September 2015
Volume 15, Issue 12
Free
Vision Sciences Society Annual Meeting Abstract  |   September 2015
Whole-brain, sub-second data collection for task-evoked fMRI studies using simultaneous multi-slice/multiband acquisition
Author Affiliations
  • Stephanie McMains
    Center for Brain Science, Harvard University
  • R. Matthew Hutchison
    Center for Brain Science, Harvard University Athinoula A. Martinos Center for Biomedical Imaging, Harvard Medical School, Massachusetts General Hospital
  • Ross Mair
    Center for Brain Science, Harvard University Athinoula A. Martinos Center for Biomedical Imaging, Harvard Medical School, Massachusetts General Hospital
Journal of Vision September 2015, Vol.15, 233. doi:https://doi.org/10.1167/15.12.233
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Stephanie McMains, R. Matthew Hutchison, Ross Mair; Whole-brain, sub-second data collection for task-evoked fMRI studies using simultaneous multi-slice/multiband acquisition. Journal of Vision 2015;15(12):233. https://doi.org/10.1167/15.12.233.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Slice-accelerated EPI using multiband (MB) RF pulses that allow for simultaneous multi-slice (SMS) acquisition of BOLD contrast images can significantly enhance the temporal and spatial resolution of fMRI by acquiring up to 8 non-contiguous slices at the same time, thus enabling whole-brain sub-second TRs. Here we studied visual cortex response at a variety of MB accelerations and TR reductions to investigate whether there were any costs associated with parameters that allowed for whole-brain, sub-second data collection at 2mm resolution. 6 subjects were scanned (3.0T Siemens Tim Trio) with a 32-ch head coil while subjects performed a fixation task and blocks of flashing checkerboards were presented to alternating visual fields. BOLD scans were acquired at 3mm and 2mm isotropic resolutions, a max TR of 3s, and MB accelerations of 0 (conventional BOLD sequence), 1, 4 and 8 (Siemens WIP 770A). Beta and t-statistics were extracted from regions localized in visual cortex. With a TR of 3s, there were 91 timepoints, while for TR= 1.25/0.75/0.7s, there were 184/307/328 timepoints. There were no significant differences in betas for any parameters, or in t-statistics for levels of MB when holding the TR constant. Shortening the TR increased t-statistics significantly. This advantage was reduced when temporal autocorrelations in the noise were modeled. An event-related study was also conducted for 2mm voxels to compare 3s TR (MB1) versus 750ms TR (MB8). Betas were larger for the MB8 scans, likely due to improved characterization of the hemodynamic response, even though stimulus onset was jittered to the TR. The results suggest that whole brain coverage with high spatial and temporal resolution can be achieved using SMS with little to no cost in terms of BOLD signal sensitivity, as measured by betas and t-statistics, even though time-series SNR decreased significantly at high MB factors.

Meeting abstract presented at VSS 2015

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×