Healthy human brain diffusion anisotropy maps derived from standard spin echo diffusion tensor imaging (DTI) were compared with those using fluid-attenuated inversion recovery (FLAIR) preparation prior to DTI to null the signal from cerebrospinal fluid (CSF). Consistent comparisons entailed development of DTI postprocessing methods, image masking based on fitting quality, and an objective region-of-interest-based method for assessment of white matter extent. FLAIR DTI achieved an extended delineation of major white-matter tracts (genu, splenium, and body of the corpus callosum) close to large CSF-filled spaces (lateral ventricles), but did not affect representation of tracts remote from CSF (internal and external capsules and coronal radiation). This result, which was detectable qualitatively (visual inspection), was verified quantitatively by analyses of the relative anisotropy (RA) distribution over white matter structures for 11 subjects. FLAIR DTI thus suppresses the CSF signal that otherwise masks underlying anisotropic parenchymal tissue through partial volume averaging. Diffusion-weighted (DW) MRI is used widely for mapping the properties of water self-diffusion in human brain parenchyma (gray and white matter) in vivo (1). However, cerebrospinal fluid (CSF) may compromise diffusion mapping because of: 1) its physical properties (large proton density, T 2 , and diffusivity), and 2) commonly used imaging parameters (long echo times and large voxel sizes due to single-shot imaging) (2). As a result, the apparent diffusivity and the trace of the diffusion tensor (DT) are in general overestimated, particularly within parenchymal areas proximal to CSF spaces (2-4). These adverse effects can be reduced by the use of CSF-suppression techniques, such as fluid-attenuated inversion recovery (FLAIR) (5), prior to DW imaging (2-4). Although CSF-suppressed DT imaging (DTI) is applied clinically (6), its effect on DT anisotropy (7) has not been investigated. Since stationary CSF is isotropic, it may affect the anisotropy of neighboring parenchyma (in particular, anisotropic white matter, such as the corpus callosum). This study investigates the effect of CSF on DT anisotropy maps by comparing standard (non-FLAIR) and FLAIR DTI in a group of 11 normal subjects. SUBJECTS AND METHODS Diffusion Tensor ImagingDTI was performed on 11 healthy volunteers (seven males, four females, age range 25-42 years) using a 1.5T clinical MRI scanner (Eclipse, Marconi Medical, Cleveland, OH) with actively shielded whole-body gradient set (maximum strength per axis of 27 mT/m, slew rate of 72 mT/m/s).

Mohammed Mustafa finds a different kind of enlightenment in the East

Mohammed Mustafa worries about the image of medical students and sets out on a journey of self discovery