Skip navigation
Please use this identifier to cite or link to this item: http://repository.iitr.ac.in/handle/123456789/17566
Title: ECG modeling and QRS detection using principal component analysis
Authors: Chawla M.P.S.
Verma H.K.
Kumar V.
Published in: Proceedings of IET Conference Publications
Abstract: Principal component analysis (PCA) is used to arrange the redundantly distributed information on functional activity from several biomedical signals. A modeling procedure for ECG using PCA has been suggested in this paper. Eigenvectors are created which form a new orthogonal basis for finding various segments of an ECG waveform. The simulation results are obtained using FFT (fast Fourier transform) and PCA. The concept of "largest variance" in PCA is chosen so as to extract the QRS complex portion only and to exclude P-wave and T-wave. Normalization of the ECG data to zero mean and unit variance can considerably improve the results of visualization of various segments of an ECG waveform. Visualization plots or screen plots showed that there was good segments separation of the considered ECG waveform.
Citation: Proceedings of IET Conference Publications, (2006), 60. Glasgow
URI: https://doi.org/10.1049/cp:20060339
http://repository.iitr.ac.in/handle/123456789/17566
Issue Date: 2006
Keywords: Covariance
Electrocardiogram
Principal component analysis
QRS complex
Shape-space
ISBN: 0863416586; 9780863416583
Author Scopus IDs: 16174096200
57204684351
25646515800
Author Affiliations: Chawla, M.P.S., Department of Electrical Engineering, Indian Institute of Technology, Roorkee 247667, India
Verma, H.K., Department of Electrical Engineering, Indian Institute of Technology, Roorkee 247667, India
Kumar, V., Department of Electrical Engineering, Indian Institute of Technology, Roorkee 247667, India
Corresponding Author: Chawla, M.P.S.; Department of Electrical Engineering, Indian Institute of Technology, Roorkee 247667, India; email: mpschawla@rediffmail.com
Appears in Collections:Conference Publications [EE]

Files in This Item:
There are no files associated with this item.
Show full item record


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.