fMRI Calibration
Tuesday 21 April 2009
Room 323ABC 10:30-12:30


Catherine E. Chang and Peter Jezzard

10:30  212. Quantification of CMRO2 and CBF Using Simultaneous NIRS and FMRI
    Sungho Tak1, Jong Chul Ye1
Bio and Brain Engineering, KAIST, Daejeon, Korea
    This paper introduces an accurate technique to estimate the cerebral metabolic rate of oxygen (CMRO2) and cerebral blood flow (CBF) using simultaneously measured near infrared spectroscopy (NIRS) and blood oxygenation level dependent (BOLD) fMRI signals. Owing to simultaneous acquisition of both fMRI and NIRS, separate hypercapnia condition or arterial spin labeling (ASL) acquisition are not necessary to quantify CMRO2 and CBF, which greatly improves the accuracy of the proposed method. The dynamic coupling ratio of CBF changes to CMRO2 changes has been also investigated. Experimental results using finger tapping task showed that the activation pattern of CBF calculated using NIRS-SPM software is more specific to the primary motor cortex than fMRI BOLD and NIRS-HbR signal. Furthermore, the dynamic couple ratio coincides with the existing results from the literature.
10:42 213. Evaluation of a New Quantitative BOLD Approach to Map Local Blood Oxygen Saturation
    Thomas Christen1,2, Benjamin Lemasson1,3, Nicolas Pannetier1,2, Régine Farion1,2, Christoph Segebarth1,2, Chantal Rémy1,2, Emmanuel L. Barbier1,2
Inserm, U836, Grenoble, F-38043, France; 2Université Joseph Fourier, Grenoble Institut des Neurosciences, UMR-S836, Grenoble, France; 3Oncodesign Biotechnology, Dijon, France
    Recently, an in vivo MR approach – quantitative BOLD – was introduced to obtain blood volume fraction (BVf) and local blood oxygen saturation (lSO2) maps. The mesoscopic effect on T2* depends on both BVf and lSO2, but these two contributions are difficult to separate during data analysis. To improve the accuracy on the determination of lSO2, we introduce a different measurement scheme which combines a steady-state BVf measurement with standard B0 and T2 mapping. The proposed scheme was evaluated in healthy rats under an O2 challenge. Value of lSO2 are consistent with data from the literature.


10:54 214.

Does Global Cerebral Oxygen Metabolism Change During Hypercapnia and Hypocapnia in Awake Humans?

    Jean J. Chen1, G. Bruce Pike1
McConnell Brain Imaging Centre, Montreal Neurological Institute, Montreal, Quebec, Canada
    The effect of CO2 on cerebral metabolism is of great interest, since hypercapnia is routinely used in calibrated BOLD-based δCMRo2 estimation and assumes that CMRo2 remains unchanged during hypercapnia-induced blood flow increase. Certain anesthetized-animal studies, however, have put this claim under question, necessitating its verification in humans and under the conditions customary of calibrated BOLD. We report, for the first time, on steady-state global δCMRo2 measurements in awake humans during graded hypercapnia and hypocapnia. Our results show that under the mild levels of end-tidal CO2 changes commonly used in calibrated BOLD, there is negligible global δCMRo2.


11:06 215. CO2 Breathing Suppresses Cerebral Metabolic Rate of Oxygen
    Feng Xu1, Uma Yezhuvath1, Matthew R. Brier2, John Hart, Jr. 3, Michael A. Kraut4, Clair Moore2, Hanzhang Lu1
University of Texas Southwestern Medical Center at Dallas, Dallas, TX, USA; 2University of Texas at Dallas, Dallas, TX, USA; 3University of  Texas at Dallas, Dallas, TX, USA; 4Johns Hopkins University, Baltimore, MD, USA
    CO2 is highly potent in modulating cerebral blood flow. However the effect of CO2 on neuronal activity and brain energy consumption is not yet known. Here we used a novel technique to measure global cerebral metabolic rate of oxygen, CMRO2, during normocapnia and hypercapnia, and showed that 5% CO2 breathing can reduce the CMRO2 by 13±5% (n=8, p<0.001). We further used electroencephalography (EEG) to investigate which component of the neuron’s energy budget has changed, and found that the delta band in the EEG signal was enhanced by 18%, suggesting that brain was switched to a low-arousal state during CO2 breathing.
11:18 216.

CBF-CMRO2 Coupling in the Default Mode Network

    Pan Lin1, Jorge Jovicich1, Simon Robinson1
Center for Mind/Brain Sciences, University of Trento, Trento, Trentino, Italy

The Default Mode is a network of brain regions which show reduced blood flow (CBF) and BOLD signal in the task state relative to baseline, independent of the nature of the task. There has been controversy over the origin of these signal fluctuations, which overlap to some extent with regions affected by respiration rate variation. We show for the first time that the coupling ratio between cerebral blood flow and cerebral metabolic rate of oxygen consumption is the same in task-independent deactivation as in activation, and thereby consistent with a neuronal basis for deactivation of the Default Mode.

11:30 217. Changes in Arterial Oxygen Tension with Evoked Stimulation in the Rat Somato-Sensory Cortex:  Implications for Quantitative FMRI
    Alberto L. Vazquez1, Mitsuhiro Fukuda1, Seong-Gi Kim1
Radiology, University of Pittsburgh, Pittsburgh, PA, USA
    In an effort to investigate the changes in vascular and tissue oxygenation with evoked brain function, arterial, tissue and venous oxygen tension (and saturation) were measured in the rat somato-sensory cortex using oxygen micro-electrodes. Significant increases in the oxygen tension and saturation of pre-penetrating pial arterioles were observed. The largest increases in oxygen saturation were observed at the sampled small emerging pial veins. The observed steep oxygen tension and saturation gradient indicates that the quantification of the relative changes in CMRO2 from high-resolution BOLD fMRI data (hundreds of microns) will be underestimated.
11:42 218.

A Comparison of Physiologic Modulators of FMRI Signals

    Hanzhang Lu1, Feng Xu1, Uma S. Yezhuvath1, Yamei Cheng1, Rani Varghese1
Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, USA
    A critical prerequisite for fMRI to be possibly used for personalized disease diagnosis is that we have to understand and account for differences in fMRI signals across healthy subjects. Recently, much attention has been focused on the physiologic markers that can explain the inter-subject variations. Here we conducted visual-stimulation fMRI in a group of young, healthy controls and compared four potential modulators in explaining the variations in fMRI signals: baseline venous oxygenation, cerebrovascular reactivity, resting state BOLD signal fluctuation, and baseline CBF. It was found that these physiologic parameters in combination can explain up to 69% of the inter-subject variations.
11:54 219. Comparison and Validation of FMRI Calibration Techniques
    Rasmus Matthias Birn1, Daniel Handwerker1, Peter A. Bandettini1
Laboratory of Brain and Cognition, National Institute of Mental Health, Bethesda, MD, USA
    The amplitude of the BOLD response across the brain depends strongly on the variations in the underlying vasculature, making it difficult to determine subtle differences in neuronal activity between regions or subjects. In this study we test various calibration techniques designed to account for spatial variability in the vasculature, and compare their ability to pull out known subtle underlying variations in neuronal activity.
12:06 220.

An Intra-Subject Investigation of the BOLD Contrast Mechanism in Response to Visual Stimulation and Breath Hold at 1.5T, 3.0T and 7.0T: Insight Into the Extravascular Sensitivity, Resolution-Dependence and Vascular Origins of BOLD Contrast

    Manus J. Donahue1, Hans Hoogduin2, Peter CM van Zijl3,4, Peter Jezzard1, Reinoud Pieter Harmen Bokkers2, Matthias JP van Osch5, Jaco J.M. Zwanenburg2, Peter Luyten2, Jeroen Hendrikse2
Clinical Neurology, The University of Oxford, Oxford, UK; 2University Medical Center Utrecht, Utrecht, Netherlands; 3Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, USA; 4FM Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, USA; 5Leiden University Medical Center, Leiden, Netherlands
    B0-dependence of gradient-echo BOLD-fMRI contrast was investigated in the same subjects at 1.5T, 3.0T and 7.0T by performing (1) BOLD-fMRI (visual) in the presence of flow-dephasing gradients (b=100 s/mm2) at multiple TEs, (2) BOLD-fMRI (visual) at high (1.5x1.5x1.5mm3) and low (3.5x3.5x3.5mm3) spatial resolution, and (3) BOLD-fMRI (breath hold). Extravascular (EV) contributions to total BOLD ΔR2* were 45±13%, 70±11% and 92±19% at 1.5T, 3.0T and 7.0T, respectively, suggesting that BOLD fMRI performed at 7.0T is almost purely EV. Caution should be exercised when comparing total BOLD reactivity at different B0 due to different EV contributions.


12:18 221. Physiological Noise Effects on the Flip Angle Selection in BOLD FMRI
    Jerzy Bodurka1, Peter Bandettini1,2
Functional MRI Facility, National Institute of Mental Health, NIH, Bethesda, MD, USA; 2Section on Functional Imaging Methods, National Institute of Mental Health, NIH, Bethesda, MD, USA

Here we have considered theoretically and experimentally the physiological noise effects on the flip angle selection. We will shown that for given TR, temporal signal to noise ratio (TSNR=mean voxel time course signal/time course standard deviation) versus flip angle plots are significantly affected by the physiological noise. For situations where available SNR is high and physiological noise dominates over system/thermal noise the selection of Ernst angle does not results in large improvements in TSNR. In fact it is possible to select much smaller flip angle and have similar TSNR as for the Ernst angle, however smaller flip angle have important benefits of reducing possible inflow effect.