A NOVEL TECHNIQUE FOR ECG DENOISING

Abstract

The Electrocardiogram is the record of the electrical activities of the heart. During the acquisition process of the cardiac signal, several noises are added to the signal, making the medical diagnosis of this critical signal difficult. Important noises induced in the signal are: power line interference, baseline wander and electromyography noise. In the case of exercise electrocardiogram, baseline wander noise is naturally present. Several works has been done in the field of noise cancellation in cardiac signals. In this thesis work, a novel approach for the removal of baseline noise has been implemented by using the hybrid of morphological filters and empirical mode decomposition technique. The proposed algorithm has been made partially adaptive to the variations of the heart rate. As, the heart rate increases, there are definite changes in the morphology of the characteristic waves of the heart signal. Thus it becomes essential to make the denoising specially in the case of baseline drift noise adaptive to these variations when using morphological filters. The results of the proposed algorithm are compared to the existing techniques and are found to perform better.