Efficient calculation of current densities in the human body induced by arbitrarily shaped, low-frequency magnetic field sources

  • Authors:
  • Andreas Barchanski;Markus Clemens;Herbert De Gersem;Thomas Weiland

  • Affiliations:
  • Technische Universität Darmstadt, Institut für Theorie Elektromagnetischer Felder, Fachbereich Elektrotechnik und Informationstechnik, Schlossgartenstrasse 8, D-64289 Darmstadt, Germany;Helmut-Schmidt-Universität, Universität der Bundeswehr Hamburg, Professur für Theoretische Elektrotechnik und Numerische Feldberechnung, P.O. Box 700833, 22008 Hamburg, Germany;Technische Universität Darmstadt, Institut für Theorie Elektromagnetischer Felder, Fachbereich Elektrotechnik und Informationstechnik, Schlossgartenstrasse 8, D-64289 Darmstadt, Germany;Technische Universität Darmstadt, Institut für Theorie Elektromagnetischer Felder, Fachbereich Elektrotechnik und Informationstechnik, Schlossgartenstrasse 8, D-64289 Darmstadt, Germany

  • Venue:
  • Journal of Computational Physics
  • Year:
  • 2006

Quantified Score

Hi-index 31.45

Visualization

Abstract

In this paper, we extend the scalar-potential finite-difference (SPFD) approach in order to consider arbitrarily shaped time-harmonic field sources. The SPFD approach is commonly used to compute the currents induced by an externally applied magnetic field in regions with weak, heterogeneous conductivities such as, e.g., the human body. We present the extended scalar-potential finite-difference (Ex-SPFD) approach as a two step algorithm. In the first step, the excitation is computed by solving the magnetoquasistatic curl-curl equation on a coarse grid that is well adapted for the field sources. In the second step, the magnetic vector potential is prolongated onto a finer grid and a divergence correction inside the conductor is applied. Using the Maxwell-grid-equations (MGEs) of the finite integration technique, a geometric discretization scheme for Maxwell's equations, this new approach has been implemented in a parallel environment in order to account for the memory-demanding high-resolution anatomy models used for the calculation of induced currents inside the human body. We demonstrate the validity and the improved numerical performance of the new approach for a test case. Finally, an application example of a human exposed to a realistic electromagnetic field source is presented.