Related Experiment Video
Updated: Feb 27, 2026
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
Artificial Intelligence Models for the Detection and Quantification of Orthodontically Induced Root Resorption Using
Carlos M Ardila1,2, Eliana Pineda-Vélez2,3, Anny M Vivares-Builes2,3
1Department of Periodontics, Saveetha Institute of Medical and Technical Sciences, Saveetha Dental College and Hospitals, Saveetha University, Chennai 600077, India.
Artificial intelligence (AI) models show high accuracy in detecting and quantifying orthodontically induced root resorption (OIRR) using cone-beam computed tomography (CBCT). These AI tools demonstrate excellent agreement with expert assessments, supporting their clinical use in orthodontics.
Area of Science:
- Dentistry
- Medical Imaging
- Artificial Intelligence
Background:
- Orthodontically induced root resorption (OIRR) is a common adverse effect of orthodontic treatment.
- Accurate detection and quantification of OIRR are crucial for patient care.
Purpose of the Study:
- To systematically review and meta-analyze the diagnostic performance of AI models for OIRR detection using CBCT.
- To evaluate AI model agreement with manual assessments and explore factors influencing performance.
Main Methods:
- A comprehensive literature search was conducted across major databases.
- Seven studies employing AI for OIRR diagnosis on CBCT were included.
- A random-effect meta-analysis and subgroup analyses were performed following PRISMA guidelines.
Main Results:
- AI models demonstrated high pooled sensitivity (0.903) and excellent specificity (82%-98%).
- Area under the ROC curve reached up to 0.96, indicating strong diagnostic capability.
- Near-perfect agreement (ICC=1.000) was observed between AI and manual quantification.
Conclusions:
- AI models applied to CBCT show excellent diagnostic accuracy for OIRR detection.
- AI exhibits high concordance with expert assessments, suggesting potential for clinical integration.
- Further research may refine AI model architecture and validation for optimal performance.
