Diagnostic accuracy of artificial intelligence systems for radiographic caries detection and high-quality dataset of annotated radiographs

Artificial intelligence algorithms require training. Each research group tests the performance of its system with its own data set, which limits extrapolation or comparison.

The project seeks to (1) evaluate and synthesize the evidence supporting the use of Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) for caries detection in intraoral radiographs, (2) generate a high quality annotated dataset of radiographic images, (3) generate a model for caries detection in intraoral radiographs using ML/DL with acceptable diagnostic accuracy for clinical use and (4) explore alternatives for the development of a viable clinical production system.