Gradient-Map-Guided Adaptive Domain Generalization for Cross Modality MRI Segmentation
This addresses the problem of local domain shifts and high data requirements in cross-modal MRI segmentation for medical diagnosis applications, representing a domain-specific incremental improvement.
The paper tackles cross-modal MRI segmentation by proposing an adaptive domain generalization framework that integrates gradient-map-based cross-domain representation and test-time adaptation, achieving consistent outperformance across six cross-modal segmentation tasks with stable performance even with limited training data.
Cross-modal MRI segmentation is of great value for computer-aided medical diagnosis, enabling flexible data acquisition and model generalization. However, most existing methods have difficulty in handling local variations in domain shift and typically require a significant amount of data for training, which hinders their usage in practice. To address these problems, we propose a novel adaptive domain generalization framework, which integrates a learning-free cross-domain representation based on image gradient maps and a class prior-informed test-time adaptation strategy for mitigating local domain shift. We validate our approach on two multi-modal MRI datasets with six cross-modal segmentation tasks. Across all the task settings, our method consistently outperforms competing approaches and shows a stable performance even with limited training data.