Accurate Reconstruction of Electrical Impedance Tomography Images Using Genetic Algorithm

One of the methods to solve the inverse problem of electrical impedance tomography is D-bar method, which is the direct and no repetition solution. In this method, a specific operator called a is used. In the D-bar integral equation, range of integration is considered a circle with the center of origin of coordinates and radius R. This integration is non-zero value only for
$k$
≤ |
$R$
|. In this equation, the used radius R, is effective in the resolution of the reconstructed image. In this paper, we used genetic algorithm to obtain the best radius to produce the optimized image with minimum mean square error. Therefore, the considered model was meshed up to three stages, by increasing the discretization nodes and implemented genetic algorithm was evaluated on it. We showed that the optimum radius was increased and mean square error was decreased at each step.