Optimized diffusion tensor encoding schemes with anisotropic diffusion weighting

J. Lee, A. L. Alexander Medical Physics, University of Wisconsin, Madison, WI, United States Background/Introduction Diffusion tensor MRI and the associated measures, such as FA, trace {D}, and eigenvector direction, are highly sensitive to image measurement noise. The main strategy to decrease noise sensitivity is to employ uniformly distributed diffusion encoding directions with a diffusion-weighting value that is nearly optimum for the average diffusivity (ADC) [1-4]. These approaches make sense for the case where the diffusion tensor distributions and directions are arbitrary or unknown. However, in the case where the diffusion tensor shape and orientation in a specific region may be estimated a priori (such as in either the corpus callosum, the corticospinal tract or the spinal cord), it may be possible to make more precise measurements in that region by using an anisotropic diffusion-weighting scheme. In this study, the diffusion-weighting was optimized for each encoding direction to minimize the error in FA measurements of corpus callosum. Theory