data transparency and provenance

Abstract

The cross-Modality Domain Adaptation (crossMoDA) challenge series, initiated in 2021 in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), focuses on unsupervised cross-modality segmentation, learning from contrast-enhanced T1 (ceT1) and transferring to T2 MRI. The task is an extreme example of domain shift chosen to serve as a meaningful and illustrative benchmark. From a clinical application perspective, it aims to automate Vestibular Schwannoma (VS) and cochlea segmentation on T2 scans for more cost-effective VS management. Over time, the challenge objectives have evolved to enhance its clinical relevance. The challenge evolved from using single-institutional data and basic segmentation in 2021 to incorporating multi-institutional data and Koos grading in 2022, and by 2023, it included heterogeneous routine data and sub-segmentation of intra- and extra-meatal tumour components. In this work, we report the findings of the 2022 and 2023 editions and perform a retrospective analysis of the challenge progression over the years. The observations from the successive challenge contributions indicate that the number of outliers decreases with an expanding dataset. This is notable since the diversity of scanning protocols of the datasets concurrently increased. The winning approach of the 2023 edition reduced the number of outliers on the 2021 and 2022 testing data, demonstrating how increased data heterogeneity can enhance segmentation performance even on homogeneous data. However, the cochlea Dice score declined in 2023, likely due to the added complexity from tumour sub-annotations affecting overall segmentation performance. While progress is still needed for clinically acceptable VS segmentation, the plateauing performance suggests that a more challenging cross-modal task may better serve future benchmarking.

Problem

• Automate Vestibular Schwannoma and cochlea segmentation on T2 MRI scans
• Address domain shift in unsupervised cross-modality medical image segmentation
• Enhance clinical relevance with evolving datasets and segmentation tasks

Innovation

• Unsupervised cross-modality segmentation learning
• Multi-institutional data integration
• Heterogeneous routine data utilization