Meta-analysis of 8 studies (7814 radiographs, 12,373 molars): deep learning models detected furcation involvement with sensitivity 0.93, specificity 0.94 and AUC 0.97, higher in mandibular molars.
Meta-analysisn = 7,814 radiographs
Atieh et al. Diagnostic accuracy of artificial intelligence-based deep learning models in detecting furcation involvement: A systematic review and meta-analysis