Table 2.

Two-class classification results: Dataset 1.

CNN	Accuracy (%)	Sensitivity (%)	Specificity (%)	Precision (%)	$F_1$ score	AUC
80% training and 20% testing
AlexNet	99.14 ± 0.62	98.44 ± 1.19	99.51 ± 0.66	99.10 ± 1.21	0.988 ± 0.009	0.999 ± 0.002
GoogLeNet	99.70 ± 0.52	100 ± 0.00	99.54 ± 0.80	99.16 ± 1.46	0.996 ± 0.007	0.999 ± 0.000
SqueezeNet	99.85 ± 0.26	100 ± 0.00	99.77 ± 0.40	99.57 ± 0.74	0.998 ± 0.004	0.999 ± 0.000
50% training and 50% testing
AlexNet	99.19 ± 0.23	98.32 ± 0.65	99.65 ± 0.14	99.35 ± 0.26	0.988 ± 0.003	0.999 ± 0.000
GoogLeNet	99.22 ± 0.46	99.14 ± 0.60	99.26 ± 0.42	98.63 ± 0.79	0.989 ± 0.007	0.999 ± 0.002
SqueezeNet	98.43 ± 2.10	95.85 ± 6.28	99.81 ± 0.32	99.66 ± 0.032	0.976 ± 0.000	0.999 ± 0.000