Multi-objective Optimization by Simulated Annealing Applied to Image Reconstruction by Electrical Impedance Tomography

The image reconstruction from the Electrical Impedance Tomography (EIT) can be approached as an optimization problem, in which is desired to minimize the Euclidean distance between the potential values measured in the cross section of the body and the calculated values, for every pattern of current applied, through the modelling of the problem by the Finite Elements Method (FEM). It’s known, a priori, that this formulation is ill-posed, which increases dependence of the EIT to the reconstruction algorithm, which must has any regularization technique to improve the conditioning of the problem. Therefore, this project proposes the use of an Multi-Objective Optimization algorithm in order to find the set of optimal solutions to the problem, aiming the minimization both the Euclidean distance and a regularization parameter.