Key result
A machine learning approach combining discrete wavelet transform, Hjorth descriptors, and entropy-based features achieved 100% accuracy in classifying atrial fibrillation, congestive heart failure, and normal sinus rhythm.
Why the study?
Noise interference in ECG signals can prevent standard Hjorth descriptor features from accurately distinguishing between AF, CHF, and normal sinus rhythm.
Population
ECG signals of AF, CHF, and normal sinus rhythm conditions
Comparison
Optimization of k-NN, SVM, RF, ANN, and RBFN classifiers using DWT-based Hjorth and entropy features
Design
Machine learning classification study
Authors
Loading...
Requires prospective clinical validation before use; leaves open real-world generalizability and outcome impact.
Absolute Event Rate: 100% vs 92%
The application of discrete wavelet transform combined with Hjorth descriptor and entropy-based features achieves up to 100% accuracy in classifying ECG signals for atrial fibrillation and congestive heart failure.
Fuadah et al. (2022) studied Atrial Fibrillation and Congestive Heart Failure (n=150). Machine learning classification using DWT, Hjorth descriptors, and entropy-based features vs. Classification using Hjorth descriptor features alone was evaluated on Classification accuracy. A machine learning approach combining discrete wavelet transform, Hjorth descriptors, and entropy-based features achieved 100% accuracy in classifying atrial fibrillation, congestive heart failure, and normal sinus rhythm.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: