Table 1.

Quantitative results in comparative study of different convolutional neural network (CNN) models.

Model	Accuracy	Sensitivity	Specificity	F1	AUC
VGG19	0.826 ± 0.029	0.975 ± 0.010	0.636 ± 0.058	0.862 ± 0.021	0.874 ± 0.021
InceptionV3	0.832 ± 0.016	0.987 ± 0.014	0.636 ± 0.036	0.868 ± 0.011	0.891 ± 0.017
Xception	0.832 ± 0.015	0.986 ± 0.012	0.636 ± 0.024	0.868 ± 0.011	0.844 ± 0.010
DenseNet121	0.838 ± 0.016	0.991 ± 0.005	0.643 ± 0.031	0.872 ± 0.012	0.861 ± 0.020