Massively Parallelized Boundary Element Simulation of Voxel-Based Human Models Exposed to MRI Fields

Author

Bottauscio, Oriano ; Chiampi, Mario ; Zilberti, Luca

Author_Institution

Ist. Naz. di Ricerca Metrol. (INRIM), Turin, Italy

Volume

50

Issue

2

fYear

2014

fDate

Feb. 2014

Firstpage

1029

Lastpage

1032

Abstract

Boundary element and hybrid boundary element-finite element approaches are applied to the computation of induced electric field and specific absorption rate in voxel-based human models undergoing magnetic resonance imaging. Due to the very large size of the algebraic system, the procedure uses an iterative GMRES solver recalculating the element matrix at each iteration. A suitable processing of the Green integrals and a massively parallelized algorithm, based on the use of graphical processing units, leads to a strong reduction of the computational time.

Keywords

Green´s function methods; bioelectric potentials; biomedical MRI; boundary-elements methods; finite element analysis; graphics processing units; iterative methods; medical computing; parallel algorithms; Green integrals; MRI fields; algebraic system; computational time reduction; element matrix; graphical processing units; hybrid boundary element-finite element approach; induced electric field; iteration; iterative GMRES solver; magnetic resonance imaging; massively parallelized algorithm; massively parallelized boundary element simulation; specific absorption rate; voxel-based human models; Coils; Computational modeling; Graphics processing units; Magnetic resonance imaging; Mathematical model; Numerical models; Radio frequency; BEMs; electromagnetic radiation effects; magnetic resonance imaging; modeling;

fLanguage

English

Journal_Title

Magnetics, IEEE Transactions on

Publisher

ieee

ISSN

0018-9464

Type

jour

DOI

10.1109/TMAG.2013.2280523

Filename

6749150