|
Abstract
|
Article Information:
Non- Linear Principal Component Analysis Neural Network for Blind Source Separation of EEG Signals
Saad A. Al-Shaban, Muaid S. Al-Faysale and Auns Qusai H. Al- Neami
Corresponding Author: Saad Alshaban
Submitted: 2009 November, 17
Accepted: 2009 December, 14
Published: 2010 March, 10 |
Abstract:
|
The complex system such as human brain generates electrical recording activity from thousands of
neurons in the brain. This activity is given as electroencephalogram (EEG) waveforms. The EEG potentials
represent the combined effect of potentials from a fairly wide region of the skull's skin (scalp). Mixing some
underlying components of brain activity presumably generates these potentials. The mixing of brain fields at
the scalp is basically linear mixture. The present study aims to design and implement an unsupervised
neurocomputing model for separating the original components of brain activity waveforms from their linear
mixture, without further knowledge about their probability distributions and mixing coefficients. This is called
the problem of "Blind Source Separation "(BSS). It consists of the recovery of unobservable original
independent sources from several observed (mixed) data masked by linear mixing of the sources, when nothing
is known about the sources and the mixture structure. The current study used recently developed source
separation method known as "Independent Component Analysis" (ICA) technique for solving blind EEG source
separation problem. The ICA is used to decompose the observed data into components that are as statistically
independent from each other as possible. The ICA algorithm that was used for linear BSS problem is the
Nonlinear Principal Component Analysis (BSS) algorithm. The proposed ICA BSS model was implemented
using the Matlab version6.1 package. The measured real EEG data signals obtained from normal and abnormal
states from the (Neurosurgery Hospital) in Baghdad. The results of the present work show the good
performance of the proposed model in separating the m ixed signals. Since the present ICA model is a reliable,
robust and effective unsupervised learning model which, enable us to separate the EEG signals from their linear
observation records, and extract several specific brain source signals that are potentially interesting and contain
useful information that help physician to diagnose the abnormality of the brain easily.
Key words: BSS, EEG, ECG Signals, neural networks, PCA, ,
|
|
Abstract
|
PDF
|
HTML |
|
Cite this Reference:
Saad A. Al-Shaban, Muaid S. Al-Faysale and Auns Qusai H. Al- Neami, . Non- Linear Principal Component Analysis Neural Network for Blind Source Separation of EEG Signals. Research Journal of Applied Sciences, Engineering and Technology, (2): Page No: 180-190.
|
|
|
|
 |
ISSN (Online): 2040-7467
ISSN (Print): 2040-7459 |
 |
| Information |
|
|
|
|
| Sales & Services |
|
|
|
|
