| Summary: | This work was focused on designing a model which could simulate the gradient echo signal evolution when spins experience spatially varying magnetic fields due to the presence of spherical particles of different magnetic susceptibility. The cases where the spins are stationary and when they undergo diffusion have both been considered. The model initially was designed to look at the effect of a singular stationary sphere with a volume magnetic susceptibility of 1x10-6 in a magnetic field of 7T. It was seen that the average frequency offset and rate of signal decay scaled with the volume fraction occupied by the sphere, in accordance to the literature. The model was then modified with a random walk algorithm. The displacement of the particles after a given time was assessed by looking at the standard deviation, σ, after each time step. A plot of σ2 against time was created and a positive linear relationship was seen between the two variables. The model was then used to simulate 10,000 perturbers moving in a magnetic field. A mono-exponential signal decay was observed and increasing the diffusion step size decreased R2* and the frequency offset.
Three patients who suffer from Alzheimer’s disease (AD) were scanned on a Philips 7T Achieva MRI scanner and a T2* and GRE scan was carried out using a Nova 2Tx/32Rx coil with TE=20 ms. Quantitative susceptibility maps were created from the resulting data in order to look at the feasibility of measuring the susceptibility in the hippocampus in AD patients in a larger study.
|
|---|