MultiSlice CAIPIRINHA Using View Angle Tilting Technique (CAIPIVAT)

Abstract

We aim to focus on improving the performance of slice parallel imaging while simultaneously correcting for spatial shift artifacts related to off-resonance. In multislice controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA), simultaneously excited slices are shifted along the phase-encoding direction by varying the radiofrequency phase for each slice, thereby obtaining virtually shifted coil sensitivity information. Meanwhile, the view angle tilting (VAT) technique provides additional shifts in the readout direction to further spread an image overlap while correcting for field inhomogeneity-induced spatial misregistration using a compensation gradient. By combining these features of CAIPIRINHA and VAT, named CAIPIVAT, the excited individual slices are shifted along both phase-encoding and readout directions. Consequently, the number of aliased voxels is reduced, and the virtual coil sensitivity information is more effectively used. Blurring due to the compensation gradient in VAT was alleviated by using a constrained least square filter. The advantages of CAIPIVAT are shown by signal-to-noise ratio simulation, phantom experiments, and in vivo experiments. Thus, CAIPIVAT can be useful for multislice parallel imaging while providing the correction of off-resonance-related spatial shift artifact.