Language:

A Novel Approach for Classification of Congestive Heart Failure Using Relatively Short-term ECG Waveforms and SVM Classifier

Lecture notes in engineering and computer science, 2015, Vol.1, p.47-50

Full text available

Citations Cited by

Title:
A Novel Approach for Classification of Congestive Heart Failure Using Relatively Short-term ECG Waveforms and SVM Classifier
Author: Liao, Ken Ying-Kai ; Chiu, Chuang-Chien ; Yeh, Shoou-Jeng
Subjects: Classification ; Classifiers ; Echocardiography ; Failure ; Stacking ; Support vector machines ; Trends ; Waveforms
Is Part Of: Lecture notes in engineering and computer science, 2015, Vol.1, p.47-50
Description: In this study, a novel approach for assessing congestive heart failure by using support vector machine (SVM) and relatively short-term ECG waveforms is presented. This approach only involved a simple data normalization, resampling, and repeti-tion of periodic data in order to obtain a good accuracy while still being a good general classifier. The ECG unit patterns were first consecutively extracted from the ECG signal. Then the unit patterns were inputted into a support vector machine for classification. Stacking three unit patterns obtained the most general classifier, while stacking five unit patterns achieved the best accuracy. In conclusion, by introducing an extra periodic pattern into the SVM, it is possible to both in-crease the accuracy and generality of the classifier. This makes short waveform classification possible, rather than looking at long trends.
Language: English
Identifier: ISSN: 2078-0958
ISBN: 9789881925329
ISBN: 9881925320
EISSN: 2078-0966
Source: ROAD: Directory of Open Access Scholarly Resources

Back to results list


INSPIRE LIBRARY - TON DUC THANG UNIVERSITY	(84-028) 37 755 057	Feedback
19 Nguyen Huu Tho St. Dist.7, HCM	thuvien@tdtu.edu.vn	Feedback

Searching Remote Databases, Please Wait