Related Experiment Video
Updated: Jul 18, 2025
05:49
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
Published on: February 23, 2024
882
Automated permanent tooth detection and numbering on panoramic radiograph using a deep learning approach
Ramadhan Hardani Putra1, Eha Renwi Astuti1, Dina Karimah Putri2
1Department of Dentomaxillofacial Radiology, Faculty of Dental Medicine, Universitas Airlangga, Surabaya, Indonesia.
Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology
|August 26, 2023
Summary
Deep learning models accurately detect and number teeth in panoramic radiographs. The YOLO v4 model significantly outperforms human speed for automated tooth detection and numbering.
Area of Science:
- Dentistry
- Radiology
- Artificial Intelligence
Background:
- Automated tooth detection and numbering in panoramic radiographs is crucial for dental diagnostics.
- Traditional methods can be time-consuming and prone to human error.
Purpose of the Study:
- To evaluate the performance of a deep learning (DL) model for automated tooth numbering in panoramic radiographs.
- To compare the efficiency of the DL model against human dentists.
Main Methods:
- A dataset of 500 panoramic images was utilized, divided into training (80%) and testing (20%) sets.
- Tooth numbering utilized the universal numbering system with 32 classes, annotated using LabelImg software.
- The You Only Look Once (YOLO) v4 deep convolution neural network model was employed for object detection and performance evaluation via a confusion matrix.
Main Results:
- The YOLO v4 model achieved high performance metrics: 88.5% accuracy, 87.70% precision, 100% recall, and 93.44% F1 score for tooth detection and numbering.
- Automated numbering using YOLO v4 took a mean of 20.58 ± 0.29 ms, significantly faster than human dentists (P < .0001).
Conclusions:
- Deep learning, specifically the YOLO v4 model, offers a viable solution for automated tooth detection and numbering in panoramic radiographs.
- This DL approach can assist dentists by providing accurate and rapid tooth identification, enhancing daily clinical practice.
