Journal of Biosciences and Medicines, 2017, 5, 64-79
http://www.scirp.org/journal/jbm
ISSN Online: 2327-509X
ISSN Print: 2327-5081
DOI: 10.4236/jbm.2017.511008 Nov. 27, 2017 64 Journal of Biosciences and Medicines
Classification of Cardiovascular Disease
Using Feature Extraction and Artificial
Neural Networks
Shalin Savalia
*
, Eder Acosta, Vahid Emamian
Department of Electrical Engineering, St. Mary’s University, San Antonio, USA
Abstract
Electrocardiogram (ECG) signals are used to identify cardiovascular disease.
The availability of signal processing and neural networks techniques for
processing ECG signals has inspired us to do research that consists of extract-
ing features of an ECG signals to identify types of cardiovascular diseases. We
distinguish between normal and abnormal ECG data using signal processing
and neural networks toolboxes in Matlab. Data, which are downloaded from
an ECG database, Physiobank, are used for training and testing the neural
network. To distinguish normal and abnormal ECG with the significant accu-
racy, pattern recognition tools with NN is used. Feature Extraction method is
also used to identify specific heart diseases. The diseases that were identified
include Tachycardia, Bradycardia, first-degree Atrioventricular (AV), and
second-degree Atrioventricular. Since ECG signals are very noisy, signal
processing techniques are applied to remove the noise contamination. The
heart rate of each signal is calculated by finding the distance between R-R in-
tervals of the signal. The QRS complex is also used to detect Atrioventricular
blocks. The algorithm successfully distinguished between normal and abnor-
mal data as well as identifying the type of disease.
Keywords
Electrocardiogram (ECG), Cardiovascular Disease, MATLAB,
Artificial Neural Network, Physiobank, R-R interval, Matlab,
QRS Complex, Atrioventricular, Tachycardia, Bradycardia
1. Introduction
An electrocardiogram (ECG) is a measure of how the electrical activity of the
heart varies with respect to time as action potentials propagate throughout the
How to cite this paper: Savalia, S., Acosta,
E. and Emamian, V. (2017) Classification of
Cardiovascular Disease Using Feature Ex-
traction and Artificial Neural Networks.
Journal of Biosciences and Medicines, 5,
64-79.
https://doi.org/10.4236/jbm.2017.511008
Received: October 19, 2017
Accepted: November 24, 2017
Published: November 27, 2017
Copyright © 2017 by authors and
Scientific Research Publishing Inc.
This work is licensed under the Creative
Commons Attribution International
License (CC BY 4.0).
http://creativecommons.org/licenses/by/4.0/
Open Access