Table 4.

Performances of cephalometric skeletal classification for DCNN-based AI model and automated-tracing AI software.

	DCNN-based AI model				Automated-tracing AI software
	Sensitivity	Specificity	Precision	Accuracy	Sensitivity	Specificity	Precision	Accuracy
Class I	0.97	0.98	0.95	0.97	0.90	0.83	0.72	0.85
Class II	0.97	0.95	0.91	0.96	0.98	0.95	0.91	0.96
Class III	0.88	0.99	0.97	0.95	0.68	1.00	1.00	0.89
Micro-average	0.94	0.97	0.94	0.96	0.85	0.93	0.85	0.90
Macro-average	0.94	0.97	0.94	0.96	0.85	0.93	0.90	0.88

DCNN, deep convolutional neural network.