GEPAR3D: Geometry Prior-Assisted Learning for 3D Tooth Segmentation

2025-07-31

AI Summary

Accurate 3D segmentation of teeth in cone-beam computed tomography (CBCT), particularly fine anatomical structures such as apices, remains challenging for orthodontic root resorption assessment. To address this, we propose an end-to-end 3D instance segmentation framework that integrates a statistical dental arch shape model with deep watershed transformation: shape priors enforce anatomical consistency, while voxel-wise boundary distance encoding and instance-aware segmentation representations jointly generate contiguous energy basins—eliminating the need for strong adjacency constraints and enabling joint multi-class segmentation and instance detection. Evaluated under a single-center training and multi-center validation paradigm, our method achieves a mean Dice score of 95.0% across multiple external test sets—outperforming the second-best approach by 2.8 percentage points—and attains a root apex recall of 95.2%, significantly improving segmentation fidelity of root structures. The code and data are publicly available.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Web Mining and Content Analysis: Web data integration and cleaning
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

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.

Problem

• Improves 3D tooth segmentation in CBCT scans
• Enhances root apex segmentation for orthodontic assessment
• Integrates geometric prior for anatomical consistency in segmentation

Innovation

• Unifies instance detection and multi-class segmentation
• Integrates Statistical Shape Model as geometric prior
• Uses deep watershed method for 3D energy basins