Why the study?
Reduced in-hospital mortality after PCI has caused class imbalance that complicates risk prediction, and machine learning models providing both high accuracy and personalized risk assessment are lacking.
Does a machine learning model using LightGBM and SMOTE accurately predict in-hospital mortality in AMI patients post-PCI?
Population
1693 patients diagnosed with AMI post-PCI
Comparison
Six machine learning algorithms using SMOTE, Boruta, and GSCV
Design
Retrospective study
Authors
Loading...
ML models may enable accurate personalized IHM prediction after AMI; leaves open prospective validation before clinical use.
Does a machine learning model using LightGBM and SMOTE accurately predict in-hospital mortality in AMI patients post-PCI?
A machine learning approach utilizing LightGBM and SMOTE for class balancing demonstrated high accuracy in predicting in-hospital mortality among AMI patients post-PCI.
Lei et al. (2025) studied this question.