07-05-2011, 03:22 PM
Abstract
Electrocardiogram (ECG) is one of the most importantparameters for heart activity monitoring. The main objective ofdigital signal processing of ECG signal is to deliver accurate, fastand reliable estimation of clinically important parameters such asthe duration of the QRS complex, the R-R interval, theoccurrence, amplitude and duration of the P, R, and T waves. Inthis paper, I have measured all these parameters by using pan-Tompkins’s algorithm. It is a real time QRS detection algorithm.It gives the number of QRS peaks for recorded ECG signals.Results of simulations in MATLAB are presented.
Keywords-component; pan-tompkin’salgorithm; band-passfilter; differentiator; integrator; moving-window.
I. INTRODUCTION
In many applications for biomedical signal processing theuseful signals are superposed by different components. So,extraction and analysis of the information-bearing signal arecomplicated, caused by distortions from interference. Usingadvanced digital signal processing this task can be solved.Pan-Tompkin’s algorithm is a real time algorithm which isconsists of band-pass filter, differentiator, integrator andmoving-window. The electrocardiogram (ECG) provides aphysician with a view of the heart’s activity through electricalsignals generated during the cardiac cycle, and measured withexternal electrodes. Its clinical importance in cardiology iswell established, being used for example to determine heartrate, investigate abnormal heart rhythms, and causes of chestpain.
II. ECG COMPONENTS
The electrocardiogram (ECG) provides a physician with aview of the heart’s activity through electrical signals generatedduring the cardiac cycle, and measured with externalelectrodes. Its clinical importance in cardiology is wellestablished, being used for example to determine heart rate,investigate abnormal heart rhythms, and causes of chest pain.As shown in Figure 1, the most important ECG signal featuresin a single cardiac cycle are labeled (along with thephysiological cause of that feature):• “P” wave - due to depolarization of the atria• “Q” wave - due to activation of the anterior septal region ofthe ventricular myocardium• “R” wave - due to depolarization of the ventricularmyocardium• “S” wave - due to activation of the posterior basal portion ofthe ventricles• “T” wave - due to rapid ventricular repolarizationBecause the QRS complex is the major feature of an ECG, agreat deal of clinical information can be derived from itsfeatures. Identification of this feature in an ECG is known inthe literature QRS detection, and it is a vital task in automatedECG analysis, portable arrhythmia monitoring, and manyother applications. Though trivial in an “ideal” ECG (as shownin Figure 1), the range in quality of real-world ECG signalsobtained from a variety of subjects under differentmeasurement conditions makes this task much more difficult.Fig. 1 An ideal ECG beat from a healthy subject (time duration equivalent toone Heartbeat).
III. PAN-TOMPKIN’S ALGORITHM DESIGN ANDSIMULATION
A. BAND-PASS FILTER
The band pass filter reduces the influence of muscle noise,60Hz interference, baseline wander, and T-wave interference.The desirable pass band to maximize the QRS energy isapproximately 5-15Hz. Our filter is a fast, real-time recursivefilter in which poles are located to cancel zeros on the unitcircle of the z-plane. This approach results in a filter designwith integer coefficients. Since only integer arithmetic isnecessary, a real-time filter can be implemented with a simpleISSN : 0975-3397 2502Sachin singh et al. / (IJCSE) International Journal on Computer Science and EngineeringVol. 02, No. 07, 2010, 2502-2505microprocessor and still has available computing power left todo the QRS recognition task.This class of filters having poles and zeros only on the unitcircle permits limited pass band design flexibility. For ourchosen sample rate, we could not design a band pass filterdirectly for the desired pass band of 5-15 Hz using thisspecialized design technique. Therefore, we cascaded the lowpassand high-pass filters described below to achieve a 3 dbpass band from about 5-12 Hz, reasonably close to the designgoal.
Download full report
http://www.enggjournalsijcse/doc/IJCSE10-02-07-63.pdf