Evaluation of Submillimeter Diffusion Imaging of the Macaque Brain Using Readout-Segmented EPI at 7T

Objective: The purpose of the present study was to achieve submillimeter-level diffusion tensor imaging (DTI) of the macaque brain by using diffusion weighted (DW) readout-segmented echo planar imaging (rsEPI) with an optimized protocol at 7T MRI. Methods: Three anesthetized macaques were included in this study. Under different scan settings, we compared signal-to-noise ratio (SNR) and geometric distortion of DW images, implemented an optimized protocol for submillimeter-level DTI acquisition, and evaluated its performance. Results: Parallel-imaging-enabled (in GRAPPA mode) monopolar or monopolar+ diffusion scheme has higher SNR vs. bipolar scheme, whereas trivial differences in SNR and image geometric distortion when using increased readout segments with monopolar and monopolar+ that did not reach statistical significance. Submillimeter-level (0.8mm isotropic) DTI data provides a sharper delineation of white matter contour than 1mm level. Conclusion: The rsEPI technique with parallel imaging enabled, and with the shortest readout segments in conjunction with monopolar/monopolar+ diffusion encoding scheme may be optimal for submillimeter-level diffusion imaging over macaque brain in vivo. Significance: rsEPI could effectively merit high-resolution DTI for in vivo macaque brain submillimeter structural architecture investigations at ultra-high field (UHF).