Related Experiment Video
Updated: Nov 4, 2025
10:23
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
3.2K
Comparison of cephalometric measurements between conventional and automatic cephalometric analysis using
Sangmin Jeon1, Kyungmin Clara Lee2
1, Gwangju, Republic of Korea.
Progress in Orthodontics
|May 31, 2021
Summary
Artificial intelligence (AI) in medical imaging enables automatic cephalometric analysis. While AI shows clinically acceptable performance, dental measurements require manual adjustment for improved accuracy.
Area of Science:
- Radiology
- Artificial Intelligence
- Medical Imaging
Background:
- Advancements in AI have led to automated identification of anatomical landmarks on radiographs.
- Cephalometric analysis is a crucial diagnostic tool in orthodontics and maxillofacial surgery.
Purpose of the Study:
- To compare the diagnostic accuracy of an automated cephalometric analysis using convolutional neural networks (CNNs) with a conventional manual approach.
- Evaluate the performance of AI in identifying skeletal, dental, and soft tissue landmarks.
Main Methods:
- Lateral cephalograms from 35 patients were analyzed using both an automated CNN program and a conventional method.
- Fifteen skeletal, nine dental, and two soft tissue measurements were compared using paired t-tests and Bland-Altman plots.
Main Results:
- Statistically significant differences were observed in the saddle angle, maxillary incisor to NA, and mandibular incisor to NB measurements between the two methods.
- All measurements fell within the limits of agreement, though dental measurements showed wider limits than skeletal measurements.
- The automated CNN approach demonstrated clinically acceptable diagnostic performance.
Conclusions:
- AI-based automated cephalometric analysis offers a promising, clinically acceptable diagnostic tool.
- Manual adjustments and careful consideration are necessary for dental measurements to enhance accuracy and performance, particularly for tooth structures.
