The forward problem of electrical impedance tomography is often linearized with respect to the conductivity and the resulting linear inverse problem is regarded as a subproblem in an iterative algorithm or as a reconstruction method as such. We introduce and numerically test a novel, accurate linearization technique based on the logarithm of the Neumann-to-Dirichlet operator. A smoothened contact conductance model is also proposed for electrode measurements, leading to improved regularity for the complete electrode model.

This presentation is part of Minisymposium “MS20 - Advances in Reconstruction Methods for Electrical Impedance Tomography (3 parts)”
organized by: Melody Alsaker (Gonzaga University) , Samuli Siltanen (University of Helsinki) .

Authors:

Nuutti Hyvönen (Aalto University)

Keywords:

bayesian methods, image reconstruction, inverse problems, nonlinear optimization, partial differential equation models