1541.     A Novel Technique for Functional MRI in Regions of Magnetic Field Inhomogeneity – Z-Shim Asymmetric Spin-Echo (ASE) Spiral

Kimberly D. Brewer^1,2, Ryan CN D'Arcy^2,3, Chris V. Bowen^2,4, Steven D. Beyea^2,4

¹Department of Physics, Dalhousie University, Halifax, Nova Scotia, Canada; ²Institute for Biodiagnostics (Atlantic), National Research Council of Canada, Halifax, Nova Scotia, Canada; ³Departments of Psychology and Radiology, Dalhousie University, Halifax, Nova Scotia, Canada; ⁴Departments of Physics, Radiology and Biomedical Engineering, Dalhousie University, Halifax, Nova Scotia, Canada

The presence of magnetic susceptibility induced field gradients (SFGs) in regions such as the orbital frontal cortex results in significant signal loss and image distortion in functional MRI studies. To overcome this problem we propose a new technique that combines an approach in which multiple spiral images are acquired asymmetric to a spin-echo, with each spiral obtained using a different z-shim value (optimized on a slice-by-slice basis). This method, called Z-shim ASE spiral, permits significant signal recovery in regions of SFGs, resulting in increased SNR and fMRI signal recovery.

                  1542.     Combined Gradient- And Spin-Echo EPI Acquisition Technique for High-Resolution FMRI

Heiko Schmiedeskamp¹, Samantha J. Holdsworth¹, Stefan Skare¹, Rexford David Newbould², Gary H. Glover¹, Roland Bammer¹

¹Lucas Center, Department of Radiology, Stanford University, Stanford, CA, USA; ²GlaxoSmithKline, London, UK

A combined gradient- and spin-echo EPI-based acquisition technique is presented for application in high-resolution BOLD-fMRI. A gradient-echo EPI readout prior to a 180° spin-echo refocusing pulse is highly sensitive to overall BOLD-signal changes, while the spin-echo readout is more confined to the microvasculature, resolving detailed maps of functional activity. The high sensitivity from gradient-echo EPI combined with high specificity through spin-echo EPI resulted in detailed maps showing more distinct areas of functional activity.

                  1543.     Comparison of Volume-Selective Z-Shim and EPI with an FMRI Memory Task

Hu Cheng¹, Srikanth Padmala¹, Rena Fukunaga¹

¹Psychological and Brain Sciences, Indiana University, Bloomington, IN, USA

Volume-selective z-shim technique applies z-shim to only those slices with large susceptibility so that the scanning efficiency is not sacrificed as much as conventional z-shim approach. An fMRI study using a memory task involving faces is performed while applying both volume selective z-shimming and normal EPI techniques. The results show that despite fewer volumes collected during volume-selective z-shimming, volume-selective z-shim is superior to EPI especially in the targeted areas of susceptibility, while little difference is observed in other areas.

                  1544.     Unaliasing of Multiband Multislice EPI and GRE Imaging with GRAPPA

Steen Moeller¹, Essa Yacoub¹, Eddie Auerbach¹, Cheryl Ohlman¹, Kamil Ugurbil¹

¹Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, USA

Whole brain fMRI at ultra high field with large reduction factors requires accurate treatment of the sparsely sampled data. While maintaining conventional techniques for unalasing, the effects of only using 5x4 GRAPPA kernel size are demonstrated for 4x4 aliasing. Larger kernels must be used to capture the more complex sensitivity profiles. The correct use of navigator echoes for ACS data is also investigated. Significant image improvements are demonstrated in rapid whole brain coverage using 16 fold acceleration, with 4 fold originating from simultaneous acquisition of multiple slices, enhancing our ability to perform rapid, whole brain fMRI at ultrahigh fields like 7T.

                  1545.     A Method to Increase Temporal Resolution in 3D-EPI FMRI Using UNFOLD

Onur Afacan^1,2, W. Scott Hoge², Dana H. Brooks¹, Jing Yuan², Ming-Long Wu³, Lawrence P. Panych², Istvan Akos Morocz²

¹ECE Dept., Northeastern University, Boston, MA, USA; ²Radiology Department, Brigham and Women's Hospital and Harvard Medical School, Boston, MA, USA; ³Brain Imaging and Analysis Center, Duke University

An fMRI study is ideally acquired at a high temporal resolution and at

                  1546.     Isotropic Sub-Millimeter FMRI of V5 in Human at 7T

Robin Martin Heidemann¹, Robert Trampel¹, Dimo Ivanov¹, Fabrizio Fasano^2,3, Josef Pfeuffer⁴, Robert Turner¹

¹Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany; ²Fondazione Santa Lucia, Rome, Italy; ³Siemens Medical Solutions, Milano, Italy; ⁴Siemens Medical Solutions, Charlestown, MA, USA

The increased SNR of ultra-high field MR scanners can be utilized to push the resolution of fMRI acquisitions. Besides the SNR, unwanted effects such as geometric distortions and blurring also scale with field strength. The use of parallel imaging can reduce those effects significantly. However, due to imperfections in the reconstruction, parallel imaging can be affected by residual aliasing artifacts. A combination of a zoomed approach and parallel imaging enables single-shot EPI acquisitions with an isotropic resolution of (0.7mm) ³. It is shown that the activation of the visual motion sensitive area V5 does not lead to signal change in an adjacent large vein.

                  1547.     Fat Suppression with Low SAR for SE EPI FMRI at 7T

Dimo Ivanov¹, Andreas Schäfer¹, Markus Streicher¹, Robert Trampel¹, Robert Turner¹

¹Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

The goal of this study was to remove the scalp fat signal present in spin-echo BOLD EPI head acquisitions without using a fat suppression pulse. This was achieved by increasing the duration of the refocussing pulse to the T2* value of fat at 7 Tesla. This method has the advantage of reducing SAR in two ways - by eliminating the need for a fat saturation pulse and by decreasing the amplitude of the SAR intensive 180 degree pulse. The decrease in SAR results in doubled brain volume coverage for the same repetition time.

                  1548.     Multiple Area B1 Shimming: An Efficient, Low SAR Approach for T2-Weighted FMRI Acquired in the Visual and Motor Cortices of the Human Brain at Ultra-High Field

Pierre-Francois Van de Moortele¹, Eddie Auerbach¹, Kamil Ugurbil¹, Johannes Ritter¹

¹CMRR/Medical School, University of Minnesota, Minneapolis, MN, USA

B1 heterogeneities are a major challenge at Ultra-High Field[1,2]. B1 shim techniques can mitigate those inhomogeneities but B1 Shim solutions aiming at uniform B1 over the whole brain generally result in poor RF efficiency because of large destructive interferences [3]. A strong case has been made for the need to address B1 inhomogeneities for T2-weighted fMRI at Ultra-High fields. It has been demonstrated that a multi region B1 shim is a very efficient approach to sample T2-weighted contrast in the visual and motor cortices.

                  1549.     A Multi-Resolution Comparison of Single-Shot EPI, 3DFFE and PRESTO for FMRI at 7T

John A. Sexton^1,2, John C. Gore^1,3, James Chris Gatenby^1,3

¹Vanderbilt University Institute of Imaging Science, Nashville, TN, USA; ²Biomedical Engineering, Vanderbilt University, Nashville, TN, USA; ³Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, USA

At high field, magnetic susceptibility effects such as distortion and signal dropout become severe, and the usefulness of single-shot EPI for fMRI becomes limited. Here we consider two alternatives to EPI for fMRI at 7T: 3D Fast Field Echo (FFE) and 3D PRESTO (Principles of Echo Shifting with a Train of Observations). We observe higher t-values and more activated voxels in EPI compared with 3DFFE in high-resolution, limited field-of-view studies (1mm3, 1.2mm3, and 1.75mm3), while PRESTO demonstrates higher t-values and more activated voxels than EPI at lower resolution, large field-of-view studies (3mm3).

                  1550.     Comparison of Single-Shot 2D EPI and Segmented 3D EVI Acquisition for FMRI at 7T

Wietske van der Zwaag^1,2, Tobias Kober^1,3, Jose Pedro Marques^1,2, Gary Glover⁴, Rolf Gruetter^1,5, Gunnar Krueger³

¹LIFMET, EPFL, Lausanne, Switzerland; ²UNIL , Lausanne, Switzerland; ³Advanced Clinical Imaging Technology, Siemens Suisse SA - CIBM, Lausanne, Switzerland; ⁴Stanford University, Stanford, USA; ⁵UNIL/UNIG, Lausanne/Geneva, Switzerland

Single shot and segmented 3D echo volumar imaging (EVI) acquisitions have a great potential for high spatio-temporal fMRI. Here, we compare multi-slice 2D EPI and segmented 3D EVI acquisitions with respect to performance in fMRI experiments and physiological noise properties at 7T. The maximum z-score and size of active regions in activation maps were found to increase significantly with 3D acquisitions. However, the asymptotic limit of temporal-SNR is lower for 3D than for 2D acquisitions. fMRI protocols using segmented EVI at high B0 should take this into account by, for example, using a small voxel size.

                  1551.     Inner Volume BOLD-FMRI at Ultra-High Spatial Resolution

Patrick Michael Heiler¹, Simon Konstandin¹, Lothar Rudi Schad¹

¹Computer Assisted Clinical Medicine, University of Heidelberg, Mannheim, Germany

Functional Magnetic Resonance Imaging at ultra high spatial resolution requires both, a volume selective excitation pulse in order to acquire small field of view sizes and multi-echo acquisition as a compromise between enough signal and acceptable measurement time.

                  1552.     Bridging the Gaps: High-Resolution Visual Field Mapping Using Passband B-SSFP FMRI

Jin Hyung Lee¹, Gary H. Glover², Dwight G. Nishimura¹, John M. Pauly¹, Mark Schira³

¹Electrical Engineering, Stanford University, Stanford, CA, USA; ²Radiology, Stanford University, Stanford, CA, USA; ³Psychology, University of New South Wales, Sydney, Australia

Accurate visual field mapping using non-invasive BOLD fMRI is an important tool for elucidating the basic scaffold of the brain¡¯s functional architecture underlying visual information processing. In this study, visual field mapping experiment using passband b-SSFP acquisition with a resolution of 1.5 mm iso-voxels and an 18 x 18 x 3.3 cm³ FOV was used to investigate it¡¯s capability to produce more reliable high-resolution functional activations.

                  1553.     Pass-Band Balanced SSFP FMRI at 7 Tesla

Jongho Lee¹, Masaki Fukunaga¹, Jeff H. Duyn¹

¹Advanced MRI/LFMI/NINDS, National Institute of Health, Bethesda, MD, USA

We demonstrate that pass-band SSFP fMRI at 7 Tesla can provide high quality images and large spatial coverage with limited signal null areas. The characteristics of the bSSFP functional contrast were also investigated using multi-echo and multi-TR experiments.

                  1554.     Monte Carlo Simulations of Phase Cycled PbSSFP FMRI Acquisitions

Steven Patterson^1,2, Steven Donald Beyea^1,3, Chris Van Bowen^1,3

¹NRC Institute for  Biodiagnostics, Halifax, NS, Canada; ²Physics, Dalhousie University, Halifax , NS, Canada; ³Physics, Biomedical Engineering, and Radiology, Dalhousie University, Halifax, NS, Canada

Monte Carlo simulations were used to investigate phase cycled pass band SSFP (pbSSFP) fMRI acquisitions. Functional contrast development and transient time reduction were explored for different catalyzation trains and inter-acquisition time delays to permit signal recovery. Functional contrast is reduced in phase cycled acquisitions but steadily increases during the data acquisition window, reaching 2/3 of that achieved in full steady state for a 2.5 s volume time acquisition. The diffusion contribution to contrast develops early in the RF train suggesting phase cycled pbSSFP fMRI acquisitions provide equal or better neurovascular coupling than observed in the fully developed steady state.

                  1555.     Interleaved Spiral In/Out B-SSFP Acquisition for Functional Imaging

Anna Leigh Rack-Gomer¹, Giedrius Buracas², Youngkyoo Jung², Thomas T. Liu^1,2

¹Bioengineering, University of California, San Diego, La Jolla, CA, USA; ²Radiology, University of California San Diego, La Jolla, CA, USA

Passband b-SSFP (pbSSFP) provides an advantageous approach for functional imaging by reducing off-resonance artifacts. Setting TE »½ TR for complete refocusing of spins further reduces the sensitivity of pbSSFP fMRI to physiological noise contamination. Spiral approaches can provide greater acquisition efficiency than EPI, but are less efficient when TE » ½ TR. We propose an interleaved spiral in/out approach that has its echo time at the spin refocusing time and maintains its high acquisition efficiency.

                  1556.     Spatial and Spectral Analysis for a Radial Sampling Balance SSFP  for FMRI

Ray F. Lee¹, Jian Xu², Karthik Prabhakaran, Michael Arcaro³

¹Neuroscience Institute, Princeton University, Princeton, NJ, USA; ²Siemens Medical Solution USA; ³Psychology, Princeton University

The spatial and spectral analysis for the balanced SSFP sequence revealed that the transition bandwidth can be significantly different between the middle and the edge of a slice, which, so far, has not been counted for in its fMRI study. The radial sampling balanced SSFP sequence can minimize the TE so that susceptibility effect can be further reduced, which allows fMRI to study some brain region that cannot be seen in EPI.

                  1557.     Measuring Task-Modulated Contrasts by Transition-Band BSSFP FMRI: A Rate-Dependence Study of Human Visual Cortical Response

Yi-Lu Tsai¹, Teng-Yi Huang¹

¹Dept. of Electrical Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan

The functional contrast of Transition-band bSSFP (TB-bSSFP)depends on the complex and nonlinear phase behavior, which is quite different from the BOLD methods. Whether the functional-contrast studies using the BOLD methods and TB-bSSFP can reach the same conclusions is still not investigated. To answer this question, in our study, we designed a visual experiment to compare the stimulus-rate dependency of the functional contrast of both BOLD-EPI and TB-bSSFP. In the result, the TB-bSSFP curve shows highly correlated to the BOLD-EPI curve. Furthermore, the averaged functional contrasts obtained by TB-bSSFP are higher than that obtained by BOLD-EPI.

                  1558.     Distinguishing Pial and Laminar Gradient-Echo BOLD Signals at 7 Tesla

Peter Jan Koopmans¹, Stephan Orzada^2,3, Markus Barth^1,2, David Gordon Norris^1,2

¹Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands; ²Erwin L. Hahn Institute for Magnetic Resonance Imaging, Essen, Germany; ³Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany

Spin-echo has been claimed as superior to gradient-echo in terms of high resolution fMRI at high fields due to reduced sensitivity to pial vein contributions. This study shows that when distortions and blurring are kept to a minimum, the pial vein contribution is spatially separable from the intracortical GE-BOLD response. Patterns of laminar activation in human V1 are demonstrated at 7 Tesla. GE allows for high resolution (laminar) functional imaging with a high sensitivity, combined with a lower SAR than SE methods, and is as such the method of choice for human high field laminar imaging.

                  1559.     Laminar Analysis of High Isotropic Resolution BOLD Activation with a Resolution Pattern Stimulus in Human V1 at 7T

Jonathan Rizzo Polimeni¹, Bruce Fischl^1,2, Douglas Greve¹, Lawrence Leroy Wald^1,3

¹A. A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; ²Computer Science and AI Lab (CSAIL), Massachusetts Institute of Technology, Cambridge, MA, USA; ³Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, MA, USA

To test the spatial resolution of the BOLD signal in primary visual cortex (V1), we designed a simple resolution stimulus that contains several easily identifiable features for measuring the spatial spread of the BOLD response across the cortical surface. The isotropic 1x1x1mm3 data could then be analyzed as a function of cortical lamina on the flattened cortex. Our measurements suggests that the majority of the BOLD spatial errors arise from near the pial surface and that avoiding these lamina can significantly improve the study of fine spatial scale activation in the visual cortex.

                  1560.     Functional MRI in the Rat Brain with Single-Shot Gradient Echo EPI at 16.4 T

David Z. Balla¹, Hannes M. Wiesner¹, Gunamony Shajan¹, Rolf Pohmann¹

¹High-Field MR Center, Max-Planck-Institute for Biological Cybernetics, Tuebingen, Germany

The feasibility of gradient echo echo-planar imaging sequence (GE-EPI) for the accurate detection of stimulation-specific BOLD activation contrast in the rat brain at 16.4 T was investigated. An experimental protocol for longitudinal fMRI studies with extensive monitoring of the animal’s physiological status was employed. It was found that parameter optimized single-shot GE-EPI detects high quality images and is suitable for fMRI studies, provided motion effects during the timeseries can be compensated by data processing. The first specific BOLD activation maps at 16.4 T are presented and methodical details are discussed.

                  1561.     Large Influence of Flow-Related Contribution on FMRI Signal: A 9.4T Study Using Hypercapnia

Xiao Wang¹, Xiao hong Zhu¹, Yi Zhang¹, Wei Chen¹

¹Center for Magnetic Resonance Research and Department of Radiology, University of Minnesota, Minneapolis, MN, USA

Although fMRI based on BOLD contrast has become one most prominent neuroimaging modality for mapping brain activity, it is still elusive whether the measured signal by fMRI is completely originated from the true BOLD or could be significantly contributed by the flow-related component, and if yes, whether the flow-related component is from macro- or micro vascular origin. The primary goals of this study are: i) to quantitatively investigate the relative contributions from BOLD and flow-related signal in the rat brain cortex at 9.4T using hypercapnia. The results indicate that: i) the flow-related contribution into the total signal change detected by fMRI is large and compatible to the BOLD contribution when a short repetition time is applied; and ii) the flow-related signal is mainly originated from perfusion change and inflow effect in small arterioles, thus, it can be utilized to further improve sensitivity and specificity of fMRI for mapping neuronal activity during brain activation.

                  1562.     Improved Separation of Tissue Oxygenation Extraction Fraction and Deoxygenated Blood Volume by Using the QBOLD Technique.

Jan Sedlacik¹, Jürgen R. Reichenbach², Claudia M. Hillenbrand¹

¹St. Jude Children's Research Hospital, Memphis, TN, USA; ²Friedrich Schiller Universität, Jena, Germany

He and Yablonskiy introduced a quantitative BOLD method for tissue oxygenation extraction fraction (OEF) and deoxygenated blood volume (DBV) estimation. However, the calculation of OEF and DBV can be erroneous or deliver ambiguous results when both parameters are unknown and have to be estimated simultaneously. The purpose of this work was to develop a model strategy for more accurate OEF and DBV fitting based on the analysis of the imaginary part of the tissue model signal which is decoupled from the signal of the tissue matrix.

                  1563.     Modulating Brain Activity Via Multi-Echo FMRI Neurofeedback

Mark Chiew^1,2, Audrey Yu Ching Kuo^2,3, Simon J. Graham^1,2

¹Rotman Research Institute, Toronto, ON, Canada; ²Medical Biophysics, University of Toronto, Toronto, ON, Canada; ³Sunnybrook Health Sciences Centre, Toronto, ON, Canada

The use of functional MRI (fMRI) to perform neurofeedback (NF) experiments is of increasing interest. Here a novel multi-echo pulse sequence designed for real-time fMRI and NF is validated by explicit demonstration of NF using the right and left primary sensorimotor cortices (SMC). Participants viewed a colour bar that quantified SMC laterality and were asked to modulate their brain activity to match the target laterality by performing a unilateral hand clenching task. In 18 experiments over 4 participants, there was 100% success in matching the target task condition.

                  1564.     White Matter FMRI: Exploring Functional Differentiation in the Corpus Callosum

Jodie R. Gawryluk^1,2, Kim Dillen¹, Kim D. Brewer^1,2, Erin L. Mazerolle^1,2, Steven D. Beyea^1,2, Ryan CN D'Arcy^1,2

¹Institute for Biodiagnostics (Atlantic), National Research Council Canada, Halifax, Nova Scotia, Canada; ²Dalhousie University, Halifax, Nova Scotia, Canada

The ability to detect fMRI activation in white matter has been a matter of considerable debate. We have recently reported the first two prospective 4T fMRI studies, which used interhemispheric transfer tasks to elicit activation in the corpus callosum. These studies used different tasks and reported activation in functionally appropriate and separable white matter regions. The current study replicated these activation results using a within subjects design and showed that task type can be used to map functional specialization in white matter.

                  1565.     Real-Time Single-Trial BOLD Response Detection for Visual Attention at 7T

Patrik Andersson¹, Jeroen C. Siero^1,2, Josien PW Pluim¹, Max A. Viergever¹, Nick F. Ramsey²

¹UMC Utrecht, Image Sciences Institute, Utrecht, Netherlands; ²UMC Utrecht, Department of Neurology and Neurosurgery, Utrecht, Netherlands

Using high-field (7T) fMRI we have investigated the possibility of using visuospatial attention as a control-method for Brain-Computer-Interfaces. A visual attention localizer task was first used to determine the cortical regions that respond to visual attention. A control signal based on the activation in these regions was subsequently used as real-time feedback to the subjects.

                  1566.     Compensation of Non-T₁-Related Artifacts in Variable TR FMRI at 1.5T

Shuowen Hu¹, Olumide Olulade¹, Gregory G. Tamer², Wen-ming Luh³, Thomas M. Talavage^1,2

¹School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA; ²Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, USA; ³National Institute of Mental Health, Bethesda, MD, USA

                  1567.     Hybrid Modelling of Axonal Magnetic Fields for Direct MR Neuronal Detection Estimation

Syed Muhammad Anwar¹, Greg G. Cook¹, Li Sze Chow², Martyn N. Paley²

¹Electronics and Electrical Engineering, University of Sheffield, Sheffield, UK; ²Academic Unit of Radiology, University of Sheffield, Sheffield, UK

A novel hybrid modelling technique is described to predict the strength of evoked axonal fields within a voxel. This is done to help corroborate the results of previously reported Direct MR Neuronal Detection (DND experiments, which test the hypothesis that such fields perturb the MR

                  1568.     Observation of Functional ADC Decrease in the Extravascular Tissue: A FMRI Study with Suppression of Intravascular Signal

Tao Jin¹, Ping Wang¹, Seong-Gi Kim^1,2

¹Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA; ²Department of Neurobiology, University of Pittsburgh, Pittsburgh, PA, USA

The diffusion weighted fMRI signal change has been found to increase at high b-values, indicating a decrease of the apparent diffusion coefficient (ADC). It is currently unclear whether this ADC decrease is predominately caused by vascular contributions. We have studied the diffusion-weighted fMRI in the cat visual cortex for high b-values (up to 2500s/mm²), with and without the suppression of intravascular blood using a contrast agent. We found that the diffusion fMRI signal after contrast agent injection increases with b-value at the parenchyma, indicating a small decrease of ADC in the extravascular space.

                  1569.     How Vascular Effects Contribute to Heavily Diffusion-Weighted FMRI Signal

Denis Le Bihan^1,2, Toshihiko Aso^1,2, Shin-ichi Urayama², Cyril Poupon¹, N Sawamoto², Kenji Aso², Hidenao Fukuyama²

¹I²BM/DSV/CEA, NeuroSpin, Gif-sur-Yvette, France; ²Human Brain Research Center, Kyoto University, Kyoto, Japan

Using heavily diffusion-sensitized MRI a transient decrease in water diffusion has been reported in the activated visual cortex of human subjects. The steep onset of the diffusion response and its temporal precedence relative to the BOLD fMRI response suggest a non-vascular origin. However, this assumption has been challenged by recent studies. The purpose of this work is to provide a theoretical framework which resolves the controversy. It is shown that during neuronal activation residual vascular effects only represent a fraction of the DfMRI signal. The diffusion component largely dominates at very large b values before hemodynamic events take place. The vascular-related tissue relaxivity component dominates the DfMRI signal at the end of the activation window.

                  1570.     Monte-Carlo Simulations of BOLD Background Gradient Contributions in Diffusion-Weighted FMRI— Comparison of Spin Echo and Twice-Refocused Echo Sequences

Andre Pampel¹, Thies H. Jochimsen, ¹, Harald E. Moeller, ¹

¹Nuclear Magnetic Resonance Unit, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

The interaction (‘cross-terms’) between diffusion-weighting gradients and susceptibility-induced background gradient fields around vessels has an impact on ADC measurements using spin echoes. Monte-Carlo (MC) simulations integrating the Bloch equations numerically for a large number of random walks in a vascular network were used to investigate to what extent such interactions could influence the extravascular signal change observed in diffusion-weighted fMRI experiments. It is shown, that the Twice Refocused Echo sequence permits sufficient suppression of such contribution under experimentally relevant conditions.

                  1571.     Signal Source in Heavily Diffusion-Weighted Functional MRI

Daigo Kuroiwa¹, Hiroshi Kawaguchi¹, Jeff Kershaw¹, Atsumichi Tachibana¹, Joonas Autio¹, Masaya Hirano², Ichio Aoki¹, Iwao Kanno¹, Takayuki Obata¹

¹Department of Biophysics, Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Japan; ²Japan Applied Science Laboratory, GE Yokogawa Medical Systems, Hino, Tokyo, Japan

The contribution of the BOLD effect in heavily diffusion-weighted fMRI is still unclear. In this study, a multiple spin-echo echo-planar-imaging sequence after motion probing gradients was used to measure the transverse relaxation rate at different b-values. The results suggest that the contribution of BOLD to DW fMRI signal changes is independent of the b-value. It is speculated that the main contribution to heavily diffusion-weighted functional MRI signal is not the BOLD effect.

                  1572.     Diffusion-Weighted TE-Dependent FMRI Signal in Rat Somatosensory Cortex at 7 T

Joonas Arttu Autio^1,2, Jeff Kershaw^1,3, Takayuki Obata¹, Sayaka Shibata¹, Daigo Kuroiwa¹, Iwao Kanno¹, Ichio Aoki¹

¹Molecular Imaging Center, National Institute of Radiological Sciences, Inage-Ku, Chiba, Japan; ²National bio-NMR facility, University of Kuopio, Kuopio, Eastern Finland, Finland; ³School of Bioscience and Biotechnology, Tokyo Institute of Technology, Midori-Ku, Yokohama, Japan

Diffusion-weighted fMRI study was performed on alpha-chloralose anaesthetized rats at 7 T with somatosensory stimulation. Measurements were made for a range of b-values 0-2000 s/mm2 and echo-times 30, 60 and 90 ms. The signal increase often observed at high b-values was not observed, nor was there any significant attenuation of the post-stimulus undershoot.

                  1573.     Evaluation of Diffusion FMRI (DfMRI) with Short Event Related Paradigms

Toshihiko Aso¹, Cyril Poupon¹, Shin-ichi Urayama², Hidenao Fukuyama², Denis Le Bihan^2,3

¹NeuroSpin, I2BM/DSV/CEA, Gif-sur-yvette, France; ²Human Brain Research Center, Kyoto University Graduate School of Medicine, Kyoto, Japan; ³NeuroSpin, I2BM/DSV/CEA, Gif-sur-yvette,, France

Visual activation can be detected with heavily sensitized diffusion-fMRI (DfMRI). The striking temporal precedence of the diffusion response to BOLD suggests a non-vascular source. However, a residual tissue T2’ component sensitive to vascular (BOLD) effects also remains present in the DfMRI signal.The purpose of this study was to build a DhRF response function optimized for DfMRI taking into account both diffusion and vascular dynamics, and to evaluate it in an ultra-short event-related stimulation mode. This DhRF was shown to better predict activation in DfMRI data than the usual HRF, offering an improved sensitivity and specifity. The overall temporal precedence of the DfMRI signal over BOLD-fMRI was confirmed with those ultra-short event-related paradigms.

                  1574.     Cortical Layer-Dependent BOLD and Arterial Blood Volume Responses Measured by MT-Varied BOLD FMRI

Tae Kim¹, Seong-Gi Kim^1,2

¹Radiology, University of Pittsburgh, Pittsburgh, PA, USA; ²Neurobiology, University of Pittsburgh

The cortical depth profile of CBVa was measured to examine spatial specificity in a cortical layer model. BOLD and CBVa change during visual stimulation was simultaneously measured by MT-varied BOLD fMRI in cats. The highest signal changes were detected at cortical surface areas in conventional BOLD fMRI, but at the middle of cortex and the cortical surface in CBVa fMRI. This suggests that both parenchymal and upstream pial arterial vessels dilate, and the CBVa can be used to improve spatial specificity of fMRI signals within the cortex.

                  1575.     Measurement of Absolute CBV Change During Brain Activation Using Grey Matter Nulled 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

MR signal measured by grey matter nulled (GMN) fMRI is affected by partial volume effects from CSF, complicating quantification of possible change in absolute CBV. We present an approach to estimate change in absolute CBV to brain activation by determining blood and CSF signals separately at voxel level. Two different TRs were used to acquire GMN fMRI data which were used for quantification of CBV. Increase in absolute CBV was found to be 16.7 ± 5.8% in the visual cortex to checkerboard stimulation, which is significantly more than the increase in raw GMN fMRI signal by 7.0 ± 1.9%.

                  1576.     A Theoretical and Experimental Investigation of Vascular-Space-Occupancy (VASO) Blood Nulling Times:  Influence of Hematocrit and Oxygenation on Null Times and CBV Quantification

Manus Joseph Donahue¹, Stefan K. Piechnik¹, Rob Tijssen¹, Daniel Gallichan¹, Karla L. Miller¹, Peter Jezzard¹

¹Clinical Neurology, The University of Oxford, Oxford, UK

Traditionally, macrovascular blood T1 is nulled in CBV-weighted VASO experiments, although CBV changes occur in microvasculature. VASO fMRI (n=8) is investigated for varying blood-nulling inversion times corresponding to a range of blood oxygenation and hematocrit. Microvascular null times produce similar CBV changes (DCBV=18.9±2.7%) to macrovascular blood null times (DCBV=23.2±3.1%) at long TR=5s. In agreement with previous reports, we show that the choice of TI predictably influences the observed sensitivity of VASO to functional stimulation but that this influence is small at long TR over a blood T1 range corresponding to typical variation in average blood oxygen saturation and hematocrit.

                  1577.     Separation of the Vascular and Tissue Contributions to the T1ρ Change Induced by Brain Activation

Tao Jin¹, Seong-Gi Kim^1,2

¹Department of Radiology, University of Pittsburgh, Pittsburgh, PA, USA; ²Department of Neurobiology, University of Pittsburgh, Pittsburgh, PA, USA

T1rho of tissue water is a physiological parameter sensitive to the tissue microenviroment as well as macromolecular composition and density. We recently reported an activation-induced change of T1rho in the parenchyma of the cat visual cortex. In order to investigate whether this T1rho contrast has significant extravascular contribution, we have studied the T1rho fMRI response with and without suppression of the intravascular blood using a contrast agent. We found that a majority of the T1rho change arose from the extravascular tissue.

                  1578.     Pseudo-Continuous Arterial Spin Labeling with Optimized Tagging Efficiency for Quantitative ASL FMRI

Youngkyoo Jung¹, Anna Leigh Rack-Gomer², Eric C. Wong^1,3, Giedrius T. Buracas¹, Thomas T. Liu¹

¹Radiology, University of California, San Diego, San Diego, CA, USA; ²Bioengineering, University of California, San Diego, San Diego, CA, USA; ³Psychiatry, University of California, San Diego, San Diego, CA, USA

Arterial spin labeling fMRI can provide quantitative measurements of functional changes in CBF that can be used either on their own or in conjunction with BOLD measures. The PCASL method offers higher SNR and therefore the potential for improved detection of activation compared to pulsed ASL. The quantification of CBF activation with PCASL has not yet been examined in detail, and may be sensitive to phase errors. Here we optimized the PCASL method by estimating and compensating the phase errors at the tagging vessels. Our result shows good agreement in quantitative CBF measures between PICORE and our optimized PCASL method.

                  1579.     High Temporal Resolution Functional MRA for Investigating the Neural Activity

Chan-A Park¹, Chang-Ki Kang¹, Seung-Taek Oh¹, Young-Bo Kim¹, Zang-Hee Cho¹

¹Neuroscience Research Institute, Gachon University of Medicine and Science, Incheon, Korea

Even though the exact mechanism of the blood oxygen level dependent (BOLD) signal is still controversial, it is certainly that the signal change is related to the combination of physiological changes, which include the blood flow/volume change. Until now, with lower sensitivity of blood flow signal, temporal resolution was not enough to compare with fMRI. However, due to high sensitivity for blood flow signal, ultra high field MRI (7T) could provide the opportunity of detecting directly the vascular response by external stimulation. In this paper, we tried to reduce the temporal resolution to 3 sec. for examining the vascular response induced by stimulation, comparable to fMRI.

                  1580.     Quantification of Parenchymal T1 in the Activated Visual Cortex in Grey Matter Nulled and VASO FMRI Images

Yuji Shen¹, Ida Mengyi Pu², Risto A. Kauppinen³, Xavier Golay⁴

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; ⁴Department of Brain Repair and Rehabilitation, Institute of Neurology, University College London, London, UK

We examined spatial overlap and T1 values of the activated voxels revealed by grey matter nulled (GMN) and vascular space occupancy (VASO) fMRI techniques in the human visual cortex at 3T. Both fMRI techniques are designed to reveal blood volume changes during brain activation. It was found that the activated voxels by GMN and VASO fMRI methods overlap only by ~10%. Baseline T1 values in activated voxels were 1608 ± 283 ms and 1247 ± 214 ms for GMN and VASO fMRI, respectively, suggesting that these fMRI techniques probe different parenchymal compartments.

                  1581.     Stimulus Induced Modulation of Low Frequency Fluctuations in BOLD FMRI of the Rat

Waqas Majeed¹, Matthew Magnuson¹, Shella Keilholz¹

¹Biomedical Engineering, Georgia Institute of Technology / Emory University, Atlanta, GA, USA

We acquired fMRI data with and without continuous unilateral forepaw stimulation, from rats. Our results show that stimulation reduces power in the low frequency band in the stimulated area in the brain as well as the area 'functionally connected' to the stimulated area.

                  1582.     Voxel-Wise Regressor Selection for Physiological Noise Correction with RETROICOR

Mark Jenkinson¹, Rob Tijssen¹, Jonathan Brooks¹, Karla Miller¹

¹Clinical Neurology, FMRIB Centre, University of Oxford, Oxford, UK

A method is proposed that uses the Bayesian Information Criterion to adaptively select the most suitable set of RETROICOR regressors at each voxel separately. This is shown to dramatically reduce the number of regressors needed per voxel, with advantages in reducing degrees of freedom loss and minimising interaction between stimulus-related regressors and physiological noise regressors. Tests are performed with resting FMRI data with typical TR (3s) and low TR (0.15s).

                  1583.     Evaluation of Cyclic and Global Retrospective Corrections of Physiological Signals on Activated FMRI

Arsène Longin Ella¹, Jochen Rick¹, Jürgen Hennig¹

¹Dept. of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany

The cyclic retrospective correction of physiological signals, introduced along with Fourier and polynomial fit methods, was shown to be an alternative to RETROICOR (global method). In this study, we evaluated and compared different approaches of physiological signal correction on activated fMRI data acquired at different TR. The results showed: i) an increase in the number of activated voxels for all correction methods ii) no significant difference between global corrections performed using histogram and linear respiratory phases, either with Fourier or polynomial iii) a significant increase of the number of activated voxels after correction with cyclic methods compared to global methods.

                  1584.     Physiological Noise in GRAPPA FMRI Time-Series

Christina Triantafyllou^1,2, Mattijs Elschot^2,3, Jonathan R. Polimeni², Lawrence L. Wald^2,4

¹A.A. Martinos Imaging Center,  McGovern Institute for Brain Research, MIT, Cambridge, MA, USA; ²A.A. Martinos Center for Biomedical Imaging, Department of Radiology, MGH, Charlestown, MA, USA; ³Department of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, Netherlands; ⁴Harvard-MIT Division of Health Sciences and Technology, MIT, Cambridge, MA, USA

Highly parallel detection of functional imaging time-series provides the potential for higher image Signal-to-Noise ratio (SNR0) as well as decreased susceptibility distortions in echo-planar imaging. In this study we examine the effect of GRAPPA (R=1,2,3,4) on the time-course SNR (tSNR) at 3T with a 32 channel head coil for a number of image resolutions common to fMRI. Our findings suggest that the acceleration penalty (gÖR) behaves like other modulators of SNR0 such as field strength, voxel volume, flip angle and coil type.

                  1585.     Real-Time Observation of Spatiotemporal Dynamics of Arterial Pulsatility with MR-Encephalography

Jürgen Hennig¹, Thimo Grotz¹, Benjamin Zahneisen¹, Arsenne Ella¹, Maxim Zaitsev, Irina Mader², Andreas Harloff³

¹ Dept. of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany; ²Neuroradiology, University Hospital Freiburg, Freiburg, Germany; ³Neurology, University Hospital Freiburg, Freiburg, Germany

The paper presents results from fast measurements of the spatiotemporal measurements of arterial pulsatility in the area of the calcarine fissure measured by MR-encephalography (MREG). Results in 7 volunteers demonstrate that the peak of the arterial pulse wave is detected first in the cortext supplied by leptomeningeal arteries followed by white matter structures supplied through the circle of Willis. The sensitivity of the measurement allows real time detection of spatiotemporal variations with potential applications in stroke, stenosis and other hemodynamically relevant pathological conditions.

                  1586.     Physiological Noise Characteristics in FMRI of the Rodent at 11.7T

Daniel Kalthoff¹, Jörg Ulrich Seehafer¹, Dirk Wiedermann¹, Mathias Hoehn¹

¹In-vivo-NMR Laboratory, Max-Planck-Institute for Neurological Research, Cologne, Germany

Temporal SNR (tSNR) in fMRI studies can be compromised by physiological noise, which prevents SNR gains from higher field strength or improved hardware to be translated into fMRI sensitivity. In this study we investigated the presence of physiological noise in BOLD fMRI protocols for the rodent at very high field strength of 11.7 Tesla. Physiological Noise contributions were spatially mapped and analyzed for various voxel sizes in rats (n=5) under Isoflurane anesthesia. Results indicate that physiological noise contribution may be on the level of thermal noise or even the dominant source of noise for common fMRI protocols in this application.

                  1587.     Quantifying the Effects of CO2 on the Resting BOLD Signal

Tingying Peng¹, Rami Niazy², Richard Wise, Stephen Payne³

¹Institute of Biomedical Engineering, University of Oxford, Oxford, Oxfordshire, UK; ²CUBRIC, University of Cardiff, Cardiff, UK; ³Institute of Biomedical Engineering, University of Oxford, Oxford, UK

Spontaneous fluctuations in arterial carbon dioxide (CO2) have previously been identified as an important source of low frequency variations in the resting-state BOLD signal. Current study investigates the underlying relationship between CO2 and BOLD variations under two circumstances: eyeopen and eyeshut by using linear regression analysis and cross-correlation analysis. It has been found out that the time delay between the two signals is significantly shorter in eyeshut than in eyeopen. Furthermore, wavelet cross-correlation analysis is used to investigate the frequency-dependency of the CO2-BOLD coupling, suggesting the effects of CO2 are concentrated in the low frequency band 0-0.08Hz.

                  1588.     Effects of Sensitivity Encoding and Physiological Noise on Temporal Signal Stability and FMRI at 7T

John A. Sexton^1,2, James Chris Gatenby^1,3, John C. Gore^1,3

¹Vanderbilt University Institute of Imaging Science, Nashville, TN, USA; ²Biomedical Engineering, Vanderbilt University, Nashville, TN, USA; ³Department of Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, USA

We investigate the effects of changing SENSE acceleration factor R on the relative contributions of physiological noise to the total noise of an fMRI time series, fMRI sensitivity, and temporal signal stability at 7T. We observe decreases in the contributions of physiological noise as R increases, and we find excellent agreement between measured TSNR data and a theoretical model. We conclude that acquiring data with R = 4 is appropriate for single-shot EPI fMRI experiments like ours at 7T, and that parallel imaging coils with many elements can significantly improve fMRI results at high field.

                  1589.     B0 Dependence of Physiological Noise in BOLD FMRI

Tomas Jonsson^1,2, Terri Lindholm¹, Pierre Vestman¹, Maria Kristoffersen Wiberg², Jerzy Bodurka³, Tie - Qiang Li¹

¹Medical Physics, Karolinska Huddinge, Stockholm, Sweden; ²CLINTEC, Karolinska Institutet, Stockholm, Sweden; ³NIMH, National Institute of Health, Bethesda, MD, USA

Founded on a number of anticipated advantages, such as, gains in SNR and improved sensitivity in detecting brain activations, the interest in conducting BOLD at higher field strength is continuously growing. However, physiological noises are also amplified with the increased signal strength. Here we propose a straightforward procedure to remove cardiac and respiratory related signal fluctuations and characterize the remaining physiological noise level as a function of hardware settings.

                  1590.     Physiological Origin of Low Frequency Drift in BOLD FMRI

Lirong Yan¹, Yan Zhuo¹, Yongquan Ye¹, Sharon Xie², Jing An³, Geoffrey Aguirre⁴, Jiongjiong Wang⁵

¹State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, CAS, Beijing, China; ²Department of Biostatistics & Epidemiology, University of Pennsylvania, Philadelphia, PA, USA; ³Siemens Mindit Magnetic Resonance Ltd., Shenzhen, China; ⁴Neurology, University of Pennsylvania, Philadelphia, PA, USA; ⁵Radiology, University of Pennsylvania, Philadelphia, PA, USA

We investigated the biophysical mechanism of low frequency drift in BOLD fMRI (0~0.01Hz), by exploring its dependence on TE and image intensity as well as relationship with task-induced BOLD activation. Cardiac and respiratory signals were concurrently recorded during MRI scanning and subsequently removed. Drifts followed a characteristic dependence on TE and signal intensity that was similar to the BOLD contrast. There was a strong positive correlation between drift effects at baseline and tasked induced BOLD activation. Our study supports brain physiology, as opposed to scanner instabilities or cardiac/respiratory pulsations, as the main source of low frequency drifts in BOLD fMRI.

                  1591.     Characterizing Physiological Noise in the Brainstem: Passband SSFP Vs. GRE-EPI

Rob Hendrikus Tijssen¹, Mark Jenkinson¹, Peter Jezzard¹, Karla Loreen Miller¹

¹FMRIB Centre, Oxford University, Oxford, Oxon, UK

Functional imaging of the brainstem is problematic when using GRE-EPI BOLD due to susceptibility artefacts, small target nuclei, and high levels of physiological noise. Passband SSFP may provide an attractive alternative as it has the potential to produce high-resolution images with reduced distortions. This work presents a characterization of different physiological noise components in pbSSFP, SPGR and single-shot GRE-EPI. It was found that pbSSFP exhibits lower signal fluctuations at cardiac and respiratory frequencies compared to conventional GRE-EPI. These findings, in conjunction with the ability to obtain high-resolution, low-distortion images, may make pbSSFP an attractive option for brainstem fMRI.

                  1592.     The Use of Neurofeedback with Real-Time Functional MRI  to Suppress Physiological Noise.

Jerzy Bodurka¹, Javier Gonzalez-Castillo², Peter Bandettini^1,3

¹Functional MRI Facility, National Institute of Mental Health, NIH, Bethesda, MD, USA; ²Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, USA; ³Section on Functional Imaging Methods, National Institute of Mental Health, NIH, Bethesda, MD, USA

In the present study, we employed rtfMRI neurofeedback to investigate whether healthy subjects can learn to self-regulate the variability of the fMRI response in areas affected by high levels of physiological noise. In particular, subjects were asked to attempt to reduce the standard deviation (SDEV) of the fMRI signal in the ventricles and in the posterior sinus. We hypothesized that learning to self-control the variability of the fMRI signal in these extra-cortical areas would be accompanied by increased fMRI time series temporal signal to noise ratio (TSNR=mean voxel time course signal/time course standard deviation) in both white (WM) and gray (GM) matter compartments. Indeed five of six subjects studied were able to self-regulate the variability of the fMRI signal. Our results show: (1) that subjects can actively reduce SDEV of the fMRI signal in the ventricles and the posterior sinus using fMRI neurofeedback; and (2) such SDEV decrease is accompanied by a significant increase of fMRI TSNR in both white and gray matter compartments.

                  1593.     BOLD Physiological Noise Reduction Using Spatio-Spectral-Temporal Correlations with NIRS

Douglas N. Greve¹, Daniel Goldenholz¹, Gayatri Kaskhedikar¹, Jonathan Polimeni¹, Bruce Fischl¹, Lawrence L. Wald^1,2, Bruce Rosen¹, Christina Triantafyllou³, David Boas¹

¹Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; ²Harvard-MIT Division of Health Sciences and Technology, Cambridge, MA, USA; ³Massachusetts Institute of Technology, Cambridge, MA, USA

The purpose of this work is to improve the detection of BOLD fMRI

                  1594.     Relationship Between Respiratory Variations and End-Tidal CO₂ in BOLD FMRI Physiological Noise

Catie Chang¹, Gary H. Glover^1,2

¹Electrical Engineering, Stanford University, Stanford, CA, USA; ²Radiology, Stanford University, Stanford, CA, USA

One major source of fMRI physiological noise relates to variations in respiration depth and rate. It has been shown that respiratory variations (RVT), computed from pneumatic belt measurements of chest expansion during breathing, explains significant variance in fMRI timeseries. While RVT is believed to relate to arterial CO2, a vasodilator, the relationship between RVT and CO2 – as well as their

                  1595.     Reduction of Physiological Noise Effects in FMRI Phase Time Series

Gisela E. Hagberg¹, Marta Bianciardi², Valentina Brainovich¹, Bruno Maraviglia³

¹Santa Lucia Scientific Foundation, Rome, Italy; ²LFMI, NINDS, NIH, Bethesda, MD, USA; ³University La Sapienza, Rome, Italy

MRI is a phase sensitive detection system and both magnitude and phase time-series are obtained in an fMRI study. In a previous work it was shown that physiologic noise has a greater impact on phase than magnitude images. It was suggested that large scale fluctuations of the static magnetic field due to respiration and heart beat are the main cause of this difference. In this work we first characterised the effect of physiologic parameters (ECG and respiration) on the magnitude and the phase signal and then investigated the effect of different post-processing methods, including spatial filtering.

                  1596.     Spatiotemporal Investigation of the FMRI Response to Brief Somatosensory Stimulation in Awake Marmosets

Yoshiyuki Hirano¹, Junjie V. Liu¹, Bojana Stefanovic², Afonso C. Silva¹

¹Cerebral Microcirculation Unit, Laboratory of Functional and Molecular Imaging, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA; ²Sunnybrook Health Sciences Centre, Toronto, ON, Canada

The unceasing perfection of MRI hardware has led to continued improvement of the spatial and temporal resolution of fMRI. Here, the spatiotemporal features of the BOLD hemodynamic response (HDR) to brief somatosensory stimulation were measured in awake marmosets. The HDR to a single 333 µs electrical pulse could be robustly detected in S1 and S2, but not in caudate. Longer stimulus durations increased both the amplitude and spatial coverage of the HDR. The onset time of the BOLD HDR was shorter in S1 than either S2, suggesting the HDR starts in S1 before progressing to higher-order brain regions.

                  1597.     75 µm High-Resolution Parallel Imaging GE-EPI BOLD FMRI  in Rats at 11.7 Tesla: New Insights Into Cortical and Thalamic Micro-Structures

Joerg Ulrich Seehafer¹, Tom Geraedts², Mathias Hoehn¹

¹In-vivo-NMR Laboratory, Max-Planck-Institute for Neurological Research, Cologne, Germany; ²Biomedical NMR, Technical University Eindhoven, Netherlands

This fMRI study shows new insights into micro-structures of cortical columns and layers in S1 as well as detailed activation of thalamus by GE-EPI BOLD response following forepaw stimulation in rats using Parallel Imaging and high spatial resolution of 75 µm x 75µm x 1mm at 11.7 Tesla. Also, repetitive GE-EPI scans show continual micro-structure of activation with less susceptibility distortions and without venous vessel contribution. For the first time the activation cluster in S1 revealed areas of activation, but also areas of no activation, which give information about the actual sub-structures of cortical areas.

                  1598.     Cortical Boundaries Revealed by T₁ Mapping: Comparison with FMRI in Awake Marmosets

Junjie V. Liu¹, Nicholas A. Bock¹, Yoshiyuki Hirano¹, Afonso C. Silva¹

¹CMU/LFMI/NINDS, National Institutes of Health, Bethesda, MD, USA

We studied spatial variations in T1 across cortical gray matter, specifically the primary (SI) and secondary (SII) somatosensory cortices, using a new T1-mapping method based on inversion recovery. SI was distinguished from abutting cortical areas by its significantly lower T1 values. The anatomical SI-SII boundary demarcated from T1 map matched the functional SI-SII boundary, which was defined by the gap between two clusters of functional MRI responses to electrical stimulation of peripheral nerves. Thus, cortical architecture revealed by T1 mapping can corroborate the fMRI brain mapping results.

                  1599.     Neurophysiological Underpinnings of Ketamine-Induced Negative BOLD Response.

Naranjargal Dashdorj¹, Mirjam I. Schubert¹, Malcolm Prior², Rob Mason³, Dorothee Auer¹

¹Academic Radiology, University of Nottingham, Nottingham, UK; ²Brain and Body Centre, University of Nottingham, Nottingham, UK; ³School of Biomedical Sciences, University of Nottingham, Nottingham, UK

In depth understanding of drug-induced BOLD response and its underlying neural activity is essential for correct interpretation of phMRI. In this study we characterized ketamine induced BOLD signal changes in the hippocampus in a parallel electrophysiology and phMRI setup in rodent brain. Ketamine 25mg/kg i.p administration evoked both neural activity decrease (multi-unit activity and local field potential) and negative BOLD response (NBR) that was moderately correlated. To the best of our knowledge this is the first study to investigate drug-induced negative BOLD changes and its underlying neuronal activity changes. Our results suggest that drug-induced BNR reflects reduced neural activity.

                  1600.     Gradient-Echo and Spin-Echo BOLD FMRI of Rat Spinal Cord – Insight Into Its Hemodynamic Response to Neuronal Activity

Fuqiang Zhao¹, Denise Welsh¹, Xiangjun Meng¹, Mangay Williams¹, Jacquelynn J. Cook¹, Alise S. Reicin¹, Richard Hargreaves¹, Donald S. Williams¹

¹Imaging Department, Merck Research Laboratories, West Point, PA, USA

The mechanisms of hemodynamic response to neuronal activity in spinal cord have not been clearly elucidated. Whether the hemodynamic response in spinal cord is similar to that in brain is unknown. In this study, spin-echo and gradient-echo BOLD fMRI were performed in the rat cervical spinal cord to answer this question. Based on different vessel size dependence of SE and GE BOLD signals, by comparing the SE and GE BOLD measured by EPI, the vasculature involved in BOLD fMRI response in the cortex has been investigated. In this study, GE and SE BOLD measurements in cervical spinal cord of rat were performed. The stimulation-induced relaxation rate ratio ÄR2* / ÄR2 for GE vs SE is 1.25 / 0.35 =3.52, which is consistent with the ratios of 3.3-4.0 observed in various cerebral cortical regions. The similarity of these ratios between spinal cord and cortex supports the concept that the hemodynamic response and the vasculature of the spinal cord are similar to those of cerebral cortex.

                  1601.     Simultaneous FMRI and Field Potential Measurements of Epileptic Seizures in Rat Using RASER Pulse Sequence

Antti Markku Airaksinen¹, Juha-Pekka Niskanen^1,2, Joanna Katariina Huttunen¹, Ryan Chamberlain³, Michael Garwood³, Asla Pitkänen^4,5, Olli Gröhn¹

¹Department of Neurobiology, A. I. Virtanen Institute for Molecular Sciences, University of Kuopio, Kuopio, Finland; ²Department of Physics, University of Kuopio, Kuopio, Finland; ³Center for Magnetic Resonance Research and Department of Radiology, University of Minnesota Medical School, Minneapolis, MN, USA; ⁴Epilepsy Research Laboratory, A. I. Virtanen Institute for Molecular Sciences, University of Kuopio, Kuopio, Finland; ⁵Department of Neurology, Kuopio University Hospital, Kuopio, Finland

fMRI of medetomidine anesthetised rats was performed using RASER and simultaneous FP measurements in response to kainic acid (KA) induced seizures. The image distortion caused by measuring electrode was clearly seen in EPI image, whereas no artefact was seen in the RASER image. Robust BOLD responses were observed in the hippocampus during KA induced seizures. The recurrent epileptic seizures were detected in the FP signal after KA injection. The presented combination of deep electrode FP measurements and fMRI in medetomidine anesthesia that is not suppressing KA induced seizures provide a unique tool for studying abnormal brain activity in rat.

                  1602.     Comparison of Functional Activation in the Temporal Lobe of Awake and Anesthetized Monkeys

Jozien Goense¹, Hellmut Merkle², Nikos Logothetis^1,3

¹Max-Planck Institute for Biological Cybernetics, Tuebingen, Germany; ²NIH/NINDS, Bethesda, MD, USA; ³University of Manchester, Manchester, UK

The fMRI response to a movie stimulus was compared in the ventral visual pathway of awake and anesthetized macaques. The ventral visual stream is essential for object recognition and memory, and in awake monkeys large swaths of the pathway are activated. In anesthetized monkeys the temporal lobe also shows large areas of activation, corresponding to the areas in awake monkeys. It is reported difficult to elicit activation beyond early sensory areas, but our results show robust activation high in the visual pathway in areas involved in object recognition. The robust activation seen here is possibly due to the higher CNR at 7T.

                  1603.     Assessment of Rat Barrel Cortex BOLD Responses After Trigeminal Nerve Stimulation. on the Linearity of BOLD FMRI Changes Against Neuronal Activity in the Barrel Cortex

Nathalie Just^1,2, Stephane Germain^1,2, Martin Lauritzen^3,4, Rolf Gruetter^1,2

¹LIFMET, CIBM, EPFL, Lausanne, Switzerland; ²Department of Radiology, UNIL and HUG, Lausanne and Geneva, Switzerland; ³Department of Clinical Neurophysiology, Glostrup Hospital, Glostrup, Denmark; ⁴Department of Medical Physiology, Panum Institute and University of Copenhafen, Copenhagen, Denmark

The present study examined the BOLD fMRI response of the rat barrel cortex upon trigeminal nerve stimulation. BOLD changes as a function of current intensity and stimulus frequency were measured under α -chloralose anesthesia. The BOLD responses showed non-linear behavior as a function of current intensity and stimulus frequency that can be compared to the non-linear patterns found for cerebral blood flow and local field potentials measured in the barrel cortex as a function of the same stimulus parameters and stimulation model. Linear coupling between the neuronal and BOLD responses cannot be generalized for every condition or brain region.

                  1604.     Unexpected BOLD Behavior at 11.7 Tesla: A Comparison of Experimental Data with the Integrative BOLD Signal Model

Joerg Ulrich Seehafer¹, Kamil Uludag², Daniel Kalthoff¹, Tracy Deanne Farr¹, Mathias Hoehn¹

¹In-vivo-NMR Laboratory, Max-Planck-Institute for Neurological Research, Cologne, Germany; ²Hochfeld Magnetresonanz Zentrum, Max-Planck-Institute for Biological Cybernetics, Tübingen, Germany

Synopsis: This fMRI study presents experimental data and theoretical consideration about the unexpected BOLD behavior at high magnetic field strengths greater than 7.0 Tesla. Experimental SE- and GE-EPI data at 7.0 and 11.7 Tesla did not show an increase of BOLD contrast with field strength. The Integrative BOLD Signal Model delivers a possible explanation, namely the decrease of extravascular contributions from venules and arterioles. Furthermore, these predictions are compared to experimental data. It is shown, that the Integrative BOLD Signal Model can describe our data and is giving new perspectives to high field fMRI, which are of definite interest to future experiments.

                  1605.     High Resolution BOLD-FMRI of the Auditory System in Rats

xin Yu¹, Artem Goloshevsky¹, Stephen Dodd¹, Afonso Silva¹, Alan Koretsky¹

¹National Institutue of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, USA

There is growing interest in using MRI to study plasticity in the rodent brain. The auditory system has been important for studies of sensory plasticity. Previously, experience-dependent plasticity changes in the auditory midbrain were characterized with activation induced manganese-enhanced MRI (MEMRI). However, cortical reorganization could not be mapped with MEMRI. In this work, the auditory cortex and midbrain were mapped with BOLD-fMRI in rats anesthetized with propofol. 3D EPI images with 200 micron isotropic resolution was obtained in reasonable times. This work will enable analysis of interaction between the auditory midbrain and cortex under conditions where plasticity occurs.

                  1606.     Multi-Parametric Classification of FMRI-Activated Voxels Using Venous Vessel-Size, BOLD Latency and Susceptibility-Weighted Imaging

Thies H. Jochimsen¹, Harald E. Möller¹

¹Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Sachsen, Germany

Several strategies to increase the spatial specifity of BOLD-based fMRI by excluding the signal of large veins were combined into a single set of experiments in order compare their effectiveness. In particular, the BOLD latency, venographic intensity and venous vessel size of activated voxels was analyzed using a cluster and correlation analysis. The relatively good mutual correlation of venous vessel size, latency and venographic intensity suggests that these parameters can be used equally well to discriminate large veins.

                  1607.     Feasibility of Detecting Differential Layer Specific Activations in Humans Using SE BOLD FMRI at 7 T

Essa Yacoub¹, Kâmil Ugurbil², Cheryl A. Olman¹

¹Center For Magnetic Resonance Research, University of Minnesota, Minneapolis , MN, USA; ²Center For Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, USA

Studying layer-dependent neural activity is important because in all species cells, cortical afferents and projection neurons are arranged with reference to different depths in the gray matter. The investigation of layer specific functional activation in humans has been demonstrated at 3 T. Differential layer specific activation has not been demonstrated in any model. High field magnets have proven beneficial for fMRI with respect to spatial specificity and functional sensitivity, offering significant advantages over lower fields. In this work we demonstrate the potential of high field fMRI to observe layer specific activations as well as possible changes in these activations.

                  1608.     High Resolution Auditory FMRI at 7 T

Wietske van der Zwaag^1,2, Eleanora Fornari³, Philippe Maeder³, Rolf Gruetter^4,5

¹LIFMET, EPFL , Lausanne, VD, Switzerland; ²Radiology, UNIL, Lausanne, VD, Switzerland; ³Radiology, CHUV, Lausanne, VD, Switzerland; ⁴LIFMET, EPFL, Lausanne, VD, Switzerland; ⁵Radiology , UNIL, UNIG, Lausanne, Geneva, Switzerland

Auditory fMRI at ultra high field can be challenging because of the extended stray magnetic field. Here, we use an actively shielded head-only 7 T system and a gradient insert to allow good quality sound delivery to the subjects. Sparce fMRI data with a resolution of 1.5*1.5*2 mm was acquired during a word recognition task. Activation in the primary auditory cortex (planum temporalis, Superior temporal gyrus, Heshl’s gyrus) and voice and sentence processing was consistently found and was spatially constricted to the gray matter.

                  1609.     Latency and Echo-Time Dependence of the GRE-BOLD Signal in High-Resolution FMRI

Stefan Hetzer¹, Toralf Mildner¹, Thies H. Jochimsen¹, Karsten Müller¹, Torsten Schlumm¹, Harald E. Möller¹

¹Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

The relation between amplitude and latency of the BOLD signal was studied at 3T by use of high-resolution fMRI employing a center-out EPI sequence that allowed a minimum TE of 2 ms. The measured relaxation rate change δR2* during visual stimulation, i. e. the slope of the linear fit of the TE dependence and the corresponding intercept were found to increase with the BOLD latency. The δR2* increase over the total range of the BOLD latency was between 40 and 70%. In general, δR2* increased with image resolution which points to an underestimation of δR2* in low-resolution fMRI studies.

                  1610.     On the Cross-Modal Relationship Between FMRI and EEG

Zhongming Liu¹, Nanyin Zhang^2,3, Cristina Rios¹, Lin Yang¹, Wei Chen^2,3, 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

To test the linearity of the neurovascular coupling requires quantifying electrophysiological and hemodynamic signals before comparing them against a linear function. The variety of ways for quantifying multimodal signals may partly account for the existing disagreements on the linearity of the neurovascular coupling. In the present study, we modeled the cascaded interactions between stimuli and neural and vascular responses, and proposed a pair of quantitative measures for assessing the relationship between electrophysiological and hemodynamic signals. The neurovascular coupling in the human primary visual cortex was investigated through visual stimulation experiments using a variable visual contrast. Our theoretical and experimental results suggest a linear relationship between the BOLD effect size and the integrated power of the neural impulse response. This is regardless of the presence of the vascular nonlinearity attributed to the vascular refractory effect, since this nonlinear effect can be compensated by using the vascular refractory tuning function.

                  1611.     A Comparison of Quantitative Perfusion Measurements and MEG Phenomena

Claire Stevenson¹, Matthew Brookes¹, Peter Morris¹, Susan Francis¹

¹SPMMRC School of Physics and Astronomy, The University of Nottingham, Nottingham, Nottinghamshire, UK

ASL techniques provide a quantitative measurement of local tissue blood flow, and this may be more directly related to neuronal activation, as measured by MEG, than BOLD. The correlation between MEG phenomena and the changes in cerebral blood flow (CBF), T2* and BOLD, are investigated for a graded visual stimulus. Spatial separation of ASL and BOLD peaks and differences in stimulus related modulations suggest a decoupling of local perfusion and activity measured with BOLD. CBF shows a monotonic increase in amplitude with increasing stimulus contrast, similar to the visual evoked field and γ activity, however β activity is not modulated.

                  1612.     A Comparison of Source Localisation Techniques in Concurrent EEG/fMRI

Matthew J. Brookes¹, Karen J. Mullinger¹, Gerða Björk Geirsdóttir¹, Claire M. Stevenson¹, Richard Bowtell¹, Peter G. Morris¹

¹Sir Peter Mansfield Magnetic Resonance Centre, School of Physics and Astronomy, University of Nottingham, Nottingham, Nottinghamshire, UK

Simultaneous EEG/fMRI is an attractive means by which to investigate human brain function in vivo. Recent work has shown that the spatial location of changes in electrical neuronal activity can be determined using EEG data recorded during simultaneous fMRI. Further, such source localisation techniques can reduce significantly the level of interference in EEG timecourses caused by the MR scanner. In this work we compare source localisation methods. We show quantitative evidence that both beamformer and dipole fitting can be used for accurate source localisation; however the beamformer is markedly more effective at reducing interference in an EEG time course estimate.

                  1613.     Simultaneous EEG/fMRI Phantom Experiments with a Realistic Neuronal Signal

Gerða Björk Geirsdóttir¹, Matthew J. Brookes¹, Karen J. Mullinger¹, Winston X. Yan¹, Peter G. Morris¹, Richard W. Bowtell¹

¹Sir Peter Mansfield Magnetic Resonance Centre, The University of Nottingham, Nottingham, Nottinghamshire, UK

The large artefacts in the EEG signal produced by rapid gradient switching and pulsatile blood flow hamper EEG/fMRI measurements.

                  1614.     Visual BOLD-FMRI with 32 Channel Phased-Array Coil at 3.0T MRI System: Comparison with 12 Channel Coil

Jianqi Li¹, Lijia Wang¹, Yi Wang^2,3

¹Shanghai Key Laboratory of Functional Magnetic Resonance Imaging and Department of Physics, East China Normal University, Shanghai, China; ²Department of Biomedical Engineering, Cornell University, Ithaca, NY; ³Department of Radiology, Weill Medical College of Cornell University, New York, USA

A 32-channel 3T receive-only phased-array head coil was recently developed for human brain imaging and showed significantly increased image signal-noise-ratio in the cortex. In this study, the activated volume in visual BOLD-fMRI was evaluated at different spatial resolutions with a 32 channel WIP phased-array coil and a commercial 12 channel coil at 3T MRI system. It was found that the 32 channel coil is beneficial for fMRI at high spatial resolution but may not provide significant benefit at lower spatial resolution.

                  1615.     Dependencies of the Negative BOLD Signal in Primary Somatosensory Cortex on Stimulation Intensity and Duration

Katharina Schaefer¹, Henrik Bo Wiberg Larsson¹, Martin Lauritzen^2,3

¹Department of Clinical Physiology and Nuclear Medicine, Glostrup Hospital, Glostrup, Denmark; ²Department of Clinical Neurophysiology, Glostrup Hospital, Glostrup, Denmark; ³Department of Medical Physiology, The Panum Institute, University of Copenhagen, Copenhagen, Denmark

Unilateral electrical stimulation of the somatosensory system was performed in high and low stimulation intensities. High intensity stimulation elicited a positive contralateral and a negative ipsilateral BOLD signal in the primary somatosensory cortex (SI). Additionally a spatially extensive negative BOLD signal was found in the parietal and occipital cortex. This negative BOLD signal might represent functional inhibition of the sensory map of the human body.

                  1616.     Further Evidence of Initial BOLD Dip Across Cortico-Thalamic Visual Network During Visual Stimulation

Nanyin Zhang¹, Xiao-Hong Zhu¹, Yi Zhang¹, Wei Chen¹

¹Radiology, University of Minnesota, Minneapolis, MN, USA

The initial dip in the BOLD signal is controversial. In this study we explored the existence and generality of initial dip using a paired-stimulus paradigm. We observed that for the pixels showing initial dip to the first stimulus, they also show obvious initial dip to the second stimulus. The delay between the two dips correlates with the delay between the stimulation pair. Moreover, our study demonstrates that initial dip also exists in sub-cortical brain areas such as thalamus, and suggests the possibility to apply initial dip for functional mapping of the entire cortico-thalamic visual network with improved specificity.

                  1617.     Integrated Magnetic Field Variation Correction for Quantitative FMRI

John David Dickson¹, Guy B. Williams², Sally Georgia Harding², Thomas Adrian Carpenter², Richard E. Ansorge¹

¹Cavendish Laboratory, Cambridge University, Cambridge, Cambs, UK; ²Wolfson Brain Imaging Centre, Cambridge University, UK

It is possible to measure the venous Cerebral Blood Volume (vCBV) and Oxygen Extraction Fraction (OEF) by quantifying the BOLD signal. If this is to be used in a routine clinical setting then the scan times must be kept low. This study shows how the need for an external fieldmapping acquisition can be avoided using an integrated algorithm based on the phase data. These methods are shown to provide realistic measures of OEF and vCBV in a healthy volunteer.

                  1618.     Susceptibility Artefacts in Experiments Involving Changes in Inspired Oxygen Level

Nicholas Paul Blockley¹, Ian D. Driver², Susan T. Francis², Joseph A. Fisher³, Penny A. Gowland²

¹SPMMRC, School of Physics and Astronomy, University of Nottingham, Nottingham, UK; ²SPMMRC, School of Physics and Astronomy,, University of Nottingham, Nottingham, UK; ³University of Toronto, Toronto, Ontario, Canada

Changes in oxygen concentration in the nasal cavity during a hyperoxic challenge cause a variation in the susceptibility gradient at the sinus-tissue interface. We found that this caused a measurable dipolar pattern of signal change in slices inferior to the corpus callosum along with more subtle changes in superior slices. This artefact will be

                  1619.     SNR Optimization of MION FMRI in the Anaesthetized Monkey Using an 8-Channel PA-Coil and Accelerated Imaging

Thomas Janssens¹, Wim Vanduffel¹, Hauke Kolster¹

¹Lab for Neuro- and Psychophysiology, KU Leuven, Leuven, Belgium

In high-resolution awake monkey fMRI studies contrast agents (MION) are used to improve CNR compared to BOLD measurements. MION, however, decreases SNR significantly. To mitigate this issue, we use an 8-channel coil, accelerated data acquisition, and regularized SENSE reconstruction. In addition to a factor of 1.5 gain in SNR in cortex due to the increased sensitivity of the PA coil, we show that for accelerated imaging a minimization of TE and TR leads to an additional gain factor of 1.6 in SNR, to a total of 2.4 above a single loop coil, which can be used to further increase resolution.

                  1620.     Modeling the Effect of Changes in Arterial Blood Volume on the BOLD Signal

Richard Buxton¹

¹University of California, San Diego, La Jolla, CA, USA

A model for the BOLD effect was extended to include the role of arterial CBV changes, and compared with the simpler Davis model that is typically used in the calibrated BOLD experiment to analyze combined CBF/BOLD data and extract an estimate of the change in CMRO2 with brain activation. Simulations indicate that the Davis model is reasonably robust, but tends to underestimate CMRO2 changes when arterial CBV changes are large. For this reason, this potential source of error cannot reconcile calibrated BOLD measurements with some PET measurements finding much smaller changes in CMRO2.

                  1621.     Quantitative Combined ASL/BOLD Imaging: Implications for the Interpretation of the BOLD Post-Stimulus Undershoot

Valerie Griffeth¹, Aaron Simon¹, Joanna Perthen¹, Richard Buxton²

¹University of California, San Diego; ²University of California, San Diego, La Jolla, CA, USA

Two theories for the mechanism of the BOLD post-stimulus undershoot seen with brain activation are a slow recovery of CBV or a slow recovery of CMRO2. Dual-echo spiral ASL measurements of responses to a visual stimulus were analyzed to determine apparent M0 and R2* values for both the total signal and the tagged component delivered by arterial flow. In addition to CBF and BOLD responses, that allowed calculation of CMRO2 with a calibrated BOLD approach, a CBV-effect (similar to VASO) was also isolated. In this study the BOLD post-stimulus undershoot was not associated with a slow recovery of CBV nor with a slow recovery of CMRO2. Instead, it was consistent with a vascular origin due to a slight undershoot of CBF.

                  1622.     Balloon Modeling of BOLD Responses in Children with Sickle Cell Anemia (SCA)

Robert Ogg¹, Ping Zou¹, Winfred Wang¹, Russel Ware¹, Kathleen Helton¹

¹St. Jude Children's Research Hospital, Memphis, TN, USA

Elevated resting cerebral blood flow in SCA patients may decrease detection of activation with fMRI. We used fMRI during visual stimulation to characterize the hemodynamic responses in children with SCA. Detection, magnitude of BOLD signal, and activation volume were significantly lower in SCA patients than control subjects. Modeling showed that differences in the BOLD response were accounted for by differences in CBF response, oxygen extraction fraction during stimulation, CBV, and vascular mechanical characteristics. These findings suggest that the balloon model is useful to relate altered BOLD response to important clinical and physiological parameters in patients with SCA.

                  1623.     Theoretical Prediction of Parameter Stability in Quantitative BOLD MRI: Dependence on SNR and Sequence Parameters

Maja Cornelia Sohlin¹, Lothar R. Schad²

¹Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany; ²Computer Assisted Clinical Medicine, Faculty of Medicine Mannheim, University of Heidelberg, Germany

A static dephasing model that analytically connects BOLD signal to hemodynamic parameters can be used to map the blood oxygenation level (Y) and venous cerebral blood volume (vCBV) in the brain. In this work, the accuracy of the method is tested by means of simulations and measurements. The result shows that accurate fitting can only be performed at high SNR (>500). A separate quantization of the vCBV would allow a stable method to quantify blood oxygenation even at low SNR (<200).

                  1624.     Comparison of Methods Used in Calibrated FMRI

Jerod Michael Rasmussen¹, Sumiko Abe¹, Liv C. McMillan¹, Josef Pfeuffer², Jessica Turner¹, Function BIRN³

¹University of California, Irvine, Irvine, CA, USA; ²Siemens Medical Solutions; ³www.nbirn.net

A comparison of calibration methods and their effect on the strength of correlations between: Cerebral Blood Flow (CBF), a hypercapnic task and a cognitive task. Calibration methods compared are the different permutations between active state CBF, resting state CBF, composite maps of hypercapnic breath holding and composite maps of a cognitive sensorimotor task. Each composite map has a highly temporally correlated interleaved ASL/BOLD and a traditional EPI BOLD acquisition version for comparison. Results demonstrate that using the active CBF ascertained from the interleaved ASL/BOLD acquisition method has the highest potential for calibration.

                  1625.     Calibration of ASL Quantification Based on Hypercapnia and Sensory Motor Activation

Sumiko Abe¹, Jerod Michael Rasmussen¹, Jessica Turner¹, L. Ch McMillan¹, Josef Pfeuffer², F. BIRN³

¹University of California, Irvine, Irvine, CA, USA; ²Siemens Medical Solutions USA Inc., USA; ³www.nbirn.net

To evaluate the feasibilities of quantifying cerebral perfusion with dynamic magnetic resonance artery spin labeling (ASL) sequence, we studied 4 subjects under BH and CB activities for evaluating the correlations of CBF with CMRO2 and M coefficient (i.e. CBV) with CMRO2. Based on analysis results, we confirmed that BH task increased CBF and it is not significant or weak effects to CMRO2 changes. CB task doesn¡¯t increase the GM CBF as much as BH task but CMRO2 is increased significantly due to CBF and M increasing in specified areas such as vision and motor areas.

                  1626.     Validation of Optical Measurements of Cerebral Blood Flow and Volume with SPION and ASL FMRI: Implications for CMRO2 Changes During Hypercapnia

Stefan A. Carp¹, David A. Boas¹, Maria A. Franceschini¹, Bruce Ro Rosen¹, Young Ro Kim¹

¹Athinoula Martinos Center for Biomedical Imaging/ Massachusetts General Hospital, Charlestown, MA, USA

Direct measurement of variations in the cerebral metabolic rate of oxygen (CMRO2) can be very useful in the study of neuro-vascular coupling in various disease states. We attempt to validate near-infrared optical measurements of cerebral blood volume and blood flow in a rat model during hypercapnia using functional MRI, with the purpose of developing a non-invasive method for measuring relative CMRO2 changes.

                  1627.     The BOLD-Specific Flow-Volume Relationship During Hypercapnia and Hypocapnia

Jean J. Chen¹, G. Bruce Pike¹

¹McConnell Brain Imaging Centre, Montreal Neurological Institute, Montreal, Quebec, Canada

Grubb’s power-law (with α =0.38) has been extensively used in fMRI, and the equivalence of this flow-volume relationship observed under neuronal activation and hypercapnia is instrumental in CMRo₂ estimation using calibrated BOLD. However, Grubb’s α was based on measurements of total δCBV instead of the BOLD-specific venous δCBV. Moreover, the comparability of the venous flow-volume relationship under focal and CO₂-induced hyperemia has not been established. We used the venous refocusing for volume estimation (VERVE) technique to measure the venous flow-volume relationship in humans under graded hypercapnia and hypocapnia, and found α to be 0.19±0.04, equivalent to findings for neuronal activation.

                  1628.     T2* Responses in Grey Matter Nulled and VASO FMRI Images During Visual Stimulation in Hypoxic Hypoxia

Yuji Shen¹, Ida Mengyi Pu², Y-C L. Ho³, R Vidyasagar⁴, Xavier Golay⁵, Risto A. Kauppinen⁶

¹School of Medicine, University of Birmingham, Birmingham, UK; ²Department of Computing, Goldsmiths, University of London, London, UK; ³Department of Neuroradiology, National Neuroscience Institute, Singapore; ⁴MARIARC, University of Liverpool, Liverpool, UK; ⁵Department of Brain Repair and Rehabilitation, Institute of Neurology, University College London, London, UK; ⁶Biomedical NMR Research Center, Dartmouth Medical School, Hanover, NH 03755, USA

Grey matter nulled (GMN) and VASO fMRI methods were used to measure CBV and T2* changes associated with brain activation in different parenchymal compartments, i.e., blood and extravascular tissues. fMRI experiments were performed in normoxia and mild hypoxic hypoxia. It was observed that while the CBV-weighted signal changes to visual stimulation remain similar in both normoxia and hypoxia, the activated areas are significantly reduced by 72±22% in GMN and 66±23% in VASO fMRI in hypoxia. Interestingly, T2* decreased in GMN active volume during hypoxia in contrast to increased T2* in normoxia, while T2* in VASO-activated areas increased in both conditions.

                  1629.     Vasoreactivity Mapping Using Dynamic (C)O₂ Enhanced MRI

Stefanie Remmele¹, Julien Sénégas¹, Andreas Müller², Ingobert Wenningmann³, Frank Träber², Hans Clusmann⁴, Hannes Dahnke⁵, Jürgen Gieseke⁵, Sebastian Flacke², Winfried A. Willinek², Hans H. Schild², Petra Mürtz²

¹Philips Research Europe, Hamburg, Germany; ²Department of Radiology, University of Bonn, Bonn, Germany; ³Department of Anesthesiology, University of Bonn, Bonn, Germany; ⁴Department of Neurosurgery, University of Bonn, Bonn, Germany; ⁵Philips Medical Systems, Hamburg, Germany

The dR2* response of a tumor to inhalation of O2- or CO2-enriched air is known to give insight into blood oxygenation and vasoreactivity. This work demonstrates that tumors may differ from normal tissue in the amplitude but also the temporal behavior (delay, velocity) of the response. We further propose an approach to robust dynamic (C)O2-enhanced MR measurement, analysis and depiction of the response in high-resolution vasoreactivity maps. The technique was evaluated in 5 patients with cerebral tumors.

                  1630.     The Ratio of CBF to CMRO2 Change with Brain Activation Increases with Increasing Stimulus Amplitude in Human Visual Cortex

Christine Liang¹, Beau Ances¹, Joanna Perthen¹, Joy Liau¹, Giedrius Buracas¹, Susan Hopkins¹, Richard Buxton²

¹University of California, San Diego; ²University of California, San Diego, La Jolla, CA, USA

A calibrated-BOLD method was used to test whether the CBF/CMRO2 coupling ratio n, defined as the ratio of the fractional changes in CBF and CMRO2 with brain activation, changes with stimulus intensity. In human visual cortex, the ratio of the CBF and BOLD responses decreased with increasing contrast of a flickering checkerboard, consistent with n increasing from ~1.7 to ~2.3. This steady increase in CBF, with a rolling off of CMRO2, may be consistent with CBF driven by the initial excitatory phase of neural activity while CMRO2 responds to the overall energy needs of the full evoked response.

                  1631.     Grey & White Matter Cerebrovascular Reactivity Response to Hypercapnia at 3 & 7T

Ian David Driver¹, Nicholas P. Blockley², Susan T. Francis¹, Penny A. Gowland¹

¹SPMMRC, School of Physics and Astronomy, University of Nottingham, Nottingham, UK; ²SPMMRC, School of Physics and Astronomy, University  of Nottingham, Nottingham, UK

The BOLD signal response to mild hypercapnia provides useful information about the BOLD haemodynamic response and has applications in cerebrovascular diseases. Typical problems in previous experiments include poor information regarding end-tidal gas concentrations and a coupling between CO₂ and O₂ levels. This study used high resolution imaging to reduce partial volume effects and a system providing independent control of end tidal CO₂ and O₂. Rapid monitoring of end-tidal CO₂ enabled correlation of the BOLD signal against actual hypercapnic level at each time point rather than prescribed level of hypercapnia. This provided sufficient sensitivity to observe both grey and white matter reactivity in response to hypercapnia.

                  1632.     Sinusoidally Modulated CO₂ Stimulus Provides New Temporal Information on Cerebrovascular Reactivity

Nicholas Paul Blockley¹, Ian D. Driver², Susan T. Francis², Joseph A. Fisher³, Penny A. Gowland²

¹SPMMRC, School of Physics and Astronomy, University of Nottingham, Nottingham, UK; ²SPMMRC, School of Physics and Astronomy,, University of Nottingham, Nottingham, UK; ³University of Toronto, Toronto, Ontario, Canada

We present a new method for generating cerebrovascular reactivity maps.

                  1633.     Feasibility of Estimating CBF/CMRO2 Coupling with a Calibrated-BOLD Method When the Driving Stimulus Is Unknown

Aaron Simon¹, Valerie Griffeth¹, Joanna Perthen¹, Richard Buxton²

¹University of California, San Diego; ²University of California, San Diego, La Jolla, CA, USA

The calibrated BOLD method provides a quantitative measurement of CBF/CMRO2 coupling during brain activation. The analysis is typically applied to a region of interest (ROI) selected based on correlation with a known stimulus. We tested the feasibility of using just the coherence of the CBF and BOLD signals measured simultaneously with a dual-echo ASL method to identify an active ROI, with no prior knowledge of the driving stimulus. In human data with a visual stimulus, the CBF/BOLD coherence approach and the standard model correlation approach gave ROI estimates of CBF/CMRO2 coupling that differed by only 7%. This opens the possibility of measuring CBF/CMRO2 coupling with more complex or naturalistic stimuli, supporting a wider range of application of the calibrated-BOLD method.

                  1634.     An Alternative Technique for Measuring Cerebrovascular Reactivity; Comparing Cued Deep Breathing Hypocapnia with Inspiration of Carbon Dioxide

Molly Gallogly Bright^1,2, Daniel P. Bulte², Peter Jezzard², Jeff H. Duyn¹

¹Advanced MRI Section, LFMI, NINDS, National Institutes of Health, Bethesda, MD, USA; ²FMRIB Centre, Department of Clinical Neurology, University of Oxford, Oxford, UK

Recently we introduced a novel respiratory technique, cued deep breathing (CDB), which induces robust BOLD signal decreases. The strong, short-lived response to CDB allows for whole-brain spatial and temporal characterization of cerebrovascular reactivity, and CDB is suitable for a broad range of clinical situations. We have directly compared the quantitative effects on end-tidal gas levels and BOLD signal of CDB and a traditional hypercapnia 4% CO2 gas challenge. CDB induced up to 7.0±0.7 mmHg decreases in end-tidal CO2 and the resulting reactivity maps show good correlation with the CO2 study, indicating CDB offers easy access to similar information.

                  1635.     Dependence of BOLD Signal Amplitude on Baseline Venous Oxygenation and Cerebral Blood Flow

Joy Liau¹, Youngkyoo Jung², Anna Leigh Rack-Gomer³, Thomas Liu⁴

¹UC San Diego Center for Functional MRI, La Jolla, CA; ²UCSD Center for fMRI, La Jolla, CA; ³UCSD Center for FMRI, La Jolla, CA; ⁴UC San Diego Center for Functional MRI, La Jolla, CA, USA

Prior studies have shown a strong correlation across a sample of healthy subjects between the BOLD signal amplitude and measures of baseline venous oxygenation. In addition, a significant correlation has been shown between baseline venous oxygenation and cerebral blood flow (CBF). However, while some studies have found a significant correlation between BOLD signal amplitude and baseline CBF, others have not. In this study we demonstrate that the correlation between BOLD signal amplitude and baseline CBF depends on the gray matter volume fraction of the voxels used to compute the baseline CBF measures.

                  1636.     Inter-Subject Variability in Hypercapnic Normalization of the BOLD FMRI Response

Joy Liau¹, Thomas Liu²

¹UC San Diego Center for Functional MRI, La Jolla, CA; ²UC San Diego Center for Functional MRI, La Jolla, CA, USA

Hypercapnic normalization has been proposed as a method for reducing inter-subject variability in the amplitude of the BOLD fMRI signal. Some prior studies have reported a decrease in variability, while others have found an increase. Here we show that the presence of an intercept term in the relation between the functional and hypercapnic BOLD responses results in a systematic bias term in the normalized response that can lead to increased inter-subject variability. An alternate approach that treats the hypercapnic BOLD response as a covariate can be used to minimize this bias term and reduce inter-subject variability in the normalized response.

                  1637.     Stress Testing the Brain at 7T with Breath Holding

Hans Hoogduin¹, Lars Bosboom¹, Mandy Conijn¹, Willem Mali¹, Peter Luijten¹

¹University Medical Center, Utrecht, Netherlands

Breath holding can provide information on Cerebrovascular Reactivity when combined with BOLD fMRI. The possibilities of the method are explored in healthy volunteers at 7T. The increased BOLD contrast of this field strength provides a step towards clinical use of the method.

                  1638.     Breath Hold Task Significantly Correlates with Cognition Across Run and Visit

Jerod Michael Rasmussen¹, Sumiko Abe¹, Liv C. McMillan¹, Josef Pfeuffer², Jessica Turner¹, Function BIRN³

¹University of California, Irvine, Irvine, CA, USA; ²Siemens Medical Solutions; ³www.nbirn.net

The hypercapnic breath hold task is a non-invasive measurement of Cerebral Blood Flow (CBF) and has been shown to reduce variance across subjects through calibration. Subject wise correlations between breath hold activity and working memory are calculated across visit and run. Correlations were found to be positive and strong demonstrating the physiological coupling between CBF and the Blood Oxygenation Level Dependent (BOLD) signal. These results suggest confounds of run and visit can be accounted for in a calibrated model.

                  1639.     Study of Tricarboxylic Acid Cycle Flux Changes in Human Visual Cortex During Two-Hemifield Visual Stimulation with Different Stimulus Frequency Using In Vivo ¹H-{¹³C} MRS and FMRI

Fei Du¹, Malgorzata Marjanska¹, Xiao-Hong Zhu¹, Anjali Kumar², Elizabeth R. Seaquist², Kamil Ugurbil¹, Wei Chen¹

¹Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, USA; ²General Clinic Research Center, University of Minnesota, Minneapolis, MN, USA

The coupling relationships among the stimulus-evoked brain activity, hemodynamic and metabolic responses are critical for understanding the mechanism underlying fMRI BOLD signal and brain function. They are, nevertheless, highly debated. One particularly interesting question is whether the increase in CMRO₂ during graded brain activation follows BOLD and CBF changes. Previously published PET functional study demonstrated that CMRO₂ changes were larger at 4 Hz reversal frequency (≈18%) than at 8 Hz (≈4%) using checkerboard visual stimulus despite the opposite trend for CBF changes (≈34% at 4 Hz and ≈46% at 8 Hz). The present study aimed to re-examine the CMRO₂-BOLD coupling relationship between 4 and 8 Hz visual stimuli by simultaneously measuring BOLD and the changes of relative oxidative CMR</sub>glc</sub> in two hemispheric visual cortical regions, which were stimulated at 4 Hz and 8 Hz, respectively. The results indicate strong CMRO₂-BOLD as well as CMRO₂-CBF coupling relationships in the human brain during graded visual stimulation.

                  1640.     Nulling the CSF Signal in Quantitative FMRI

John David Dickson¹, Guy B. Williams², Sally Georgia Harding², Thomas Adrian Carpenter², Richard E. Ansorge¹

¹Cavendish Laboratory, Cambridge University, Cambridge, Cambs, UK; ²Wolfson Brain Imaging Centre, Cambridge University, UK

Quantifying the BOLD signal allows the measurement of local blood oxygenation. This could provide a useful tool for the analysis of cerebral vascular disorder; however the presence of Cerebral Spinal Fluid (CSF) can confound this process. This study shows how realistic measures of blood oxygenation can be made in voxels of partial CSF volume by nulling the CSF signal. This technique, combined with parallel acquisition allows useful data to be acquired within clinically realistic scan times.

                  1641.     Identification of Functional Changes in Early Alzheimer Disease Using Resting-State FMRI

Xiaowei Song^1,2, Ryan D'Arcy^1,3, Andrew Ryan McIntyre^1,4, Alma Major^1,2, Steven Beyea^1,5, Kenneth Rockwood^2,6

¹Institute for Biodiagnostics - Atlantic, National Research Council, Halifax, NS, Canada; ²Medicinie, Dalhousie University, Halifax, NS, Canada; ³Psychology and Neuroscience, Dalhousie University, Halifax, NS, Canada; ⁴Faculty of Computer Science, Dalhousie University, Halifax, NS, Canada; ⁵Physics, Dalhousie University, Halifax, NS, Canada; ⁶Centre for Health Care for the Elderly, QEII Health Sciences Centre, Halifax, NS, Canada

In this study, we investigated the low-frequency fMRI signal fluctuations during a resting phase to identify changes in the brain associated with AD, using an efficient un-supervised SOM-clustering method. Results indicate altered spontaneous fMRI fluctuations that suggest increased functional activation in certain cortical regions and decreased global functional synchronization connecting many such regions.

                  1642.     Social Network Theory Applied to Resting-State FMRI Connectivity Data in the Analysis of Epilepsy Networks

Xiaohui Zhang¹, Fuyuze Tokoglu², R. Todd Constable²

¹Department of Diagnostic Radiology, School of Medicine , Yale University, New Haven, CT, USA; ²Department of Diagnostic Radiology, School of Medicine, Yale University, New Haven, CT, USA

Epilepsy is a brain disorder with the essential basis of abnormal cortical and/or subcortical networks. Currently, most of the evidence of seizure networks come from ictal EEG observations. Resting-state functional connectivity studies can be helpful to localize abnormal networks and widen the array and approach of therapeutic options. This study aims to classify patients¡¯ data from control subjects by characterizing the interictal epilepsy connectivity networks using social network topology in functional MRI. An average sensitivity of the 87.5% and specificity of 78.9% were achieved in the given data set.

                  1643.     Functional Subdivision of Corticostriatal Interconnections Visualized by FMRI

Julia M. Boehm^1,2, Veit Roessner², Peter Dechent¹, Juergen Baudewig¹

¹MR-Research in Neurology and Psychiatry, University of Goettingen, Goettingen, Germany; ²Department of Child and Adolescent Psychiatry, University of Goettingen, Goettingen, Germany

Corticostriatal interconnections are thought to be organized in anatomically and functionally segregated circuits. We attempted visualization of the pure motor and the associative loop by using a self-paced finger tapping task performed with either hand separately. Brain regions involved in both, left and right unimanual movements were identified via conjunction analysis, a contrast analysis revealed regions behaving movement side specifically. Coactivation of anterior SMA, pre-SMA and anterior striatum allowed assignment to associative parts, coactivation of posterior SMA, SMC and posterior striatum to purely executive parts of corticostriatal interconnections. We provide a new in-vivo visualization of corticostrial interconnections using BOLD fMRI.

                  1644.     Impaired Thalamic Functional Connectivity in Vegetative State Detected by FcMRI

Jingsheng Zhou^1,2, Chunming Xie¹, Wenjun Li¹, Feng Ling², Shi-jiang Li¹

¹Biophysics Department, Medical College of Wisconsin, Milwaukee, WI, USA; ²Xuanwu Hospital Capital Medical University, BJ, Beijing, China

                 1677.     Validation of Awake Rat Model for Measures of Pain and Analgesia with FMRI

Lino Becerra^1,2, Pei-Ching Chang¹, James Bishop¹, Eric Crown³, Yang Tsai³, Mark Urban³, Michael Klimas³, Richard Hargreaves³, Smriti Iyengar⁴, Rosa Maria Simmons⁴, Steve C. Peters⁴, Adam James Schwarz⁴, David Bleakman⁴, David Borsook^1,2

¹Imaging Consortium for Drug Development, McLean Hospital, Belmont, MA, USA; ²Athinoula A. Martinos Center, Massachusetts General Hospital, Charlestown, MA, USA; ³Merck and Co, West Point, PA, USA; ⁴Eli Lilly and Co., Indianapolis, IN, USA

Functional imaging studies of rodents generally use anesthetics and paralyzers to control stress levels and motion. For pain studies, however, the use of anesthetics introduces confounds that hamper our ability to measure pain responses in the brain and their modulation by analgesics. An awake animal fMRI pain model, with proper control of stress and motion, is necessary. Here, we evaluate an awake pain model and compare it to traditional behavioral pain studies to determine the optimal conditions for imaging.

                  1678.     Opposite Parietal and Midbrain BOLD Patterns During Inflammatory Pain in the Rat: A High-Field FMRI Study

Ariel Graff-Guerrero^1,2, Juan Manuel Ortega-Legaspi³, Camilo De la Fuente-Sandoval⁴, Francisco Pellicer³, Desiree López⁵, Alvino Moreno⁵, Rodrigo Martín⁵, Sergio Solis⁵, Silvia Hidalgo⁵, Alfredo O. Rodriguez⁵

¹Instituto de Neurobiologia, Universidad Nacional Autonoma de Mexico, Mexico, DF, Mexico; ²PET Centre, Centre for Addiction and Mental Health, Toronto, Ontario, Canada; ³Laboratorio de Neurofisiología Integrativa, Instituto Nacional de Psiquiatria Ramon de la Fuente, Mexico; ⁴Laboratorio de Psiquiatria Experimental, Instituto Nacional de Neurologia y Neurocirugia, Mexico; ⁵Electrical Engineering Department, Universidad, Autónoma Metropolitana-Iztapalapa, Mexico, DF, Mexico

The development of high-field magnetic resonance scanners, for small animals, have redirected the study of pain using fMRI in preclinical models. While human studies have advantages to study cognitive-psychological processes during pain perception, animal studies allow to correlate fMRI with basic mechanisms.

                  1679.     Spin Versus Gradient Echo Sequence: What Is Better for Pain Research with Small Animal FMRI?

Lubos Budinsky¹, Silke Kreitz¹, Marina Sergejeva¹, Nicole Motzkus¹, Anna Pamberg¹, Andreas Hess¹

¹Institute of Pharmacology, FAU Erlangen-Nuernberg, Erlangen, Germany

The aim of this study was to compare GE and SE EPI for the purposes of pain research in anesthetized rats.

                  1680.     A Peripheral Nerve Repair Model Using FMRI in Rats

Maida A. Parkins¹, Rupeng Li², Hani S. Matloub¹, Ji-Geng Yan¹, James S. Hyde², Christopher P. Pawela^1,2

¹Plastic and Reconstructive Surgery, Medical College of Wisconsin, Milwaukee, WI, USA; ²Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA

The purpose of this study was to create a model for peripheral nerve injury and repair. This was done in a rat model using BOLD fMRI to track cortical recovery of the repaired nerve. The median nerve repair data show an increase in activation in the primary sensory forepaw region after just two weeks. This increase in activation is expected to continue in time as healing of the repaired nerve progresses. This model is exciting because it can be used as a vehicle for evaluating different interventions that could improve nerve healing, like application of growth factors or mechanical stress.

                  1681.     Assessing Alcohol-Induced Liver Damage in Chronic Alcoholic Rats by Gradient Echo FMRI

Manfred Brauer¹, Melissa Yau¹, Leslie M. Foley²

¹Molecular and Cellular Biology, University of Guelph, Guelph, Ontario, Canada; ²Pittsburgh NMR Ctr. for Biomedical Research, Carnegie Mellon University, Pittsburgh, PA, USA

Long-term alcohol consumption causes hypoxic liver damage. Functional MRI can measure tissue oxygenation changes in situ. Here we report gradient echo fMRI studies of liver oxygenation changes in chronic ethanol-treated rats showing that alcohol causes a greatly decreased and slower response to hyperoxia, hypoxia and carbogen challenge vs. controls. Liver signal intensities doubled with carbogen inhalation in controls, with no significant increase in alcoholic rats. TE-dependence studies confirm that both SE and GE fMRI studies are consistent with BOLD contrast. Low-volume hemoximetry showed that chronic ethanol decreases blood oxygen levels in the portal vein, hepatic artery and hepatic vein under all challenge conditions. Morphometric histology showed that alcoholic liver vasculature was compressed and less responsive to systemic challenge, confirming the fMRI results. This shows that fMRI can noninvasive monitor organ oxygenation status.

                  1682.     Potentiation of the Metabotropic Glutamate Receptor Subtype 2 Blocks Phencyclidine-Induced Brain Activation: A PhMRI Study

Nellie E. Byun^1,2, Elizabeth A. Hackler^1,2, Carrie K. Jones³, Jason M. Williams^1,2, Malcolm Avison^2,3, P. Jeffrey Conn³, John C. Gore^1,2

¹Radiology & Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, USA; ²Vanderbilt University Institute of Imaging Science, Nashville, TN, USA; ³Pharmacology, Vanderbilt University Medical Center, Nashville, TN, USA

We utilized pharmacologic MRI to evaluate the therapeutic potential of a metabotropic glutamate receptor subtype 2 potentiator, biphenyl indanone-A, in the phencyclidine (PCP) rat model predictive of schizophrenia. Here we show that mGluR2 potentiation attenuates the PCP-induced BOLD response in anesthetized rats in specific brain regions, including the prefrontal cortex, caudate-putamen, and thalamus, which correlate with behavioral data. These findings bolster the growing body of evidence that mGluR2 is a viable target for the treatment of schizophrenia.

                  1683.     The Effects of Methylphenidate on Resting-State Brain Activity in Normal Adults: An FMRI Study

Biao Jiang¹, Yihong Zhu¹, Pingzhen Guo², Yichao Deng¹, Hongjian He¹, Feiyan Chen¹, Yufeng Zang³

¹Zhejiang University, Hangzhou, Zhejiang, China; ²Memorial Sloan-Kettering Cancer Center , USA; ³Beijing Normal University, China

Objective:By using resting-state fMRI and analytic approach, regional homogeneity (ReHo), the study focused on the MPH effect on the intra-regional synchronization of spontensou brain activity in a group of normal adults. Methods: 1)Eighteen healthy males subjects.. 2) handedness questionnaire and the Wechsler Adult Intelligence Scale were tested. 3) After taking either MPH or placebo,the subjects were scanned a resting-state fMRI 4) The data analysis Results: MPH increases activity mainly in bilateral superior temporal lobe, right superior parietal lobe, and bilateral cuneus. MPH decreased the activity mainly in bilateral superior frontal lobes, bilateral orbital-frontal lobes, left middle frontal lobe. Conclusions: Single dose of MPH can change the resting activity in most resting-state related areas.

                  1684.     Anesthesia Effects on BOLD and RCBV Responses Induced by L-Tetrahydropalmatine

Xiping Liu¹, Zheng Yang², Rupeng Li¹, Qian Yin¹, Shi-Jiang Li¹

¹Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA; ²Beijing Institute of Basic Medical Science, Beijing, China

L-tetrahydropalmatine (l-THP), purified from Chinese herb Stephanie, was recently demonstrated to be effective in attenuating heroin craving and relapse in heroin addicts. Also, it inhibited cocaine¡¯s rewarding effects in animal models. As interactions of drug and anesthetic agent may affect the phMRI response to the drug of interest and complicate the interpretation of the data. We investigated the l-THP-induced BOLD and rCBV brain responses using three anesthesia conditions (isoflurane, urethane, and medetomidine) at high field to assess the confounding effects and interaction of different anesthetics on l-THP.

                  1685.     Hyperalgesic Effects of Low Dose Lidocaine Detected by BOLD FMRI in Mice

Simone Claudia Bosshard¹, Christof Baltes¹, Hanns Ulrich Zeilhofer², Markus Rudin^1,2

¹Institute for Biomedical Engineering, ETH and University of Zurich, Zurich, Switzerland; ²Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland

Effects of the local anesthetic Lidocaine on the BOLD signal in the somatosensory cortices and the thalamus of mice following electrical stimulation of the forepaw have been studied. Pretreatment of the stimulated forepaw with low concentrations of Lidocaine led to an increased BOLD response in the corresponding brain areas, in line with increased sensitivity observed in behavioral tests, indicating a hyperalgesic effect of Lidocaine on the C-type sensory afferents which process noxious stimuli. High concentration of Lidocaine abolished the BOLD response after electrical stimulation. This approach potentially allows distinguishing noxious from innocuous stimuli using BOLD fMRI in mice.

                  1686.     Statistical Pitfalls in Pharmacological MRI

Michael Luchtmann¹, Katja Jachau², Claus Tempelmann³, Johannes Bernarding¹

¹Institute for Biometry and Medical Informatics, Otto-Guericke-University of Magdeburg, Magdeburg, Germany; ²Institute of Forensic Medicine, Otto-Guericke-University of Magdeburg; ³Clinic for Neurology II, Otto-Guericke-University of Magdeburg

Despite potential effects of the examined drugs on the mechanisms of neurovascular coupling, most pharmacological MRI (phMRI) studies are analyzed under the premise of an unchanged hemodynamic response function (HRF). This assumption may lead to false negative activation changes when the HRF underlies considerable alterations induced by the employed drugs. This statistical pitfall is demonstrated on an exemplary phMRI study and show how this assumption may lead to incorrect conclusions if HRF changes are neglected in statistical analysis methods.

                  1687.     Characterization of a 5-HT2C Antagonist Profile of Agomelatine Using Challenge PhMRI in the Rat

Inna Linnik¹, Shane McKie², Jennifer A. Stark³, Simon Luckman³, Bill Deakin², Steve R. Steve R. Williams¹

¹Imaging Science and Biomedical Engineering, Manchester University, Manchester, UK; ²Neuroscience and Psychiatry Unit, Manchester University, Manchester, UK; ³Faculty of Life Science, Manchester University, Manchester, UK

Agomelatine is a novel clinically active antidepressant drug. The antidepressant activity of agomelatine needs the combination of melatonin agonist activity and 5-HT2C serotoninergic antagonist properties.

                  1688.     Quantification of Neural Activation in Spinal Cord by Blood Volume-Weighted FMRI on a Run-By-Run Basis – a Viable Pain Assay for Analgesics Development

Fuqiang Zhao¹, Mangay Williams¹, Denise Welsh¹, Xiangjun Meng¹, Amy Ritter¹, Catherine Abbadie¹, Jacquelynn J. Cook¹, Alise S. Reicin¹, Richard Hargreaves¹, Donald S. Williams¹

¹Merck Research Laboratories, West Point, PA, USA

Spinal cord fMRI offers an excellent opportunity to quantify nociception using neuronal activation induced by painful stimuli. Measurement of the magnitude of stimulation-induced activation, and its suppression with analgesics can provide objective measures of pain and efficacy of analgesics. In this study, a high sensitivity pain fMRI technique by combining BV-weighted fMRI with optimum electrical stimulus was developed. Its performance is demonstrated by measuring the effect of systemic lidocaine on noxious electrical stimulation-induced activation in spinal cord. The study also revealed that systemic lidocaine, which is clinically used for the treatment of neuropathic pain, and believed to only block the neural activity originating from the damaged peripheral nerves, also blocks noxious electrical stimulation-induced activity.

                  1689.     Imaging Brain Activity: Longitudinal FMRI in Mice Using Medetomidine Sedation

Joanna Magdalena Adamczak¹, Tracy Deanne Farr¹, Joerg Ulrich Seehafer¹, Mathias Hoehn¹

¹In-Vivo-NMR, Max-Planck-Institute for Neurological Research, Cologne, Germany

In this study we present the successful establishment of a non-invasive medetomidine based protocol for longitudinal fMRI studies of brain activity in mice. The optimal dose of medetomidine was evaluated with regard to sufficient sedation, tolerance by the animal and preservation of electrical activity in the brain. For this purpose transcutaneous CO₂ and somatosensory evoked potentials (SSEP) were measured. Following electrical forepaw stimulation we observed brain activity in the contralateral somatosensory cortex using blood oxygenation level-dependent (BOLD) contrast with gradient echo EPI images at 7.0T and 11.7T.

                  1690.     Effect of Isoflurane Anesthesia on BOLD Response to Somatosensory Stimulation: Results from FMRI Experiments in Conscious Rats

Denise Welsh¹, Alexandre Coimbra², Andrew Danziger², Al Rauch², Christopher Regan², Joseph J. Lynch², Donald S. Williams²

¹Merck & Co., Inc, West Point, Pa, USA; ²Merck & Co.,Inc

Translation of fMRI applications from preclinical to clinical experiments is confounded by the need for anesthesia in animal experiments. Successful implementation of fMRI of sensory stimulation in conscious rats using a combined coil and restraint system, and a pneumatically controlled mechanical stimulation device is described, and cerebral activation following sensory stimulation in awake and anesthestized rats is compared. Hind paw stimulation evoked statistically significant signal intensity increases in the contralateral primary somatosensory cortex of rats in the conscious state, but not in the anesthetized state.

                  1691.     Functional MRI Reveals the Critical Role of Broca’s Area in Speech Sound Disorders

Xu Chen^1,2, Barbara Lewis³, Amy Hansen³, Lisa Freebairn³, Jean Tkach^1,2

¹Radiology, Case Western Reserve University, Cleveland, OH, USA; ²Case Center For Imaging Research, CWRU, Cleveland, OH, USA; ³Communication Sciences, CWRU, Cleveland, OH, USA

Speech sound disorders (SSD) are the largest group of communication disorders observed in children requiring special education. However, the neurological origin of SSD has not been researched extensively. To investigate the neural substrate of SSD, we conducted an functional MRI study using the HUSH (Hemodynamics Unrelated to Sounds from Hardware) technique on 6 participants with SSD history in comparison to 9 age-matched controls during an overt non-word repetition task. Preliminary results demonstrate significant under- activation in Broca’s area –an area known to be crucial for speech production– for the patient group, suggesting the critical role Broca’s area plays in SSD.

                  1692.     Broca’s Area BA 45 Mediates the Cognitive and Emotional Interference Resolution in Healthy Adolescents

Adina Maria Mincic^1,2, Patricia A. McGrath³, Karen D. Davis^4,5

¹Division of Brain, Imaging and Behavior – Systems Neuroscience, Toronto Western Research Institute,University Health Network, Toronto, ON, Canada; ²Faculty of Medicine, University of Oradea, Oradea, Romania; ³Department of Anesthesia, Hospital for Sick Children, Toronto, ON, Canada; ⁴Division of Brain, Imaging and Behavior – Systems Neuroscience, Toronto Western Research Institute, University Health Network, Toronto, ON, Canada; ⁵Department of Surgery and Institute of Medical Science, University of Toronto, Toronto, ON, Canada

To date it is unclear to what extent the mechanisms engaged in cognitive control are similar in an emotional, compared to non-emotional context. We used variants of Stroop task and fMRI, to compare the activation patterns during cognitive and negative emotional interference, and to identify areas specifically engaged in processing selection (i.e. commonly showing activation in task vs. various baseline conditions - fixation, motor control, neutral words). Cognitive interference elicited activation in a bilateral fronto-parietal circuit; emotional interference engaged a mainly left lateralized fronto-temporo-occipital network. Broca’s area, BA 45, was found to be specifically involved in both types of interference.

                  1693.     Evaluation of Language Processing in FMRI at 7 T

Nina Koschnicke^1,2, Stefan Maderwald^1,2, Jens Matthias Theysohn^1,2, Katja Großkurth¹, Mark E. Ladd^1,2, Elke Ruth Gizewski^1,2

¹Department of Diagnostic and Interventional Radiology and Neuroradiology, University Hospital Essen, Essen, Germany; ²Erwin L. Hahn Institute for Magnetic Resonance Imaging, University Duisburg-Essen, Essen, Germany

The aim of this study was the evaluation of language regions (Wernicke’s and Broca’s area) by comparing fMRI at 1.5 and 7 T using a “word generation” and “synonym detection” test in twelve right-handed, volunteers. The speech regions were reliably and consistently revealed in all volunteers at both field strengths. In direct comparison, we acquired a 2 times higher BOLD signal and a 5-6 times higher voxel level at 7 T. In further studies, this advantage at 7 T may lead i.e. to a constant preoperative detection of speech areas in patients with brain pathologies and restricted powers of concentration.

                  1694.     Lateralization of the Mirror Neuron System During Observation and Execution of Object-Related Actions: A Functional Magnetic Resonance Study in Right and Left-Handed Subjects

Monia Cabinio¹, Valeria Blasi¹, Marcella Montagna², Paola Borroni³, Antonella Iadanza¹, Giuseppe Scotti¹, Andrea Falini¹, Gabriella Cerri²

¹Neuroradiology, San Raffaele Hospital, Milan, Italy; ²Istituto di Fisiologia Umana II, Università degli Studi di Milano, Milan, Italy; ³Dipartimento di Medicina, Chirurgia e Odontoiatria, Università degli Studi di Milano, Milan, Italy

Mirror neurons are visuomotor neurons active during both oservation and execution of object-related hand movements. It’s hypotesized their involvement in action understanding. Aim of this study is to evaluate if the activation of the mirror neuron system is lateralized in humans. We performed a Functional Magnetic Resonance Imaging study on right and left-handed normal subjects during observation and execution of grasping movements done with the left or right hand.

                  1695.     Diverting Attention Suppresses Human Amygdala Responses to Faces

Carmen Morawetz^1,2, Juergen Baudewig¹, Stefan Treue^2,3, Peter Dechent¹

¹MR-Research in Neurology & Psychiatry, Medical Faculty, Georg August University, Goettingen, Germany; ²Cognitive Neuroscience Laboratory, German Primate Center, Goettingen, Germany; ³Bernstein Center of Computational Neuroscience, Goettingen, Germany

Here we investigated the effects of high/low attentional load and different stimulus locations on face processing in the amygdala using fMRI. The display presented a rapid stream of letters and digits in each visual field quadrant and pairs of faces at one of three eccentricities. Participants had to either attend to the streams performing tasks of high/low attentional load or to the faces matching their gender/expression. The results revealed a strong attenuation of amygdala activity when the attentional load was high indicating that the processing of emotional stimuli in the amygdala is strongly dependent on the availability of attentional resources.

                  1696.     Functional Connectivity of the Insula in Smokers

Peter S. LaViolette¹, William Collier², M D. Verber³, Kathleen M. Schmainda^1,4, L Piacentine³, K L. Douville, S A. Claesges³, S J. Durgerian^3,5, Alan S. Bloom^3,6

¹Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA; ²Pharmacology, Medical College of Wisconsin, Milwaukee, WI; ³Psychiatry and Behavioral Medicine, Medical College of Wisconsin, Milwaukee, WI, USA; ⁴Radiology, Medical College of Wisconsin, Milwaukee, WI, USA; ⁵Neurology, Medical College of Wisconsin, Milwaukee, WI, USA; ⁶Pharmacology, Medical College of Wisconsin, Milwaukee, WI, USA

Nicotine is a highly addictive substance, and smoking is a leading worldwide cause of preventable death. Nicotine has been shown to increase fmri bold activation when administered, and much research has been done in fMRI and brain metabolism studies to localize specific brain anatomy involved. Functional connectivity of these regions, on the other hand, has received little attention. This study shows that regions that actively respond to nicotine administration are functionally connected at rest, and the strength of this network is paradoxically increased in smokers compared to non-smokers.

                  1697.     Neurofeedback Real-Time FMRI for the Regulation of Motor Areas in Chronic Stroke Patients

Jong-Hwan Lee¹, Lee A. Kearse², Richard Hughes³, Yi Tang¹, Lawrence P. Panych¹, Joel Stein³, Ferenc A. Jolesz¹, Randie M. Black-Schaffer³, Seung-Schik Yoo¹

¹Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; ²Anesthesiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; ³Spaulding Rehabilitation Hospital, Boston, MA, USA

We applied the developed fMRI neurofeedback method to chronic stroke patients with unilateral hand motor deficits, with an aim to demonstrate that real-time feedback of an individual's regional brain activation, especially from the motor-related areas that are affected by a stroke, will help the patients to gain a degree of voluntary regulation of the activation from the same areas.

                  1698.     Human MT Complex: Selectivity to Motion Flow Stimuli and to Retinotopic and Spatiotopic Input

Laura Biagi¹, Sofia Crespi^2,3, Giovanni D'Avossa⁴, Maria Concetta Morrone^1,5, Michela Tosetti¹

¹Stella Maris Scientific Institute, Pisa, Italy; ²Università degli Studi Milano-Bicocca, Milan, Italy; ³Università Vita Salute San Raffaele, Milan, Italy; ⁴Bangor University, Bangor, Wales, UK; ⁵Università degli Studi di Pisa, Pisa, Italy

A recent fMRI study demonstrated the existence of a region within MT complex with a spatiotopic selectivity. However this contrasts with other results that showed that MT represents the stimuli manly in retinotopic coordinate with a small or absent gaze modulation. The aim of this work is to understand whether the portion in MT complex that shows spatiotopic properties does belong to the retinotopic portion of MT and its specificity to the flow motion. Different kinds of visual stimuli (retinotopy, motion localizers and spatiotopic selectivity) were used to test this hypothesis. The region showing a spatiotopic selectivity is partially overlapping with the classical MT region and is strongly activated by all types of flow motion.

                  1699.     Acute Blockade of 5HT2a Receptors Reduces Orbitofrontal Cortex Response to Angry and Fearful Faces

Bettina Hornboll^1,2, Jon S. Wegener^1,2, Olaf B. Paulson^1,2, James B. Rowe^2,3, Gitte Moos Knudsen^2,4, Hartwig Siebner^1,2, Julian Macoveanu^1,2

¹Danish Recearch Center for MR, Copenhagen University Hospital, Hvidovre, Hvidovre, Denmark; ²Center for Integrated Molecular Brain Imaging, Copenhagen, Denmark; ³Department of Clinical Neurosciences, Cambridge University, Cambridge, UK; ⁴Neurobiology Research Unit,, Copenhagen University Hospital, Copenhagen, Denmark

In 17 healthy volunteers, we used pharmacological functional MRI to

                  1700.     Modulation of the Brain Activities During 3D Object Perception from 2D Optic Flow

Sunao Iwaki¹, John W. Belliveau²

¹National Institute of Advanced Industrial Science and Technology (AIST), Ikeda, Osaka, Japan; ²Massachusetts General Hospital, Boston, MA, USA

Two-dimensional optic flow is a strong cue to reconstruct the shape of the 3D object in motion. Here, we parametrically controlled the coherence of the random-dot motion to elucidate the changes of the brain activities between different 3D perception conditions in human. Significant increases of BOLD signal were observed both in the dorsal and ventral visual pathways in the conditions which generate robust 3D precepts. The results suggest that the perception of 3-D object from 2-D motion includes both perception of global motion and 3-D mental image processing that are accomplished by cooperative activation in the two distinct visual pathways.

                  1701.     Studying Familiarity of Different Stimulus Types

Esin Karahan¹, Ozlem Ozmen Okur¹, Ozlem Alkan², Tulin Yildirim², Cengizhan Ozturk¹

¹Institute of Biomedical Engineering, Bogazici University, Istanbul, Turkey; ²Department of Radiology, Baskent University Medical School, Ankara, Turkey

Neural correlates of familiarity to faces has been studied in fMRI studies suggesting that it is associated with increased neural activity in restrosplenial cortex and precuneus. However experience based familiarity has not been studied comparatively in detail. We aim to investigate whether brain regions associated with face familiarity are also activated with different types of stimuli. fMRI is obtained from two different groups (radiologists and volunteers) while they are viewing radiological images and natural scenes. Fusiform gyrus and frontal lobe regions were activated in radiologists' group while they are viewing radiological images whereas in volunteers group inferior occipital gyrus and posterior cingulate were activated during the same stimulus, suggesting that familiarity depends on the stimulus type.

                  1702.     Gender Differences in Correlations of Functional Brain Activation with Intelligence Factor Scores

Emily Eaves¹, Cheuk Ying Tang^1,2, Johnny Ng¹, David Carpenter¹, David Schroeder³, Chris Condon³, Roberto Colom⁴, Richard J. Haier⁵

¹Radiology, Mount Sinai School of Medicine, New York, NY, USA; ²Psychiatry, Mount Sinai School of Medicine, New York, NY, USA; ³Johnson O'Connor Research Foundation, Chicago, IL, USA; ⁴Universidad Autónoma de Madrid, Madrid, Spain; ⁵. School of Medicine (emeritus), UC Irvine, Irvine, CA, USA

We sought to determine if brain activity during a working memory task correlated with various intelligence measures and what, if any, effect gender had on these correlations. Forty subjects completed a battery of cognitive tests prior to completing an N-back working memory task where fMRI data was collected. Percent activations in areas of increased activity during the task were extracted for each subject. Correlation analysis of intelligence scores and percent activations were different for males and females. All correlations were negative suggesting reduced activity with increased performance, providing support for the efficiency model of brain function.

                  1703.     Neural Differences Between Intrinsic Motivation and Incentive Motivation

Woogul Lee¹, Johnmarshall Reeve², Yiqun Xue³, Jinhu Xiong³

¹Department of Psychological and Quantitative Foundations, University of Iowa, Iowa city, IA, USA; ²Department of Educational Psychology, University of Wisconsin-Milwaukee, Milwaukee, WI, USA; ³Department of Radiology, University of Iowa, Iowa city, IA, USA

We examined the neural differences underlying intrinsic motivation from incentive motivation. Brain activations were scanned when participants decided whether they wanted to do activities while reading sentences describing intrinsic and incentive motivation. In intrinsic motivation, there were unique activations in the anterior precuneus associated with judgments of self-relevance. In contrast, in incentive motivation, the nucleus accumbens and the anterior cingulate gyrus were activated, related to the anticipation of rewards and the making of decisions on the basis of evaluation of rewards respectively. We conclude that the neural bases of intrinsic motivation can be distinguished from those of incentive motivation.

                  1704.     System for Computerized Writing and Drawing During FMRI

Fred Tam¹, Nathan Churchill^1,2, Stephen C. Strother^1,2, Simon J. Graham^1,2

¹Rotman Research Institute, Toronto, Ontario, Canada; ²Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada

Writing and drawing are understudied with fMRI, partly for lack of a device that enables realistic approximation of this behavior, including task performance feedback and quantitative behavioural logging in the confines of the magnet. Therefore, we developed a tablet based on touchscreen technology that is accurate, reliable, and relatively inexpensive. In a preliminary fMRI experiment, we examined the neural correlates of a widely used pen-and-paper neuropsychological assessment, the Trail Making Test. The results were similar to the major results of a previous group study, which encourages us to extend the experiment and to continue developing the tablet.

                  1705.     Functional Magnetic Resonance Imaging of Human Brain Exposed to a 60Hz, 1800µT Magnetic Field

Alexandre G. Legros^1,2, Julie Weller¹, John Robertson^1,2, Michael Corbacio^1,2, Daniel Goulet³, Robert Stodilka^1,2, Frank S. Prato^1,2, Alex W. Thomas^1,2

¹Lawson Health Research Institute, London, Ontario, Canada; ²University of Western Ontario, London, Ontario, Canada; ³Hydro-Québec, Montréal, Québec, Canada

A simple finger tapping task is known to activate the controlateral Supplementary Motor Area and the controlateral Primary Motor Area. Interestingly, it has been shown that exposure to Extremely Low Frequency Magnetic Fields (ELF MF, below 300Hz) decreases the amplitude of spontaneous motor activity of healthy subjects. Moreover, a few results suggest a decrease of brain electrical activity with exposure. We hypothesize that for an index finger vs. thumb opposition task realized at natural frequency, thirty minutes of ELF MF exposure will induce a decrease of BOLD activation in the involved brain structures.

                  1706.     Multimodality Imaging in Cerebro-Cerebeller Verbal Working Memory

Wen-Yang Chiang¹, Jing-Syun Yu², Shen-Hsing Annabel Chen², Wen-Yih Isaac Tseng³

¹Research Institute, The Methodist Hospital, Houston, TX, USA; ²Department of Psychology, National Taiwan University, Taipei, Taiwan; ³Center for Optoelectronic Biomedicine, National Taiwan University College of Medicine, Taipei, Taiwan

                  1708.     Two Local Constrained Canonical Correlation Analysis Methods for FMRI

Mingwu Jin¹, Rajesh Nandy², Dietmar Cordes¹

¹Radiology, University of Colorado Denver, Aurora, CO, USA; ²Biostatistics and Psychology, UCLA, Los Angeles, CA, USA

There are no constraints of the spatial weights employed in conventional canonical correlation analysis (CCA) leading to model overfitting and a decrease of specificity. We propose two different constrained CCA (cCCA) methods to improve the detection power of fMRI data analysis and compare results with the GLM method without and with fixed Gaussian spatial smoothing. Quantitative results from pseudo-real data as well as qualitative results from real data show that both novel cCCA methods can detect activations more accurately for noisy fMRI data without losing specificity.

                  1709.     Effectiveness of Gaussian Smoothing on Spatially Correlated Noise: A 3T Case Analysis

Mingwu Jin¹, Rajesh Nandy², Dietmar Cordes¹

¹Radiology, University of Colorado Denver, Aurora, CO, USA; ²Biostatistics and Psychology, UCLA, Los Angeles, CA, USA

Gaussian spatial smoothing using a fixed FWHM is a common preprocessing step in fMRI data analysis. In this work, we investigate the effects of spatially correlated noise and the size of activation patterns to detect activations with and without spatial smoothing. The detection power is measured using ROC curves on simulated data and activation maps of real 3T fMRI data. Results indicate that spatial smoothing with any fixed FWHM is far less effective when the noise is spatially correlated as in 3T resting-state data and more advanced locally adaptive smoothing kernels should be applied.

                  1710.     Robust, Unbiased General Linear Model Estimation of PhMRI Data in the Presence of Variance in the Temporal Response Profile

Gautam Vithal Pendse^1,2, Adam James Schwarz^2,3, Richard Baumgartner^2,4, Alexandre Coimbra^2,5, David Borsook^1,2, Lino Becerra^1,2

¹Imaging and Analysis Group (IMAG), McLean Hospital, Harvard Medical School, Belmont, MA, USA; ²Imaging Consortium for Drug Development (ICD), Belmont, MA, USA; ³Translational Imaging Group, Lilly Research Laboratories, USA; ⁴Biometrics Research, Merck Research Laboratories, USA; ⁵Imaging Department, Merck Research Laboratories, USA

PhMRI can elucidate brain circuits underlying pharmacological action and provide a translatable, pharmacodynamic biomarker of CNS activity for novel compounds in the early phases of drug development. As applications increasingly shift to compounds whose direct effects on brain activity may be more subtle – including new chemical entities and novel target profiles – there is increasing demand on accurate and robust estimation of the phMRI signal. We present a simple method to generate parsimonious design matrices that accurately estimate, within a GLM framework, the phMRI response in the presence of confounding signals and variability in temporal profile.

                  1711.     A Stochastic Linear Model of Neurovascular Dynamics in the BOLD Signal

Leigh A. Johnston^1,2, Maria Gavrilescu², Eugene P. Duff³, Gary F. Egan^2,4

¹Electrical and Electronic Engineering and NICTA Victorian Research Laboratory, University of Melbourne, Parkville, VIC, Australia; ²Howard Florey Institute, Melbourne, VIC, Australia; ³FMRIB, Oxford University, UK; ⁴Centre for Neuroscience, University of Melbourne, Australia

A stochastic linear model (SLM) of the BOLD signal is presented in which neurovascular dynamics are modelled by an autoregressive exogenous input signal, embedded in parametrically modelled noise. The unknown SLM states and parameters are estimated by an iterative coordinate descent algorithm based on the Kalman smoother, from which novel activation weights are calculated. We demonstrate, through application to a motor task fMRI dataset, that the SLM produces more robust and consistent activation estimates than the general linear model. The stochasticity of the SLM embodies sufficient flexibility to account for observed variations in the BOLD signal.

                  1712.     Deconvolving Haemodynamic Response Function in FMRI Under High Noise by Compressive Sampling

Christine Law¹, Gary Glover¹

¹Stanford University, Stanford, CA, USA

A simple technique to deconvolve haemodynamic response function (HRF) from fMRI data using 1-norm minimization is introduced. The true HRF is typically sparse after wavelet transform, but we find the proposed technique to be robust w.r.t relative sparsity. HRF is deconvolved via convex optimization which has the flexibility to impose local HRF monotonicity and smoothness in the time domain. Real fMRI data under low SNR (-10dB) confirms reliability of this technique.

                  1713.     A Novel Test Statistic Allowing a General Linear Contrast Vector for Local Canonical Correlation Analysis in FMRI

Mingwu Jin¹, Rajesh Nandy², Dietmar Cordes¹

¹Radiology, University of Colorado Denver, Aurora, CO, USA; ²Biostatistics and Psychology, UCLA, Los Angeles, CA, USA

Local canonical correlation analysis (CCA) is a multivariate method that has been proposed to determine more accurately activation patterns in fMRI data. One drawback of CCA is that, unlike the general linear model (GLM), an arbitrary linear contrast of the temporal regressors has not been incorporated in the CCA formalisms. In this research we show how to extent CCA so that an arbitrary linear contrast of the temporal regressors can be computed similar to a t-statistic in GLM.

                  1714.     A Quantitative Comparison of Different Group Analysis Methods in Dynamic Causal Modeling: Dependence on Noise and Inter-Subject Variability

Christian H. Kasess^1,2, Andreas Weissenbacher¹, Lukas Pezawas², Ewald Moser^1,3, Christian Windischberger^1,3

¹MR Center of Excellence, Medical University Vienna, Vienna, Austria; ²Division of Biological Psychiatry, Department of Psychiatry and Psychotherapy, Medical University Vienna, Vienna, Austria; ³Center for Biomedical Engineering and Physics, Medical University Vienna, Vienna, Austria

A number of different approaches have been proposed to infer group statistics for dynamic causal models: (a) classic random effects (RFX) second level analysis based on the mean parameter estimates ignoring intra subject variance, (b) Bayesian averaging, combining multivariate single subject posterior parameter distributions according to Bayes’ theorem and (c) temporal averaging. Here, we compared these methods based on simulated data. We found that RFX and temporal averaging analysis are more robust against heterogeneity in the study population than Bayesian averaging which, for a heterogeneous population, can yield results that only partly resemble the input statistics.

                  1715.     Elucidating the Impulse Response Function of SEEP Contrast in the Human Spinal Cord  by Means of Event-Related Spin-Echo Spinal FMRI

Chase R. Figley¹, Jordan K. Leitch¹, Celina Nahanni¹, Patrick W. Stroman^1,2

¹Centre for Neuroscience Studies, Queen's University, Kingston, Ontario, Canada; ²Diagnostic Radiology and Physics, Queen's University, Kingston, Ontario, Canada

Event-related fMRI is now a prevalent technique in both systems and cognitive neuroscience, but event-related spinal cord fMRI has not yet been reported. However, with recent methods to reduce motion-related confounds, the sensitivity and specificity of spin-echo spinal fMRI methods have been dramatically improved. Herein, we have capitalized on these improvements to investigate peristimulus signal intensity changes following brief periods of cold thermal stimulation (1 second durations of 22 °C) applied to the palm of the hand. The results of this study have refined our understanding of the SEEP (i.e., ‘signal enhancement by extravascular water protons’) impulse response function in the human spinal cord, and represent the first successful demonstration of event-related spinal fMRI.

                  1716.     An Analytic Magnitude and Phase FMRI Activation Model Applied to ASL

Daniel B. Rowe¹, Luis Hernandez-Garcia²

¹Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA; ²FMRI Laboratory, University of Michigan, Ann Arbor, MI, USA

A computationally fast high tSNR magnitude and phase activation model is presented then applied to an ASL fMRI visual experiment. In fMRI, Fourier encoded k-space measurements, inverse Fourier reconstructed images, and voxel time series are complex-valued (real and imaginary or magnitude and phase). Nearly all fMRI studies derive functional activation from magnitude-only time series while discarding the phase time series. A GLM magnitude and/or phase activation model has been introduced and shown to have higher sensitivity. However, the existing method to compute magnitude and phase activation utilizes an iterative MLE algorithm while here we use a large tSNR exact solution.

                  1717.     Optimisation of Support Vector Machine Classifier Algorithms for Use as a Brain Computer Interface in Real-Time FMRI Applications

Guy B. Williams¹

¹Wolfson Brain Imaging Centre, University of Cambridge, Cambridge, UK

Support Vector machines have recently been developed for application to the analysis of fMRI data, and have great potential for use in fMRI based brain computer interfaces. In this work, we explore how different acquisition and processing strategies can increase the reliability of the classifier, while constrained by the fact that results must be reported rapidly and in time-order.

                  1718.     Granger Causality Via Vector Auto-Regression Tuned for FMRI Data Analysis

Gang Chen¹, J. Paul Hamilton², Moriah E. Thomason², Ian H. Gotlib², Ziad S. Saad¹, Robert W. Cox¹

¹Scientific and Statistical Computing Core, NIMH, National Institutes of Health, Bethesda, MD, USA; ²Mood and Anxiety Disorders Laboratory, Department of Psychology, Stanford University, Stanford, CA, USA

We present a platform-independent modeling tool that performs multivariate Granger causality analysis particularly tuned for FMRI data. With an extended vector auto-regressive modeling strategy that accounts for confounds such as baseline drift, head motion parameters, tasks of no interest, physiological measurements, time breaks and signal irregularities, our program provides various model fine-tuning tools including order selections and various diagnosis tests, and identifies the causality at each lag among the pre-selected regions. We also propose a valid group analysis per lag based on signed path coefficients to reveal the network at the group level. The software is in open-source R and available for download.

                  1719.     High Temporal Resolution Estimation of Hemodynamic Response from Event-Related FMRI

Mark A. Smith¹, Mark J. Lowe¹

¹Imaging Institute, Cleveland Clinic, Cleveland, OH, USA

A method for deconvolution of event-related fMRI scans to extract information on the hemodynamic response is described in detail. The method can be applied to whole-brain data sets acquired with standard echo planar imaging techniques. The method allows for slice-specific solutions which can have a temporal resolution much better than the TR of the scan. Demonstration of the method on real data will be given.

                  1720.     Towards Quality Control of Subject Movement in FMRI Studies

Daniel Cromb^1,2, Andy Simmons^1,3

¹Centre for Neuroimaging Sciences, Institute of Psychiatry, London, UK; ²Kent Institute for Medicine and Health Sciences, University of Kent, Canterbury, Kent, UK; ³Clinical Neuroscience, Institute of Psychiatry, London, UK

Large subject movements during fMRI can lead to unanalyzable data. Despite this, subject movement remains an understudied area and there are no guidelines for quality control of subject movement. We report the development of a software package for investigating subject movement and a novel overall movement parameter (OMP). The OMP is shown to be affected by the amount of time spent in the scanner and varies between patient groups. Characteristic movement patterns and directions have been identified. The OMP shows promise both for quality control of subject movement and for the refinement of fMRI study designs to minimise subject movement.

                  1721.     Induced Correlation in FMRI Magnitude Data from K-Space Preprocessing

Daniel B. Rowe¹, Andrew S. Nencka¹

¹Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA

Correlations between image-space voxels over time have been used to identify functionally connected regions of the cortex of subjects in the “resting state.” The common pre-processing techniques that are applied to fMRI k-space data before image reconstruction and before magnitude image formation can induce image-space voxel correlations. Since connectivity inferences are based upon image-space correlations, one must quantify and account for such pre-processing induced correlations. This framework allows the quantification and removal of these induced correlations. This will enhance and refine all future fMRI connectivity studies.

                  1722.     Development of the Expert System for Quantification of Brain Activation (QUBA) in FMR Images: Differential Activation Analysis

Tae-Hoon Kim¹, Gwang-Woo Jeong¹, Byung-In Min¹, Hyeong-Jung Kim¹, Sundaram Thirunavukkarasu¹, Han-Su Baek¹, Gwang-Won Kim¹, Heoung-Keun Kang², Moo-Suk Lee³

¹Biomedical engineering, Chonnam National University Medical School, Gwang-ju, Korea; ²Radiology, Chonnam National University Medical School, Gwang-ju, Korea; ³Psychiatry, Chonnam National University Medical School, Gwang-ju, Korea

In fMRI, labels for brain activation foci are expressed by the x, y, z coordinates of Talairach¡¯s brain space using the Talairach Demon software. However, this method needs to use the MNI2TAL function of SPM, which converted MNI brain space to Talairach¡¯s brain space, as well as to confirm with comparison of activation maps overlaid on MNI-normalized structural MRI images.

                  1723.     Shape Invariant Modelling of Trial Based FMRI Data

Kristoffer Hougaard Madsen^1,2, Lars Kai Hansen², Morten Mørup²

¹Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital Hvidovre, Hvidovre, Greater Copenhagen, Denmark; ²DTU Informatics, Technical University of Denmark, Kgs. Lyngby, Greater Copenhagen, Denmark

To overcome poor signal-to-noise ratios in fMRI, data sets are often acquired over repeated trials that form a three-way array of space × time × trials. As fMRI data contain multiple inter-mixed signal components blind signal separation and decomposition methods are frequently invoked for exploratory analysis and as a preprocessing step for signal detection. Here we extend multi-linear decomposition to account for general temporal modelling within a convolutional representation. We demonstrate how this alleviates degeneracy and helps to extract physiologically plausible components. The resulting convolutive multi-linear decomposition can model realistic trial variability as demonstrated in fMRI data.

                  1724.     Classification Technique to Detect Activation Patterns in Pain FMRI Data

Loan Vo^1,2, Harish A. Sharma³, Y. Michelle Wang^1,4, Dirk B. Walther¹, Arthur F. Kramer^1,4, William Olivero⁵

¹Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, USA; ²Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, USA; ³Biomedical Imaging Center, University of Illinois at Urbana-Champaign, Urbana, IL, USA; ⁴Department of Psychology, University of Illinois at Urbana-Champaign, Urbana, IL, USA; ⁵Carle Foundation Hospital, Urbana, IL, USA

Traditional univariate GLM (General Linear Model) in fMRI analysis has its weakness as not taking into account the relationship between data from different (but adjacent or physiology related) voxels. In this back-pain study, we use classification method i.e multi-voxel pattern analysis to find the brain activation regions involved in pain processing. Using this technique we successfully detect the activations in the motor cortex, insula and thalamus region in the patients with low back pain.

                  1725.     Time Course Information Extracted by the Sliding Window Analysis Detected the Decrease of T-Values in Hippocampal Areas During a Paired Associative Learning of Novel Faces and Names

Kayako Matsuo^1,2, Epifanio Tila Bagarinao³, Tetsuya Iidaka⁴, Akinori Takeda⁵, Toshiharu Nakai²

¹Dept. Psychology, National Taiwan University, Taipei, Taiwan; ²Dept. Gerontechnology, National Center for Geriatrics and Gerontology, Ohbu, Aichi, Japan; ³Neuroscience Research Institute, National Institute of Advanced Industrial Science and Technology, Tsukuba, Ibaraki, Japan; ⁴Dept. Psychiatry, Nagoya University, Nagoya, Aichi, Japan; ⁵Ministry of Health, Labour and Welfare, Chiyoda-ku, Tokyo, Japan

A new method �gSliding Window Analysis (SWA)�h was applied to extract a decreasing trend of t-values in hippocampal areas during fMRI. The SWA moves a statistical window to examine how the t-value changes during the session. We designed a paired associative learning of novel faces and names that were compared with famous faces and names. Young participants, specifically those with high scores, demonstrated a downward trend of the t-values, whereas the trends were obscure in the elderly participants. The decrease in the t-values suggested the establishment of the associative learning because the hippocampus is specifically involved in the memory encoding.

                  1726.     The Application of the Sliding Window Analysis: The Brain Location and the Window Width

Kayako Matsuo¹, Epifanio Tila Bagarinao², Tetsuya Iidaka³, Akinori Takeda⁴, Toshiharu Nakai⁵

¹Dept. Psychology, National Taiwan University, Taipei, Taiwan; ²Neuroscience Research Institute, National Institute of Advanced Industrial Science and Technology, Tsukuba, Ibaraki, Japan; ³Dept. Psychiatry, Nagoya University, Nagoya, Aichi, Japan; ⁴Ministry of Health, Labour and Welfare, Chiyoda-ku, Tokyo, Japan; ⁵Dept. Gerontechnology, National Center for Geriatrics and Gerontology, Ohbu, Aichi, Japan

A new method �gSliding Window Analysis (SWA)�h was applied to extract a decreasing trend of t-values in hippocampal areas during fMRI. The SWA moves a statistical window to examine how the t-value changes during the session. We designed a paired associative learning of novel faces and names that were compared with famous faces and names. A detailed analysis using the SWA demonstrated that a locus around the hippocampus was specifically involved in the associative learning process.

                  1727.     An Efficient Global Clustering Algorithm for the Identification of Functional Connectivity from BOLD FMRI

Andrew Ryan McIntyre^1,2, Xiaowei Song^1,3, Evangelos E. Milios², Malcolm I. Heywood², Alma Major¹, Ryan D'Arcy^1,4, Kenneth Rockwood^3,5

¹Institute for Biodiagnostics - Atlantic, National Research Council, Halifax, NS, Canada; ²Faculty of Computer Science, Dalhousie University, Halifax, NS, Canada; ³Medicinie, Dalhousie University, Halifax, NS, Canada; ⁴Psychology and Neuroscience, Dalhousie University, Halifax, NS, Canada; ⁵Centre for Health Care for the Elderly, QEII Health Sciences Centre, Halifax, NS, Canada

A clustering approach is proposed for globally identifying functional connectivity patterns from resting state fMRI data. The Self-Organizing Map algorithm is directly applied to time course data to reduce the input space for efficient application of the k-means algorithm. Hard clusters are selected and correlated time course patterns are assigned to each cluster. Results on two benchmark data sets indicate that the approach is able to efficiently and accurately identify embedded target time course patterns with low variation between initializations.

                  1728.     The Nintendo Wii Remote as an MR-Compatible Interface to Cognitive Studies Using FMRI

Jerod Michael Rasmussen^1,2, Karen Ayre Rasmussen³

¹Pediatric Neuroimaging Research Consortium, Cincinnati Children's Hospital, Cincinnati, OH, USA; ²Dept. of Neurology, Cincinnati Children's Hospital, Cinicinnati, OH, USA; ³Long Beach Memorial Medical Center

The Nintendo Wii remote controller was investigated to determine fMRI-compatibility inside of a Bruker 3T scanner. Time variant and time invariant statistics were used to quantify the impact on T2* weighted imaging. A comparison with and without the controller present showed little effect on the ability to capture Echo Planar Images. Using freely available software and inexpensive materials (less than US $80) the Wii remote can be easily implemented to provide a novel interface to human cognition safely and free of artifact.

                  1729.     A Virtual Radial Arm Maze for the Study of Multiple Memory Systems

Dongrong Xu^1,2, Xuejun Hao^1,2, Zhishun Wang^1,2, Yunsuo Duan^1,2, Rachel Marsh^1,2, Shan Yu^1,2, Bradley S. Peterson^1,2

¹Psychiatry, Columbia University, New York, NY, USA; ²New York State Psychiatric Institute, New York, NY, USA

An increasing number of functional brain imaging studies are employing computer- based virtual reality (VR) for the study of dynamic brain activity during the performance of high-level psychological and cognitive tasks. We report the development of a standardized VR radial arm maze that adapts for human use in a scanning environment the same general experimental design of the Winstay and Winshift tasks that have been used with remarkable effectiveness for the study of multiple memory systems in rodents. The software platform is independent of computer hardware and operating system that aims to provide shared access to this technology by the research community. The details of its design and implementation are reported, and the system is now available for downloading for research and test purposes.

                  1730.     Optimalization of Event-Related FMRI Design for BOLD Deconvolution

Johan Martijn Jansma¹, Jacco A. de Zwart², Peter van Gelderen², Wayne C. Drevets¹, Maura L. Furey

¹SNMAP/NIMH, NIH, Bethesda, MD, USA; ²AMRI/NINDS, NIH, Bethesda, MD, USA

Because BOLD response timing can vary across regions and subjects, model free fMRI signal analysis through ‘deconvolution’ (a.k.a. FIR) analysis can be preferred above methods that apply an a-priori BOLD model. A disadvantage is the larger number of regressors, which can increase regressor dependency. The effect of regressor dependency (expressed as the average tolerance or ‘aTOL’) on sensitivity and reliability of multi-regressor-based ‘deconvolution’ analysis is investigated here. Comparison, through simulations and experimental data, of 128 random regressors with a pseudo-random m-sequence design indicated that performance of the design scaled with aTOL and that the m-sequence consistently outperformed all random designs.

                  1731.     Real-Time Rejection of Gradient and Pulse Related Artefact (GRA and PRA) from Electroencephalographic Signals Recorded During Functional Magnetic Resonance Imaging (FMRI)

Yves Leclercq¹, Andre Luxen¹, Evelyne Balteau¹, Pierre Maquet¹, Christophe Phillips¹

¹Cyclotron Research Centre, University of Liege, Liege, Belgium

Simultaneous EEG and fMRI acquisition offer new possibilities to study brain dynamics. Though EEG signal get polluted by 2 artefacts, related to the MR scanner (“gradient related artifact”, GRA) and to the heart beat (“pulse related artifact”, PRA). GRA is easily dealt with but not PRA. We recently introduced a “constrained ICA” (cICA) that outperforms other current methods in rejecting PRA. Here we introduce an algorithm also based on cICA to correct the PRA in real time, allowing online monitoring and neuro-feedback during fMRI scanning. The approach was tested on a patient resting awake, eyes closed in the scanner.

                  1732.     Analyzing Task Activated BOLD FMRI Signal Voxel Area and Intensity Measurements with Bootstrap Power Analysis

Christopher Paul Pawela^1,2, B Douglas Ward², Rupeng Li², Bharat B. Biswal³, James S. Hyde²

¹Department of Plastic Surgery, Medical College of Wisconsin, Milwaukee, WI, USA; ²Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, USA; ³Department of Radiology, University of Medicine and Dentistry of New Jersey, Newark, NJ, USA

The number of subjects needed to determine information about BOLD signal changes during task performance is a matter of debate. Every fMRI experiment is different; however, information can be learned by studying sensory stimulation in a highly controlled animal model. In this study, BOLD activation voxel area and intensity are measured in response to different frequencies of forepaw stimulation. The results demonstrate that the area of activation is a more sensitive parameter for this task than the area-under-the-curve for the BOLD signal, and indicates that a larger sample size may be needed than was previously assumed.

                  1733.     Reverse BOLD Improves Efficiency for Brain Response to Faces

Christopher Glielmi¹, George A. James¹, Zhihao Li¹, Xiaoping Hu¹

¹Department of Biomedical Engineering, Georgia Institute of Technology / Emory University, Atlanta, GA, USA

Simultaneous electroencephalography (EEG) and fMRI are advantageous because they provide concurrent measures of neural electrical activity at high temporal resolution and hemodynamics at high spatial resolution. While EEG-only sessions can present trials every 1-3 seconds for efficient averaging, event-related fMRI typically requires much longer interstimulus intervals. We propose a new stimulation protocol in which long periods of trials are interleaved with missing trials, leading to a “reverse BOLD” response and increasing EEG trials per reversed BOLD trial. We confirm the integrity of the reverse BOLD response by comparing it to the regular BOLD response for visual processing of faces.

                  1734.     Gradient Non-Linearity Correction Relocates Normalised Group Activation Hotspot

Arnold Skimminge^1,2, Mark Schram Christensen^1,3

¹Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital, Hvidovre, Denmark; ²DTU Informatics, Technical University of Denmark, Lyngby, Denmark; ³Department of Exercise and Sport Sciences, University  of Copenhagen, Denmark

High-speed MRI gradients tend to be nonlinear outside the center of the magnetic field. This introduces structural deformations which may influence not only the shape of structural images but also the location of activation in fMRI studies. In worst case scenarios it could have serious impact for presurgical planning fMRI studies if activation gets mislocalised. Here we show that even in a best case scenario where functional images are normalised, smoothed and group activation maps created, there is still a significant difference in activation localisation with and with a correction for the gradient nonlinearities.

                 1735.     Background Suppression Does Not Reduce Physiological Noise in Myocardial ASL Perfusion Imaging

Zungho Zun¹, Eric C. Wong², Krishna S. Nayak³

¹Department of Electrical Engineering , University of Southern California, Los Angeles, CA, USA; ²Departments of Radiology and Psychiatry, University of California, La Jolla, CA, USA; ³Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA

Background suppression (BGS) is widely used in brain arterial spin labeling (ASL) as a means to reduce physiological noise due to subject motion and metabolic fluctuation. We performed pulsed ASL myocardial blood flow (MBF) experiments with and without BGS. Using BGS, myocardial signal was reduced by 82%, but the change in physiological noise and MBF measurement confidence was statistically insignificant. This study demonstrates that static myocardial tissue signals are not significant contributors to physiological noise in myocardial ASL.

                  1736.     Assessment of Myocardial Blood Flow in Humans Using Arterial Spin Labeling: Feasibility and Noise Analysis

Zungho Zun¹, Eric C. Wong², Krishna S. Nayak³

¹Department of Electrical Engineering , University of Southern California, Los Angeles, CA, USA; ²Departments of Radiology and Psychiatry, University of California, La Jolla, CA, USA; ³Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA

We demonstrate the feasibility of myocardial blood flow (MBF) measurement in humans using arterial spin labeling (ASL) and present an analysis of thermal and physiological noise and their impact on MBF measurement error. A desired measurement error of ¡¾ 0.1 ml/ml/min could be achieved with 85.5% confidence with thermal noise alone, but with 18.2 to 94.0 % confidence based on physiological noise measured in a fifteen subjects. We demonstrate that MBF can be assessed in humans using ASL, and that SNR improvement and the reduction of physiological noise are key areas for future development.

                  1737.     Myocardial ASL Perfusion Imaging Using Pulsed 2D Tagging of the Proximal Aorta

Zungho Zun¹, Eric C. Wong², Krishna S. Nayak³

¹Department of Electrical Engineering , University of Southern California, Los Angeles, CA, USA; ²Departments of Radiology and Psychiatry, University of California, La Jolla, CA, USA; ³Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA

Current methods for myocardial ASL perfusion imaging suffer from a large spurious ASL signal in the ventricular blood pool, due to inadvertent tagging of blood in the LV and LA. This signal is 50-100x larger than the ASL signal in myocardium, and is a potential source of artifact, physiological noise, and MBF measurement error. We demonstrate the feasibility of pulsed tagging of the proximal aorta using a B1-insensitive 2D selective inversion pulse without disturbing the LA or LV. This scheme achieves a 67% inversion efficiency, and reduces the apparent ASL signal in the LV by 78%.

                  1738.     Assessment of Myocardial Perfusion Reserve with Blood Oxygen Level-Dependent Cardiovascular Magnetic Resonance Imaging

Jacqueline Allison Flewitt¹, Matthias Voehringer^1,2, Jordin Daniel Green^1,3, Matthias G. Friedrich¹

¹Stephenson CMR Centre, Calgary, Alberta, Canada; ²Robert-Bosch-Krankenhaus, Stuttgart, Germany; ³Siemens Healthcare, Calgary, Canada

We hypothesize that new Blood Oxygen Level Dependent (BOLD) Cardiovascular MRI sequences have the image quality and spatial resolution to quantitatively asses myocardial perfusion. In a canine model, flow, coronary sinus saturation and BOLD signal intensity (SI) changes were correlated during graded adenosine infusion. In healthy volunteers, flow and BOLD SI were acquired before and during adenosine infusion. In dogs, adenosine-induced blood flow changes agreed strongly with coronary venous saturation changes (r2=0.94, p<0.001) and coronary venous saturation showed a strong linear correlation with BOLD-SI changes (r2=0.80, p<0.001). In volunteers, adenosine infusion resulted in a significant myocardial perfusion increase (416±69%, p<0.001).

                  1739.     Blood Oxygen Level-Dependent Cardiovascular Magnetic Resonance Imaging (BOLD-CMR) in Assessment of Transmural Perfusion Effect

Jacqueline Allison Flewitt¹, Matthias Voehringer^1,2, Jordin Daniel Green^1,3, Matthias G. Friedrich¹

¹Stephenson CMR Centre, Calgary, Alberta, Canada; ²Robert-Bosch-Krankenhaus, Stuttgart, Germany; ³Siemens Healthcare, Calgary, Alberta, Canada

Blood Oxygen Level Dependant (BOLD) cardiovascular MR imaging can detect perfusion changes in the myocardium and has the potential to detect transmural changes across the myocardium. A study of healthy control subjects and patients with suspected coronary artery disease was conducted using SSFP BOLD sensitive imaging techniques. The signal intensity was calculated during rest and the increase with adenosine infusion. A significant difference between the epicardium and endocardium was detected in the patients whereas no significant difference occurred in the healthy subjects.

                  1740.     Short Inversion Time Gadolinium-Enhanced Myocardial Infarct Imaging Yields Improved Infarct to Blood Contrast

James W. Goldfarb¹, Sheeba Arnold¹, Margeurite Roth¹

¹Saint Francis Hospital, Roslyn, NY, USA

Short inversion time myocardial infarct imaging (LGE-HYPO) provides improved contrast between the LV bloodpool and infarcted myocardium at the expense of decreased contrast between viable myocardium and both the LV bloodpool and infarcted myocardium. A combination of LGE-HYPER and LGE-HYPO may provide the best image contrast for detection of myocardial infarction.

                  1741.     Myocardial Viability Imaging Using HYPR-Based MRI Techniques

Orhan Unal^1,2, Julia Velikina¹, Kevin M. Johnson¹, Charles A. Mistretta^1,2

¹Department of Medical Physics, University of Wisconsin, Madison, WI, USA; ²Department of Radiology, University of Wisconsin, Madison, WI, USA

The goal was to investigate the potential of HYPR-based delayed-enhancement magnetic resonance imaging (DE-MRI) techniques for 2D / 3D myocardial viability imaging that allow retrospective selection of inversion time (TI) to null myocardium signal.

                  1742.     Validation of Cine-FLASH as a Method to Image Late Gd Enhancement in Mice

Andrea Protti¹, Alexander Alexander Sirker², Tobias Schaeffter³, Ajay Shah², Rene Botnar³

¹Cardiovascular Division and Division of Imaging Sciences, King’s College London BHF Centre of Excellence, London, UK, UK; ²Cardiovascular Division, King’s College London BHF Centre of Excellence, London, UK; ³Division of Imaging Sciences, King’s College London BHF Centre of Excellence, London, UK

Late Gadolinium Enhancement (LGE) cardiovascular magnetic resonance (CMR) imaging is usually performed by a T1-weighted sequence with an inversion recovery (IR) pre-pulse. Although the images achieve good contrast from areas of Gd agent uptake, there are several drawbacks such as absence of contrast between Gd infarcted areas and blood; image blurring due to ECG and respiration instability; anatomical information lost. In order to avoid such limitations, a cine-FLASH is suggested as a more reliable method of imaging LGE in mice 2 days after MI.

                  1743.     Estimation of the Myocardial Extracellular Volume Fraction from Dynamic Contrast-Enhancement Measurements and Comparison with Partition Coefficient Measurements

Michael Jerosch-Herold¹, Ray E. Hershberger², Craig Broberg³

¹Radiology, Brigham & Women's Hospital, Boston, MA, USA; ²Medicine, University of Miami Miller School of Medicine, Miami, FL, USA; ³Medicine, Oregon Health & Science University, Portland, OR, USA

T1 measurements after contrast administration have been shown to provide a novel marker of diffuse fibrosis and remodeling of the extracellular matrix. In this study we tested the hypothesis that multiple T1 measurements for determination of the partition coefficient can be replaced with a single dynamic contrast enhancement study for determination of the extracellular volume fraction. The two methods were found to agree well. Multi-slice measurements of myocardial contrast enhancement by dynamic imaging provide an efficient method to determine the distribution volume for an extracellular contrast agent, compared to multiple T1 measurements.

                  1744.     Intra-Individual Comparison of 0.1 Mmol/kg Versus 0.2 Mmol/kg Gadobutrol for Magnetic Resonance Late Enhancement Imaging at 3 Tesla

Frank Grothues¹, Hagen Boenigk¹, Martin Kanowski², Joachim Graessner³, Katrin Christel Sprung³, Martin Rohrer⁴, Ruediger Christian Braun-Dullaeus¹

¹Cardiology, University Hospital Magdeburg, Magdeburg, Germany; ²Neurology, University Hospital, Magdeburg, Germany; ³Siemens Medical Solutions, Erlangen, Germany; ⁴Bayer Schering Pharma AG, Berlin, Germany

Two doses of gadobutrol (0.1 mmol/kg or 0.2 mmol/kg) were evaluated qualitatively and quantitatively for late enhancement (LE) imaging at 3T in an intra-individual comparison in 15 patients with a history of myocardial infarction. Qualitative judgement by blinded observers revealed an in tendency better image quality with fewer artifacts and less noise with the higher contrast dosage. Quantitative measurements of infarct size and contrast-to-noise ratio between infarcted and remote myocardium were significantly higher with the 0.2 mmol/kg dosage and should therefore be preferred.

                  1745.     Free-Breathing Semi-Automated Quantitation of Area-At-Risk Size in Patients with Acute Myocardial Infarction

Samuel Alberg Kock¹, Anders Koustrup Niemann¹, Jacob Thorsted Sørensen², Jens Flensted Lassen², Christian Juul Therkelsen², Hans Erik Bøtker², Torsten Toftegaard Nielsen², Won Yong Kim^1,2

¹MR-Center, Aarhus University Hospital Skejby, Aarhus, DK, Denmark; ²Dept. of Cardiology B, Aarhus University Hospital Skejby, Aarhus, DK, Denmark

T2-weighted MRI scans can depict areas at risk of suffering irreversible due to myocardial infarction but are prone to segmentation difficulties due to insufficient suppression of slow blood-flow which is often present in trabeculated left ventricles. B-SSFP sequences are known for their excellent blood-to-tissue contrast and may be employed for accurate segmentation purposes. Combining B-SSFP with T2-weighted scans may improve diagnostic accuracy and facilitate quantitative measurements of the size of non-viable myocardium and areas-at-risk.

                  1746.     Structural and Functional Consequences of Ischemia and Reperfusion in the Heart - A Long-Term Follow Up Via Cardiac MRI in a Rodent Model

Florian Andreas Bönner^1,2, Stefan Weber³, Karl Friedrich Kreitner, Wolfgang Günther Schreiber³, Georg Horstick²

¹1st Medical Clinic, Division of Cardiology, Pulmonary Disease and Vascular Medicine, University Hospital RWTH Aachen, Aachen, Germany; ²2nd Medical Clinic, Mainz University Medical School, Mainz, Germany; ³Section of Medical Physics, Department of Radiology, Mainz University Medical School, Mainz, Germany

The objective of this study was to noninvasively highlight global and local effects of ischemia and reperfusion (I/R) in a rodent model by using contrast enhanced (CE) and cine MRI based on a 1.5 Tesla clinical scanner system as a preliminary study. Rats received a more or less prolonged cardiac ischemia. SR-FLASH and SR-Turbo-FLASH sequences were used to generate perfusion, CE and cine MRI 48h and 6 weeks after I/R. Images were taken with sufficient resolution to have a valid analysis of contrast agend distribution and myocardial func-tion. This setup promises good practicability for further pathophysiological studies.

                  1747.     Observer Performance with State-Of-The-Art T2 Weighted Sequences (T2p SsSSFP, T2w ACUTE, T2 DB TIRM, T2 DB BLADE) for the Detection of Area-At-Risk (AAR) in Acute Coronary Syndromes (ACS).

Magalie Viallon¹, Nathan Mewton², Didier Revel², Peter Kellman³, Pierre Croisille²

¹Radiology, Geneva University hospital, GENEVA, Switzerland; ²Radiology, Hospices Civils de Lyon, Université de lyon, Lyon, France; ³Radiology, Laboratory of Cardiac Energetics, NIH/NHLBI, Bethesda, MD, USA

No clear consensus exists on the most appropriate sequence to characterize AAR. Our objective is to determine the capacity of each technique to clearly identify region of hyper/hypo enhanced area (edema/no reflow) as a surrogate to AAR.

                  1748.     Relationship of Infarct Size and Cardiac Functions in the Hyper-Acute Phase of Myocardial Ischaemia-Reperfusion in Rats

King Kenneth Cheung^1,2, Anthony Neil Price², Johannes Riegler², Roger J. Ordidge¹, Mark Francis Lythgoe²

¹Department of Medical Physics and Bioengineering, University College London (UCL), London, UK; ²Centre for Advanced Biomedical Imaging (CABI), Department of Medicine and Institute of Child Health, UCL, London, UK

MRI measurement of in vivo cardiac function has become a popular method to assess treatment effects in small animal models of myocardial infarction (MI). In chronic MI, infarct size has been shown to correlate closely to deterioration in cardiac functions; however, the relationship between infarct size and cardiac contractility in the hyper-acute phase (i.e. hours) following MI is unclear. We aimed to investigate this relationship using cardiac MRI with Gd-DTPA delayed-enhancement and ex vivo myocardial viability assay. Our study suggests a discrepancy between infarct size and reduction of ejection fraction during the early phases of MI. The observed variability in the hyper-acute phase of MI may reflect the presence of myocardial stunning.

                  1749.     3D Black-Blood T1 Mapping of the Mouse Heart Using IntraGate FLASH and DESPOT1 Analysis

Bram Fransiscus Coolen¹, Tessa Geelen¹, Arno Nauerth², Larry de Graaf¹, Klaas Nicolay¹, Gustav J. Strijkers¹

¹Biomedical NMR, Department of Biomedical Engineering, Eindhoven University of Technlogy, Eindhoven, Netherlands; ²Bruker BioSpin MRI GmbH, Ettlingen, Germany

A method for fast and robust 3D T1 quantification of the mouse heart is presented. This is based on a IntraGate FLASH retrospectively triggered steady-state sequence in combination with DESPOT1 analysis. By additionally incorporating blood saturation, flow artifacts are reduced and blood-myocardial contrast is enhanced. With this method, very homogeneous T1 maps of the mouse myocardium were obtained. This steady-state approach allowed quantification of T1 through the entire mouse heart within reasonable acquisition times.

                  1750.     Temporal Cardiac Manganese-Enhanced MRI with T1-Mapping in Mice Following Myocardial Infarction

fang liu^1,2, Ben Waghorn^1,2, Jimei Liu¹, Tom C. - C. Hu^1,2

¹Small Animal Imaging, Department of Radiology, Medical College of Georgia, Augusta, GA, USA; ²Nuclear and Radiological Engineering / Medical Physics Programs, George W. Woodruff School, Georgia Institute of Technology, Atlanta, GA, USA

A temporal manganese-enhanced MRI T1-mapping method has been implemented to monitor the remodeling process following myocardial infarction in the mouse heart. The study demonstrates that this T1-mapping method is sensitive to detect the uptake of manganese dynamically in the myocardial cells and that reduced Mn2+ uptake can be an indication of the viability of the injured myocardial cells. Similarly, the quantification of Mn2+ content enabled by this T1-mapping method can be applied to estimate the progression of myocardial infarction, which may be useful for potential therapeutic interventions.

                  1751.     Cardiac Function and Fat Composition in STZ-Induced Diabetic Rats

Jun Lu¹, Beau Pontre², Stephen Pickup³, Yee Soon Choong, Anthony Phillips, Hong Xu⁴, Garth JS Cooper, Alistair Young⁵

¹School of Biological Sciences, University of Auckland, Auckland, New Zealand; ²Centre for Advance MRI, University of Auckland; ³Dept of Radiology, University of Pennsylvania, Philadelphia, PA, USA; ⁴Chemistry & Chemical Engineering, Shenzhen University, Shenzhen, Guangdong, China; ⁵Dept of Physiology, University of Auckland

End stage diabetes is always associated with heart failure, which is the leading cause of death. We are the first to investigate the heart function and fat composition in STZ-induced diabetic rats using High Field MRI. Gradient echo cine method was used to determine cardiac function and fast spin echo Dixon method was used to detect cardiac fat composition. We found that the cardiac injection fraction of every rat decreased with prolonged diabetic status and cardiac fat disappeared after 8 weeks of diabetes. The results suggest that heart function impairment occur with long term diabetes in STZ rat model.

                  1752.     Preservation of Cardiac Function in Transient Receptor Potential V4 Knockout Mice After Myocardial Infarction Measured by Magnetic Resonance Imaging

Gregory Turner¹, Weike Bao, Beat Jucker, Kevin Thorneloe, Robert Willette

¹GlaxoSmithKline, King of Prussia, PA, USA

The Transient Receptor Potential V4 (TRPV4), expressed in endothelium throughout the cardiovascular system, is a cation channel that contributes to intracellular Ca++ homeostasis and cell volume regulation . Pharmacologic activation of TRPV4 has been associated with circulatory collapse and failure of the endothelial barrier. In the present study, the effects of TRPV4 deletion (TRPV4-/-) on cardiac function was examined in a myocardial infarction (MI) model. MRI revealed preservation of cardiac function in the TRPV4-/- mouse 1 week post-MI. These results indicate that TRPV4 deletion produces a cardio-protective phenotype and suggests an important relationship between the endothelium and cardiac function.

                  1753.     Assessment of Left Ventricular Myocardial T1 Hyperintesity in Patients with Suspected Cardiac Amyloidosis Using MRI

Kirsi M. Taimen¹, MIchael C. Yang¹, Karin Dill¹, John Sheehan¹, James C. Carr¹

¹Department of Radiology, Northwestern University, Chicago, IL, USA

Amyloidosis is a systemic disease caused by extracellular deposition of insoluble protein. Cardiac involment is common and often fatal form of disease. Cardiovascular magnetic resonance exhibits late enhancement and shorter T1 in cardiac amyloidosis. We assessed T1 signal in cardiac amyloidosis using T1 weigheted gradient echo (GRE) technique and compared findings in age-matched control group. Results show that left ventricular myocardium is T1 hyperinterse on pre-contrast T1-GRE imaging in patients with suspected amyloidosis. They also have a shorter TI compared to controls following contrast administration.

                  1754.     Pre-Contrast Imaging of the Area at Risk in Acute Myocardial Infarction: T2 Yields Higher Contrast Than T1 MRI

Anthony Homer Aletras¹, Peter Kellman¹, Li-Yueh Hsu¹, Daniel Groves¹, Robert F. Hoyt Jr. ², Andrew Ernest Arai¹

¹Laboratory of Cardiac Energetics, National Heart, Lung and Blood Institute, Bethesda, MD, USA; ²Office of the Scientific Director, National Heart, Lung and Blood Institute, Bethesda, MD, USA

We hypothesized that, due to the intrinsic T1 and T2 associated with myocardial edema, higher contrast between the AAR and normal myocardium would be measured with pre-gadolinium T2 MRI than with pre-gadolinium T1 MRI. We also hypothesized than this would also be true for contrast between the infarct and peri-infarct zones within the AAR. Our in-vivo results suggest that T2 MRI can yield better contrast than T1 MRI for differentiating infarct, peri-infarct and remote myocardium. These results do not reflect limitations set by a particular pulse sequence but rather by the physiology itself. These results are based on the intrinsic T1 and T2 changes as a result of different water content, water mobility and interactions with the lattice within the three regions.

                  1755.     Accurate Assessment of Myocardial Infarction in Mice Using 3D Inversion Recovery Gradient Echo MRI

Steffen Bohl^1,2, Craig A. Lygate¹, Hannah Barnes¹, Debra Medway¹, Lee-Anne Stork¹, Jeanette Schulz-Menger², Stefan Neubauer³, Juergen E. Schneider¹

¹Cardiovascular Medicine, BHF Experimental MR Unit, Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, UK; ²Cardiology, Franz Volhard Klinik, Charite University Medicine, HELIOS Klinikum, Berlin, Germany; ³Cardiovascular Medicine, John Radcliffe Hospital, University of Oxford, Oxford, UK

In-vivo myocardial gadolinium pharmacokinetics were investigated in infarcted mice using rapid T1-measurements. In a second step fast high resolution three-dimensional late Gadolinium enhancement MRI was implemented using tailored latencies and inversion times for both intravenous and intraperitoneal injections of the contrast agent. Infarct sizes derived from MRI were related to the histological standard of reference.

                  1756.     Impact of Myocardial Infarct on Regional Left Ventricular Function

Ramsey Al-Hakim¹, James Sayre¹, Kalyanam Shivkumar¹, Paul Finn¹, Carissa Fonseca¹

¹David Geffen School of Medicine at UCLA, Los Angeles, CA, USA

We utilized Cine and Delayed Contrast Enhanced MRI to measure the extent of functional impairment in the infarct, infarct-adjacent, and remote myocardial regions in patients with previous left ventricular myocardial infarction (MI). The infarct-adjacent region was functionally distinct from the infarct and remote regions as measured by segmental myocardial thickening, ejection fraction, and end systolic volume. Furthermore, the decline in thickening from the infarct-adjacent to infarct region was significantly greater than from the remote to infarct-adjacent region. This suggests a non-linear relationship exists between post-MI myocardial function and distance from infarct, which has important clinical implications in revascularization therapy.

                  1757.     Validation of Infarct Measurements by Inversion Recovery Delayed-Enhancement MRI During the Hyper-Acute Phase of Myocardial Infarction in Rats

King Kenneth Cheung¹, Anthony Neil Price², Roger J. Ordidge¹, Mark Francis Lythgoe²

¹Department of Medical Physics and Bioengineering, University College London (UCL), London, UK; ²Centre for Advanced Biomedical Imaging (CABI), Department of Medicine and Institute of Child Health, UCL, London, UK

Delayed-enhancement (DE)-MRI is a well-established method for assessing in vivo infarct size in chronic and sub-chronic myocardial infarction (MI); however its application in small animal models has been limited to sub-optimal MR sequences, and its accuracy in the hyper-acute phase (i.e. hours) of MI remains ill-defined. This abstract describes a validation of inversion recovery DE-MRI for measuring infarct size during the hyper-acute phase of MI in rats. Our results demonstrate excellent agreement between DE-MRI and TTC staining. DE-MRI can offer accurate infarct measurement at the hyper-acute phase of MI, which may serve as a valuable early marker for prognostic predictions and assessment of therapeutic outcome.

                  1758.     The Combination of Low Dose Dobutamine and Delayed Enhancement CMR Is More Predictive of Infarct Zone and Left Ventricular Functional Recovery Than Either Element Alone.

Anne Elizabeth Scott¹, Scott Ian Kay Semple², Thomas W. Redpath³, Graham S. Hillis¹

¹Cardiology, University of Aberdeen, Aberdeen, Grampian, UK; ²Medical Physics, University of Edinburgh, Edinburgh, Midlothian, UK; ³Radiology, University of Aberdeen, Aberdeen, Grampian, UK

This study investigated the relative and combined utility of delayed enhancement (DE) CMR and low dose dobutamine (LDD) CMR for the prediction of both infarct zone and global left ventricular functional recovery following myocardial infarction. Adding LDD to DE CMR required an additional fifteen minutes and significantly improved prediction of functional recovery. A comprehensive CMR examination with both DE and LDD allowed more accurate prediction of both infarct zone and global functional recovery than either component alone.

                  1759.     Myocardial Fat Deposition in Segements with Infarction Detected with Water-Fat Separation Imaging

James W. Goldfarb¹, Margeurite Roth¹, Jing Han¹

¹Saint Francis Hospital, Roslyn, NY, USA

There is a high prevalence of fat deposition in chronic myocardial infarction detected by water-fat separation MR imaging. MR imaging is capable of fat and infarction detection and could be useful in determining the functional cause and significance of fatty deposition after left ventricular myocardial infarction.

                  1760.     Myocardial T1, T2 and T2* Measurements for in Vivo Assessment of Iron Overload in Thalassemia

Taigang He¹, John-Paul Carpenter¹, Guang-Zhong Yang², Dudley J. Pennell¹, David N. Firmin¹

¹Cardiovascular MR Unit, Royal Brompton Hospital and Imperial College, London, UK; ²Department of Computing, Imperial College London

This study is to compare myocardial T1, T2, and T2* measurements and their intrinsic relationship in thalassemia patients (N=48). Multi-echo T1, T2, and T2* images were acquired in the myocardium. The results demonstrate that both T1 and T2 correlate linearly with T2* in patients with iron overload, but not in patients with normal iron. The study also demonstrates a linear correlation between T1 and T2 in all the patients. This study suggests that T1, T2, and T2* can be used for assessing myocardial iron overload. For subjects with normal iron, T1 and T2 can potentially provide improved myocardial tissue characterization.

                  1761.     Comparing in Vivo T2* and T2 Measurements in Tissues of Liver and Heart in Thalassemia

Taigang He He¹, Gillian C. Smith¹, Raad H. Mohiaddin¹, David N. Firmin¹, Dudley J. Pennell¹

¹Cardiovascular MR Unit, Royal Brompton Hospital and Imperial College, London, UK

This study aims at determining the intrinsic T2*/T2 relationship in thalassemia patients (N=50) and compare this relationship in the tissues of liver and heart. Multi-echo T2* and T2 images were separately acquired in the heart and consequently in the liver. It demonstrates that hepatic T2 correlated linearly with T2* in patients with iron overload, suggesting that both can be used for assessing iron overload. This relationship between T2* and T2 in the liver, however, appears different from that in the myocardium, suggesting the calibration results of T2*/T2 against biopsy in the liver may not be directly transferred to the heart.

                  1762.     Twelve Weeks of Chronic Mitral Regurgitation Alters Myocardial Structure as Measured by Diffusion Tensor MRI

Daniel Bruce Ennis¹, Tom C. Nguyen², Gabriel R. Barnard¹, Aki Itoh², Stefan Skare³, Wolfgang Bothe², Neil B. Ingels², D C. Miller²

¹Department of Radiological Sciences, David Geffen School of Medicine, University of California, Los Angeles, CA, USA; ²Department of Cardiothoracic Surgery, Stanford University, Palo Alto, CA, USA; ³Department of Radiology, Stanford University, Palo Alto, CA, USA

Left ventricular (LV) dysfunction in chronic mitral regurgitation (MR) may result, in part, from changes in LV microstructure and diffusion tensor magnetic resonance imaging (DTMRI) is well suited for the evaluation of microstructural organization. The <B>objective</B> of this study was to determine whether twelve-weeks of chronic MR alters the LV myocardial microstructure as measured with DTMRI. Comparisons between LV microstructural measures in normal sheep and chronic MR sheep demonstrate a significant (P<0.000001) increase in the rate of water diffusion (trace) and a decrease in fractional anisotropy (FA). These changes may have a direct impact on ventricular systolic and diastolic function.

                  1763.     Cardiac Diffusion Tensor Magnetic Resonance Imaging for High Resolution Reconstruction of Individual Whole Hearts

Patrick William Hales¹, Fleur Mason², Rebecca Burton², Christian Bollensdorff², Martin Bishop³, Gernot Plank³, Peter Kohl², Jurgen Schneider¹

¹Department of Cardiovascular Medicine, Oxford University, Oxford, Oxon, UK; ²Department of Physiology, Anatomy and Genetics, Oxford University, Oxford, Oxon, UK; ³Computational Biology Group, Oxford University, Oxford, Oxon, UK

The ‘3D Heart’ project aims to develop an efficient methodology for the acquisition of cardiac anatomical MRI data, DTI data, and histology, and to develop techniques for automated segmentation and registration of these data sets to create high resolution computer models of individual whole hearts. We present here the progress to date of this project, focussing on methodological developments for the acquisition of the high resolution DTI data sets. The challenges posed by the processing and combination of large, multiple data sets into a single comprehensive model are also discussed.

                  1764.     Temporal Changes in Diffusion Tensor Imaging Parameters in the Ex-Vivo Rat Heart

Patrick William Hales¹, Fleur Mason², Rebecca Burton², Christian Bollensdorff², Peter Kohl², Jurgen Schneider¹

¹Department of Cardiovascular Medicine, Oxford University, Oxford, Oxon, UK; ²Department of Physiology, Anatomy and Genetics, Oxford University, Oxford, Oxon, UK

We have monitored the changes in the DTI parameters of ex-vivo rat hearts over time, in order to determine the ‘useful lifetime’ of these samples in long term studies. Initial results have shown that after a period of approximately 50 hours post excision, the mean ADC and FA values in the myocardium start to change. By monitoring changes in orientation of the primary eigenvector over time, on a voxel-by-voxel basis, we are able to examine regional changes in the structure and composition of the myocardium.

                  1765.     Myocardial Vasculature: A Major Contributor to Cardiac MR Diffusion

MinSig Hwang¹, Melvin Clark¹, John R. Forder¹

¹University of Florida, Gainesville, FL, USA

Since the coronary circulation is a pivotal component of the myocardium, vascular spaces may need to be weighted as much as interstitial and intracellular spaces in order to understand the physiological compartments responsible for the myocardial DTI signals. In this experimental study using 11T magnet, we demonstrated vascular compartment is a significant contributor to the myocardial MR diffusion characteristics observed in isolated rabbit heart by replacing the vascular space with perfluorocarbon-emulsion (PFC) and by changing the vascular flow rate of a modified St. Thomas¡¯ Hospital cardioplegic solution (STH).