Automated Classification of Third Molar Development Stages Using Deep Learning
Accurate assessment of tooth development is crucial in various fields, including dentistry, orthodontics, and forensic science. Traditional methods, such as the Demirjian system, rely on visual inspection by dental professionals, which can be subjective and prone to inter-observer variability. This study aimed to develop a fully automated system for classifying third molar (wisdom teeth) development stages using deep learning, offering a more objective and efficient approach. The research utilized a large dataset of orthopantomograms (OPGs), which were meticulously labeled by dental experts according to the Demirjian system. The dataset comprised images of both left and right lower jaws, resulting in a substantial collection for model training and evaluation. To enhance the accuracy and robustness of the system, several key techniques were employed: The results demonstrated the superior performance of the EfficientNet model, achieving a classification accuracy of 83.7%. This significant … Read more