Related Experiment Video
Updated: Jun 19, 2026
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
Published on: August 5, 2021
Towards clinically applicable automated mandibular canal segmentation on CBCT.
Fang-Duan Ni1, Zi-Neng Xu2, Mu-Qing Liu1
1Department of Oral & Maxillofacial Radiology, Peking University School & Hospital of Stomatology, Beijing 100081, China; National Center for Stomatology & National Clinical Research Center for Oral Diseases, Beijing 100081, China; National Engineering Research Center of Oral Biomaterials and Digital Medical Devices, Beijing 100081, China; Beijing Key Laboratory of Digital Stomatology, Beijing 100081, China.
A new AI system accurately segments the mandibular canal in cone beam computed tomography (CBCT) scans. This automated segmentation aids dental procedures, improving efficiency and precision in clinical workflows.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oral and Maxillofacial Surgery
Background:
- Accurate segmentation of the mandibular canal is crucial for dental procedures like implant placement and third molar extraction.
- Cone Beam Computed Tomography (CBCT) is widely used, but manual segmentation can be time-consuming and prone to variability.
- Developing automated segmentation tools can enhance the precision and efficiency of dental diagnostics and treatment planning.
Purpose of the Study:
- To develop and validate a deep learning-based system for precise, robust, and fully automated segmentation of the mandibular canal on CBCT images.
- To evaluate the system's performance on both internal and external datasets, ensuring generalizability.
- To assess the potential clinical utility of the AI system in dental practice.
Main Methods:
- A three-step deep learning strategy was employed, utilizing 2D and 3D U-Net architectures.
- The system involved region of interest extraction, global mandibular canal segmentation, and refinement.
- The model was trained on 536 CBCT scans and validated on an additional 89 scans from multiple centers.
Main Results:
- The AI system achieved high accuracy in mandibular canal segmentation on internal datasets (Dice Similarity Coefficient [DSC] 0.952, Intersection over Union [IoU] 0.912).
- Excellent performance was also demonstrated on the external validation set (DSC 0.960, IoU 0.924).
- Quantitative metrics including Average Symmetric Surface Distance (ASSD) and Hausdorff Distance 95% (HD95) confirmed the system's precision.
Conclusions:
- The developed AI system provides accurate and automated segmentation of the mandibular canal from CBCT images.
- The system's robust performance across different datasets indicates its potential for clinical application.
- This technology can significantly facilitate clinical workflows and contribute to more efficient and precise dental automation systems.
