Related Experiment Video
Updated: Mar 28, 2026
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
A new mandible-specific landmark reference system for three-dimensional cephalometry using cone-beam computed
Pisha Pittayapat1,2, Reinhilde Jacobs3, Michael M Bornstein3,4
1*OIC, OMFS-IMPATH Research Group, Department of Imaging & Pathology, Faculty of Medicine, University of Leuven and Oral & Maxillofacial Surgery, University Hospitals Leuven, Belgium, p.pittayapat@gmail.com.
Objectives:
To develop a novel 3D landmark reference system that is specific for mandibular midline cephalometric landmarks and to assess its repeatability and reproducibility.
Methods:
Cone-beam computed tomography (CBCT) scans (3D Accuitomo® 170) were performed on 26 dry human skulls. The CBCT data were exported into DICOM files and imported to Maxilim® software to create 3D surface models. Two observers identified five landmarks to create a specific mid-sagittal mandibular plane: two mandibular foramina, two molar landmarks and one interincisive landmark. On this mid-sagittal mandibular plane, four mandibular cephalometric landmarks were marked: Point B, Pogonion, Gnathion and Menton. All observations were repeated by the two observers after an interval of 4 weeks. The coordinates (x, y, z) of each landmark were exported, and statistical analyses were performed to evaluate inter- and intra-rater precision.
Results:
The intra-observer median precision in locating all landmarks ranged between 0.17 and 0.61mm. The intra-observer repeatability was generally good with a precision under 1mm in more than 50 per cent. The overall median inter-observer precision was 0.26-2.30mm. The mandibular foramina showed the best inter-observer reproducibility. The general inter-observer reproducibility was moderate to good, except for Pogonion and Point B.
Limitations:
Dry human skulls may not represent anatomical conditions found in living patients, thus the system should be validated using patients' data.
Conclusion:
The novel reference system offered good precision and generally good to moderate repeatability and reproducibility for mandibular midline cephalometric landmark identification in three dimensions. These findings will be useful for further improvement of 3D cephalometric systems.
