TABLE 2.

Performance of the proposed network with different backbones.

Backbone	Accuracy	Specificity	Sensitivity	AUC
ResNet-18	0.9817	0.9800	0.9833	0.9960
ResNet-50	0.9733	0.9700	0.9767	0.9967
VGG-11	0.9833	0.9833	0.9833	0.9984
VGG-16	0.9783	0.9833	0.9783	0.9983
DenseNet-121	0.9767	0.9767	0.9767	0.9976
DenseNet-161	0.9767	0.9867	0.9667	0.9971

The bold figures represent the maximum value of each evaluation index.