Figure 1.

Sequences employed for: (A) 3D Spin Echo EPI DTI acquisitions, incorporating a phase-encoding gradient for resolving the slab dimension, a variable gradient along the low-bandwidth (phase) dimension for scan interleaving, and double sampling (double sampling involves acquiring each readout line twice in opposite directions while removing the PE blip between the oscillating gradients, in order to obtain images that are free of half-FOV ghost artifacts; Yang et al., 1996). Illustrated in red are the bipolar gradient pulses encoding the diffusion tensor. (B) 3D SPEN DTI acquisitions, including a 180° chirp pulse acting in the presence of a gradient that encodes the more artifact-prone, low bandwidth dimension, and a pre-encoding T_a/2 delay introduced for achieving full-refocusing. Timings are illustrated for fulfilling SPEN's full-refocusing condition; similar interleaving and phase-encoding procedures as in the EPI counterpart were also used here. (C) 2D SPEN DTI sequence used to scan zoomed regions in juvenile mice, possessing similar DTI- and interleaved elements as (B) but including a final, 2D echo-based reacquisition of a common region in the readout/ SPEN domain for all interleaves, for correcting excessive motions in these experiments.