Skip to main content

Table 3 Performance of machine learning and deep learning models in predicting KILLIP class (four-class classification). Fivefold cross-validation was performed in all the 1574 patients

From: Interpretable artificial intelligence model for predicting heart failure severity after acute myocardial infarction

 

F1

Precision

AUPRC

AUROC

RF

0.786 ± 0.022

0.788 ± 0.009

0.674 ± 0.025

0.797 ± 0.012

XGboost

0.738 ± 0.023

0.761 ± 0.011

0.663 ± 0.027

0.783 ± 0.006

MLP

0.771 ± 0.022

0.775 ± 0.008

0.634 ± 0.028

0.814 ± 0.009

TabNet

0.783 ± 0.024

0.787 ± 0.012

0.684 ± 0.030

0.827 ± 0.005

  1. Note: The values for all evaluation metrics are calculated using weighted averages. Results are presented as mean ± standard deviation across 5 stratified folds (random seed = 42)
Back to article page

BMC Cardiovascular Disorders

ISSN: 1471-2261

Contact us