| Magnetic resonance (MR) diffusion tensor imaging (DTI) has emerged as a unique technique to reveal small anatomical structures of brain by characterizing the diffusion process of water molecules in image voxels. The DTI data can be further used to reveal and visualize the orientation of the white matter fibers and the connectivity network between different brain areas when they are analyzed with fiber tractography algorithms. A large number of DTI applications ranges from studies of brain development, functions, aging, and diagnoses of various white matter abnormalities such as Alzheimer's disease, HIV, and multiple sclerosis among many others.;EPI (Echo-Planar Imaging) is a fast imaging technique that is commonly used in DTI and fMRI. EPI suffers from the field inhomogeneity artifacts due to the susceptibility difference and eddy current. The fidelity and accuracy of the DTI can be deteriorated due to these artifacts. Therefore, susceptibility artifacts correction is a necessary and an important step for DTI studies. In this thesis, the effects of the susceptibility artifacts on DTI were analyzed, and several correction schemes for DTI improvements were proposed to correct the susceptibility artifacts.;By incorporating the proposed correction scheme and other optimization strategies, an optimized high-resolution in-vivo monkey DTI scan procedure on a 3T human clinical scanner was proposed. Data acquired with this procedure is appropriate for accurate diffusion tensor quantification and fiber tractography in rhesus monkey brains, and is accessible within an acceptable scan time. We investigated in detail the effects of spatial resolution and SNR on diffusion tensor derived quantities and fiber tractography. Our results should be of general utility for implementation of in vivo non-human primate DTI studies with proper spatial resolution. |
