Simulation of the MRI Measurement of Diffusion in Realistic Neural Tissue Models

In principle, the sensitivity of MRI to the diffusion of water in neural tissues provides a method for inferring local characteristics of the tissue structure and physiology by sampling the spatial (diffusion tensor imaging, DTI) and spectral (q-space imaging) variations of the diffusion weighted imaging (DWI) signal and relating this to a model of the signal behavior generated by a specific pulse sequence. Unfortunately, in practice the complexity of neural tissues precludes the formulation of analytical solutions for the DWI signal behavior, making investigation of diffusion in real neural tissues problematic. In this paper we describe a novel approach using smoothed particle hydrodynamics (SPH) in combination with ray tracing and complex tissue generation using parametric curves. Initial results in a standard geometry are shown to be consistent with theory