Table 3.

Performance comparison for the identification of image modality.

Architectures	Accuracy	Sensitivity	Specificity	Precision	F-measure	G-mean
AlexNet70	0.98520	0.97851	0.99500	0.99500	0.98625	0.98545
VGG-1932	0.98451	0.97632	0.99320	0.99530	0.98480	0.98785
SqueezeNet71	0.98475	0.97540	0.99885	0.99635	0.98645	0.98890
GoogLeNet72	0.99535	0.98965	0.99845	0.99845	0.99450	0.99480
ResNet-10173	0.99550	0.98975	0.99750	0.99750	0.99548	0.99565
Proposed	0.99835	0.99565	0.99885	0.99885	0.99750	0.99750

Best result for each metric is shown in bold.