Noise robust automatic heartbeat classification system using support vector machine and conditional spectral moment
- PDF / 831,701 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 103 Downloads / 164 Views
SCIENTIFIC PAPER
Noise robust automatic heartbeat classification system using support vector machine and conditional spectral moment Pratik Singh1 · Gayadhar Pradhan2 Received: 9 August 2020 / Accepted: 12 November 2020 © Australasian College of Physical Scientists and Engineers in Medicine 2020
Abstract Heartbeat classification is central to the detection of the arrhythmia. For the effective heartbeat classification, the noise-robust features are very significant. In this work, we have proposed a noise-robust support vector machine (SVM) based heartbeat classifier. The proposed classifier utilizes a novel noise-robust morphological feature which is based on the conditional spectral moment (CSM) of the heartbeat. In addition to the proposed CSM feature, we have also employed the existing RR interval, the wavelets, and the higher-order statistics (HOS) based temporal and morphological feature sets. The noiserobustness test of the proposed CSM and all the studied feature sets is performed for the SVM based heartbeat classifier. Further, we have studied the significance of combining these temporal and morphological features on the final classification performance. For this purpose, the individual SVMs were trained for each of the feature set. The final classification is based on the ensemble of these individual SVMs. Various combining scheme such as sum, majority, and product rules are employed to ensemble the result of the individually trained SVMs. The experimental results show the noise-robustness of the proposed CSM feature. The proposed classifier gives improved overall performance compared to the existing heartbeat classification systems. Keywords Electrocardiogram · Conditional spectral moment · Morphological feature · Support vector machine · Heartbeat classification
Introduction The automatic heartbeat classification system is primal for the detection of arrhythmia and to perform the cardiac diagnosis [1]. Heartbeat extracted from the electrocardiogram (ECG) signal is processed through several stages to get the final classification result for the Arrhythmia detection. The key component of this heartbeat classification system includes the preprocessing stage, the feature extraction stage, and the classifier stage. Every individual stage contributes
* Pratik Singh [email protected] Gayadhar Pradhan [email protected] 1
Department of Electronics and Communication Engineering, National Institute of Technology, Tadepalligudem, Andhra Pradesh 534101, India
Department of Electronics and Communication Engineering, National Institute of Technology, Patna 800005, India
2
towards the improvement of the accuracy of the heartbeat classification system. In the preprocessing stage, the ECG signals from the training and testing datasets are denoised for further processing. The noises responsible for corrupting these ECG signals are powerline interference (PLI), baseline wander (BW), muscle artifact (MA), electrode motion (EM), and additive white Gaussian noise (AWGN) [1]. These noises affect different frequen
Data Loading...
