Water-fat separation in diffusion-weighted EPI using an IDEAL approach with image navigator.

PURPOSE: Echo planar-based diffusion-weighted MRI (DW-MRI) requires robust suppression of fat signal. Fat suppression techniques such as inversion recovery or spectrally selective excitation with subsequent gradient spoiling can extend scan time or perform suboptimally in the presence of strong main field inhomogeneities. Chemical shift-encoded water-fat separation using iterative decomposition of water and fat with echo asymmetry and least-squares estimation (IDEAL) is robust in areas of main field inhomogeneity but requires accurate phase information, which can be distorted by patient motion during diffusion-weighting gradients. A method is proposed to overcome this with the use of image navigators. THEORY AND METHODS: A spin echo planar imaging (SE-EPI) diffusion-weighted sequence was modified to incorporate IDEAL acquisition in combination with an image navigator to correct for patient motion-induced phase effects. Images were acquired in phantoms and in healthy volunteers in brain, pelvic, and abdominal regions. RESULTS: Without navigator, diffusion-weighted IDEAL created artifacts in areas of motion. These were removed when the two-dimensional navigator was used to correct the phase, resulting in correct water-fat separation. CONCLUSION: DW-EPI with IDEAL and an integrated image navigator allows for robust water and fat separation in different body areas and are a time-efficient alternative to standard fat-suppression techniques in DW-MRI.

Distribution and fibre field similarity mapping of the human anterior commissure fibres by diffusion tensor imaging.

OBJECT: The anterior commissure is a critical interhemispheric pathway in animals, yet its connections in humans are not clearly understood. Its distribution has shown to vary greatly between species, and it is thought that in humans it may convey axons from a larger territory than previously thought. The aim was to use an anatomical mapping tool to look at the anterior commissure fibres and to compare the distribution findings with published anatomical understanding. MATERIALS AND METHODS: Two different diffusion-weighted imaging data sets were acquired from eight healthy subjects using a 3 Tesla MR scanner with 32 gradient directions. Diffusion tensor imaging tractography was performed, and the anterior commissure fibres were selected using three-dimensional regions of interest. Distribution of the fibres was observed by means of registration with T2-weighted images. The fibre field similarity maps were produced for five of the eight subjects by comparing each subject's fibres to the combined map of the five data sets. RESULTS: Fibres were shown to lead into the temporal lobe and towards the orbitofrontal cortex in the majority of subjects. Fibres were also distributed to the parietal or occipital lobes in all five subjects in whom the anterior commissure was large enough for interhemispheric fibres to be tracked through. The fibre field similarity maps highlighted areas where the local distances of fibre tracts were displayed for each subject compared to the combined bundle map. CONCLUSION: The anterior commissure may play a more important role in interhemispheric communication than currently presumed by conveying axons from a wider territory, and the fibre field similarity maps give a novel approach to quantifying and visualising characteristics of fibre tracts.