BAYES IAN BASED BRAIN SOURCE LOCALIZATION TECHNIQUE USING EEG SIGNALS

Brain Source localization from EEG/MEG is an ill-posed inverse problem with high uncertainty in the solution. This source localization information is used to diagnose various brain disorders such as epilepsy, schizophrenia, stress, depression and Alzheimer. Different algorithms are proposed for t...

Full description

Main Author: JATOI, MUNSIF ALI
Format: Thesis
Language: English
Institution: Universiti Teknologi Petronas
Record Id / ISBN-0: utp-utpedia.21912 /
Published: 2016
Subjects:
Online Access: http://utpedia.utp.edu.my/21912/1/2016%20-ELECTRICAL%20%26%20ELECTRONIC%20-%20BAYESIAN%20BASED%20BRAIN%20SOURCE%20LOCALIZATION%20TECHNIQUE%20USING%20EEG%20SIGNALS%20-%20MUNSIF%20ALI%20JATOI.pdf
http://utpedia.utp.edu.my/21912/
Tags: Add Tag
No Tags, Be the first to tag this record!
Summary: Brain Source localization from EEG/MEG is an ill-posed inverse problem with high uncertainty in the solution. This source localization information is used to diagnose various brain disorders such as epilepsy, schizophrenia, stress, depression and Alzheimer. Different algorithms are proposed for the solution of this ill-posed problem which include minimum norm estimation (MNE), second order Laplacian based low resolution brain electromagnetic tomography (LORET A), standardized LORETA (sLORETA), exact LORETA, subspace based multiple signal classifier (MUSIC), Beamformer and Bayesian framework based multiple source priors (MSP). The solution provided by each of the algorithms mentioned above is characterized by various parameters which include the accuracy, computational complexity and localization error. The existing algorithms suffer from low resolution (LORETA family), high computational time (subspace algorithms and FOCUSS, WMNLORETA) and no validation.