Skip to main content

Table 2 Results of model output indicators in all participants

From: A machine learning-based model for predicting paroxysmal and persistent atrial fibrillation based on EHR

Model

ACC (95%CI)

Precision

Recall

AUC (95%CI)

F1 Score (95%CI)

SEN (95%CI)

SPE (95%CI)

GradientBoost

0.801 (0.775–0.821)

0.730

0.716

0.870 (0.858–0.882)

0.722 (0.686–0.754)

0.716 (0.676–0.755)

0.849 (0.816–0.872)

AdaBoost

0.796 (0.785–0.817)

0.727

0.700

0.858 (0.836–0.877)

0.713 (0.694–0.747)

0.700 (0.673–0.747)

0.851 (0.844–0.858)

XGBoost

0.782 (0.764–0.802)

0.701

0.697

0.858 (0.849–0.872)

0.699 (0.665–0.732)

0.697 (0.648–0.747)

0.831 (0.815–0.843)

  1. ACC, Accuracy; AUC, Area under curve; CI, Confidence interval; SEN, Sensitivity; SPE, Specificity