U-Mamba2-SSL for Semi-Supervised Tooth and Pulp Segmentation in CBCT
This addresses a domain-specific problem for dental clinicians by improving segmentation accuracy with limited labeled data, though it is incremental as it builds on existing models.
The paper tackles the problem of automating tooth and pulp segmentation in CBCT scans, which is time-consuming and expertise-intensive, by proposing U-Mamba2-SSL, a semi-supervised learning framework that achieved an average score of 0.789 and a DSC of 0.917, winning first place in a 2025 challenge.
Accurate segmentation of teeth and pulp in Cone-Beam Computed Tomography (CBCT) is vital for clinical applications like treatment planning and diagnosis. However, this process requires extensive expertise and is exceptionally time-consuming, highlighting the critical need for automated algorithms that can effectively utilize unlabeled data. In this paper, we propose U-Mamba2-SSL, a novel semi-supervised learning framework that builds on the U-Mamba2 model and employs a multi-stage training strategy. The framework first pre-trains U-Mamba2 in a self-supervised manner using a disruptive autoencoder. It then leverages unlabeled data through consistency regularization, where we introduce input and feature perturbations to ensure stable model outputs. Finally, a pseudo-labeling strategy is implemented with a reduced loss weighting to minimize the impact of potential errors. U-Mamba2-SSL achieved an average score of 0.789 and a DSC of 0.917 on the hidden test set, achieving first place in Task 1 of the STSR 2025 challenge. The code is available at https://github.com/zhiqin1998/UMamba2.