Home

Menu

Loading wiki pages...

View
Wiki Version:
**Introduction** Multi-band can dramatically accelerate acquisition of EPI-based MRI data (e.g. DWI and fMRI sequences). While many research groups use 32 or 64 channel head coils, these are impractical for many clinical applications. Specifically, the small internal size does requires one to exclude individuals with large heads. Here data is provided from a Siemens Prisma using both the 32-channel head coil as well as the 20-channel head(16)/neck(4) coil. Siemens describes the 20-channel coil as having a [design with 20 integrated pre-amplifiers, two rings of 8 elements each and one ring with 4 elements][1]. For more details on the motivation see the [practiCal fMRI][2] web site. Note that the repetition time (TR) was kept constant for all acquisitions (2000ms). Normally, one exploits multi-band to acquire images more rapidly. However, this would change the T1 recovery. The intention of this data is to explore the viability of the 20-channel head coil for multi-band. **Methods** The data was acquired on a Siemens Prisma-Fit E11C using the [CMRR Multi-Band Accelerated EPI Pulse Sequences][3]. Data is provided in DICOM format and can be converted to NIfTI format using [dcm2niix][4]. Note that the number of slices acquired varies with multiband. As Ben Inglis describes it: "To get symmetric excitation and relaxation across the slice dimension, the number of slices divided by MB factor should be odd, and you've got 48/2 = 24 and 48/3 = 16 and 48/4 = 12. If you can set these to 50, 51 and 52, respectively, you'll reduce the amount of striping in the event of significant head movement along the slice direction (It's a similar argument to using contiguous versus interleaved slices for regular multislice EPI). See Fig 3 of this review from [Barth et al.][5]" A flip-angle of 52-degrees was used. This is less than [many][6] suggest, and about 2/3 typical [Ernst angle estimates][7]. However, [Gonzalez-Castillo et al. (2011)][8] have noted the benefits of using a low flip angle. Six series are provided. Each series provides 100 volumes at the specified multi-band level plus a single-band reference image. 1. 20-Channel MB=2 2. 20-Channel MB=3 3. 20-Channel MB=4 4. 20-Channel MB=5 5. 32-Channel MB=2 6. 32-Channel MB=5 **Results** Temporal signal-to-noise was estimated using SPM12 with the provided Matlab script. An image of the results is provided. No gross multi-band aliasing artifacts were observed. tSNR seemed equivalent across multiband levels. The 32-channel head coil appears to provide better tSNR near the cortex and similar tSNR for subcortical regions relative to the 20-channel head/neck coil. ![SNR image][9] In general, the 20-channel coil performs well with these settings. From personal experience, things tend to fall apart with smaller voxel sizes (for both this coil and the 32-channel coil). The methods provided here provide a nice method for empirically investigating the performance of your sequence. [1]: https://usa.healthcare.siemens.com/magnetic-resonance-imaging/options-and-upgrades/coils/head-neck-20/features [2]: https://practicalfmri.blogspot.com/2019/02/using-multi-band-aka-sms-epi-on-on-low.html [3]: https://www.cmrr.umn.edu/multiband/ [4]: https://github.com/rordenlab/dcm2niix/releases [5]: https://www.ncbi.nlm.nih.gov/pubmed/26308571 [6]: https://www.umass.edu/ials/sites/default/files/hmrc_tn_otimum_rf_angle_0.pdf [7]: https://github.com/rordenlab/spmScripts/blob/master/tr_flipangle.m [8]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3020268/ [9]: https://files.osf.io/v1/resources/q4d53/providers/osfstorage/5c7326f18d5d98001c32c474?mode=render
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
Accept
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.
Accept
×

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.