Table 7 Performance comparison among the original and truncated ResNets on the testing set in the classification of aggressive and non-aggressive EC

From: Deep learning for endometrial cancer subtyping and predicting tumor mutational burden from histopathological slides

Backbone

Acc.

Sens.

Spec.

MeanSS

AUROC

ResNet50

 

0.89

0.93

0.84

0.89

0.97

ResNet101

Truncated

0.84

0.80

0.89

0.84

0.86

ResNet152

 

0.86

0.83

0.90

0.87

0.93

ResNet50

 

0.78

0.87

0.67

0.77

0.87

ResNet101

Original

0.80

0.84

0.74

0.79

0.85

ResNet152

 

0.74

0.75

0.71

0.73

0.80

  1. MeanSS: Mean of sensitivity and specificity
Back to article page