A modular LVQ neural network with fuzzy response integration for arrhythmia classification

Author

Amezcua, Jonathan ; Melin, Patricia

Author_Institution

Div. of Grad. Studies, Tijuana Inst. of Technol., Tijuana, Mexico

fYear

2014

fDate

24-26 June 2014

Firstpage

1

Lastpage

5

Abstract

In this paper, the development of a fuzzy system as the integrating unit in a classification model based on modular learning vector quantization (LVQ) neural networks is presented. The method uses a modular approach and is applied for the classification of different types of arrhythmias. The architecture is composed by three modules, each one is working with five different types of arrhythmias; the MIT-BIB arrhythmia dataset, composed by 15 classes, was used for this work. Simulation results show that the modular LVQ with fuzzy response integration is a good arrhythmia classification model.

Keywords

cardiology; fuzzy set theory; medical diagnostic computing; neural nets; MIT-BIB arrhythmia dataset; arrhythmia classification; fuzzy response integration; modular LVQ neural network; modular learning vector quantization; Accuracy; Computer architecture; Fuzzy systems; Neural networks; Support vector machine classification; Vector quantization; Vectors; LVQ; arrhythmias; classification; clustering methods; fuzzy system; unsupervised neural networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Norbert Wiener in the 21st Century (21CW), 2014 IEEE Conference on

Conference_Location

Boston, MA

Type

conf

DOI

10.1109/NORBERT.2014.6893884

Filename

6893884