Key result
An artificial neural network model predicted the occurrence of tachyarrhythmia after acute myocardial infarction with an AUC of 0.597, demonstrating better performance than the GRACE variable set model.
Why the study?
Early identification of arrhythmia in acute myocardial infarction patients is essential for clinical decision-making, motivating the use of machine learning methods to build predictive models.
Does a machine learning model improve the prediction of tachyarrhythmia during admission in patients with acute myocardial infarction compared to the GRACE risk score variable set?
Observational (n=2,084)
No
Does a machine learning model improve the prediction of tachyarrhythmia during admission in patients with acute myocardial infarction compared to the GRACE risk score variable set?
Absolute Event Rate: 0.597% vs 0.575%
Background: Early identification of the occurrence of arrhythmia in patients with acute myocardial infarction plays an essential role in clinical decision-making. The present study attempted to use machine learning (ML) methods to build predictive models of arrhythmia after acute myocardial infarction (AMI). Methods: A total of 2084 patients with acute myocardial infarction were enrolled in this study. The primary outcome is whether tachyarrhythmia occurred during admission containing atrial arrhythmia, ventricular arrhythmia, and supraventricular tachycardia. All data is randomly divided into training set (80%) and internal testing set (20%). Three machine learning algorithms (including decision tree, random forest, and artificial neural network) learn from the training set to build a model, use the testing set to evaluate the prediction performance, and compare it with the model built by the variable set involved GRACE risk score. Results： Three ML models predict the occurrence of tachyarrhythmia after AMI. After variable selection, the artificial neural network (ANN) model achieves the highest accuracy of 0.654 (95% CI, 0.625--0.683). The area under the value of the curve (AUC) is 0.597 (95% CI, 0.568-0.626). The highest accuracy of the model built using the Grace variable set is 0.627 (95% CI, 0.598-0.656), and the AUC value is 0.574 (95% CI, 0.545-0.603). Conclusions： We used advanced machine learning methods to build prediction models for tachyarrhythmia after AMI for the first time (especially the ANN model has the best performance). The current study can supplement the current AMI risk score, provide a reliable evaluation method for the clinic, and broaden the new horizons of ML and clinical research. Trial registration： Clinical Trial Registry No.: ChiCTR2100041960.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2021) conducted an observational in Acute myocardial infarction (n=2,084). Artificial Neural Network (ANN) model vs. GRACE variable set model was evaluated on Area under the curve (AUC) for predicting tachyarrhythmia (95% CI 0.568-0.626). An artificial neural network model predicted the occurrence of tachyarrhythmia after acute myocardial infarction with an AUC of 0.597, demonstrating better performance than the GRACE variable set model.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: