A Deep Neural Network classifier utilizing hybrid T-wave features demonstrated high accuracy (98.33%) in detecting arrhythmias from ECG signals, offering a potential automated tool to assist physicians.
Hybrid feature ECG classification may aid automated arrhythmia detection; leaves open prospective clinical validation before practice change.
An Electrocardiogram (ECG) is used as one of the important diagnostic tools for the detection of the health of a heart. An automatic heart abnormality identification methods sense numerous abnormalities or arrhythmia and decrease the physician's pressure as well as share their work load. In ECG analysis, the main focus is to enhance degree of accuracy and include a number of heart diseases that can be classified. In this research paper, arrhythmia classification is proposed using Hybrid features of T-wave in ECG. The classification system consists of majorly three phases, windowing technique, feature extraction and classification. In feature extraction phase various features are used such as Differential Entropy (DE), Peak-Magnitude RMS ratio, Auto Regressive feature based Yule Walker, Burgs method. In classification phase Deep Neural Network (DNN) classifier is used. This classifier categorizes the normal and abnormal signals efficiently. The experimental analysis showed that, the Hybrid features Arrhythmia classification performance of accuracy approximately 98.3%, Specificity 98.0% and Sensitivity 98.6% using MIT-BIH database.
No takes yet. Share an insight, caveat, or question.
Nanjundegowda et al. (2018) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: