Table 9.

The results of different models on the testing set in the RS dataset + DCGAN dataset during the 10-fold cross-validation (mean ± standard deviation).

Models	Evaluation Metrics			Training Time (mins:secs)
	Acc	Se	Sp
VGG-16	0.981 ± 0.003	0.978 ± 0.004	0.979 ± 0.003	494:56
VGG-19	0.964 ± 0.004	0.956 ± 0.005	0.961 ± 0.005	194:78
ResNet-18	0.957 ± 0.004	0.941 ± 0.006	0.949 ± 0.005	314:34
ResNet-50	0.968 ± 0.005	0.952 ± 0.006	0.951 ± 0.006	263:71
DenseNet-121	0.936 ± 0.005	0.922 ± 0.007	0.913 ± 0.006	286:45
AlexNet	0.962 ± 0.004	0.959 ± 0.005	0.955 ± 0.005	381:24
Proposed CNN	0.987 ± 0.001	0.986 ± 0.002	0.988 ± 0.002	130:13

RS dataset + DCGAN dataset = 33,093 images in the LVDD group and 43,227 images in the control group.