Simulation research on three dimension electrical impedance tomography and evaluation of image reconstruction quality / 生物医学工程学杂志

ABSTRACT

Three dimensional electrical impedance tomography (3D-EIT) became an important branch of EIT recently. It is important to research imaging and image quality evaluation methods for single targets of different positions and multi-targets in 3D field. Using finite element subdivision method, 3D-EIT field was dispersed into cube unit in the present study for models with single target located in the center of field, middle of field, and near to the edge, respectively. For models with two targets and four targets near to the field edge, Tikhonov-Noser algorithm was adopted in image reconstruction. Imaging error function ER and structure similarity degree function SSIM were introduced to evaluate the reconstructed images. For the models with signal target, with the movement of the target from the center to the edge of the field, the value of ER increased and SSIM decreased, and reconstruction quality decreased. For the models with multi-targets near to the field edge, ER and SSIM increased and decreased respectively with the increase of target number, mage quality also decreased. Tikhonov-Noser algorithm is an effective 3D-EIT algorithm. ER and SSIM are adaptive for the characteristic of 3D-EIT images, and it can quantitatively evaluate the 3D-EIT imaging effect from the two perspective of imaging error and structure quality.