AI-Based Dental Disease Detection Using OPG Images

Panoramic dental radiographs, known as orthopantomograms (OPGs), give a wide view of the teeth and surrounding jaw structures. Interpreting them can be complicated by image quality, anatomical overlap and variation between imaging devices. This paper proposes an AI-based workflow for the preliminary analysis of OPG images using a convolutional neural network (CNN) or a transfer-learning model. The planned system comprises image validation, preprocessing, model inference and a simple mobile interface through which a user can upload an image and view the result. Dataset selection, annotation quality, patient privacy and class-wise evaluation are treated as core design requirements, and performance will be assessed with accuracy, precision, recall, F1-score and a confusion matrix rather than accuracy alone. The work is currently at the design stage, so no trained-model performance or clinical results are claimed. The intended outcome is a research prototype that can support preliminary review while leaving diagnostic decisions to qualified dental professionals.