Detection of Heart Abnormalities Using Signal Processing

International Journal of Research and Innovation in Applied Science (IJRIAS) | Volume VI, Issue XI, November 2021|ISSN 2454-6194

Detection of Heart Abnormalities Using Signal Processing

Robinson, Mbato and Kabari, Ledisi G.
Ignatius Ajuru University of Education, Port Harcourt, Nigeria

Abstract: The heart is the center of life. It pumps and distributes blood to every other part of the body. Thus, it holds a strategic position in the body and must be in perfect condition at all times to perform these operations. The Electrocardiogram (ECG) is used to demonstrate the circuit activity of the heart. However, ECG signals can be difficult to interpret especially from non-health professionals. In this work, we developed a model that can detect and interpret the characteristics of an ECG signal, hence, identifying non-linearity of the heart. Fast Fourier Transform was used to filter our ECG readings dataset and remove unwanted signals, before the signals were used for classification and calculation of heart rate using peak values/intervals. The dataset contained about 218,000 ECG readings, including gender and age grades of the patients. Object Oriented Analysis and Design Methodology (OOADM) was adopted in this approach. The system was implemented using MATLAB software. The overall efficiency of the model is 95%, which outperforms other existing models. This system could be beneficial to the research community on signal processing.

I. INTRODUCTION

Due to the gravely important functions of the heart in the human body such as pumping of blood, maintaining blood pressure etc, it is always important to ensure it is kept in the best condition possible at all time. The electrocardiogram (ECG) is a medical test that is used to detect cardiac or heart abnormalities. The ECG is measured in terms of a voltage against time graph of the electrical activity of the heart. The normal reading of the heart is 120 – 200 m/s. While the normal heart rate is 60 – 100 beats per minute. The ECG is also used to track some heart disorders such as reduced blood flow, high blood flow, delayed or fast beats per second etc. The ECG signal is generated by the expansion and contraction of the heart.

Interpreting or detecting the ECG signal can be a tedious task [1]. To overcome this limitation, we proposed fast and accurate classifier that simulates the diagnosis of the cardiologist to classify the ECG signals into normal and abnormal from a single lead ECG signal and better than other well-known classifiers.

The ECG signal is normally noisy therefore it is important to denoise the signal before applying the classification models to it. Some of the noises associated with ECG signal include Baseline Wander, Powerline interference, EMG noise and electrode motion [2]. While making a choice for a denoising technique, we should be concerned about retaining the validity of the signal after the noise has been removed.

Read More