Claudio Santelli1, 2, Sebastian Kozerke1, 2, Tobias Schaeffter2
1Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland; 2Division of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom
Fourier velocity encoding (FVE) resolves the distribution of velocities within a voxel by acquiring a range of kv-points. The The long acquisition times, however, have excluded the method from clinical use so far. SPIRiT provides a very general reconstruction framework for non-Cartesian undersampled data. Prior assumption of Gaussian velocity spectra additionally allows undersampling along the velocity encoding dimensions. We extended non-Cartesian SPIRiT to include the temporal dimension thereby additionally exploiting temporal correlations in k-t space. The k-t method is applied to non-uniformly undersampled kv-encodes to reconstruct mean and standard deviation of the velocity spectra for each voxel in aortic flow measurements.