Key result
A support vector machine classifier using heart rate variability indices classified NYHA functional classes with 84.0% accuracy and 86.4% AUC, outperforming the CART method (81.4% accuracy).
Why the study?
This study evaluated the effect of heart rate variability indices on NYHA functional classification in congestive heart failure and tested the performance of different machine learning algorithms.
Does the combination of HRV indices and machine learning algorithms accurately classify NYHA functional classes in patients with congestive heart failure?
Comparison
Support vector machine vs classification and regression tree classifiers
Design
Database-derived comparative algorithm development and validation study
Authors
Loading...
SVM-HRV classification may aid NYHA assessment in heart failure; leaves open prospective validation before clinical adoption.
Observational (n=29)
Does the combination of HRV indices and machine learning algorithms accurately classify NYHA functional classes in patients with congestive heart failure?
Absolute Event Rate: 84% vs 81.4%
Machine learning algorithms, particularly SVM, combined with heart rate variability indices can accurately classify NYHA functional classes in patients with congestive heart failure.
Qu et al. (2019) conducted an observational in Congestive heart failure (n=29). Support vector machine (SVM) classifier vs. Classification and regression tree (CART) classifier was evaluated on Classification accuracy of NYHA functional classes. A support vector machine classifier using heart rate variability indices classified NYHA functional classes with 84.0% accuracy and 86.4% AUC, outperforming the CART method (81.4% accuracy).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: