Title
Three-Dimensional Electrical Impedance Tomography: A Topology Optimization Approach.
Abstract
Electrical impedance tomography is a technique to estimate the impedance distribution within a domain, based on measurements on its boundary. In other words, given the mathematical model of the domain, its geometry and boundary conditions, a nonlinear inverse problem of estimating the electric impedance distribution can be solved. Several impedance estimation algorithms have been proposed to solve this problem. In this paper, we present a three-dimensional algorithm, based on the topology optimization method, as an alternative. A sequence of linear programming problems, allowing for constraints, is solved utilizing this method. In each iteration, the finite element method provides the electric potential field within the model of the domain. An electrode model is also proposed (thus, increasing the accuracy of the finite element results). The algorithm is tested using numerically simulated data and also experimental data, and absolute resistivity values are obtained. These results, corresponding to phantoms with two different conductive materials, exhibit relatively well-defined boundaries between them, and show that this is a practical and potentially useful technique to be applied to monitor lung aeration, including the possibility of imaging a pneumothorax.
Year
DOI
Venue
2008
10.1109/TBME.2007.912637
IEEE Trans. Biomed. Engineering
Keywords
Field
DocType
inverse problems,medical imaging,numerical simulation,biomedical imaging,topology optimization,tomography,finite element method,electrical impedance,three dimensional,mathematical model,linear program,finite element analysis,finite element,boundary condition,electrical impedance tomography,linear programming
Boundary value problem,Computer science,Dual impedance,Electrical impedance,Finite element method,Electronic engineering,Linear programming,Inverse problem,Topology optimization,Electrical impedance tomography
Journal
Volume
Issue
ISSN
55
2
0018-9294
Citations 
PageRank 
References 
6
1.11
4
Authors
5