Skip to main content

Table 2 Evaluation of our model and transferred model

From: Accuracy of machine learning for differentiation between optic neuropathies and pseudopapilledema

 

Our Model

Inception V3

ResNet

VGG

 

Ensemble Accuracy

AUROC

Ensemble Accuracy

AUROC

Ensemble Accuracy

AUROC

Ensemble Accuracy

AUROC

Training Data

100%

1.0

100%

1.0

100%

1.0

100%

1.0

Validation Data

96.35%

0.989

98.18%

0.993

98.18%

0.996

97.81%

0.996

Test Data

95.89%

0.992

96.35%

0.997

98.63%

0.999

96.80%

0.999