Wavelet Transform Method for the Analysis of ECG Signals for Ambulatory Cases

Abstract

The electrocardiograph (ECG) is a common clinical and biomedical tool used for diagnosis of heart patients. The thesis is aimed towards the development of beat detection algorithm with high level of accuracy for ambulatory monitoring of arrhythmia patients. The thesis has been inspired by the need to find an efficient method for ECG signal analysis which is simple and has good accuracy. The initial task for efficient analysis is the removal of effect of noise. It actually involves the use of wavelet filters which extract the required cardiac components by rejecting the background noise and the second task is that of R peak detection. Efficiency of the method is measured in terms of sensitivity and positive predictivity. The development, simulation and the evaluation of the methodology is done in MATLAB environment and the database of MIT-BIH is used for the purpose of the evaluation. The accuracy of the algorithm is evaluated against the MIT-BIH arrhythmia database, giving an average sensitivity of 99.71% and positive predictivity of 99.64% respectively.