[论文解读] SemiTooth: a Generalizable Semi-supervised Framework for Multi-Source Tooth Segmentation

Muyi Sun, Yifan Gao+5 more|arXiv (Cornell University)|Mar 12, 2026

Dental Radiography and Imaging被引用 0

一句话总结

SemiTooth 提出一个带多教师监督的多分支半监督框架，并使用 Stricter Weighted-Confidence 约束以实现跨多源 CBCT 数据的通用牙齿分割，在 MS 3 Toothset 上展示了前沿性能。

ABSTRACT

With the rapid advancement of artificial intelligence, intelligent dentistry for clinical diagnosis and treatment has become increasingly promising. As the primary clinical dentistry task, tooth structure segmentation for Cone-Beam Computed Tomography (CBCT) has made significant progress in recent years. However, challenges arise from the obtainment difficulty of full-annotated data, and the acquisition variability of multi-source data across different institutions, which have caused low-quality utilization, voxel-level inconsistency, and domain-specific disparity in CBCT slices. Thus, the rational and efficient utilization of multi-source and unlabeled data represents a pivotal problem. In this paper, we propose SemiTooth, a generalizable semi-supervised framework for multi-source tooth segmentation. Specifically, we first compile MS3Toothset, Multi-Source Semi-Supervised Tooth DataSet for clinical dental CBCT, which contains data from three sources with different-level annotations. Then, we design a multi-teacher and multi-student framework, i.e., SemiTooth, which promotes semi-supervised learning for multi-source data. SemiTooth employs distinct student networks that learn from unlabeled data with different sources, supervised by its respective teachers. Furthermore, a Stricter Weighted-Confidence Constraint is introduced for multiple teachers to improve the multi-source accuracy.Extensive experiments are conducted on MS3Toothset to verify the feasibility and superiority of the SemiTooth framework, which achieves SOTA performance on the semi-supervised and multi-source tooth segmentation scenario.

研究动机与目标

• 解决跨多机构的完整标注 CBCT 牙齿数据稀缺问题。
• 构建一个多源半监督框架以提升对源的泛化能力。
• 发布 MS 3 Toothset 以促进多源牙齿分割研究。
• 通过区域层面的 Stricter Weighted-Confidence 约束提升伪标签质量。

提出的方法

• 从三个来源构建具有不同注释的 MS 3 Toothset。
• 实现一个多分支半监督框架（三个学生、两个教师），让每个学生从对应的教师学习。
• 通过教师参数的指数滑动平均（EMA）更新教师。
• 引入 Stricter Weighted-Confidence（SWC）约束，对伪标签的区域与体素级置信度进行门控和加权。
• 定义 SWC 损失并与监督损失结合用于总优化。

实验结果

研究问题

• RQ1 如何将半监督学习扩展为鲁棒地处理多源 CBCT 数据以实现牙齿分割？
• RQ2 相比单源或单教师基线，采用多教师多学生结合 SWC 的方案是否能提升跨源泛化和分割精度？
• RQ3 区域级别的伪标签门控是否能降低噪声、提高三维 CBCT 牙齿分割的边界质量？
• RQ4 源之间的分布差异对半监督牙齿分割性能有何影响？

主要发现

• SemiTooth 在半监督多源牙齿分割（MS 3 Toothset）上对比现有方法取得更优性能。
• 三学生、两教师的设置并通过 EMA 更新，相较单教师或非教师的多学生基线，在形态和边界质量上更佳。
• 更严格的加权置信约束通过区域门控和体素级加权提升伪标签的可靠性，减少边界噪声。
• 消融研究显示加入 SWC 及完整的 SemiTooth 配置带来逐步增益。
• t-SNE 可视化表明 SemiTooth 后跨源特征聚类更强，反映对源之间的一致化泛化有所提升。