Skip to main content

Table 1 Comparison of model performance

From: Implementing deep learning models for the classification of Echinococcus multilocularis infection in human liver tissue

Model performance parameters

DL neural models

VGG19_bn

Squeezenet

ResNet-18

Validation entropy loss

0.054

0.057

0.079

Validation accuracy

0.98

0.98

0.97

Validation weighted accuracy

0.98

0.98

0.97

Validation weighted precision

0.98

0.98

0.97

Validation weighted recall

0.98

0.98

0.97

Validation weighted f1

0.98

0.98

0.97

Median test set accuracy

0.94

0.92

0.9

Slide level test set AUC

1

1

1

Training time (in min) (15 epochs)

217

120

151

  1. AUC Area under the curve, bn batch normalization, DL deep learning