JanusNet: Hierarchical Slice-Block Shuffle and Displacement for Semi-Supervised 3D Multi-Organ Segmentation

TL;DR

JanusNet introduces a novel data augmentation framework for semi-supervised 3D medical image segmentation that preserves anatomical continuity and emphasizes hard-to-segment regions, leading to significant performance improvements.

Contribution

It proposes a dual-stage augmentation method combining slice-block shuffling and confidence-guided displacement to enhance semi-supervised 3D segmentation.

Findings

Achieves a 4% DSC improvement on Synapse dataset with 20% labeled data.

Outperforms state-of-the-art methods on Synapse and AMOS datasets.

Effective in preserving anatomical structure while focusing on challenging regions.

Abstract

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.