Related Experiment Video
Updated: Jun 6, 2025
Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
Published on: September 8, 2023
Comparative Analysis of 3D Cephalometry Provided with Artificial Intelligence and Manual Tracing
Zurab Khabadze1, Oleg Mordanov1, Ekaterina Shilyaeva1
1Department of Therapeutic Dentistry, Peoples' Friendship University of Russia Named after Patrice Lumumba (RUDN University), 6 Miklukho-Maklaya St., 117198 Moscow, Russia.
Objectives:
To compare 3D cephalometric analysis performed using AI with that conducted manually by a specialist orthodontist.
Methods:
The CBCT scans (a field of view of 15 × 15 cm) used in the study were obtained from 30 consecutive patients, aged 18 to 50. The 3D cephalometric analysis was conducted using two methods. The first method involved manual tracing performed with the Invivo 6 software (Anatomage Inc., Santa Clara, CA, USA). The second method involved using AI for cephalometric measurements as part of an orthodontic report generated by the Diagnocat system (Diagnocat Ltd., San Francisco, CA, USA).
Results:
A statistically significant difference within one standard deviation of the parameter was found in the following measurements: SNA, SNB, and the left interincisal angle. Statistically significant differences within two standard deviations were noted in the following measurements: the right and left gonial angles, the left upper incisor, and the right lower incisor. No statistically significant differences were observed beyond two standard deviations.
Conclusions:
AI in the form of Diagnocat proved to be effective in assessing the mandibular growth direction, defining the skeletal class, and estimating the overbite, overjet, and Wits parameter.
