Table 2 Class-level performance metrics using the Enhanced-Xception Network.

From: MRI Compatibility: Automatic Brain Shunt Valve Recognition using Feature Engineering and Deep Convolutional Neural Networks

 

N

Precision

Recall

F1-score

 

Strata II - NSC

164

0.98

0.96

0.97

***p < 0.001

Codman-Hakim

106

0.92

0.98

0.95

***p < 0.001

Sophysa Polaris SPV

82

0.98

1.00

0.99

***p < 0.001

Codman-Certas

42

0.95

0.86

0.90

***p < 0.001

Miethke proGAV

22

1.00

0.95

0.98

***p < 0.001

  1. N = number of samples per class. (Precision is also known as positive predictive value; recall is also known as sensitivity; F1-score is the harmonic mean of precision and recall). The p-value is computed using the two-tailed Mann–Whitney U test to reject the null hypothesis that each valve is indistinguishable from the others in a 1 vs all framework.
Back to article page