Modified Born Iterative Method in Medical Electromagnetic Tomography Using Magnetic Field Fluctuation Contrast Source Operator

Abstract

The portability of electromagnetic tomography (EMT) systems provides the possibility for exploiting them in onsite medical emergencies, like stroke or sport and road accidents. However, such a possibility is currently hindered by the high computational time required to retrieve accurate tomographic images, as the governing equations of EMT are generally nonlinear. One available approach to accelerate the computational time of EMT is the Born iterative method (BIM) which linearizes the EMT equations through some approximations. These approximations, nevertheless, noticeably reduce the image accuracy. To retrieve more accurate images within the medical emergency time frame, a modified BIM is presented. In this method, rather than the conventional electric field analysis, magnetic fields are analyzed and processed. The fundamental difference between the magnetic and electric field vector equations enables us to decompose the contrast source of the biological imaged tissue into two subsources that are: the average contrast source, which is usually known, and the fluctuation contrast source, which is unknown and should be retrieved. This, in turn, establishes a first-order vector partial differential operator for the dielectric fluctuations of biological tissues. During image reconstruction process, and under the same BIM linearization approach, this operator simultaneously considers the additional effect of the fluctuation gradients and the induced currents, resulting in an improved tomographic accuracy with respect to the conventional BIM. The merits of the proposed approach are verified through imaging the realistic head phantoms in simulations and experiments.