A COMPARATIVE STUDY ON ATRIAL FIBRILLATION ECG WITH THE NORMAL LIMITS

Transcription

1 Volume 119 No , ISSN: (on-line version) url: A COMPARATIVE STUDY ON ATRIAL FIBRILLATION ECG WITH THE NORMAL LIMITS Thomas Benyamin, Raghunath, Mohammad Mishal, S.Sathish Department of Biomedical Engineering, Vel Tech Multi Tech Dr.Rangarajan Dr.Sakunthala Engineering College, Chennai, Tamil Nadu, India. sathishdir@gmail.com ABSTRACT Atrial fibrillation is a shuddering or sporadic heartbeat, where the upper council of heart beat sporadically as opposed to pulsating viably to move blood into the ventricles. The unevenness of cardiovascular movement influences the ordinary sinus pulse that prompts heart arrhythmias. Among them the issues related with atria are at most extreme which causes stroke, blood cluster, AV square which is serious and dangerous in India. The event of conflicting P-wave is noted from the recorded electrical action of heart, it is because of the imperfections that show up in atrial depolarization stage. Keeping in mind the end goal to look at P-Ta portion PC based procedure are proposed to gauge with high affectability (>99%) and specificity (>99%) and contrasted and those of regular ECG. MatLab is executed continuously by signal average. Furthermore, can effectively recognize P-Ta portion for a wide assortment of ECG shapes and confused signals that prompt the treatment took after by conveying low vitality atrial defibrillation streams.. Index Terms: Inconsistent P-waves, stroke, signal average, P-Ta segment, low energy atrial defibrillation current. 1. INTRODUCTION An electrocardiogram (ECG) records the electrical activity of heart. It provides vital information about the wide range of Cardiac disorders depending upon the deviations from normal ECG pattern. The ECG signal is characterized by five peaks 497

2 and valleys labeled by the letters P, Q, R, S, T.In general, the frequency range of an ECG signal varies between Hz with the dynamic range between 1 10 mv. The QRS complex is the most prominent waveform in the electrocardiographic (ECG) signal, with normal duration from 0.06 s to 0.1 s.1[9] ECG analysis depends directly on the heart beat segmentation results. Data obtained from the ECG signals provides valuable tools for diagnosing cardiac disorders. TERMINOLGY Paroxsyal Persistent CLINICAL FEATURES Spontaneous Terminating Not Self Terminating ARRYHYTHMIA PATTERN Recurrent Recurrent Among them the major problem seen in patient is with atrial fibrillation. The atrial fibrillation is characterized predominantly by uncoordinated atrial activation with consequent deterioration of atrial mechanical function. [2][6] 1.1 Classification of atrial fibrillation AF has heterogenic clinical presentation. It may occur in the present and absent of detectable heart disease and also with its symptoms. An AF may be self-terminating or require medical intervention for termination. Although the pattern of arrhythmia may change with respect to time, so the clinical value helps to characterize it.[2] The types of atrail fibrillation occurring is briefly explained in the table 1. Permenent Terminating But Relapsed Established Table 1: Types of Atrial fibrillation 2. PREPROCESSING: Generally the recorded ECG signal is often contaminated by different types of noises and artifacts which may change the characteristics of ECG signal. The signal is generally corrupted by unwanted interface such as motion artifacts, muscle noise, electrode artifacts, base line drift line noise, and respiration.[6] 2.1 Power Line Interference It is an environmental noise in the ECG that results from poorly grounded ECG recording machine. Because of the alternating current feature, this noise appears at 60Hz and its harmonics. 498

3 2.2 Baseline Wander Noise Baseline drift is a biological noise that appears in ECG signal. The fig 1 shows ECG signal with the base line wandering removed. Fig1: Removal of Baseline Drift 2.3 Electromyography Noise Electromyography noise is caused by the contraction of other muscles besides the heart. When other muscles in the vicinity of the electrodes contract, they generate depolarization and repolarization waves that can also be picked up by the ECG. The frequency of this EMG noise is in between Hz. 2.4 Motion Artifacts Noise The motion artifacts noise is caused by the improper positioning of the leads. Due to this the movement of the leads produces some error. The fig 2 shoes the artifacts signal that can interfere while picking the ECG signals. (a) (b) Fig2: Artifact Signals 3. METHODOLOGY The collected ECG samples are to be diagnosed whether the patient is suffering from atrial fibrillation. So they are inserted in the MATLAB software for analysis in the form of txt format (.txt). The principle to find the abnormality is achieved by signal averaging technique. So for this process the methods undertaken are- Normalisation, Detrending, Smoothening, Wavelet filtering, Windowing.[11] Normalization: The mathematical function for importing ECG samples in Matlab. The function involved is nyquist frequency which helps in the selection of required subsets.[9] Detrending: This is done to remove the baseline wander noise from the standard ECG signal. Smoothening: This is use to filter the subset ECG signal with the help of savitzki-golay filter.[10] Wavelet filtering: this helps in finding the atrial peak of the ECG signal for accurate diagnosis.[11] Windowing: This helps in denoising the signal with the help of empirical mode decomposition technique[12]. 499

4 3.1 Signal averaging The fig 3 shows the signal averaging of the signal. In this method the noise bandwidth is completely separated from the signal bandwidth. So the noise can easily be discarded by applying a linear FFT filter. On the other hand, the noise bandwidth overlaps the signal bandwidth, and the noise amplitude is larger than the signal. For this situation, a FFT Filter would need to discard some of the signal energy in order to remove the noise, thereby distorting the signal.[12] This is done by the process of Thresholding. Subject Measure ment Amplitude (mv) Time Period(ms) Normal Abnormal Normal Abnormal 1. P P-Ta P P-Ta P P-Ta P P-Ta Fig3: signal averaging of ecg 3.2 Analysis The waveforms were analyzed using Matlab software.[14] By the method of signal averaging the threshold values is been maintained with the normal ECG samples. When the difference of threshold is been noted the occurrence of fibrillation is conformed. The fig 4 represents the 5. P P-Ta

5 from the amplitude of P wave taken for both normal and mimic conditions than 12 lead system. The table 3 explains about the mean and standard deviation of P wave and P-Ta interval. The mean age of the subjects chosen was 21 and the mean and standard deviation values are measured Table 2: Amplitude And Time Period Of 5 Male Samples Of Mean Age 20 Fig4: processing of ECG signal. Initially 5 male subjects are chosen and the time period and amplitude of normal condition and mimic condition is noted using the Modified limb lead system[14]. Because the normal lead system focus on the ventricular components rather than the ventricular component. The mimic is given by various exercise like push up and running The values are noted in a table below. From the table 2 it is clearly inferred that the time duration of the P-Ta segment decreases in the case of mimic condition than the normal. Usage of MLL enhanced the atrial components which could be seen And found that the P-Ta amplitude is 124 ms with a standard deviation of 23ms for normal cases and for abnormal cases it ranges from 135 with standard deviation of 30.7 ms. And the amplitufde of the p wave increases, which was about 164 µv with standard deviation of ±21.9 µv for normal condition and 184 µv with a standard deviation of 25.7 µv. AMP LITU DE (mv) Measur ement Mean Nor mal Abno rmal Standard deviation Normal Abnor mal P P-Ta

6 TIME PERI OD (ms) P P-Ta Table 3: Mean And Standard Deviation Of 5 Samples The fig 5 shows is the input signal that is taken from the subject using modified limb lead sytem which shows a enhanced P-Ta interval when compared to conventional 12 lead system. fig 5: ECG Wave Obtained Using Modified Limb Lead System The fig 6 shows the first step of the processing a signal is preprocessing which includes filtering and the normalization. The filtering is done to remove the noises like EMG interference, power line interference, base line wandering and baseline drift. The normalization of the signal is done to bring the values of the signal to the manageable limits Fig 7: The Filtered And Normalized ECG Signal 5. RESULT: The atrial Ta wave is clearly enhanced using Modified limb lead system. The windowing and signal averaging technique shows segmented P- Ta interval. For the normal subjects the time interval is 124 ± 23 ms and of the fibrillated ECG is 134 ± 30 ms. By the comparing the time intervals of normal and atrial fibrillation patients there is significant change in time duration. It is clearly seen that the time interval of P-Ta interval abnormally varies with atrial fibrillation patients when compared with normal subjects. A database created using this can be used for the clinical diagnosis in the hospitals for the diagnosis of atrial fibrillation. 6. REFERENCES 1. Einthoven W. The different forms of the human electrocardiogram 502

7 and their signification. Lancet 1912; 1: Atrial fibrillation in a tertiary care institute A prospective study2012 Sep; 64(5): K. Konings, C. Kirchhof, J. Smeets, H. Wellens, et al., "Highdensity mapping of electrically induced atrial fibrillation in humans, Circulation, vol. 89, pp , ECG Acquisition, Storage, Transmission, and Representation Gari D. Clifford and Matt B. Oefinger. 5. G. B. Moody, R. G. Mark, and A. L. Goldberger, PhysioNet: A w web based resource for the study of physiological signals, IEEE Engineering in Medicine and Biology Mag., vol. 20, no. 3, pp , Abboud, S., and O. Barnea, Errors Due to Sampling Frequency of Electrocardiogram in Spectral Analysis of Heart Rate Signals with Low Variability, Computers in Cardiology, Vol. 22, September 1995, pp Enhancement of ECG Signal by using Digital FIR Filter B. Jagadiswara Rao1, A. Usharani2., International Journal of Science and Research (IJSR)., India. 8. Sivaraman J, Uma G, Venkatesan S, Umapathy M, Dhandapani VE. A novel approach to determine atrial repolarization in electrocardiograms. J Electrocardiol 2013; 46: e1 9. Oppenheim, A. V., and R.W. Schafer, Discrete-Time Signal Processing, Englewood Cliffs, NJ: Prentice-Hall, Clifford, G. D., and P. E. McSharry, Method to Filter ECGs and Evaluate Clinical Parameter Distortion Using Realistic ECG Model Parameter Fitting, Computers in Cardiology, Vol. 32, C. Li, C. Zheng, and C. Tai, Detection of ECG characteristic points using wavelet transforms, IEEE Trans. Biomed. Eng., vol. 42, no. 1, Jan Sivaraman J, Uma G, Venkatesan S, Umapathy M, Keshav Kumar N. A study on atrial Ta wave morphology in healthy subjects: Anapproach using P wave signalaveraging method. J Med Imaging Health Inf 2014; 4: Joseph Ackora-Prah, Anthony Y. Aidoo An Artificial ECG Signal Generate Function in MATLAB, Applied Mathematical Sciences, Vol.7, 2013, no. 54, HIKARI Ltd 14. Sivaraman J, Uma G, Venkatesan S, Umapathy M,Dhandapani VE. Normal limits of ECG measurements 503

8 related to atrial activity using a modified limb lead system. Anatol J Cardiol 2014 Feb26. Epub ahead of print. 15. Fujiki, H. Nagasawa, M. Sakabe, K. Sakurai, et al., "Spectral characteristics of human atrial fibrillation waves of the right atrial free wall with respect to the duration of atrial fibrillation and effect of class I antiarrhythmic drugs," Jpn Circ J, vol. 65, pp ,

9 505

10 506