Preliminary Study on the Feasibility of Reconstructing Anatomically Complex Numerical Brain Phantoms with Limited Prior Information

In this paper, we present a preliminary study of brain imaging by using MWI algorithms when limited prior information is given. By using the distorted Born iterative method with the fast iterative shrinkage thresholding algorithm, reconstructions are performed for different kinds of head models based on the anatomically complex numerical Zubal head phantom. Results show that the quality of the reconstructed images decreases when less prior information is given.