MATHEMATICAL MODELLING OF ECG SIGNAL

Abstract

Electrocardiogram (ECG) is extensively used by cardiologists in clinical interpretations for knowing the functioning of the heart. The electrocardiogram has good correlation with the mechanical activity of the heart which is responsible for the flow of blood throughout the circulatory system. ECG also gives a clear picture about alignments of the - heart. The first section in this dissertation deals with the accurate representation of the ECG signal by a model and the later section deals about the diagnosis of the disease with the extracted parameters from the reconstructed signal. In the present, work the ECG signal has been separated out into its various segments. So the earlier part of the first section deals about the identification and separation of segments of an ECG signal by slope line criteria. Then additional parameter like FPR and EPA angle are evaluated for better representation of the ECG signal. Two diffefent types of basis functions namely polynomial function and special mathematical functions have been used in the model for the representation of various segments and sub segments of an ECG wave. These basis functions have specified relationships. The coefficients of the expressions defining the various basis functions for standard cases are computed and stored once for all. The main variations in the ECG parameters like horizontal and/or vertical elongation or contraction of various segments and also its relative 11 phase shifts among the frontal plane leads have been taken into account. In the second section dealing with the diagnosis of the disease, twenty two parameters have been extracted from the reconstructed ECG signal for the analysis purpose. These parameters are expressed by composite binary codes. The parameters are represented by a binary code 11 for values within normal range, 01 for values below normal range and 10 for values above normal range. This binary codes are *compared with three specified categories of symptom patterns namely, right ventricular hypertrophy. iii (RVH), left ventricular hypertrophy (LVH) and myocardial infraction (MI). Depending upon the matching or mismatching of a symptom of a case with the specified symptom, the symptom multiplier are assumed as either 1, 0 or -1. Weight factors are also used to differentiate among different categories. The weighted sum for each particular disease is expressed as the sum of multiplication of the symptom multiplier with respective weight factor for all the symptoms. iv