Key result
The proposed DASMcC model using XG Boost achieved an overall accuracy of 93.0% and an F1 score of 93.0% in classifying four main types of cardiovascular diseases and normal ECGs.
Why the study?
Cardiovascular disease is a leading cause of global mortality, and noninvasive 12-lead ECG can be leveraged to predict multiple types of cardiovascular diseases.
Does the DASMcC model improve the prediction of cardiovascular diseases from 12-lead ECGs compared to other machine learning classifiers?
Population
12-lead ECG signals
Comparison
Five classifiers: Random Forest, KNN, Gradient Boost, Adda Boost, and XG Boost
Design
Machine learning model development and validation study
Authors
Loading...
May support ML-based ECG classification in smart healthcare; leaves open prospective clinical validation before adoption.
Does the DASMcC model improve the prediction of cardiovascular diseases from 12-lead ECGs compared to other machine learning classifiers?
A novel multi-class classifier using data augmentation and SMOTE on 12-lead ECG features was developed for predicting various cardiovascular diseases.
Sinha et al. (2023) studied Cardiovascular Diseases (n=18,885). DASMcC (Data Augmented SMOTE Multi-class Classifier) using XG Boost vs. Other machine learning classifiers (Random Forest, Cat-Boost, Gradient Boost, KNN) was evaluated on Overall accuracy. The proposed DASMcC model using XG Boost achieved an overall accuracy of 93.0% and an F1 score of 93.0% in classifying four main types of cardiovascular diseases and normal ECGs.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: