Two-dimensional Tissue Image Reconstruction Based on Magnetic Field Data

This paper introduces new possibilities within two-dimensional reconstruction of internal conductivity distribution. In addition to the electric field inside the given object, the injected current causes a magnetic field which can be measured either outside the object by means of a Hall probe or ins...

Full description

Bibliographic Details
Main Authors: J. Dedkova, K. Ostanina
Format: Article
Language:English
Published: Spolecnost pro radioelektronicke inzenyrstvi 2012-09-01
Series:Radioengineering
Subjects:
Online Access:http://www.radioeng.cz/fulltexts/2012/12_03_0917_0922.pdf
Description
Summary:This paper introduces new possibilities within two-dimensional reconstruction of internal conductivity distribution. In addition to the electric field inside the given object, the injected current causes a magnetic field which can be measured either outside the object by means of a Hall probe or inside the object through magnetic resonance imaging. The Magnetic Resonance method, together with Electrical impedance tomography (MREIT), is well known as a bio-imaging modality providing cross-sectional conductivity images with a good spatial resolution from the measurements of internal magnetic flux density produced by externally injected currents. A new algorithm for the conductivity reconstruction, which utilizes the internal current information with respect to corresponding boundary conditions and the external magnetic field, was developed. A series of computer simulations has been conducted to assess the performance of the proposed algorithm within the process of estimating electrical conductivity changes in the lungs, heart, and brain tissues captured in two-dimensional piecewise homogeneous chest and head models. The reconstructed conductivity distribution using the proposed method is compared with that using a conventional method based on Electrical Impedance Tomography (EIT). The acquired experience is discussed and the direction of further research is proposed.
ISSN:1210-2512