JanusNet: Hierarchical Slice-Block Shuffle and Displacement for Semi-Supervised 3D Multi-Organ Segmentation
This work addresses the challenge of semi-supervised segmentation in medical imaging, offering a plug-and-play solution that improves performance for tasks like organ segmentation, though it is incremental as it builds on existing teacher-student schemes.
The paper tackles the problem of limited training data for 3D multi-organ segmentation by proposing JanusNet, a data augmentation framework that preserves anatomical continuity while focusing on hard-to-segment regions, resulting in a 4% DSC gain on the Synapse dataset with only 20% labeled data.
Limited by the scarcity of training samples and annotations, weakly supervised medical image segmentation often employs data augmentation to increase data diversity, while randomly mixing volumetric blocks has demonstrated strong performance. However, this approach disrupts the inherent anatomical continuity of 3D medical images along orthogonal axes, leading to severe structural inconsistencies and insufficient training in challenging regions, such as small-sized organs, etc. To better comply with and utilize human anatomical information, we propose JanusNet}, a data augmentation framework for 3D medical data that globally models anatomical continuity while locally focusing on hard-to-segment regions. Specifically, our Slice-Block Shuffle step performs aligned shuffling of same-index slice blocks across volumes along a random axis, while preserving the anatomical context on planes perpendicular to the perturbation axis. Concurrently, the Confidence-Guided Displacement step uses prediction reliability to replace blocks within each slice, amplifying signals from difficult areas. This dual-stage, axis-aligned framework is plug-and-play, requiring minimal code changes for most teacher-student schemes. Extensive experiments on the Synapse and AMOS datasets demonstrate that JanusNet significantly surpasses state-of-the-art methods, achieving, for instance, a 4% DSC gain on the Synapse dataset with only 20% labeled data.