Accuracy of artificial intelligence technology in detecting the number of canals in human maxillary first molars indicated for retreatment : Diagnostic accuracy experimental study /

Abstract

Statement of Problem: Missed canals are one of the main causes of failure of primary root canal treatment. CBCT is considered the gold standard in morphology detection. Problems of using CBCT include high radiation dose and practitioner inability to interpret images. Artificial Intelligence (AI) technology may help overcome these problems. Aim: The aim of this study was to evaluate the accuracy of novel AI software in detecting the number of canals in 36 maxillary first molars indicated for retreatment, as well as, to compare it with accuracy of CBCT and clinical assessment. Materials and Methods: 36 Patients referred to MIU dental clinic for retreatment of upper first molars underwent pre-treatment CBCT, while only pretreatment periapical radiograph will be taken to aid in access cavity preparation in the clinical stage. The study included 3 stages: CBCT Stage: Pre-operative CBCT scans of the patients were taken and randomly assigned to 2 co-supervisors who upon scan segmentation recorded the number of canals on a pre-formed information guide. Clinical Stage: A clinical stage where the enrolled patients were randomly distributed upon 6 researchers. Practitioners then performed access cavities on the teeth under DOM. The number of orifices found was recorded on a preformed information guide. AI stage: CBCT images will be uploaded to AI software by primary investigator and Number of canals found by the software were recorded. Results of the first two stages were then compared to the findings of the third stage to determine software accuracy. Cases with missed canals by the AI software underwent further evaluation of tooth morphological features, to determine the reason for the software’s detection failure.