Label-Efficient Cross-Modality Generalization for Liver Segmentation in Multi-Phase MRI
This work addresses the challenge of liver fibrosis assessment in clinical imaging by enabling cross-modality generalization under real-world conditions with limited annotations, though it is incremental as it builds on existing methods like foundation models and co-training.
The paper tackled the problem of accurate liver segmentation in multi-phase MRI with scarce and unevenly distributed labeled data by proposing a label-efficient approach that integrates a foundation-scale 3D segmentation backbone, co-training with cross pseudo supervision, and standardized preprocessing, achieving robust segmentation performance across MRI phases and vendors without requiring spatial registration.
Accurate liver segmentation in multi-phase MRI is vital for liver fibrosis assessment, yet labeled data is often scarce and unevenly distributed across imaging modalities and vendor systems. We propose a label-efficient segmentation approach that promotes cross-modality generalization under real-world conditions, where GED4 hepatobiliary-phase annotations are limited, non-contrast sequences (T1WI, T2WI, DWI) are unlabeled, and spatial misalignment and missing phases are common. Our method integrates a foundation-scale 3D segmentation backbone adapted via fine-tuning, co-training with cross pseudo supervision to leverage unlabeled volumes, and a standardized preprocessing pipeline. Without requiring spatial registration, the model learns to generalize across MRI phases and vendors, demonstrating robust segmentation performance in both labeled and unlabeled domains. Our results exhibit the effectiveness of our proposed label-efficient baseline for liver segmentation in multi-phase, multi-vendor MRI and highlight the potential of combining foundation model adaptation with co-training for real-world clinical imaging tasks.