Related Experiment Video
Updated: May 6, 2026
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
1.8K
Performance of an Artificial Intelligence-Based Automated System for Identifying Primary and Permanent Teeth in Mixed
Everton Flaiban1, Elaine Dinardi Barioni1, Lana Ferreira Santos2
1Dentomaxillofacial Radiology and Imaging Laboratory, Postgraduate Program in Dentistry, Cruzeiro Do Sul University (UNICSUL), Rua Galvão Bueno, Liberdade, São Paulo, SP, 86801506-000, Brazil.
Journal of Imaging Informatics in Medicine
|May 4, 2026
Summary
This study evaluated an AI system for identifying teeth in dental X-rays of children. The artificial intelligence tool demonstrated high accuracy in detecting and classifying primary and permanent teeth, aiding dental imaging interpretation.
Area of Science:
- Dentistry
- Medical Imaging
- Artificial Intelligence
Background:
- Accurate identification of primary and permanent teeth in mixed dentition is crucial for pediatric dental care.
- Panoramic radiography is a common imaging modality in pediatric dentistry.
- Automated analysis of dental radiographs can potentially improve efficiency and accuracy.
Purpose of the Study:
- To evaluate the diagnostic performance of the DIO Inteligência® artificial intelligence (AI) system.
- To assess the AI system's ability to automatically detect and classify primary and permanent teeth in panoramic radiographs of patients in mixed dentition.
- To compare the AI system's performance against expert radiologist consensus.
Main Methods:
- Retrospective diagnostic accuracy study involving 110 digital panoramic radiographs from patients aged 6-12 years.
- AI system automatically identified and classified teeth using FDI notation.
- Comparison of AI output with a gold standard established by consensus of two dentomaxillofacial radiologists.
- Calculation of diagnostic performance metrics: accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).
Main Results:
- The AI system achieved an overall accuracy of 91%, sensitivity of 92%, and specificity of 72%.
- High positive predictive value (PPV) of 99% was observed, with a negative predictive value (NPV) of 29%.
- Consistently high performance was noted for permanent teeth (accuracy ~96%, PPV ~100%), with slightly lower metrics for third molars. Primary teeth also showed favorable classification performance.
Conclusions:
- The DIO Inteligência® AI system demonstrates robust performance in detecting and classifying primary and permanent teeth in mixed dentition panoramic radiographs.
- The system shows potential as a reliable adjunct tool for pediatric dental imaging interpretation.
- Further validation may be warranted, particularly for specific tooth groups like third molars.
