Detection of Spikes with Multiple Layer Perceptron Network Structures

Author

Kutlu, Y. ; Isler, Y. ; Kuntalp, D.

Author_Institution

Elektrik ve Elektronik Muhendisligi Bolumu, Dokuz Eylul Univ., Izmir

fYear

2006

fDate

17-19 April 2006

Firstpage

1

Lastpage

4

Abstract

In this work, the spikes in the electroencephalogram (EEG) signals are analyzed by using artificial neural networks (ANN). Multiple layer perceptron (MLP) networks utilizing between 3 and 15 hidden neurons are used in the network architecture. For training the MLP network backpropagation algorithm, backpropagation with adaptive learning rate, Levenberg-Marquardt (LM) algorithm, early stopping and regularization methods are used. Principal components of feature vectors obtained from 41 consecutive sample values of each peak are used for training the networks. Performances of classifiers are examined for two cases depending on both sensitivity-specificity and sensitivity-selectivity properties

Keywords

backpropagation; electroencephalography; multilayer perceptrons; ANN; EEG signal; Levenberg-Marquardt algorithm; MLP network structure; adaptive learning rate; artificial neural network; electroencephalogram; feature vector; hidden neuron; multiple layer perceptron; network backpropagation algorithm; sensitivity-selectivity property; sensitivity-specificity property; spike detection; Artificial neural networks; Backpropagation algorithms; Electroencephalography; Epilepsy; Microstrip; Neurons; Reactive power; Signal analysis; Testing; Tin;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Communications Applications, 2006 IEEE 14th

Conference_Location

Antalya

Print_ISBN

1-4244-0238-7

Type

conf

DOI

10.1109/SIU.2006.1659693

Filename

1659693