VCDP: Variation-Conditioned Distributional Proxy Learning for Semi-Supervised Medical Image Segmentation
For researchers in medical image segmentation, VCDP offers a plug-and-play regularization module that enhances existing semi-supervised methods without inference cost, addressing the bottleneck of handling intra-class variation in complex anatomical structures.
VCDP improves semi-supervised 3D medical image segmentation by learning class-specific Gaussian distributions and variation prototypes to better organize feature space for small and ambiguous organs, achieving consistent gains over multiple baselines on multi-organ benchmarks.
Semi-supervised 3D medical image segmentation reduces the need for dense voxel-level annotations by exploiting unlabeled volumes. Although existing methods such as consistency regularization, pseudo-labeling, and co-training improve prediction-level robustness, they often provide insufficient feature-space organization for anatomically complex structures, especially small organs and ambiguous boundary regions with large intra-class variations. To address this issue, we propose Variation-Conditioned Distributional Proxy Learning (VCDP), a plug-and-play training-only regularization module for semi-supervised 3D medical image segmentation. VCDP represents each class with a learnable Gaussian distribution for shared class semantics and multiple variation prototypes for fine-grained intra-class patterns. A unified variation-conditioned compatibility score is further formulated to fuse distributional similarity and soft variation aggregation, guiding voxel embeddings to align with both global organ identity and local anatomical variations. VCDP is attached to decoder features during training and removed during inference, introducing no additional inference cost. Experiments on multi-organ segmentation benchmarks show that VCDP improves most evaluated baselines, particularly for small, ambiguous, and highly variable organs. Our anonymous code is released at https://anonymous.4open.science/r/VCDP_code-41ED.