Classifying multichannel ECG patterns with an adaptive neural network

被引：41

作者：

Barro, S ^{[1
]}

Fernandez-Delgado, M ^{[1
]}

Vila-Sobrino, JA ^{[1
]}

Regueiro, CV ^{[1
]}

Sanchez, E ^{[1
]}

机构：

[1] Univ Santiago de Compostela, Dept Elect & Comp, Santiago De Compostela 15706, Spain

来源：

IEEE ENGINEERING IN MEDICINE AND BIOLOGY MAGAZINE | 1998年 / 17卷 / 01期

关键词：

D O I：

10.1109/51.646221

中图分类号：

R318 [生物医学工程];

学科分类号：

0831 ;

摘要：

A new artificial neural network model has been developed to classify the heartbeats detected on a multichannel electrocardiography (ECG) signal. The network, called MART, is based on a model which features unsupervised online learning properties. MART is able to update morphological logic templates that match the input beats. It can also create new morphological classes as they appear, and remove those classes that remain obsolete. In addition, MART adapts the discrimination capacity of each morphological class detected to the variability of those input patterns associated with it.

引用

页码：45 / 55

页数：11

共 24 条

[1]

*AAMI, 1987, TEST REP RES VENTR A

[2] Fuzzy pattern classification and the connectionist approach [J].

Bortolan, G ;

Silipo, R ;

Marchesi, C .

PATTERN RECOGNITION LETTERS, 1996, 17 (06) :661-670

[3]

CARPENTER G, 1987, APPL OPTICS, V26

[4] THE ART OF ADAPTIVE PATTERN-RECOGNITION BY A SELF-ORGANIZING NEURAL NETWORK [J].

CARPENTER, GA ;

GROSSBERG, S .

COMPUTER, 1988, 21 (03) :77-88

[5] FUZZY ARTMAP - A NEURAL NETWORK ARCHITECTURE FOR INCREMENTAL SUPERVISED LEARNING OF ANALOG MULTIDIMENSIONAL MAPS [J].

CARPENTER, GA ;

GROSSBERG, S ;

MARKUZON, N ;

REYNOLDS, JH ;

ROSEN, DB .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1992, 3 (05) :698-713

[6]

CASHMAN P, 1978, COMPUT BIOMED RES, P311

[7]

CASWELL SA, 1993, COMPUT CARDIOL, P181

[8]

CHOWDHURY E, 1994, COMPUT CARDIOL, P549

[9]

DESA JPM, 1994, COMPUT CARDIOL, P545

[10]

FELDMAN CL, 1971, COMPUT BIOMED RES, V3, P666

← 1 2 3 →