Indian Journal of Science and Technology

Abstract

Objectives: This paper aims at development of efficient ECG diagnosing system for detection of MI within small span, using novel filtering technique to remove the external noises present in ECG signal. Methods/Statistical Analysis: The medical experts study the electrical activity of the human heart in order to detect heart diseases from the electrocardiogram (ECG) of the heart patients. A Myocardial Infarction (MI) or Heart Attack is a heart disease that occurs due to a block (blood clot) in the pathway of one or more coronary blood vessels (arteries) which supplies blood to the heart muscle. The abnormalities in the heart can be identified by the changes in the ECG signal. The conventional approaches are time consuming & require too much time for the analysis of ECG signal. In this paper new technique for filtering is being introduces for removing the external noises present in ECG signal. Findings: The proposed approach evaluates the Power Spectral Density of noise filtered bands and then classification is done by using classifier. The classifier performs the comparison between the features of query database and the features of sample database and reveals the type of heart disease. By using proposed technique in this paper, the diagnosing accuracy is increased up to 96.82%. Application/Improvements: This proposes technique is best suited for all modern ECG instrument for better accuracy and analysis of ECG signal.

Keywords: ECG diagnosis, Myocardial Infarction (MI), PSD, Spectral Bands