Estimating Regional Conductivity in 2D Disc Head Model by Multidemensional Support Vector Regression

Author

Xueqin Shen ; Hui Yan ; Weili Yan ; Lei Guo

Author_Institution

Hebei Univ. of Technol., Tianjin

fYear

2007

fDate

22-26 Aug. 2007

Firstpage

4915

Lastpage

4918

Abstract

In this paper, we introduce multidimensional support vector regression (MSVR) with iterative re-weight least square (IRWLS) based procedure to estimating the regional conductivity in 2D disc head model. The results show that the method is capable of determining for the regional location of the disturbed conductivity in the 2D disc head model with single tissue and estimating for the tissue conductivities in the 2D disc head model with four kinds of tissue. The estimation errors are all within a few percent.

Keywords

bioelectric phenomena; biological tissues; electric impedance imaging; electrical conductivity; iterative methods; least squares approximations; medical computing; neurophysiology; physiological models; regression analysis; support vector machines; 2D disc head model; electrical impedance tomography; iterative re-weight least square; multidimensional support vector regression; regional conductivity estimation; tissue conductivities; Automation; Conductivity; Electrodes; Iterative algorithms; Least squares approximation; Magnetic heads; Multidimensional systems; Scalp; Support vector machines; Tomography; Algorithms; Body Composition; Computer Graphics; Electric Conductivity; Electric Impedance; Head; Humans; Least-Squares Analysis; Models, Biological;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2007. EMBS 2007. 29th Annual International Conference of the IEEE

Conference_Location

Lyon

ISSN

1557-170X

Print_ISBN

978-1-4244-0787-3

Type

conf

DOI

10.1109/IEMBS.2007.4353442

Filename

4353442