A New Algorithm for Electrical Impedance Tomography Inverse Problem

This paper proposes new technique for solution of electrical impedance tomography inverse problems. Usually, a set of voltage measurements is acquired from the boundaries of an investigated volume, whilst this is subjected to a sequence of low-frequency current patterns. In principle, measuring both the amplitude and the phase angle of the voltage can result in images of the electric conductivity and permittivity in the interior of a body. Alternating current patterns are preferred to DC to avoid polarization effects. In the usual frequency range (below 1 MHz) the field can be considered a steady current field, which is governed by the Laplace equation. It is well known that while the forward problem is well-posed, the inverse problem is nonlinear and highly ill-posed. The recently described methods are often based on deterministic or stochastic approach to solve mainly 2D problems. The aim of this paper is to present a new way for a successful image reconstruction to obtain high-quality reconstruction in electrical impedance tomography problems. Numerical results of an image reconstruction based on new technique are presented and compared.