International Journal of Computer Applications (0975 – 8887) Volume 119 – No.10, June 2015 11 Arrhythmia Detection Technique using basic ECG Parameters Mohammad Rakibul Islam, Rifad Hossain, Md. Ziaul Haque Bhuiyan, Tahmeed Ahmed Margoob, Md. Taslim Reza, Kazi Khairul Islam Department of Electrical and Electronic Engineering, Islamic University of Technology, Boardbazar, Gazipur-1704, Bangladesh. ABSTRACT A condition in which the heart beats with an irregular or abnormal rhythm is known as Arrhythmia. This paper presents a procedure to extract information from Electrocardiogram (ECG) data & determine types of Arrhythmias. The decisions were achieved by determining different intervals such as PR Interval, RR Interval, Heart Rate (HR) etc. and those intervals were compared with the ideal intervals. During the whole process MATLAB was used & ECG signals were taken from PhysioBank ATM. In this process Savitzky–Golay filter was used to reduce the noise of the signal. Tachycardia, Bradycardia, Heart Block, Junctional Arrhythmia, Premature Articular Contraction were detected during this analysis. The results show simplified detection of arrhythmia with 90%accuracy. General Terms Biomedical Signal Processing Keywords Electrocardiogram, Arrhythmia, PR Interval, RR Interval, Heart Rate. 1. INTRODUCTION Arrhythmia is the abnormal rhythm of heart. It is also known as dysrhythmia. It causes the heart to pump less effectively. There are a lot of changes in the shape of the heart wave because of arrhythmia. ECG is a common term in the diagnosis of cardiac diseases. It provides information about the electrical activity of the heart. We can detect different kinds of heart diseases by analyzing the ECG signal. Higher efficiency in classifying ECG signal is very important nowadays. Detection of actual type of heart diseases is very important for further treatment. Heart disease is very dangerous for a human being. So, proper treatment is a must in this case. There are different sorts of arrhythmias. Different kinds of arrhythmias can be detected in different parts of the heart. Heart pumps blood in a regular way. But when it is affected by arrhythmia, it can’t pump blood normally. Right bundle branch block (RBBB), left bundle branch block (LBBB), premature ventricular contractions (PVC), ventricular fibrillation (VF) are some serious arrhythmias. However ECG being a non-stationary signal, the irregularities may not be periodic and may not show up all the time, but would manifest at certain irregular intervals during the day. So, continuous ECG monitoring permits observation of cardiac variations over an extended period of time, either at the bad side or when patients are ambulatory, providing more information to physician. The heart rate and the morphology reflect the cardiac health of human heart beat [1]. It is a noninvasive method which means this signal is measured on the surface of human body, which is used in identification of the heart diseases [2]. Electrocardiography is the recording of the electrical activity of the heart. A typical ECG tracing of the cardiac cycle (heartbeat) consists of a P wave, a QRS complex, a T wave, and a U wave is shown in Fig. 1. Fig. 1 Typical ECG Signal For Arrhythmia detection we went through the Time Domain based technique. The technique is followed by ECG signal processing, determination of PR Interval, QRS Interval, QT Interval, ST segment, RR Interval (To determine Heart Rate) followed by Arrhythmia detection via some decision making rules. Table 1 shows normal ECG signal characteristics for different parameters. This paper is organized as follows. Section II describes different arrhythmia detection techniques where as in section III discuss problem formulation. ECG based arrhythmia detection technique is proposed in section IV. Results are discussed in section V whereas Section VI concludes the paper. Table 1: Normal ECG Signal Characteristics Component Characteristics Heart Rate 60-100 bpm PR Interval 0.12–0.20 sec QRS Interval 0.06–0.10 sec QT Interval Less than half of the R-R interval ST segment 0.08 sec