Title: ECG signal denoising by Functional Link Artificial Neural Network (FLANN)
Author: Nibedit Dey; Tripada Prasad Dash; Sriram Dash
Larsen & Toubro Limited, Integrated Engineering Services, KIADB Industrial area, Hootagalli, Mysore 570018, India.
Trident Academy of Technology, Bhubaneswar 751024, India.
Manipal Centre of Information Sciences, Manipal – 576104, Karnataka, India
Journal: Int. J. of Biomedical Engineering and Technology, 2011 Vol.7, No.4, pp.377 - 389
Abstract: Nowadays, Electrocardiogram (ECG) plays an important role in the primary diagnosis, prediction and survival analysis of cardiac diseases. Electrocardiography has had a profound influence on the practice of medicine. The ECG signal contains an important amount of information that can be exploited in different manners. The ECG signal allows for the analysis of anatomic and physiologic aspects of the whole cardiac muscle. Noise reduction in ECG signals is one of the main problems, which appear during analysis of electrical activity of the heart. Such noises are difficult to remove using typical filtering procedures. Efficient analytical tool, which allows increasing signal to noise ratio, is a technique of averaging of cardiac cycles. Effectiveness of this method strictly depends on stable sinus rhythm. Different ECG signals are used to verify the proposed method using MATLAB platform. In this paper, we have proposed Functional Link Artificial Neural Network (FLANN) for the denoising of the ECG signal.
Keywords: ECG signals; electrocardiograms; FLANN; functional link ANNs; artificial neural networks; signal denoising; electrocardiography; noise reduction; cardiac cycles; heart electrical activity.