An intelligent paradigm for denoising motion artefacts in ECG preprocessing: smart filters

Abstract

A well-recorded ECG contains complete information of various heart diseases. The detection of cardiac diseases encompasses ECG signal pre-processing, feature extraction and classification of the cardiac disease from the detected abnormality. A minute change in ECG due to noise signals changes the characteristics of ECG which results in a wrong diagnosis of the heart disease. The main focus of this paper is on developing and implementing a novel smart filter design for filtering the core noise signals that distort the original ECG. The core noise signals are identified as power line interference (PLI), baseline wander and electromyography (EMG). The comparison of the proposed filter design in terms of signal to noise ratio and power spectral density indicates that the proposed filter design has a good response characteristic for noise filtration. The eradication of these noise signals helps to achieve accurate identification of heart disease and makes the life of physicians easier.