Related Experiment Video
Updated: Jul 6, 2025
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
865
Deep learning driven segmentation of maxillary impacted canine on cone beam computed tomography images
Abdullah Swaity1,2, Bahaaeldeen M Elgarba1,3, Nermin Morgan1,4
1OMFS IMPATH Research Group, Department of Imaging and Pathology, Faculty of Medicine, KU Leuven, & Department of Oral and Maxillofacial Surgery, University Hospitals Leuven, Leuven, Belgium.
Scientific Reports
|January 3, 2024
Summary
This study introduces a convolutional neural network (CNN) for automated segmentation of impacted maxillary canines in CBCT scans. The CNN model offers a fast, precise, and consistent alternative to manual methods in digital dentistry.
Area of Science:
- Digital Dentistry
- Medical Imaging
- Artificial Intelligence in Healthcare
Background:
- 3D segmentation of dentomaxillofacial structures is vital for digital dental workflows.
- Manual and semi-automated segmentation methods are time-consuming and prone to observer bias.
Purpose of the Study:
- To develop and evaluate a convolutional neural network (CNN)-based online platform for automated segmentation of impacted maxillary canines.
- To compare the performance and efficiency of the automated CNN approach against semi-automated methods.
Main Methods:
- A CNN model was trained using 50 CBCT images with maxillary canine impactions.
- The trained model was tested on a separate set of 50 CBCT images.
- Segmentation performance was assessed using voxel- and surface-based comparisons and time efficiency was calculated.
Main Results:
- The automated CNN tool achieved high segmentation performance with a Dice Similarity Coefficient of 0.99 ± 0.02.
- The CNN-based segmentation was 24 times faster than the semi-automated approach.
- The model demonstrated fast, consistent, and precise segmentation of impacted canines.
Conclusions:
- The proposed CNN model provides an efficient and accurate solution for automated segmentation of impacted maxillary canines.
- This automated approach can significantly improve the speed and reliability of digital dental workflows.
- The online cloud platform offers a practical tool for clinical application.
