Intraoral cameras as a computer-aided diagnosis tool for root canal orifices.

The aim of this study was to evaluate the diagnostic advantage of a new software tool in combination with an intraoral camera for automatic detection of root canal orifices in life videos via the access cavity of extracted human molars. The performance of a predoctoral dental student analyzing the images of the camera (without automatic detection) was compared with that of an experienced observer. Sensitivity and confidence intervals were provided and compared with histological slices of 200 teeth used for evaluation. The software's sensitivity for detection of root canal orifices was 0.957 (95 percent confidence interval: 0.936 to 0.972). The sensitivity for the observer was 0.906 (95 percent confidence interval: 0.877 to 0.929) compared to 0.847 (95 percent confidence interval: 0.813 to 0.877) achieved by the predoctoral student. The tested software reaches a high sensitivity for automatic real-time detection of root canal orifices with intraoral cameras in direct comparison to histological images. The system might be a useful help for both pre-and postdoctoral students as an aid for the detection of second mesiobuccal root canal orifices.