Analysis and Parameter Extraction of P

The objective of this paper is to develop an efficient P wave detection algorithm based on the morphology characteristics of arrhythmias using correlation and regression in ECG signal. Subjects for experiments included normal subjects, patients with atrial fibrillation, ventricular tachycardia, and patients with change of the artifactuale amplitude. After the step of the detection of R peak using the pantompkins algorithm, the correlation and regression were utilized to calculate the similarity factors between a studied P wave and the reference one. The correlation coefficient can indicate the kind of arrhythmia diseases. The algorithm was tested using MIT-BIH arrhythmia database where every P wave was classified. The results are presented in terms of correlation coefficient. Then some parameters have been extracted in order to classify the arrhythmias. The correlation coefficient results of the system are 1,0.07 and –0.92 for normal beats, atrial fibrillation and change of the artifactuale amplitude, respectively. The extracted parameters are closely similar to the expert values given by the cardiologist. The results reveal that the system is accurate and efficient to detect and classify arrhythmias resulted from atrial fibrillation or change of the artifactuale amplitude.