37
https://pubmed.ncbi.nlm.nih.gov/38094539
This study developed a deep learning-based tool, CAST, that can automatically annotate and segment clinically significant teeth in intraoral photographs and radiographs, with a mean average precision of 77.4%.