Assessment of Heart Rate Variability using Independent Component Analysis
S. Thulasi Prasad1, S. Varadarajan2

1S. Thulasi Prasad, Department of ECE, Sree Vidyanikethan Engineering College, Tirupati, India.
2Dr. S. Varadarajan, Department of ECE, Sri Venkateswara University College of Engineering, Tirupati, India.
Manuscript received on 06 March 2015 | Revised Manuscript received on 26 March 2015 | Manuscript Published on 30 March 2015 | PP: 48-52 | Volume-4 Issue-10, March 2015 | Retrieval Number: J20010341015/15©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: According World Health Organisation reports, it is understood that cardiovascular diseases are increasing at an alarming rate and becoming main cause for more deaths. The early detection of cardiac related deceases is essential to save a patient from death. The ECG signal plays a key role in the early detection and diagnosis. In recent years there have been wide-ranging studies on Heart rate variability in ECG signals and Digital Signal Processing is becoming as an essential and effective pedagogical approach to solve a problem of detecting selected arrhythmia conditions from a patient’s electrocardiograph (ECG) signals. Normally the Heart rate variability is studied based on RR interval and used analyse the sympathetic-parasympathetic autonomic stability, the risk of unpredicted cardiac death. Even there are several methods to analsye the ECG signal, the Blind Source Separation (BSS) approach is very useful and successful in extracting a cleaned ECG signal from a ECG which is mutilated badly by noise. The BSS approach, it is intended to estimate a set of underlying source signals of physiological activity from the sole observation of unknown mixtures of the sources. In this paper, first we addressed Independent Component Analysis (ICA) to remove noise and artifacts from ECGs. In the second step the noise free ECG signal is reconstructed from desired Independent Components. Finally QRS complexes, R peaks, RR intervals, and HR were found using suitable algorithms and performed a statistical analysis to finding HR Variability (HRV). This method is tested on ECG signals from in MIT-BIH Arrhythmia database.
Keywords: Arrhythmia, AV node, ECG, HRV, ICA, MATLAB, QRS, RR interval, SA node

Scope of the Article: ECG signals