Adapted Random 3 D Sampling Patterns for CE-MRA

Introduction: Variable density 3D random sampling trajectories which were introduced in the context of compressed sensing [1] have great potential for subsampled CE-MR angiography techniques which deliver data sets with high contrast to noise ratio. The goal of this work was to present a parameter-free method to construct variable density sampling patterns which are tailored to angiography. Sampling patterns are generated with the use of a probability density function (pdf) that is constructed by using measured k-space data as a reference, which automatically ensures an appropriate distribution of sample points. It is also shown that these data sets can be used together with a nonlinear parallel imaging method [2, 3]. This combination allows the use of very high acceleration factors while still yielding images with excellent image quality.