Skip to main content
. 2020 May;8(9):581. doi: 10.21037/atm.2020.04.39

Figure 3.

Figure 3

Performance of CNNs for the validation. (A) ROC curves with sensitivity as x-axis and specificity as y-axis. The ROC curves and the AUC value of our model, Inception V3, SE_ResNet101, and SE_ResNeXt101-32x4d have been shown. (B) The accuracy, specificity and sensitivity of the validation task results are shown in the following table (n=50); Acc, accuracy; Sen, sensitivity; Spe, specificity; CNN, convolutional neural network; AUC, area under the raw current curves.