Table 2 Diagnostic performances of the three models implemented.

From: Automatic diagnosis of true proximity between the mandibular canal and the third molar on panoramic radiographs using deep learning

 

Accuracy (%)

Sensitivity (%)

Specificity (%)

RetinaNet

59.55

30.34

88.76

YOLOv3

73.03

82.02

64.04

EfficientDet-D4

78.65

82.02

75.28