fMRI:  Acquisition Methods
Monday 20 April 2009
Room 323ABC 11:00-13:00


Luis Hernandez-Garcia and Hanzhang Lu

11:00  12. Functional MRI Using Arteriolar Cerebral Blood Volume Changes

Jun Hua1, Qin Qin1, Manus J. Donahue2, Jinyuan Zhou1, James J. Pekar1, Peter CM van Zijl1
Dept. of Radiology, The Johns Hopkins University, Baltimore, MD, USA; 2Dept. of Clinical Neurology, University of Oxford, Oxford, UK

    Vascular changes during functional brain activity occur predominantly in the arterioles, and an MRI method sensitive to such changes is expected to improve spatial and temporal specificity for fMRI. Vascular-space-occupancy (VASO) fMRI is a blood-nulling approach assessing total CBV changes. We introduce an approach called “inflow VASO” or “iVASO”, which nulls only blood flowing into the slice. By using a blood-nulling time comparable to arterial transit times (~700ms), iVASO signal was sensitized to predominantly arteriolar blood volume effects. This arteriolar character was subsequently reflected in an effectively immediate hemodynamic response for iVASO when studying visual activation.
11:12 13. Neuronal Activity-Induced Cerebral Blood Volume Changes in Humans: Measurements with VASO and VERVE
    Claire Cohalan1, Jean J. Chen1, G. Bruce Pike1
1McConnell Brain Imaging Center, Montréal Neurological Institute, McGill University, Montréal, Québec, Canada
    The vascular-space-occupancy (VASO) technique targets changes in total cerebral blood volume (CBV), whereas the venous-refocusing for volume-estimation (VERVE) technique measures changes in venous CBV, which is more relevant for BOLD. In this work, ΔCBV measurements acquired in healthy humans using both techniques were compared. VASO produced a higher contrast-to-noise ratio and larger ΔCBV values than VERVE, as expected since VERVE measures only venous CBV changes. VERVE-based activation was more correlated with BOLD activation, since BOLD is sensitive to the venous compartment. Though the VASO technique is easier to implement, its signal potentially has many contributions other than CBV, and eliminating these contaminants is difficult, but necessary.
11:24 14.

Whole-Brain Non-Invasive Hemodynamic Imaging, Enabled by a Novel CBV-Weighted Single-Shot 3D VASO-FLAIR GRASE Sequence Combined with CBF-Weighted ASL and BOLD FMRI, Identifies Regional Hemodynamic and Metabolic Discrepancies


Manus J. Donahue1, Jakob U. Blicher2, Bradley J. MacIntosh1, Karla L. Miller1, Leif Ostergaard2, David A. Feinberg3,4, Matthias Guenther3, Peter Jezzard1
Clinical Neurology, The University of Oxford, Oxford, UK; 2Center for Functionally Integrative Neuroscience, Arhus University Hospital, Arhus, Denmark; 3Advanced MRI Technologies, Sebastopol, CA, USA; 4University of California at Berkeley, Berkeley, CA, USA

    VASO-FLAIR magnetization preparation, previously limited to single-slice imaging, is appended to a single-shot 3D-GRASE readout to generate whole-brain CBV-weighted maps. CBV-weighted courses are compared to BOLD and CBF-weighted ASL during and following motor and visual stimulation. The 3D-GRASE VASO-FLAIR approach gives similar CBV traces to those found from single-slice techniques and corresponds well with BOLD and ASL. Following stimulation, the BOLD post-stimulus undershoot is larger and endures longer in visual cortex compared to motor cortex, whereas CBV and CBF returns to baseline at the same time.
11:36  15.

3D Single-Shot VASO FMRI Using a Maxwell-Gradient Compensated GRASE Sequence


Benedikt Andreas Poser1,2, David G. Norris1,2
Erwin L. Hahn Institute for Magnetic Resonance Imaging, University Duisburg-Essen, Essen, Germany; 2Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands

    The VASO sequence was recently proposed as first fMRI method capable of detecting activation related CBV changes without the need for a contrast agent. We here present a new whole-brain VASO technique based on a parallel-accelerated single-shot 3D GRASE sequence. A flow-compensated correction scheme for concomitant Maxwell gradients is introduced, and shown to be an essential feature for 3D GRASE sequences at 3T if smearing artifacts due to violation of the CPMG condition in off-resonance excitation are to be avoided. The effectiveness of the new method demonstrated in fMRI studies with visuo-motor stimulation, and a cognitive Stroop task paradigm.
11:48  16.

Studies of BOLD Signal Characteristics Using a Modified HASTE Sequence with GRAPPA


Yongquan Ye1, Yan Zhuo1, Rong Xue1, Dehe Weng1,2, Xiaohong Joe Zhou3
State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, CAS, Beijing, China; 2Siemens Mindit Magnetic Resonance Ltd., Shenzhen, China; 3University of Illinois Medical Center, Chicago, IL, USA

    Turbo spin echo (TSE) has been proposed as an alternative to echo planar imaging (EPI) for fMRI. We have used a modified HASTE (mHASTE) sequence with GRAPPA to investigate TSE signal characteristics in fMRI and compared the results with gradient-echo and spin-echo EPI. mHASTE exhibited reliable and consistent activation with higher SNR than EPI and minimal artifacts. More interestingly, we have observed evidence suggesting that BOLD signals in mHASTE can be dominated by extravascular contributions around microvascular networks, which offers more accurate localization of neurofunctional activities.
12:00 17.

 A Parallel Transmission Method for Improved BOLD FMRI


Weiran Deng1, Cungeng Yang1, Vijayanand Alagappan2, Lawrence L. Wald2, V A. Stenger1
University of Hawaii JABSOM, Honolulu, HI, USA; 2Harvard University Martinos Center for Biomedical Imaging, Charlestown, MA

    Susceptibility-induced gradients at the air/tissue interface above the sinus regions create signal voids in axial slices of the orbitofrontal cortex (OFC) in BOLD fMRI. We present a parallel transmission technique to recover signal in the OFC with a customized four-channel TR array. A slice-selection pulse with a unique time shift is applied into each channel. Signal recovery and increased BOLD activation in the OFC is demonstrated during a breath-holding task at 3T.
12:12 18.

Comparison of Template and Individual-Based Gradient Compensated EPI in Regions Affected by Local Susceptibility-Induced Signal Loss


Jochen Rick1, Oliver Speck2, Olaf Dössel3, Jürgen Hennig1, Maxim Zaitsev1
Dept. of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany; 2Biomedical Magnetic Resonance, Otto-von-Guericke University, Magdeburg, Germany; 3Institute of Biomedical Engineering, University Karlsruhe (TH), Karlsruhe, Germany

    Anatomy-related susceptibility gradients in the human head lead to artefacts in echo planar imaging (EPI). The use of a slice-dependent common gradient compensation template improves fMRI sensitivity in areas affected by strong susceptibility gradients. This study evaluates the concept through a comparison between four groups (no compensation, common template (member), common template (not member), individual). In general the signal improvement of the three compensated cases is about 35%. No significant variations are present between these cases. Thus, it seems possible to use this method for functional experiments without repeating the calibration individually, thus saving adjustment and calculation time.
12:24 19.

Robust Detection of Functional Activation in the Superior Colliculus Without ECG-Triggering


Ute Goerke1, Kamil Ugurbil1
Radiology, CMRR/University of Minnesota Medical School, Minneapolis, MN, USA

    In fMRI, high spatial resolution is usually achieved by segmenting the echo-planar imaging (EPI) acquisition scheme. However, such images are susceptible to ghosting due to pulsatile flow of blood and cerebrospinal fluid (CSF), especially in regions near the brain stem. We propose a novel post processing technique, the spectral side band analysis (SSBA), to detect activation in deep brain structures without the need of ECG-triggering. This is demonstrated in a high resolution fMRI study of the superior colliculus stimulated with a visual paradigm.
12:36 20.

Simultaneous Monitoring of Tongue Tip Movements in Functional MRI Motor Tasks for Speech and Swallowing Studies


Bradley P. Sutton1,2, Charles A. Conway1, David P. Kuehn3
1Bioengineering Department, University of Illinois at Urbana-Champaign, Urbana, IL, USA; 2Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, USA; 3Speech and Hearing Science, University of Illinois at Urbana-Champaign, Champaign, IL, USA

    A pulse sequence is developed that acquires images of dynamic movement of the tongue during speech or swallowing simultaneously with the acquisition of functional MRI data. A single midsagittal dynamic slice is acquired at 16 frames per second while several oblique axial functional slices are acquired with lower temporal resolution and functional MRI contrast. The acquisition allows for real-time monitoring of task performance without the need for, or interference from, additional monitoring hardware. The sequence is shown to detect similar activations of the primary motor cortex in a self-paced compared to a cued tongue-tapping task.
12:48 21.

Investigating the Whole Brain with 1.5mm Isotropic Resolution and 1.5s TRs Using Highly Accelerated High-Field FMRI


Cheryl A. Olman1,2, Steen Moeller2, Jennifer F. Schumacher3, Serena K. Thompson3, Edward J. Auerbach2, Kamil Ugurbil2, Essa Yacoub2
Psychology, University of Minnesota, Minneapolis, MN, USA; 2Radiology, University of Minnesota, Minneapolis, MN, USA; 3Neuroscience, University of Minnesota, Minneapolis, MN, USA

    Neuroscientists who want to take advantage of the higher spatial resolutions afforded by the increased SNR and CNR at 7 Tesla do not want to sacrifice temporal resolution in exchange for whole-brain coverage with high spatial resolution. A multi-band acquisition at 7T (simultaneous excitation of 4 coronal slices) permits whole-brain coverage with higher spatial and temporal resolutions than previously feasible. In this study we measure whole-brain activation patterns during a visual object recognition task with 1.5 mm spatial resolution and 1.5 s temporal resolution. Whole-brain, high-resolution fMRI is therefore possible with temporal resolutions sufficient for event-related designs.