Using a priori information to improve on localization of conductivity changes in brain in electrical impedance tomography

Abstract

We perform computer simulations of electrical impedance tomography to detect and localize the conductivity changes in brain. The reconstruction algorithm is derived from Newton’s One Step Error Reconstructor (NOSER) and it minimizes algorithm execution time by pre-computing many quantities. It builds dynamic images which only display the conductivity changes in brain. We primarily presume that the head is a homogeneous unity according to NOSER algorithm. Then we find that the conductivity changes in brain are detected with a positional shift, which is mainly relative to the low conductivity of the skull. However, with a priori conductivity distribution information we assume that the conductivity distribution of the major head tissues is known and the pre-computed quantities are modified. Thus the conductivity changes in brain are detected with a relatively accurate localization. Future work in obtaining a more accurate a priori information of the head is expected to improve more on localization of the conductivity changes in brain.