Electrical Impedance Tomography imaging using Gauss-Newton algorithm

Electrical Impedance Tomography (EIT) is a noninvasive tomographic imaging modality, where cross-sectional images of conductivity and/or permittivity distribution inside of a volume are reconstructed. Electrical Impedance and Diffused Optical Reconstruction Software (EIDORS) is used in this tomographic imaging. In this paper, simulation is carried out in two different conditions and using simulated data images of change of conductivity distribution within the volume are generated with Gauss-Newton (GN) algorithm in EIDORS. Priors like Laplace, Tikhonov, Gaussian High Pass Filter (HPF) and NOSER prior are used in reconstructing images. The percentage of error in reconstructing conductivity and 1D conductivity profile are also calculated. The difference images thus obtained are compared with respect to shape, size, and spatial location of the impurity, calculated error and 1D profile to evaluate performance of GN algorithm with different priors. The comparison is done in noise free and noisy (SNR 30dB) case.