Skip to main content
Insights into Imaging

Table 4 Metric results for each class of the classification model

From: An active learning approach to train a deep learning algorithm for tumor segmentation from brain MR images

Class

Sensitivity

Specificity

Positive predictive value

F-score

AUC

Poor Quality

(4 cases)

0.750

1.000

1.000

0.857

0.995

Acceptable with Adjustments

(11 cases)

0.545

0.938

0.316

0.400

0.810

Acceptable Quality

(85 cases)

0.859

0.455

0.936

0.896

0.877

Back to article page