Segmented Thick-Slab 3D Brain DWI with Motion-Compensated Diffusion Encoding Evaluated on an MRI System with Ultrahigh Gradient Performance.

Motivation: Performing diffusion imaging with a segmented 3D acquisition rather than with a 2D multi-slice acquisition promises higher SNR and higher spatial resolution, provided phase variations that arise from motion during diffusion encoding can be controlled. Goal(s): The goal is to assess the effectiveness of gradient moment nulling and ultrahigh gradient performance to combat motion-related phase shifts produced by diffusion encoding during segmented 3D scanning. Approach: A 3D diffusion imaging sequence with integrated 1D navigators for phase analysis was tested in normal subjects. Results: The combination of gradient moment nulling and ultrahigh gradient effectively reduces motion-related phase variations and enables in-vivo 3D diffusion imaging. Impact: Gradient moment nulling combined with ultrahigh gradient effectively suppresses motion-related phase shifts produced by diffusion encoding that otherwise interfere with spatial encoding during segmented scanning, potentially enabling artifact-free in-vivo 3D DWI at unprecedented spatial resolution.