Skip to main content

Table 4 Accuracy of the models on the five-fold cross-validation test

From: Ultrasound-based radiomics machine learning models for diagnosing cervical lymph node metastasis in patients with non-small cell lung cancer: a multicentre study

Models

Average

Fold-1

Fold-2

Fold-3

Fold-4

Fold-5

Semantic-LR

0.878

0.839

0.873

0.889

0.839

0.952

Radiomics-LR

0.901

0.968

0.952

0.905

0.825

0.855

Combineda-LR

0.907

0.871

0.873

0.952

0.903

0.937

Radiomics-RF

0.894

0.935

0.952

0.825

0.871

0.889

Combined-RF

0.898

0.967

0.889

0.855

0.857

0.921

  1. All statistics were validated in the test set of one institution
  2. AUC the area under the curve, CI confidence interval, LR logistic regression, RF random forest
  3. asemantic-radiomics combined
Back to article page

BMC Cancer

ISSN: 1471-2407

Contact us