Simulation of 3D Electromagnetic Bio-Impedance Measurement Using Edge Finite Elements
Soleimani, M. and Lionheart, W., 2002. Simulation of 3D Electromagnetic Bio-Impedance Measurement Using Edge Finite Elements. In: 13th International Conference on Biomagnetism (BioMag 2002), 2002-08-10 - 2002-08-14, Jena.
Related documents:This repository does not currently have the full-text of this item.
You may be able to access a copy if URLs are provided below.
Electromagnetic impedance measurement is a useful technique in physiological measurement as measured impedance is a function of conductivity, premittivity and permeability distribution and anatomy. To analyse and design such a system one needs to solve both the inverse and the forward problems. In this paper we discuss modelling of coil-coil measurement systems. In addition of the contactless property of coil -coil measurements compared with an electrode-electrode system, we argue that a coil -coil system has better penetration of current. Computation of the induced voltage on sensing coil from the current on exciting coil, geometry and material properties will be demonstrated based on 3D eddy current modelling by edge finite elements. We use an ungauged A- A formulation, ( A is magnetic vector potential ) and complex and underdetermined linear system of equation is solved using the incomplete Cholesky conjugate gradient (ICCG) method. A very important tool for design of such systems is sensitivity analysis. The sensitivity of the electromagnetic bio impedance system to changes in parameters such as the position of the coils and the material distribution are investigated. Sensitivity to changes in material properties can be efficiently calculated using the integral form of the Maxwell's equation. In this study we will derive this sensitivity based on the edge finite element and A-A formulation.
|Item Type||Conference or Workshop Items (Paper)|
|Creators||Soleimani, M.and Lionheart, W.|
|Departments||Faculty of Engineering & Design > Electronic & Electrical Engineering|
Actions (login required)