Automated Tooth Detection and Numbering in Panoramic Radiographs Using YOLO
Alison Corrêa Mendes, Darlan B. P. Quintanilha, Alexandre Pessoa, Anselmo Cardoso de Paiva, Pedro de Alcantara dos Santos Neto
Abstract
Panoramic radiography is widely used in dentistry due to its low radiation exposure, quick image acquisition time, and patient comfort. This research provides an in-depth analysis of the application of detection techniques in dental panoramic radiographs using the YOLO (You Only Look Once) model, an object detection algorithm known for its speed and accuracy. The study demonstrates the potential of the YOLO model to enhance dental diagnostics and treatment planning by automating tooth detection. Accurate tooth numbering is essential in dental practice for documenting issues, creating detailed records, and monitoring long-term oral health. The experiments showed promising results, with a precision of 0.9283, recall of 0.9327, mAP50 of 0.9450, and mAP50-95 0.5781. These results suggest that the YOLO model can be an effective tool for automatic tooth detection in panoramic radiographs, potentially transforming dental diagnostics and treatment planning by providing rapid and accurate analyses of radiographic images.