fMRI data analysis using a novel clustering technique

Author

Segovia, F. ; Górriz, J.M. ; Ramírez, J. ; Salas-González, D. ; Illán, I.A. ; López, M. ; Chaves, R. ; Puntonet, C.G. ; Lang, E.W. ; Keck, I.R.

Author_Institution

Dept. of Signal Theor., Networking & Commun., Univ. of Granada, Granada, Spain

fYear

2009

fDate

Oct. 24 2009-Nov. 1 2009

Firstpage

3399

Lastpage

3403

Abstract

This paper marks the beginning of a new way of analyzing fMRI images. The idea is to model these images with a mixture of Gaussians, that allows to carry out some complex tasks more easily. One application of this approach is the artifacts subtraction. It consists of removing certain high-intensity voxels that are not relevant to the analysis of the images. On the other hand, this approach provides a way to perform the feature extraction process that avoids the small sample size problem in classification tasks.

Keywords

Gaussian processes; biomedical MRI; feature extraction; image registration; medical image processing; pattern clustering; Gaussian mixture; artifact subtraction; clustering technique; fMRI data analysis; feature extraction; high-intensity voxels; image registration; Alzheimer´s disease; Brain modeling; Data analysis; Feature extraction; Gaussian processes; Image analysis; Image segmentation; Magnetic analysis; Magnetic resonance imaging; Nuclear and plasma sciences; Artefact extraction; Gaussian Mixture Model; fMRI; fMRI classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Nuclear Science Symposium Conference Record (NSS/MIC), 2009 IEEE

Conference_Location

Orlando, FL

ISSN

1095-7863

Print_ISBN

978-1-4244-3961-4

Electronic_ISBN

1095-7863

Type

conf

DOI

10.1109/NSSMIC.2009.5401767

Filename

5401767