The Accuracy of Detection of Artificial Intelligence Second Mesio-buccal Canal of Maxillary First Molars on CBCT Images

Artificial Intelligence

CAD systems are computer applications that assist in the detection and/or diagnosis of diseases by providing an unbiased second opinion to the image interpreter, aiming at improving accuracy and reducing time for analysis. With the rapid growth of Deep Learning (DL) algorithms in image-based applications, CAD systems can now be trained by DL to provide more advanced capability (ie, the capability of artificial intelligence \[AI\]) to best assist clinicians.

Countless studies and discussions have been based on the existence of a second canal in the mesiobuccal (MB) root of the maxillary molars , since it is strongly believed that one of the foremost reasons for endodontic failure in maxillary first molars is the difficulty of detecting and treating those second mesiobuccal (MB2) canals .The literature reveals that although MB2 canals of maxillary first molars have been found in more than 70% of in vitro studies , they were detected clinically in less than 40% of cases . Cone beam computed tomography (CBCT) is an imaging modality in the field of endodontics that has several advantages, including the ability to perform three-dimensional (3D) imaging of root canal systems with lower radiation doses, higher resolution, and no superimposition . Researchers have evaluated the efficiency of CBCT when it comes to identifying MB2 canals, and CBCT has been suggested to be a reliable method for the detection of these canals. However, in clinically relevant situations, such a smaller lesions on root-filled teeth, CBCT accuracy is greatly reduced (sensitivity 0.63, specificity 0.69) . Moreover, clinician dependent interpretation of CBCT imaging still suffers from low inter- and intra-observer agreement. Computer-aided detection and diagnosis (CAD) has been widely applied to biomedical image analysis outside of dentistry .

Egypt