Related Experiment Video
Updated: Jul 16, 2026
10:23
Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
An approach for the automatic cephalometric landmark detection using mathematical morphology and active appearance
Sylvia Rueda1, Mariano Alcañiz
1Medical Image Computing Laboratory, Universidad Politécnica de Valencia, UPV/ETSIA, Camino de Vera s/n, 46022 Valencia, Spain. silruelo@degi.upv.es
Medical Image Computing and Computer-Assisted Intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
|March 16, 2007
Summary
This study introduces an automated system for cephalometric analysis using Active Appearance Models (AAMs), improving accuracy and efficiency in orthodontic diagnosis. The method offers a reliable solution for clinical applications, overcoming limitations of manual landmark identification.
Area of Science:
- Medical Imaging
- Computer Vision
- Orthodontics
Background:
- Cephalometric analysis of lateral head radiographs is crucial in orthodontics.
- Manual landmark identification is time-consuming, tedious, and prone to errors.
Purpose of the Study:
- To develop and clinically validate an automated system for cephalometric analysis using Active Appearance Models (AAMs).
- To address limitations of previous automated methods, including small datasets and lack of variability consideration.
Main Methods:
- An automated system based on Active Appearance Models (AAMs) was developed.
- A top-hat transformation was employed to correct radiograph intensity inhomogeneity.
- The AAM was trained on 96 hand-annotated images and validated using a leave-one-out scheme.
Main Results:
- The automated system achieved an average accuracy of 2.48mm.
- The proposed method demonstrated clinical suitability and robustness.
- Intensity correction using top-hat transformation ensured a consistent training set.
Conclusions:
- Active Appearance Models combined with mathematical morphology offer a suitable and accurate method for clinical cephalometric analysis.
- The automated system significantly enhances efficiency and reliability in orthodontic diagnosis.
- This approach overcomes previous limitations in automated cephalometric analysis validation.
