Table 1.

Pretrained models with performance metrics along with SD.

Metrics	VGG16	ResNet50	GoogLeNet	Inception-v3	Inception-v4	IR-v2
Sensitivity	0.99 ± 0.012	0.96 ± 0.008	0.95 ± 0.011	0.97 ± 0.011	0.97 ± 0.019	0.94 ± 0.032
Specificity	0.92 ± 0.013	0.96 ± 0.019	0.97 ± 0.011	0.96 ± 0.014	0.95 ± 0.003	0.96 ± 0.023
Precision	0.97 ± 0.013	0.96 ± 0.019	0.97 ± 0.011	0.96 ± 0.008	0.95 ± 0.003	0.97 ± 0.025
Accuracy	0.95 ± 0.00	0.96 ± 0.004	0.96 ± 0.00	0.96 ± 0.010	0.96 ± 0.008	0.95 ± 0.005
F-score	0.95 ± 0.00	0.96 ± 0.004	0.96 ± 0.00	0.97 ± 0.005	0.96 ± 0.008	0.95 ± 0.005
MCC	0.94 ± 0.014	0.93 ± 0.005	0.91 ± 0.015	0.93 ± 0.005	0.93 ± 0.018	0.89 ± 0.003
AP	0.91 ± 0.003	0.95 ± 0.013	0.96 ± 0.003	0.95 ± 0.009	0.95 ± 0.018	0.96 ± 0.01