Recent advances in Parallel Imaging for MRI: Wave-CAIPI technique

Abstract

The aim of this work is to illustrate the principles and advantages of the modern technique of Parallel Imaging (PI), Wave-CAIPI. Wave-CAIPI applies two sinusoidal gradients, Gy and Gz, during frequency encoding, with a phase shift of one quarter of a full cycle. In combination with the 2D-CAPIRINHA effect, the result is the origin of sinusoidal waves, staggered between them, that incur in "corkscrew" trajectories inside the k-space, that cross the layers, "slices per slice". The result is reduced acquisition times (Tacq), with an "aliasing" well distributed over all three dimensions of space, and an adequate signal/noise ratio (SNR). In the literature, scientific studies have shown a greater efficiency of the Wave-CAIPI than the 2D-CAIPI and GRAPPA methods. In fact, compared to a 2D-CAIPI with equal acceleration factor (R) and Tacq, the SNR is greater, thanks to geometric factor values (g) almost perfect. Similarly, compared to the GRAPPA technique, with the same number of coils channels and R, lower Tacq are described. Among the techniques of PI, the Wave-CAIPI is a technique that allows an effective acceleration of Tacq maintaining an adequate SNR, a fundamental prerequisite for clinical applications and research, in particular in the field of Neuro-MRI.