Skip to main content

Table 3 The performance of the three models in the test set

From: Prediction of all-cause mortality in coronary artery disease patients with atrial fibrillation based on machine learning models

Models

Sensitivity (95% CI)

Specificity (95% CI)

Accuracy

AUC (95% CI)

Regularization logistic regression

0.725 (0.561–0.854)

0.699 (0.660–0.737)

0.936

0.732 (0.649–0.816)

Random forest

0.750 (0.588–0.873)

0.663 (0.622–0.701)

0.935

0.728 (0.642–0.813)

Support vector machines

0.675 (0.509–0.814)

0.668 (0.628–0.706)

0.935

0.712 (0.630–0.794)