Table 4 Comparison matrix of AUC for the five models in training set and validation set.

From: Development of hypertension models for lung cancer screening cohorts using clinical and thoracic aorta imaging factors

Models

AIMeasure model

BasicClinical model

TotalClinical model

AIBasicClinical model

AITotalClinical model

AIMeasure model

1.000

0.608

0.172

0.616

0.054

BasicClinical model

0.066

1.000

0.176

0.968

0.114

TotalClinical model

 < 0.001

 < 0.001

1.000

0.156

0.432

AIBasicClinical model

0.031

0.047

 < 0.001

1.000

0.101

AITotalClinical model

 < 0.001

 < 0.001

0.207

 < 0.001

1.000

  1. In the bottom left corner, the chart illustrates the differences in AUC among the five models in the training set, while the italics in top right corner depicts the differences in AUC among the five models in the validation set. The bold in diagonal line represents self-comparisons of the five models. And Delong test was applied for all comparisons.
Back to article page