Related Experiment Video
Updated: May 16, 2026

Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
Accuracy of computerized automatic identification of cephalometric landmarks by a designed software
Sh Shahidi1, S Shahidi, M Oshagh
1Shiraz Biomaterial [corrected] Research Center, Department of Oral and Maxillofacial Radiology, Faculty of Dentistry, Shiraz University of Medical Sciences, Shiraz, Iran. [corrected].
This study developed accurate software for automatically locating cephalometric landmarks on dental X-rays. The new system shows improved precision compared to existing automated methods.
Area of Science:
- Dentistry
- Medical Imaging
- Computer Science
Background:
- Cephalometric analysis is crucial for diagnosing and planning orthodontic treatment.
- Accurate landmark identification is essential for reliable cephalometric measurements.
- Manual landmark identification is time-consuming and prone to inter-observer variability.
Purpose of the Study:
- To design and evaluate novel software for automated cephalometric landmark localization.
- To assess the accuracy of the developed software in identifying key cephalometric landmarks.
Main Methods:
- Developed software using Delphi and Matlab, incorporating template matching and edge enhancement techniques.
- Utilized 40 digital cephalometric radiographs with 16 predefined landmarks.
- Compared automated landmark identification against manual identification by three expert orthodontists.
Main Results:
- The software achieved a total mean error of 2.59 mm for landmark localization.
- 12.5% of landmarks were identified with mean errors under 1 mm.
- 43.75% of landmarks had mean errors below 2 mm, with most landmarks showing accuracy under 4 mm.
Conclusions:
- The developed software demonstrates significant accuracy for cephalometric landmark localization.
- This automated system offers a promising alternative to manual methods in cephalometric analysis.
- The software's accuracy surpasses that of previous model-based and knowledge-based automated systems.
More Related Videos
05:49Reliability of Artificial Intelligence-Based Cone Beam Computed Tomography Integration with Digital Dental Images
Published on: February 23, 2024
08:03Midface Hypoplasia and Cranial Base Morphology in Syndromic Craniosynostosis: A Comparative Analysis Study Using a Predictive Regression Model
Published on: November 4, 2025