Electrocardiogram reconstruction using support vector regression

This paper presents a new method to apply support vector regression (SVR) to reconstruct chest lead electrocardiogram (ECG) signals. The reconstructed V2, V3, V4, and V5 signals are obtained from SVRs using Lead 1, Lead II, Vl, and V6 signals as the input features. Only QRS complex, T wave, and P wa...

Full description

Saved in:

Bibliographic Details
Main Authors:	Yodjaiphet A., Theera-Umponl N., Auephanwiriyakul S.
Format:	Conference or Workshop Item
Published:	2015
Online Access:	http://www.scopus.com/inward/record.url?partnerID=HzOxMe3b&scp=84889069230&origin=inward http://cmuir.cmu.ac.th/handle/6653943832/39025
Tags:	Add Tag No Tags, Be the first to tag this record!
Institution:	Chiang Mai University

id	th-cmuir.6653943832-39025
record_format	dspace
spelling	th-cmuir.6653943832-390252015-06-16T08:01:14Z Electrocardiogram reconstruction using support vector regression Yodjaiphet A. Theera-Umponl N. Theera-Umponl N. Auephanwiriyakul S. Auephanwiriyakul S. This paper presents a new method to apply support vector regression (SVR) to reconstruct chest lead electrocardiogram (ECG) signals. The reconstructed V2, V3, V4, and V5 signals are obtained from SVRs using Lead 1, Lead II, Vl, and V6 signals as the input features. Only QRS complex, T wave, and P wave of ECGs are selected to ensure the inclusion of useful information and to reduce the size of training set. We use the 4-fold cross validation to select the best SVR models based on their regression performances. The root mean square (RMS) error of less than 0.2 mV is achieved by the SVR-based models on the test sets. © 2012 IEEE. 2015-06-16T08:01:14Z 2015-06-16T08:01:14Z 2012-12-01 Conference Paper 2-s2.0-84889069230 10.1109/ISSPIT.2012.6621299 http://www.scopus.com/inward/record.url?partnerID=HzOxMe3b&scp=84889069230&origin=inward http://cmuir.cmu.ac.th/handle/6653943832/39025
institution	Chiang Mai University
building	Chiang Mai University Library
country	Thailand
collection	CMU Intellectual Repository
description	This paper presents a new method to apply support vector regression (SVR) to reconstruct chest lead electrocardiogram (ECG) signals. The reconstructed V2, V3, V4, and V5 signals are obtained from SVRs using Lead 1, Lead II, Vl, and V6 signals as the input features. Only QRS complex, T wave, and P wave of ECGs are selected to ensure the inclusion of useful information and to reduce the size of training set. We use the 4-fold cross validation to select the best SVR models based on their regression performances. The root mean square (RMS) error of less than 0.2 mV is achieved by the SVR-based models on the test sets. © 2012 IEEE.
format	Conference or Workshop Item
author	Yodjaiphet A. Theera-Umponl N. Theera-Umponl N. Auephanwiriyakul S. Auephanwiriyakul S.
spellingShingle	Yodjaiphet A. Theera-Umponl N. Theera-Umponl N. Auephanwiriyakul S. Auephanwiriyakul S. Electrocardiogram reconstruction using support vector regression
author_facet	Yodjaiphet A. Theera-Umponl N. Theera-Umponl N. Auephanwiriyakul S. Auephanwiriyakul S.
author_sort	Yodjaiphet A.
title	Electrocardiogram reconstruction using support vector regression
title_short	Electrocardiogram reconstruction using support vector regression
title_full	Electrocardiogram reconstruction using support vector regression
title_fullStr	Electrocardiogram reconstruction using support vector regression
title_full_unstemmed	Electrocardiogram reconstruction using support vector regression
title_sort	electrocardiogram reconstruction using support vector regression
publishDate	2015
url	http://www.scopus.com/inward/record.url?partnerID=HzOxMe3b&scp=84889069230&origin=inward http://cmuir.cmu.ac.th/handle/6653943832/39025
_version_	1681421579825709056

Electrocardiogram reconstruction using support vector regression

Similar Items