An electrocardiogram classification method based on cascade Support Vector Machine

Author

Zhu, Jiangchao ; Shen, Mi ; Zhu, Kanjie

Author_Institution

Software Eng. Inst., East China Normal Univ., Shanghai, China

Volume

3

fYear

2011

fDate

15-17 Oct. 2011

Firstpage

1640

Lastpage

1644

Abstract

In this paper, a method based on cascade Support Vector Machine (SVM) to classify electrocardiogram (ECG) has been proposed. First, we extract features by threshold based method and Independent Component Analysis (ICA) method. And then we discuss the construction of the model. When using SVM, we focus on how to choose the parameters, how to structure these sub-classifiers, and how to filter data which is diagnosed by former classifier. At last, experiments which used the practical multi-lead data collected from patients of remote medical center are presented. For 2-classification experiment, the accuracy of testing data is 91.59%.

Keywords

cascade systems; electrocardiography; filtering theory; independent component analysis; patient diagnosis; support vector machines; ECG; SVM; cascade support vector machine; data filtering; electrocardiogram classification method; independent component analysis; medical center; multilead data collection; patient diagnosis; Accuracy; Correlation; Electrocardiography; Feature extraction; Support vector machines; Testing; Training; Cascade Classifier; Multi-lead ECG classification; Support Vector Machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Biomedical Engineering and Informatics (BMEI), 2011 4th International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-9351-7

Type

conf

DOI

10.1109/BMEI.2011.6098559

Filename

6098559