Optimizing BOLD sensitivity in the 7T Human Connectome Project resting-state fMRI protocol using plug-and-play parallel transmission

Vincent Gras, Benedikt A. Poser, Xiaoping Wu, Raphaël Tomi-Tricot, Nicolas Boulant

Research output: Contribution to journalArticlepeer-review

16 Scopus citations

Abstract

The Human Connectome Project (HCP) has a 7T component that aims to study the human brain's organization and function with high spatial and temporal resolution fMRI and diffusion-weighted acquisitions. For whole brain applications at 7T, a major weakness however remains the heterogeneity of the radiofrequency transmission field (B 1 + ), which prevents from achieving an optimal signal and contrast homogeneously throughout the brain. In this work, we use parallel transmission (pTX) Universal Pulses (UP) to improve the flip angle homogeneity and demonstrate their application to highly accelerated multi-band EPI (MB5 and GRAPPA2, as prescribed in the 7T HCP protocol) sequence, but also to acquire at 7T B 1 + -artefact-free T 1 - and T 2 -weighted anatomical scans used in the pre-processing pipeline of the HCP protocol. As compared to typical implementations of pTX, the proposed solution is fully operator-independent and allows ”plug and play” exploitation of the benefits offered by multi-channel transmission. Validation in five healthy adults shows that the proposed technique achieves a flip angle homogeneity comparable to that of a clinical 3 T system. Compared to standard single-channel transmission, the use of UPs at 7T yielded up to a two-fold increase of the temporal signal-to-noise ratio in the temporal lobes as well as improved detection of functional connectivity in the brain regions most strongly affected by B 1 + inhomogeneity.

Original languageEnglish (US)
Pages (from-to)1-10
Number of pages10
JournalNeuroImage
Volume195
DOIs
StatePublished - Jul 15 2019

Bibliographical note

Funding Information:
The research leading to these results has received funding from the European Research Council under the European Unions Seventh Framework Program ( FP7/2013–2018 ), ERC Grant Agreement n. 309674 . BAP is funded by the Netherlands Organisation for Scientific Research ( NWO 016.Vidi.178.052 ) and the National Institutes of Health ( R01MH111444 , PI Feinberg). XW is supported by grants from National Institutes of Health ( U01 EB025144 and P41 EB015894 , PI Ugurbil). The authors wish to thank Bertrand Thirion and Florent Meyniel for valuable discussions regarding the resting-state fMRI analysis, Alexandre Vignaud for valuable discussions on dielectric padding, Franck Mauconduit for his contribution to the MPRAGE and SPACE sequences and for setting up the B 1 + calibration protocol, and Desmond Tse for his contributions to the pTX MB-EPI sequence. Data were acquired at Scannexus BV in Maastricht, The Netherlands.

Funding Information:
The research leading to these results has received funding from the European Research Council under the European Unions Seventh Framework Program (FP7/2013–2018), ERC Grant Agreement n. 309674. BAP is funded by the Netherlands Organisation for Scientific Research (NWO 016.Vidi.178.052) and the National Institutes of Health (R01MH111444, PI Feinberg). XW is supported by grants from National Institutes of Health (U01 EB025144 and P41 EB015894, PI Ugurbil). The authors wish to thank Bertrand Thirion and Florent Meyniel for valuable discussions regarding the resting-state fMRI analysis, Alexandre Vignaud for valuable discussions on dielectric padding, Franck Mauconduit for his contribution to the MPRAGE and SPACE sequences and for setting up the B

Publisher Copyright:
© 2019

Keywords

  • Calibration
  • Human connectome project
  • Multi-band EPI
  • Parallel transmission
  • RF shimming
  • Ultra-high field
  • Universal pulse

Center for Magnetic Resonance Research (CMRR) tags

  • BFC
  • MRE
  • P41

Fingerprint

Dive into the research topics of 'Optimizing BOLD sensitivity in the 7T Human Connectome Project resting-state fMRI protocol using plug-and-play parallel transmission'. Together they form a unique fingerprint.

Cite this