Related Experiment Video
Updated: May 24, 2025
10:23
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
2.6K
Development of Artificial Intelligence-Supported Automatic Three-Dimensional Surface Cephalometry
Chihiro Tanikawa1, Hiroyuki Nakamura2, Takaaki Mimura3
1Department of Orthodontics and Dentofacial Orthopedics, Graduate School of Dentistry, Osaka University, Suita, Osaka, Japan.
Orthodontics & Craniofacial Research
|March 4, 2025
Summary
An artificial intelligence (AI) system automates 3D surface cephalometry by identifying craniofacial landmarks. This AI-driven approach with mesh fitting demonstrates clinically acceptable accuracy for analyzing patient anatomy.
Area of Science:
- Biomedical Engineering
- Medical Imaging
- Orthodontics
Background:
- Surface-based 3D cephalometry offers detailed craniofacial analysis.
- Automating this process can improve efficiency and accuracy in clinical settings.
Purpose of the Study:
- To develop an automated 3D surface cephalometry system using AI-identified landmarks and mesh fitting.
- To evaluate the accuracy of this novel system for craniofacial structure analysis.
Main Methods:
- Utilized 185 CBCT images from adult Japanese patients for system training and evaluation.
- Developed an AI system (PointNet++) to identify 3D landmarks on cranial and mandibular surfaces.
- Performed mesh fitting using AI-identified landmarks and evaluated fitting errors.
Main Results:
- The system achieved mean errors of 0.80 ± 0.57 mm for the maxilla and 1.45 ± 0.34 mm for the mandible.
- These results indicate clinically acceptable accuracy for the automated cephalometry system.
Conclusions:
- An AI-based system for landmark identification and mesh fitting in 3D surface cephalometry was successfully developed.
- The system demonstrates clinically acceptable accuracy, enabling efficient quantification and visualization of craniofacial structures for clinical applications.
