Support Vector Machine based error filtering for Holter electrocardiogram analysis

被引：3

作者：

Kigawa, Yasushi ^{[1
]}

Oguri, Koji ^{[1
]}

机构：

[1] Aichi Prefectural Univ, Grad Sch Informat Sci & Technol, Nagakute, Aichi 4801108, Japan

来源：

2005 27TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-7 | 2005年

关键词：

D O I：

10.1109/IEMBS.2005.1615306

中图分类号：

R318 [生物医学工程];

学科分类号：

0831 ;

摘要：

Holter electrocardiogram data is analyzed by a computer, however, there is a detection of non-heartbeat as a heartbeat. This study dealt with reduction of the incorrect detection using Support Vector Machine (SVM). By exploiting the power of SVM and human like information processing, the data was classified to heartbeat class or non-heartbeat class. The performance of the proposed method was verified in several experiments and comparing with SVM and neural network, and 96% of accuracy was achieved.

引用

页码：3872 / 3875

页数：4

共 6 条

[1]

Byun H, 2002, LECT NOTES COMPUT SC, V2388, P213

[2]

Kitamoto A, 2002, LECT NOTES COMPUT SC, V2388, P237

[3] Clustering ECG complexes using Hermite functions and self-organizing maps [J].

Lagerholm, M ;

Peterson, C ;

Braccini, G ;

Edenbrandt, L ;

Sörnmo, L .

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2000, 47 (07) :838-848

[4]

Mukherjee N, 2002, LECT NOTES COMPUT SC, V2388, P1

[5]

Shin KS, 2004, LECT NOTES COMPUT SC, V3316, P1254

[6]

VAPKIN V, 2000, NATURE STAT LEARNING

← 1 →