fMRI: Spatial & Temporal Signal Characteristics
Wednesday 22 April 2009
Room 323ABC 16:00-18:00

Moderators:

 Laura M. Parkes and Kamil Uludag
 
16:00 508. From Two-Photon Microscopy to BOLD-FMRI: Association of an Undershoot of Arteriolar Diameter with the BOLD Post-Stimulus Undershoot
    Anna Devor1,2, Richard Buxton3, Peifang Tian1, Ivan Teng1, Kun Lu1, Larry May1, Ron Kurz1, Anders Dale1
1University of California, San Diego; 2Martinos Center for Biomedical Imaging, Harvard; 3University of California, San Diego, La Jolla, CA, USA
    Although the origins of the BOLD post-stimulus undershoot have been debated since the early days of fMRI, there is still no consensus on whether it is a neural, metabolic or vascular effect, and experimental evidence has been presented for each interpretation. To test directly whether vascular dynamics are evident during the post-stimulus undershoot, we performed BOLD-fMRI and 2-photon microscopic dynamic vascular diameter measurements under the same conditions in rat primary somatosensory cortex (SI). Although there was no evidence of a venous balloon effect, arteriolar diameter exhibited a post-stimulus undershoot with a similar time scale to the BOLD post-stimulus undershoot, suggesting a vascular rather than metabolic origin for the undershoot.
     
16:12 509. Revisit Nonlinearity in Blood-Oxygenation-Level-Dependent Signal
    Nanyin Zhang1, Essa Yacoub1, Xiao-Hong Zhu1, Kamil Ugurbil1, Wei Chen1
1Radiology, University of Minnesota, Minneapolis, MN, USA
    Literature studies suggest that there are significant nonlinear characteristics in the BOLD signal in response to evoked neuronal activity. We demonstrated previously that when eliminating contributions from large vessels, the nonlinearity in the BOLD signal becomes much less significant. In this study we have investigated (non)linear characteristics of the BOLD signal from the microvasculature using a combination of SE fMRI and a paired-stimulus paradigm. We found that that SE BOLD signal acquired at high magnetic fields, which is sensitive mainly to the microvasculature, is primarily a linear system.
     
16:24 510. Neural and Vascular Response Nonlinearity in Human Visual Cortex
    Zhongming Liu1, Cristina Rios1, Lin Yang1, Nanyin Zhang2,3, Wei Chen2,3, Bin He1
1Department of Biomedical Engineering, University of Minnesota, Minneapolis, MN, USA; 2Center for Magnetic Resonance Research, University of Minnesota Medical School, Minneapolis, MN, USA; 3Department of Radiology, University of Minnesota Medical School, Minneapolis, MN, USA
    The BOLD nonlinearity may reflect a nonlinear neural response to stimuli and/or a nonlinear vascular response to neural activity. To pinpoint the origin of the BOLD nonlinearity, we investigated the nonlinear effects of neural and vascular responses to sustained visual stimuli with variable temporal frequency ranging from 1/6 to 25 Hz. Using scalp EEG, we found the neural response nonlinearity existed only at very short ISI (<200 ms). Using fMRI, we observed the BOLD nonlinearity for ISI between 0.25 and 4 sec. Such nonlinearity has an exclusively vascular origin and is attributed to the vascular refractory effect instead of the vascular saturation effect.
     
16:36 511. Measurement of Parenchymal T2* Changes During Visual Stimulation Using Grey Matter Nulled and VASO FMRI
    Yuji Shen1, Ida Mengyi Pu2, Risto A. Kauppinen3
1School of Medicine, University of Birmingham, Birmingham, UK; 2Department of Computing, Goldsmiths, University of London, London, UK; 3Biomedical NMR Research Center, Dartmouth Medical School, Hanover, NH 03755, USA
    It is feasible to use MRI for quantification of blood oxygen saturation (Y) and oxygen extraction fraction (OEF) in vivo from transverse relaxation data. In this study we used grey matter nulled (GMN) and vascular space occupancy (VASO) fMRI techniques to measure T2* in brain parenchyma at baseline and during visual stimulation at 3T. The change in T2* obtained by GMN fMRI was 1.8±0.8 ms, which was much greater than that detected by VASO fMRI (0.6±0.3 ms). The venous oxygen saturation (Yv) during activation and OEF derived from VASO data were found to be 0.77±0.02 and 0.22±0.02, respectively.
     
16:48 512. BOLD Imaging of Inhibition and Facilitation Induced by Paired-Pulse Transcranial Magnetic Stimulation:  Feasibility and Reproducibility
   

Juergen Baudewig1, Carsten Schmidt-Samoa1, Peter Dechent1
1MR-Research in Neurology and Psychiatry, University of Goettingen, Goettingen, Germany
    It is still a matter of controversy whether local hemodynamic changes detected with BOLD-fMRI are equally correlated to inhibitory and excitatory processes. Paired-pulse TMS is a well established method to induce inhibition or facilitation by varying the interstimulus intervals. We applied paired TMS pulses during simultaneous fMRI recordings at 3T in order to visualize the TMS-induced modulation of cortical excitability. EMG data acquired directly prior to the fMRI experiment proved the excitatory or inhibitory effect of long or short interstimulus intervals. Excitatory TMS pulse pairs resulted in increased BOLD responses in comparision to inhibitory stimulus intervals.
     
17:00 513. Cerebral Blood Flow (CBF) and Cerebral Blood Volume (CBV) Coupling Differs During Neuronal and Vascular Tasks
    Manus Joseph Donahue1,2, Robert D. Stevens3, Jun Hua1,2, Alan Huang1,2, James J. Pekar1,2, Peter CM van Zijl1,2
1Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, USA; 2FM Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, USA; 3Anesthesiology and Critical Care Medicine, The Johns Hopkins University School of Medicine, Baltimore, MD, USA
    A common BOLD quantification approach employs a hypercapnic stimulus to evaluate CBF contributions. This method assumes that the neuronal hemodynamic response is due to increased metabolism causing release of vasodilatory CO2/H+ and that CBV/CBF coupling is equal for neuronal and vascular stimuli. Using breath hold as vascular stimulus and visual activation as neuronal stimulus, we found (n = 9) a similar CBF response and a doubling of the CBV response for breath hold versus visual stimulation. These results challenge the assumption of invariance in CBF/CBV coupling and suggest BOLD calibration should account for both CBF and CBV changes.
     
17:12 514.

Transcortical BOLD Impulse Response Functions: Implications for Layer-Specific CMRO2 Calculation
    Peter Herman1,2, Basavaraju G. Sanganahalli1, Hal Blumenfeld3,4, Fahmeed Hyder1,5
1Diagnostic Radiology, Yale University, New Haven, CT, USA; 2Institute of Human Physiology and Clinical Experimental Research, Semmelweis University, Budapest, Hungary; 3Neurology, Yale University, New Haven, CT, USA; 4Neurobiology, Yale University, New Haven, CT, USA; 5Biomedical Engineering, Yale University, New Haven, CT, USA
    Oxidative neuroenergetics with calibrated fMRI has become a popular technique. Measurements of BOLD, CBV, CBF and/or neural activity are needed to calculate CMRO2. Transcortical selection of signals can be an important factor as these parameters may vary across cortical layers. We calculated transfer functions in three cortical layers to explore transcortical differences in BOLD and LFP signals. The amplitude of the BOLD and LFP signals decreases toward the deeper layers, but the normalized signals show very high correlation between layers. The transfer function analysis revealed, while the upper and middle layers interchangeable, the lower layer is needed for calibration separately.
     
17:24 515. Layer Specific BOLD Activation in Human V1 at 3 Tesla
    Peter Jan Koopmans1, Markus Barth1,2, David Gordon Norris1,2
1Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands; 2Erwin L. Hahn Institute for Magnetic Resonance Imaging, Essen, Germany
    This study shows cortical layer specific activation measured in human visual cortex using gradient echo acquisition at 3 Tesla. The BOLD point-spread function reported in literature is too broad for such measurements. Based upon the known laminar blood-vessel distribution we conclude that the PSF is much smaller in the through-cortex direction and that the intrinsic spatial resolution of the GE-BOLD-fMRI signal is in the sub-millimeter range.
     
17:36 516. BOLD FMRI of Forepaw Stimulation at Different Amplitudes in Mice
    Simone Claudia Bosshard1, Christof Baltes1, Thomas Mueggler1, Markus Rudin1,2
1Institute for Biomedical Engineering, ETH Zurich, Zurich, Switzerland; 2Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland
    fMRI in mice is challenging due to high demands in sensitivity. Using a cryogenic RF transceiver probe, BOLD fMRI of electrical forepaw stimulation was performed in mice under isoflurane anesthesia. The BOLD signal comprised a slow and a fast component, which may reflect different processing pathways. The two signal contributions were analyzed separately in the S1 somatosensory cortex and the thalamus and were shown to correlate well with the current amplitude of the stimulus. The segregation of the signal into two components might help to understand the underlying physiological processes.
     
17:48 517. Predicting Human Decisions in a Social Interaction-Scenario Using Real-Time Functional Magnetic Resonance Imaging (Rt-FMRI)
   

Maurice Hollmann1, Sebastian Baecke1, Charles Mueller1, Johannes Bernarding1
1Institute for Biometry and Medical Informatic, University Magdeburg, Magdeburg, Sachsen-Anhalt, Germany
    Making decisions in a social context is fundamental in our daily life. Is it possible to predict decisions by investigating activation patterns of the brain? We used a paradigm from economic behavioral research: the Ultimatum Game (UG). In the UG, two players split a sum of money. One player decides how the money should be split. The responder can accept or reject this offer. If rejected, neither player receives anything. In our study an rtfMRI-system was used to predict the decision of the responder 1-2 seconds before the subject conveys the decision. Classification accuracy reached 70% averaged over seven subjects.