Description
Diffusion tensor magnetic resonance imaging (DTI) is a powerful Magnetic Resonance Imaging (MRI) technique that provides a map of the three-dimensional diffusion of water within the human brain. Diffusion refers to the dispersion of water molecules in the brain by Brownian motion. Isotropic diffusion of water molecules in the brain is only observed in areas where the water molecules are unrestricted and diffuse spherically, such as in cerebral spinal fluid (CSF). However in the white matter regions of the brain, molecules encounter boundaries and the actual diffusion observed is anisotropic diffusion. Anisotropic diffusion in the white matter regions of the brain can be modeled as a symmetric second rank tensor. Several indices can then be determined from this tensor, which explain the degree of anisotropic diffusion and the underlying microstructure of the white matter anatomy. DTI has been utilized to study microstructure changes in the white matter regions associated with neurological disorders, such as Alzheimer's disease (AD). DTI acquisitions were performed in clinically acceptable scan times and thus had a low signal-to-noise. The DTI data was first denoised with a total variation regularization algorithm, which was followed by affine and nonlinear registration to generate a common reference frame for the image volumes of all subjects. We utilize the DTI derived indices of apparent diffusion coefficient (ADC) and fractional anisotropy (FA) to investigate 14 normal subjects with the specific genes that increase (APOE _2) and that decrease (APOE _4) the age of onset of AD. Region of interest (ROI), voxel-based analysis (VBA) and tract based spatial statistics (TBSS) were performed on the aligned ADC and FA maps to identify differences in the cohorts segregated by these alleles as well as segregated by mean age of the group. The effect of isotropic and anisotropic smoothing along with tensor denoising on the outcomes of the VBA analysis was evaluated and compared to the TBSS method. VBA on the denoised tensor data identified regions of reduced FA (primarily in the genu) in the APOE _4 cohort compared to the APOE _2 cohort. The most consistent results were obtained for VBA using the denoised tensors and anisotropic smoothing prior to statistical testing. In contrast, isotropic smoothing identified regional differences for small filter sizes alone, emphasizing that this method introduces bias in FA values. TBSS did not identify any regional differences in the white matter tracts and this may be related to the small localized area of change (found by VBA and ROI methods) that may not be detectable when analyzing voxels averaged normal to the tract.
