Related Experiment Video
Updated: Feb 28, 2026
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
1.6K
CBCT-Based Orthodontic Classification Using Commercial AI: Completeness and Accuracy in Independent Validation.
Natalia Kazimierczak1, Nora Sultani2, Szymon Krzykowski2
1Kazimierczak Clinic, Dworcowa 13/u6a, 85-009 Bydgoszcz, Poland.
Journal of Clinical Medicine
|February 27, 2026
Summary
The Diagnocat artificial intelligence (AI) platform shows poor diagnostic reliability for orthodontic diagnoses from cone-beam computed tomography (CBCT) scans, failing to generate sufficient data for clinical use.
Area of Science:
- Orthodontics
- Artificial Intelligence
- Radiology
Background:
- Artificial intelligence (AI) tools are emerging for orthodontic diagnosis using CBCT.
- Limited evidence exists on AI performance in CBCT-based orthodontic assessments.
- The Diagnocat platform's diagnostic reliability for CBCT data needs evaluation.
Purpose of the Study:
- To assess the diagnostic reliability of the Diagnocat AI platform.
- To evaluate categorical orthodontic diagnoses derived from CBCT examinations.
Main Methods:
- Fifty-nine patients with CBCT and cephalograms were analyzed.
- CBCT scans were processed using the Diagnocat platform (v1.0).
- AI outputs were compared to manual cephalometric analyses (reference standard).
Main Results:
- AI platform generated skeletal and vertical classifications for <10% of patients.
- Fair agreement (Cohen's kappa = 0.324) was observed for overbite categorization.
- Low data completeness (<10% for skeletal parameters) indicates poor usability.
Conclusions:
- The Diagnocat platform demonstrated insufficient diagnostic reliability and low data completeness.
- The AI tool is currently unsuitable for independent clinical decision-making in orthodontics.
- Further development is needed to improve AI performance in CBCT-based assessments.
