Artificial intelligent-driven decision-making for automating root fracture detection in periapical radiographs

Abstract

Root fracture detection in CBCT images is vital for precise dental treatment planning. This study aims to evaluate the performance of an artificial intelligence (AI)-driven decision-making system utilizing a VGG19-based convolutional neural network (CNN) for automated root fracture identification. A dataset comprising 50 CBCT images was used, split into 25 fractured and 25 non-fractured cases. The model achieved an overall accuracy of 92%, with sensitivity and specificity rates of 90% and 93%, respectively. These results underscore the potential of AI in enhancing diagnostic accuracy, efficiency, and reliability in dental radiology, paving the way for its integration into clinical workflows.