GEPAR3D: Geometry Prior-Assisted Learning for 3D Tooth Segmentation
This work addresses the problem of accurate root segmentation for orthodontists, enabling better assessment of root resorption, though it is incremental as it builds on existing segmentation methods with a novel integration of geometric priors.
The paper tackled the challenge of tooth segmentation in CBCT scans, particularly for fine root structures critical in orthodontics, by introducing GEPAR3D, which achieved a Dice Similarity Coefficient of 95.0% and improved recall by 9.5% over the second-best method.
Tooth segmentation in Cone-Beam Computed Tomography (CBCT) remains challenging, especially for fine structures like root apices, which is critical for assessing root resorption in orthodontics. We introduce GEPAR3D, a novel approach that unifies instance detection and multi-class segmentation into a single step tailored to improve root segmentation. Our method integrates a Statistical Shape Model of dentition as a geometric prior, capturing anatomical context and morphological consistency without enforcing restrictive adjacency constraints. We leverage a deep watershed method, modeling each tooth as a continuous 3D energy basin encoding voxel distances to boundaries. This instance-aware representation ensures accurate segmentation of narrow, complex root apices. Trained on publicly available CBCT scans from a single center, our method is evaluated on external test sets from two in-house and two public medical centers. GEPAR3D achieves the highest overall segmentation performance, averaging a Dice Similarity Coefficient (DSC) of 95.0% (+2.8% over the second-best method) and increasing recall to 95.2% (+9.5%) across all test sets. Qualitative analyses demonstrated substantial improvements in root segmentation quality, indicating significant potential for more accurate root resorption assessment and enhanced clinical decision-making in orthodontics. We provide the implementation and dataset at https://github.com/tomek1911/GEPAR3D.