Table 1 The performance metrics of CNN models for the validation dataset images.

From: Deep convolutional neural networks for automated scoring of pentagon copying test results

CNN model

Accuracy

Precision

Recall (sensitivity)

Specificity

AUROC

GoogLeNet

0.877

0.947

0.878

0.875

0.931

VGG-16

0.930

0.951

0.951

0.875

0.922

ResNet-50

0.789

0.872

0.829

0.688

0.784

Inception-v3

0.789

0.939

0.756

0.875

0.864

  1. AUROC, area under the receiver operating characteristic curve; CNN, convolutional neural network.
Back to article page