Fig. 1. Progression flowchart of this study. In some OPG images, the mandibular dentition was set as the ROI, and an object detection model was developed based on this. The model was then applied to the entire dataset. The extracted mandibular dentition images were used to train an age estimation model utilizing an ensemble backbone of ResNet50 and VGG19. ROI, region of interest; YOLO, You Only Look Once; CNN, convolutional neural network; OPG, orthopantomogram.

J Oral Med Pain 2024;49:118-23 https://doi.org/10.14476/jomp.2024.49.4.118

© 2024 J Oral Med Pain