Boundary element method and simulated annealing algorithm applied to electrical impedance tomography image reconstruction

TL;DR

This paper introduces electrical impedance tomography for medical imaging, demonstrating how boundary element method and simulated annealing can accurately reconstruct internal structures in a conductive wire.

Contribution

It combines boundary element method with simulated annealing to solve the inverse problem in electrical impedance tomography, providing an effective educational example.

Findings

Accurate retrieval of non-conductive inclusion demonstrated

Boundary element method effectively solves the direct problem

Simulated annealing successfully optimizes the inverse problem

Abstract

Physics has played a fundamental role in medicine sciences, specially in imaging diagnostic. Currently, image reconstruction techniques are already taught in Physics courses and there is a growing interest in new potential applications. The aim of this paper is to introduce to students the electrical impedance tomography, a promising technique in medical imaging. We consider a numerical example which consists in finding the position and size of a non-conductive region inside a conductive wire. We review the electric impedance tomography inverse problem modeled by the minimization of an error functional. To solve the boundary value problem that arises in the direct problem, we use the boundary element method. The simulated annealing algorithm is chosen as the optimization method. Numerical tests show the technique is accurate to retrieve the non-conductive inclusion.