Related Experiment Video
Updated: Jul 24, 2025
10:23
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
2.8K
Automatic recognition of cephalometric landmarks via multi-scale sampling strategy.
Congyi Zhao1,2, Zengbei Yuan1,2, Shichang Luo1,2
1College of Medical Imaging, Jiading District Central Hospital Affiliated Shanghai University of Medicine and Health Sciences, Shanghai, 201318, China.
Heliyon
|July 7, 2023
Summary
This study introduces Multi-Scale YOLOV3 (MS-YOLOV3), an automatic algorithm for detecting cephalometric landmarks. MS-YOLOV3 improves accuracy and efficiency in cephalometric analysis for orthodontic and orthognathic surgery.
Area of Science:
- Medical Imaging
- Computer Vision
- Orthodontics
Background:
- Accurate identification of cephalometric landmarks is crucial for maxillofacial tissue localization in orthodontic and orthognathic surgery.
- Current methods for landmark identification often suffer from low accuracy and complex procedures.
Purpose of the Study:
- To propose an automatic target recognition algorithm, Multi-Scale YOLOV3 (MS-YOLOV3), for enhanced cephalometric landmark detection.
- To address the limitations of existing methods in terms of accuracy and efficiency.
Main Methods:
- Developed the MS-YOLOV3 algorithm incorporating multi-scale sampling strategies and a spatial pyramid pooling (SPP) module.
- Evaluated MS-YOLOV3 against the classical YOLOV3 algorithm using public lateral and undisclosed anterior-posterior (AP) cephalogram datasets.
Main Results:
- MS-YOLOV3 demonstrated robust performance with successful detection rates (SDR) of 80.84% within 2mm for lateral cephalograms and 85.75% within 2mm for AP cephalograms.
- Achieved high SDRs of 93.75% (lateral) and 92.87% (AP) within 3mm, and 98.14% (lateral) and 96.66% (AP) within 4mm.
Conclusions:
- The proposed MS-YOLOV3 algorithm offers a robust and accurate solution for labeling cephalometric landmarks.
- This automated approach has significant potential for clinical applications in orthodontics and orthognathic surgery.
