Electronic Posters : Functional MRI
Click on to view the abstract pdf and click on to view the video presentation.
fMRI & Functional Connectivity Analysis

Monday May 9th
Exhibition Hall  14:00 - 16:00 Computer 51

14:00 3544.   Inter-Subject Correlations between Resting-State Spontaneous Fluctuations and Fractional Volume of Gray Matter 
Qihong Zou1,2, Wanyong Shin1,3, Hong Gu1, Xiujuan Geng1, Wang Zhan4, Yufeng Zang2, and Yihong Yang1
1Neuroimaging Research Branch, National Institute on Drug Abuse, National Institutes of Health, Baltimore, MD, United States, 2State Key Laboratory of Cognitive Neuroscience and Learning, Beijing Normal University, Beijing, Beijing, China, People's Republic of, 3Imaging institute, Cleveland Clinic, Cleveland, OH, United States,4Center of Imaging for Neurodegenerative Diseases, University of California San Francisco, San Francisco, CA, United States

Fractional amplitude of low-frequency fluctuations (fALFF) has been used to measure the strength of spontaneous fluctuations of resting-state fMRI signal. However, its underlying mechanisms and inter-subject variations are unknown. In this study, voxel-wise correlations between fALFF and fractional volume of gray matter (fv,GM) across subjects are investigated. Our results show that fALFF and fv,GM are positively correlated over large parts of the brain and fv,GM can be used as a covariate to remarkably reduce inter-subject variations. These findings suggest that fALFF is related to tissue fraction and its inter-subject variations can be reduced by including fv,GM in group analyses.

14:30 3545.   Quantitative BOLD using a Diffusive Model 
John David Dickson1, Dmitriy A Yablonskiy2, Alex L Sukstanskii2, Tom WJ Ash3, Guy B Williams3, and Richard E Ansorge1
1Department of Physics, Cambridge University, Cambridge, Cambridgeshire, United Kingdom, 2Mallinckrodt Institute of Radiology, University of Washington in St Louis, St Louis, MO, United States, 3Wolfson Brain Imaging Centre, Cambridge University, United Kingdom

Previous work in quantitative BOLD has made use of a static model of baseline BOLD signal. In order to account for spin diffusion we introduce a recently created phenomenological diffusive BOLD model into the quantitative BOLD method. Applying this to real MRI data we show that estimates of hemodynamic properties in the healthy brain are closer to literature values than achieved when using the static model.

15:00 3546.   Voxel-wise fMRI group analysis using fractional volume of gray matter as a covariant 
Wanyong Shin1, Hong Gu2, Qihong Zou2, Thomas Ross2, and Yihong Yang2
1Radiology, Imaging Institute, Cleveland Clinic, Cleveland, OH, United States, 2National Institute on Drug Abuse, Baltimore, MD, United States

We present a method of voxel-wise group analysis using fractional volume of gray matter (fv,GM) as a covariant in a fMRI study with visual stimulation. The results show that a) Consistent functional activation is observed between conventional t-test and the t-test with fv,GM as a covariant, b) The size of the activated ROI and average t-score in the activated ROI are increased when using the covariate t-test with the same statistical threshold, and c) The slope of â with respect to fv,GM is observed higher in V1 compared to V2, V3 and thalamus areas

15:30 3547.   Characterization of spatial variation of BOLD-associated neuronal activity in fMRI 
Yu Li1, and Hu Cheng2
1Radiology, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, United States, 2Psychological and Brain Sciences, Indiana University at Bloomington, Bloomington, Indiana, United States

The work presented here introduces a novel approach to visualizing neuronal activity using two contrast mechanisms that arise from the spatial difference of BOLD-induced changes in local B0 field homogeneity and strength. The new contrast is generated by a complex signal calculated from original fMRI signals. This signal, as it characterizes BOLD-associated changes of B0 field variation in image space, is called "field variation signal". In this work, it is demonstrated that fMRI data analysis using field variation signals has the potential to visualize spatial variation of BOLD-associated neuronal activity.

Tuesday May 10th
  13:30 - 15:30 Computer 51

13:30 3548.   A support vector machine based real-time fMRI communication channel 
Tom Ash1, Adrian Carpenter1, and Guy Williams1
1Wolfson Brain Imaging Centre, University of Cambridge, Cambridge, United Kingdom

We present a support vector machine based technique for allowing subjects in a scanner to communicate yes / no answers to questions, with answers available to experimenters in real time (less than 1 TR). The technique uses SVMs to detect subject brain state as they perform mental imagery to reply to questions. Tests in healthy volunteers show 19% of subjects answers were not distinct enough to be confidently labeled. Of the remaining 109, 107 answers were correctly interpreted by the classifier. This technique shows promise for use as a real-time communication channel for behaviorally vegetative, internally conscious brain injury patients.

14:00 3549.   Combination of SVM and ROI Approaches for Real-Time fMRI Neurofeedback 
Vadim Zotev1, Raquel Phillips1, Ruben Alvarez1, W Kyle Simmons1, Pat Bellgowan1, Wayne Drevets1, and Jerzy Bodurka1
1Laureate Institute for Brain Research, Tulsa, OK, United States

The support vector machines (SVM) approach to decode patterns of whole-brain activity can be utilized for real-time fMRI neurofeedback. We propose a combined SVM and region-of-interest (ROI) neurofeedback approach. A custom rtfMRI system was used to compute an SVM classifier and measure fMRI activation in the left amygdala ROI, and to provide neurofeedback as a variable-height bar. Three healthy male subjects were asked to feel happy so as to raise the bar. We observed that ROI-based neurofeedback improves SVM performance, while SVM-based neurofeedback increases ROI activation. Combination of these two approaches benefits both and enhances rtfMRI neurofeedback training.

14:30 3550.   Online Learning for Real Time fMRI Classification 
Hao Xu1, Yongxin Taylor Xi1, Ray Lee2, and Peter J. Ramadge1
1Electrical Engineering, Princeton University, Princeton, NJ, United States, 2Princeton Neuroscience Institute, Princeton University, Princeton, NJ, United States

We propose a real-time conjugate gradient(rtCG) algorithm as an efficient solution for online real-time fMRI(rtfMRI) system. rtCG leans to classify brain states as data is being collected. It has a close connection with well-established partial least squares(PLS) algorithm when applied to the quadratic problem of interest. Real fMRI data tests show that rtCG can process high dimensional fMRI data within one TR and reach high prediction accuracy.

15:00 3551.   Real-time BOLD functional MRI neuro-feedback: connectivity changes observed in an imagery task 
Silvina G Horovitz1, Brian D Berman1,2, and Mark Hallett1
1HMCS, NINDS - NIH, Bethesda, MD, United States, 2Neurology, School of Medicine University of Colorado Denver, Aurora, CO, United States

Functional connectivity resting state based on BOLD functional MRI (fMRI) has been used mainly to observe brain networks while subjects ‘do nothing’. The same principle, however, can be used for any other steady-state brain condition. In this study, we compare the connectivity of the motor area during hand movement imagery with and without real time neuro-feedback of BOLD signal of the hand motor area. We observed connectivity changes during neuro-feedback runs; these changes were maintained through a subsequent run without neuro-feedback.

Wednesday May 11th
  13:30 - 15:30 Computer 51

13:30 3552.   Connectivity Analysis through Structural Vector Auto-Regressive (SVAR) Modeling 
Gang Chen1, Ziad S. Saad1, J. Paul Hamilton2, Ian H. Gotlib2, and Robert W. Cox1
1SSCC/DIRP/NIMH, National Institutes of Health, Bethesda, Maryland, United States, 2Mood and Anxiety Disorders Laboratory, Department of Psychology, Stanford University, Stanford, California, United States

Between the two popular methods in connectivity analysis, vector auto-regression (VAR) faces a challenging issue in data sampling rate while structural equation modeling (SEM) usually suffers from the violation of the assumption that no lagged correlation is considered within and across regions. With the synthesis of both methods, structural vector auto-regressive (SVAR) modeling accounts for both contemporaneous and delayed effects among the regions, and provides a more powerful and robust tool for network modeling than VAR and SEM when they are applied alone. Here we present an SVAR program that is platform-independent and open source.

14:00 3553.   A Combined Dynamic Causal Modeling and Functional MRI Study to Assess Visuospatial Symmetry Judgment in Healthy Subjects 
Manisha Bhattacharya1, Shilpi Modi1, Memita Devi1, Namita Singh Saini1, Rajendra Prasad Tripathi1, and Subash Khushu1
1NMR Research Centre, INMAS, Delhi, Delhi, India

Dynamic Causal Modeling along with fMRI was used to quantify effective connectivity within a visuospatial symmetry processing network in each cerebral hemisphere. Twelve different models consisting of Occipital Gyrus, Superior Parietal Lobule (SPL(7A)) and Middle Frontal Gyrus (BA6) were defined. Our results provide evidence in support of the direct effect of the task on SPL and its modulatory effect on the connectivity between SPL and BA6 for both hemispheres. However, SPL is excitatory in its action on BA6 in the left hemisphere while it is inhibitory in the right hemisphere demonstrating the greater role of left hemisphere in symmetry processing.

14:30 3554.   Conditional Granger Causality analysis of fMRI data shows a direct connection from LGN to hMT+ bypassing V1 
Anna Gaglianese1, Mauro Costagli2, Giulio Bernardi1, Lorenzo Sani1, Emiliano Ricciardi1, and Pietro Pietrini1
1Laboratory of Clinical Biochemistry and Molecular Biology, Pisa, Italy, Italy, 2Laboratory for Cognitive Brain Mapping, RIKEN - Brain Science Institute, Wako, Japan

The human middle temporal complex (hMT+) is devoted to motion perception. To determine whether motion-related neural information may reach hMT+ directly from the thalamus, by-passing the primary visual cortex (V1), we measured effective connectivity in an optic flow fMRI experiment in humans. Conditional Granger Causality analysis was employed to measure direct influences between the lateral geniculate nucleus (LGN) and hMT+, discarding indirect effects mediated by V1. Results indicated the existence of a bilateral alternative pathway for visual motion processing that allows for a direct flow of information from LGN to hMT+. This direct link may play a role in blindsight.

15:00 3555.   Stimulus entrained dynamic effective connectivity analysis of fMRI 
Gopikrishna Deshpande1, Simon Lacey2, Henrik Hagtvedt3, Venessa Patrick4, Amy Anderson2, Randall Stilla2, João Ricardo Sato5, Srinivas Reddy6, K Sathian2, and Xiaoping Hu7
1AU MRI Research Center, Department of Electrical and Computer Engineering, Auburn University, Auburn, AL, United States, 2Department of Neurology, Emory University, Atlanta, GA, United States, 3Carroll School of Management, Boston College, Chestnut Hill, MA, United States, 4C. T. Bauer College of Business, University of Houston, Houston, TX, United States, 5Center of Mathematics, Computation and Cognition, Universidade Federal do ABC, Santo André, Brazil, 6Centre for Marketing Excellence, Singapore Management University, Singapore, 7Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, United States

Hemodynamic variability can affect the validity of inferences obtained from Granger causality (GC) analysis of fMRI. Also, it is difficult to obtain context-dependent and/or dynamic connectivity from traditional GC analysis of short fMRI time series data. In order to alleviate these problems, we developed a stimulus-entrained, dynamic GC approach which not only models the time-varying connectivity but also determines whether the dynamics are entrained to external stimuli. Using simulations, we show that this approach is not affected by hemodynamic variability. Also, we demonstrate the experimental utility of this approach using a visual art paradigm.

Thursday May 12th
  13:30 - 15:30 Computer 51

13:30 3556.   A Correlation-Matrix-Based Clustering Method for Extracting Correlation Patterns of Spontaneous BOLD Fluctuations 
Xiao Liu1,2, Xiao-Hong Zhu1, Yi Zhang1, Peihua Qiu2, and Wei Chen1
1CMRR, Radiology, University of Minnesota, Minneapolis, MN, United States, 2Statistics, University of Minnesota, Minneapolis, MN, United States

In this study, we introduce a novel correlation-matrix-based clustering method for extracting correlation patterns in spontaneous BOLD fluctuations and for identifying multiple resting-state networks. This method has merits beyond commonly-used seed-based correlation mapping and spatial independent component analysis (ICA): no priori information required, easy interpretation of outcomes, easy for the group level analysis, and effective in identifying multiple resting networks with clear and robust patterns at one time. It could be a powerful tool for investigating resting-state brain networks detected by spontaneous BOLD fluctuations.

14:00 3557.   Eigenvector centrality mapping based on low-frequency phase alignment 
Gabriele Lohmann1, Maren Grigutsch1, Daniel Margulies1, Annette Horstmann1, Burkhard Pleger1, Joeran Lepsien1, Dirk Goldhahn1, Haiko Schloegl2, Michael Stumvoll2, Arno Villringer1, and Robert Turner1
1Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2Department of medicine, University of Leipzig, Leipzig, Germany

We have previously introduced a new analysis method for fMRI data called “eigenvector centrality mapping (ECM)” (Lohmann et al, 2010). In ECM, each voxel receives a rank describing its centrality within the brain using a method similar to Google's PageRank algorithm. In this context, we have previously used spectral coherence as a similarity metric. However, this ignores phase shifts so that time series may receive high coherence values even though they are separated by large phase shifts. Here, we investigate whether changes in brain states manifest themselves not only in centrality changes of spectral coherence but also of phase alignment.

14:30 3558.   BBCA Analysis of Functional and Structural Networks 
Alex Kenneth Smith1, David J. Madden1, Pooja Gaur1, and Nan-Kuei Chen1
1Brain Imaging and Analysis Center, Duke University, Durham, North Carolina, United States

The medial cortical regions have been shown to exhibit significantly decreased brain activity during task conditions. It has been proposed that these regions form the nodes of a default mode network that is continuously and spontaneously active, even in the absence of an external task. Changes in the functioning of the default mode network have been reported for healthy older adults, relative to younger adults. The following analysis describes a behavior-based connectivity analysis method, in which whole-brain data are used to identify behaviorally-relevant intrinsic and functionally connected networks and the fractional anisotropy that indicates the level of connection between areas.

15:00 3559.   Atlas-based analysis of resting state functional connectivity MRI 
Andreia Vasconcellos Faria1,2, Suresh Joel1,3, Xiaoying Tang4, Peter vanZijl1,3, Michael Miller4, James Pekar1,3, and Susumu Mori1
1Radiology, Johns Hopkins University, Baltimore, MD, United States, 2Radiology, State University of Campinas, Campinas, SP, Brazil, 3FM Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, 4Biomedical Engineering, Johns Hopkins University, Baltimore, MD, United States

Resting state functional connectivity MRI (rsfc-MRI) is becoming widely-used for neuroscience studies. However, the identification of corresponding cortical areas across subjects is not straightforward and the pixel-to-pixel time-domain correlation is inherently very noisy. An Atlas-Based Approach (ABA), where an automated 3D segmentation is applied in each individual, reduces the dimensionality of the data and can be an alternative to evaluate functional connectivity. In this study we report on initial findings in functional brain connectivity and inter-session intra-subject reproducibility of the results obtained by applying an ABA on rsfc-MRI data acquired in two sessions from 21 normal volunteers

Electronic Posters : Functional MRI
Click on to view the abstract pdf and click on to view the video presentation.
Function Connectivity: Physiology & Application

Monday May 9th
Exhibition Hall  14:00 - 16:00 Computer 52

14:00 3560.   Spontaneous fMRI activity reflects a dynamic image of brain state 
Marta Bianciardi1, Masaki Fukunaga1, Jacco A de Zwart1, and Jeff H Duyn1
1Advanced MRI Section, LFMI, NINDS, National Institutes of Health, Bethesda, MD, United States

The origin of spontaneous fMRI activity (SA-fMRI) of the human brain is still poorly understood, for example if it represents ongoing sensory processing or homeostatic/cognitive functions that depend on brain state. To investigate this, we measured the amplitude of SA-fMRI in the visual cortex and of fluctuations in amplitude of magnetoencephalographic-MEG activity and electro-oculogram-activity, varying independently the brain-attentive-state (eyes-open/closed) and the visual input (presence/absence of light). Amplitude of SA-fMRI changed dynamically with brain state, and was not modulated by visual input. Our results suggest that the level of SA-fMRI may depend on the level of arousal and oculo-motor activity.

14:30 3561.   The association between pulse wave velocity, as a marker of sympathetic tone, and resting state BOLD signals 
Kevin Murphy1, James Coulson1,2, Ashley D Harris1, Marija Fjodorova1, and Richard G Wise1
1CUBRIC, School of Psychology, Cardiff University, Cardiff, Wales, United Kingdom, 2Wales Heart Research Institute, Cardiff University, Cardiff, Wales, United Kingdom

Dynamic changes in both blood pressure and arterial CO2 concentration lead to variations in the BOLD signal. Fluctuations in arterial blood pressure are controlled by the sympathetic nervous system through changes in arterial stiffness. Pulse wave velocity (PWV) alters with arterial stiffness and can be used as a surrogate measure of sympathetic activity. In this study, significant variance in resting state BOLD signals was explained by simultaneously recorded PWV traces. A lack of delay between the two signals suggests that the related BOLD fluctuations are associated with flow changes arising from dynamic variations in blood pressure.

15:00 3562.   Investigating the neural basis of fcMRI 
Matthew Jon Brookes1, Joanne Hale1, Claire Stevenson1, Johanna Zumer1, Gareth Barnes2, Julia Owen3, Susan Francis1, Srikantan Nagarajan3, and Peter Morris1
1Sir Peter Mansfield Magnetic Resonance Centre, University of Nottingham, Nottingham, United Kingdom, 2Wellcome Trust Centre for Neuroimaging, University College London, London, United Kingdom, 3Biomagnetic Imaging Laboratory, University of California San Francisco, San Francisco, California, United States

In this study we employ ultra high field (7T) functional connectivity (fc) MRI and magnetoencephalography (MEG) to explore, in detail, the electrophysiological basis of haemodynamic measures of functional connectivity. We show good agreement between motor cortex connectivity measured independently using these two disparate neuroimaging modalities. We employ three different MEG based functional connectivity metrics to investigate how neural oscillations mediate functional connectivity. Finally, we investigate the temporal dynamics of connectivity, showing that marked changes occur on a timescale accessible to fMRI. Our study has implications to those developing fcMRI methodologies, and to those employing fcMRI to understand functional connectivity.

15:30 3563.   T2* Modulation of Functional Connectivity using a Multi-echo Strategy 
Changwei W Wu1, Hong Gu1, Qihong Zou1, Hanbing Lu1, Elliot A Stein1, and Yihong Yang1
1Neuroimaging Research Branch, National Institute on Drug Abuse, Baltimore, Maryland, United States

Spatial and spectral effects of T2* were assessed on low-frequency spontaneous fluctuations acquired at 3T. Using a mutlti-echo sequence, broad local (ipsilateral) connections but minimal long-distance (contralateral) connections were found at an ultra-low TE (7.7 ms) in the default-mode network. In frequency domain, the uniform spectral power at low TE indicated minimal T2*-weighted connectivity and the elevated spectral power with long TEs in the frequency range of 0–0.05 Hz, likely due to T2*-mediated neuronal activities. These results characterized the effects of transverse relaxation times on functional connectivity, which would be useful for the interpretation of resting-state fMRI studies.

Tuesday May 10th
  13:30 - 15:30 Computer 52

13:30 3564.   Caffeine Tightens the Coupling Between Resting-State Blood Flow and Metabolic Fluctuations 
Anna Leigh Rack-Gomer1, and Tom T Liu1
1Bioengineering and Center for Functional MRI, UC San Diego, La Jolla, CA, United States

Prior work has shown that caffeine decreases the amplitude and correlation of resting-state BOLD fluctuations. However, the physiological mechanisms by which caffeine alters spontaneous BOLD fluctuations remain unclear. In this study, we show that the ratio of the amplitudes of resting-state BOLD and cerebral blood flow (CBF) fluctuations in the motor cortex is reduced by caffeine. This finding indicates that caffeine tightens the coupling between resting-state fluctuations in CBF and oxygen metabolism. Tighter flow-metabolism coupling is consistent with the caffeine-induced decrease in the amplitude of spontaneous BOLD fluctuations and may also contribute to the reduced resting-state BOLD connectivity.

14:00 3565.   Resting-State Functional Connectivity Modification by Non-invasive Electrical Stimulation of the Brain 
Steven Roys1, Gadi Alon2, George Makris3, and Rao Gullapalli1
1University of Maryland, School of Medicine, Baltimore, MD, United States, 2Physical Therapy, University of Maryland, School of Medicine, 3University of Maryland, School of Medicine

Intracranial electrical brain stimulation has been shown to provide favorable clinical outcomes in various CNS conditions including Parkinson’s disease, stroke, clinical depression and pain. However the mechanisms related to these outcomes are less well understood. In this study we report changes in the resting state motor network following electrical stimulation of the motor cortex using both pulsed and direct current stimulation. We observe persistent, significant changes in the resting state motor network following pulsed electric stimulation. These results suggest that functional connectivity MRI studies can help elucidate the mechanisms responsible for therapeutic effects of electric stimulation.

14:30 3566.   The Relation between Drug-induced Effects on Resting State Brain Connectivity and Cerebral Blood Flow 
Najmeh Khalili-Mahani1,2, Mathiass J. P. Van Osch1, and Serge A. R. B. Rombouts1,2
1Radiology, Leiden University Medical Center, Leiden, Netherlands, 2Leiden Institute for Brain and Cognition, Institute of Psychology, Leiden, Netherlands

We have previously shown that different CNS drugs induce distinct and network specific changes in resting state network connectivity. Here we have examined the relationship between changes in resting state connectivity and changes in cerebral blood flow measured in the same subjects.

15:00 3567.   Influence of Sevoflurane on Regional CBF and Functional Connectivity and Implications Related to Brain/Behavior during General Anesthesia 
Maolin Qiu1, Ramachandran Ramani2, Roberto Martuzzi1, Xiaohui Zhang1, and R Todd Constable1,3
1Diagnostic Radiology, Yale University School of Medicine, New Haven, CT, United States, 2Anesthesia, Yale University School of Medicine, New Haven, CT, United States,3Biomedical Engineering, Neurosurgery, Yale University School of Medicine, New Haven, CT, United States

Effects of sevoflurane on rCBF are observed across the brain cortices but are highly region-specific, with increases limited primarily to subcortical structures and insula and decreases observed primarily in neocortical regions. The functional connectivity within sub-neural networks appears to be much less affected by sevoflurane. With the assumption that both the resting state rCBF and functional connectivity are associated with the same underlying neuronal processes, we wanted to test whether a decrease in CBF during anesthesia would lead to a decrease in functional connectivity, and vice versa. Discrepancies between the changes in rCBF and connectivity caused by sevoflurane might suggest the differences in the neuronal processes with which they are associated. This work examined the relationship between local changes in rCBF and BOLD based functional connectivity in the awake and anesthetized state in humans. Regional CBF was significantly altered by sevoflurane across the brain while the network properties as measured using BOLD functional connectivity remained largely undisturbed. These discrepancies in the responses between rCBF and functional connectivity suggest that they are in general not associated with the same functional components of underlying neuronal processes.

Wednesday May 11th
  13:30 - 15:30 Computer 52

13:30 3568.   Functional networks in the macaque brain revealed by independent component analysis of resting-state fMRI 
R Matthew Hutchison1, L Stan Leung1, Seyed M Mirsattari1, Joseph S Gati2, Ravi S Menon2, and Stefan Everling2
1University of Western Ontario, London, Ontario, Canada, 2Robarts Research Institute, London, Ontario, Canada

Independent component analysis of the resting macaque brain revealed homologous large-scale network organization at multiple levels of processing with humans. These included higher-order networks facilitating executive functioning, attentional processing, reward evaluation, and default-mode activity as well as lower-order networks responsible for vision, audition, sensorimotor integration, and motor control. The consistency of RSNs between macaques and humans suggests the same gross topological cortical organization, thereby providing strong support for their use as an animal model in the study of normal and abnormal functional connectivity.

Paola Valsasina1, Maria Assunta Rocca1, Gianna Riccitelli1, Andrea Falini2, Giancarlo Comi3, and Massimo Filippi1
1Neuroimaging Research Unit, Institute of Experimental Neurology, Division of Neuroscience, San Raffaele Hospital, Milan, Milan, Italy, 2Department of Neuroradiology, San Raffaele Hospital, Milan, Milan, Italy, 3Department of Neurology, San Raffaele Hospital, Milan, Milan, Italy

In this study, we assessed gender-related differences in the entity of resting state (RS) fluctuations in a large cohort of healthy subjects (48 males, 56 males, mean age=23.1 years). Voxel-based morphometry was also run to investigate the correspondence between functional and structural differences. RS activity was found to be increased in males vs. females in several regions of the temporal and parietal lobes, whereas females had a higher RS activity than males in the frontal lobes, the cerebellum and several visual and auditory areas. Gender-related RS functional differences had only a minimal overlap with gray matter volume differences.

14:30 3570.   Intrinsic Resting State Functional Connectivity of Default Mode Network Predicts Attention Task Performance 
Pan Lin1, Nicola De Pisapia1, and Jorge Jovicich1,2
1Center for Mind Brain Sciences,University of Trento, Mattarello, Trento, Italy, 2Department of Cognitive and Education Sciences,University of Trento, Rovereto, Trento, Italy

The default mode network (DMN) is an intrinsic brain system that participates in internal modes of cognition. Neural activity and connectivity within the default network , which are correlated with cognitive ability even at rest. However, what remain unclear is the key issue of whether the inter-region functional connectivity within DMN is related to task performance. Here, we hypothesized that the strength of the functional connectivity (FC) within DMN is an index of the degree of regulation of task performance during attention task state. We found that some inter-region FC within DMN showed the significant negative correlations to reaction time during attention task. These results would suggest the important role of the inter-region FC within DMN can predict attention task performance.

15:00 3571.   Applying Resting-State Functional MRI to Study Impact of Attention Training on Healthy Highly Educated Subjects 
Bob L Hou1, Alison Smith2, Jason Chong2, Julie Brefczynski-Lewis1, and Marc Haut2
1Radiology, West Virginia University, Morgantown, WV, United States, 2Behavioral Medicine & Psychiatry, West Virginia University, Morgantown, WV, United States

We presented our study applying rfMRI data to evaluate the impacts of attention training on healthy highly educated subjects and to examine our hypothses: 1) Brain attention function is linked with rest-state brain attention network; 2) the improvement of attention due to ~{!0~}training~{!1~}results from the increase in the network connectivity.

Thursday May 12th
  13:30 - 15:30 Computer 52

13:30 3572.   Wavelet analysis of the small-world human brain functional network in adolescents prenatally exposed to cocaine 
Lei Jiang1, Zhihao Li1, Claire Coles2, Mary Lynch2, and Xiaoping Hu1
1Department of Biomedical Engineering, Emory University and Georgia Institute of Technology, Atlanta, GA, United States, 2Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, Atlanta, GA, United States

Children and adolescents prenatally exposed to cocaine are at high risk not only for attention/arousal dysregulation and possible inefficiencies in some cognitive functions, but also for problems such as antisocial behavior, substance abuse, and emotional disorders. Because functional brain networks detected in resting-state fMRI have a small-world architecture that reflects a robust functional organization of the brain, here we examined whether this functional organization is disrupted in prenatal cocaine exposure (PCE) by employing a wavelet analysis method. The results show that dysfunctional integrations occur in the brains of PCE individuals during the resting state. Differences between sub-bands were also observed in the small-world analysis. Our findings highlight the need to consider different frequency bands and the usefulness of wavelets in functional connectivity analyses of resting state fMRI.

14:00 3573.   Resting-State fMRI Multi-Spectral Connectivity Networks for Classification of Mild Cognitive Impairment Patients 
Chong-Yaw Wee1, Pew-Thian Yap1, Kevin Denny2, Lihong Wang2, and Dinggang Shen1
1Radiology, University of North Carolina, Chapel Hill, North Carolina, United States, 2Brain Imaging and Analysis Center, Duke University Medical Center, Durham, North Carolina, United States

We introduce an effective network-based multivariate classification algorithm, using multi-spectral connectivity networks derived from resting-state functional MRI, to accurately identify MCI patients from normal controls. Classification accuracy given by our approach is 86.5%, which is at least an 18.9% increment from methods using a single frequency band. The AUC value of our method is 0.863, indicating good diagnostic power. Significant improvements and promising results indicate that the proposed framework can potentially serve as a complementary approach to clinical diagnosis of alteration in brain functions associated with cognitive impairment, especially at early stages.

14:30 3574.   Resting state network abnormalities in amyotrophic lateral sclerosis mirror those of frontotemporal dementia 
Elisa Canu1, Federica Agosta1, Paola Valsasina1, Nilo Riva2, Alessandro Prelle3, Giulia Longoni1, Giancarlo Comi2, and Massimo Filippi1
1Neuroimaging Research Unit, Institute of Experimental Neurology, Division of Neuroscience, Scientific Institute and University Hospital San Raffaele, Milan, Italy,2Department of Neurology, Scientific Institute and University Hospital San Raffaele, Milan, Italy, 33Ospedale Fatebenefratelli e Oftalmico, Milan, Italy

Using resting state functional MRI (RS fMRI), the default mode (DMN), fronto-parietal, executive, and salience networks were explored in 16 patients with amyotrophic lateral sclerosis (ALS) with no cognitive impairment and 15 healthy controls. Compared to controls, ALS patients showed a decreased connectivity of the DMN and fronto-parietal networks in the frontal cortex, but enhanced connectivity in the parietal regions. In non-demented ALS patients, the pattern of network abnormalities mirrors that observed in patients with frontotemporal dementia (FTD). RS fMRI may contribute to shed light on to the overlap between ALS and FTD.

15:00 3575.   Changes in thalamus connectivity in mild cognitive impairment: Evidence from resting state fMRI 
zhiqun wang1, xiuqin jia1, peipeng liang1, and kuncheng li1
1radiology department, xuanwu hospital of Capital Medical University, Beijing, Beijing, China, People's Republic of

Problem: Recently, resting-state functional MRI (fMRI) has attracted increasing attention. Most studies have focused on the hippocampus and posterior cingulate cortex (PCC) connectivity regarding their crucial roles in cognitive function, while less attention has been devoted to the potential role of subcortical nuclei such as thalamus. Methods: Resting state fMRI was used to examine changes in thalamus connectivity in mild cognitive impairment (MCI), which presented a neuro-disconnection syndrome. Results: Functional connectivity between the thalamus and a set of regions was decreased in MCI. We also found increased functional connectivity between the left and the right thalamus in MCI. Conclusions: This study offered a clue to the reduced integrity and compensation in thalamus-related network in MCI.

Electronic Posters : Functional MRI
Click on to view the abstract pdf and click on to view the video presentation.
fMRI: High Temporal & Spatial Resolution

Monday May 9th
Exhibition Hall  14:00 - 16:00 Computer 53

14:00 3576.   Echo-Shifted Multi-Slice EPI compared with GE-EPI in Median Nerve Stimulation at 7T 
Gerda Bjork Grimnisdottir1, Natalia Petridou1,2, and Richard Bowtell1
1Sir Peter Mansfield MR Centre, Physics and Astronomy, The University of Nottingham, Nottingham, Nottinghamshire, United Kingdom, 2UMC, Utrecht, Netherlands

The performance of (fast echo-shifted multi-slice) ES-MS EPI and conventional GE-EPI (short/long TR) is compared in fMRI during median nerve stimulation at 7T. Active voxels were found to be more numerous in short TR (0.33s) ES-MS-EPI data, compared with data acquired using standard GE EPI with a long (2s) TR. The tSNR values are similar for all the short TR scans, when the different echo and repetition times are accounted for. ES-MS-EPI enables scanning with more coverage than GE-EPI of the same TR and spatial resolution. Higher temporal resolution facilitates the capture of transient features of the fMRI signal.

14:30 3577.   Slice-direction SENSE: A Sensitive Acquisition Method for Detecting Neuronal Current MRI Signal Induced by Epilepsy 
Qingfei Luo1, and Gary H. Glover1
1Department of Radiology, Stanford University, Stanford, CA, United States

MRI can potentially be used to detect the neuronal currents induced by epilepsy. However, to capture the brief epiletiform spikes, the MR images need to be acquired with a high temporal resolution (~ tens of ms). In this study, we propose to achieve the high temporal resolution by using the Slice-direction SENSE (Sl-SENSE) acquisition method, which simultaneously excites and acquires multiple image slices. Through a simulation experiment, we demonstrated that Sl-SENSE provides superior detection sensitivity than the conventional pulse sequences, and it may become a promising acquisition method for detecting the neuronal current MRI signal induced by epilepsy.

15:00 3578.   GRASE functional MRI with asymmetric spin-echo 
Lirong Yan1, Robert P Spunt2, Emily Kilroy1, Matthias Gunther3, Matthew D Lieberman2, and Danny JJ Wang1
1Department of Neurology, University of California Los Angeles, Los Angeles, CA, United States, 2Department of Psychology, University of California Los Angeles, Los Angeles, CA, United States, 3Fraunhofer MEVIS-Institute for Medical Image Computing, Bremen, Germany

Although gradient-echo EPI is widely used for BOLD fMRI, it suffers from susceptibility artifacts, and the temporal resolution is generally on the order of 2 to 3s for whole brain coverage. In the present study, we used 3D GRASE with asymmetric spin-echo (ASE) for fMRI with a temporal resolution of 1s per brain volume. Compared to standard 2D EPI, GRASE with ASE not only achieved superior sensitivity in visual cortex activation, but also uniquely detected prefrontal activation during a Go/No-Go task.

15:30 3579.   Quantitative evaluation of RSN functional contrast in low-TR FMRI 
Stephen Smith1, Karla Miller1, Christian Beckmann1,2, Steen Moeller3, Kamil Ugurbil3, Essa Yacoub3, and David Feinberg4,5
1FMRIB, Oxford University, Oxford, Oxon, United Kingdom, 2Donders Institute, Radboud University, Nijmegen, Netherlands, 3Center for Magnetic Resonance Research, University of Minnesota Medical School, MN, United States, 4Advanced MRI Technologies, Sebastopol, CA, United States, 5Helen Wills Institute for Neuroscience, UC Berkeley, CA, United States

We present quantitative evaluation of the functional contrast (effective CNR) in low-TR data that has been acquired by combining two different EPI accelerations, generating whole-brain FMRI images as rapidly as 0.4s. We use ICA and multiple-regression to identify ~62 RSNs in each of 3 subjects, and find that, while peak Z-stat is roughly constant across the 3 TRs for single-regression analyses (seed-based correlation), when deriving the functional parcellation through a multiple-regression, the lowest TR data had peak Z-stat increased by 60% and RSN spatial extent increased by 100%, compared with the unaccelerated data.

Tuesday May 10th
  13:30 - 15:30 Computer 53

13:30 3580.   Event-related functional MRI at high spatial and temporal resolution using UNFOLD 
Sebastian Domsch1, Patrick Heiler1, and Lothar Rudi Schad1
1Computer Assisted Clinical Medicine, Heidelberg University, Mannheim, Germany

UNFOLD has been recently applied to high resolution fMRI using a 3D-spiral sequence. The presented combination of UNFOLD with a simple Cartesian imaging sequence can be easily implemented and does not require regridding of k-space data. In this study, a modified multi-echo FLASH sequence was accelerated by UNFOLD to obtain high resolution 3D images covering half of the brain within 3 seconds. This multi-echo 3D-UNFOLD-FLASH was applied to an event-related finger tapping experiment. The goal was to validate feasibility of this technique for fMRI requiring both, high temporal and spatial resolution.

14:00 3581.   Imaging Cognitive Latencies with Ultrafast 7T fMRI 
Allen T Newton1, Jascha D Swisher2, and John C Gore1,3
1Radiology and Radiological Sciences, Vanderbilt Universithy, Nashville, TN, United States, 2Department of Psychology, Vanderbilt Universithy, Nashville, TN, United States,3Biomedical Engineering, Vanderbilt Universithy, Nashville, TN, United States

Ultrahigh fields (7T) have traditionally been used in fMRI to improve spatial resolution and sensitivity to activation. Here, we demonstrate the potential for these systems to instead image with extremely high temporal resolution, while maintaining significant BOLD contrast, nearly full brain coverage and spatial resolutions common to lower field strengths. We then present an example of how sequences like this can be used to image cognitive latencies between distant brain regions in a simple reaction time task.

14:30 3582.   The limit of relative timing accuracy of BOLD fMRI in human visual cortex 
Fa-Hsuan Lin1,2, Jonathan R. Polimeni2, Kevin Wen-Kai Tsai1, Thomas Witzel2, Wei-Tang Chang1, Wen-Jui Kuo3, and John W. Belliveau2
1Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan, 2Martinos Center, Massachusetts General Hospital, Charlestown, MA, United States,3Institute of Neuroscience, National Yang-Ming University, Taipei, Taiwan

BOLD-contrast fMRI has been widely considered with a poor temporal resolution. Here we investigate the limit of temporal resolution accuracy of BOLD fMRI in human visual system using magnetic resonance inverse imaging (InI) with 10 Hz volumetric sampling rate. Specifically, we interrogate if the inter-hemispheric visual cortex hemodynamic response latency can correctly follow the latency of the visual stimulation between hemifields. We also hypothesize that relative latency in the visual cortex can reliably distinguish between lateralized preceding and the delayed stimuli. Our results suggested that the relative timing accuracy of BOLD response can be as high as 100 ms.

15:00 3583.   Whole Brain Segmented Echo-Volumar-Imaging increases fMRI Sensitivity compared to Multi-Slice Echo-Planar-Imaging 
Stefan Posse1,2, Radu Mutihac1,3, Elena Ackley4, Jochen Rick5, Akio Yoshimoto6, Maxim Zaitsev5, and Oliver Speck7
1Neurology, University of New Mexico, Albuquerque, NM, United States, 2Electrical and Computer Engineering, University of New Mexico, Albuquerque, NM, United States,3Electricity & Biophysics, Faculty of Physics, University of Bucharest, Bucharest, Romania, 4Neurology, University of New Mexico, United States, 5Radiology - Medical Physics, University Medical Center Freiburg, Freiburg, Germany, 6Polytechnic Institute of New York University, New York, United States, 7Biomedical Magnetic Resonance, Otto-von-Guericke-University Magdeburg, Magdeburg, Germany

In this study we developed whole brain EVI on a clinical 3 T scanner using multiple-slab excitations and 3D spatial encoding within each slab to reduce geometrical distortion. Maximum t-score, extent of activation and percent BOLD signal change were significantly higher with EVI at TR: 280 ms compared to EPI, both using TurboFIRE software (p<0.05) in real-time and retrospectively, and using SPM8 (p<0.05). Excellent activation and strong signal changes in BA17/18 were measured at TR: 135 ms. Temporal pre-whitening procedures are required to remove temporal data correlation by autoregressive models, to improve statistical modeling.

Wednesday May 11th
  13:30 - 15:30 Computer 53

13:30 3584.   Mapping of midbrain nuclei connectivity networks using Time-domain Phase-REgularized Parallel (T-PREP) reconstruction of high-resolution fMRI 
Nan-kuei Chen1
1Brain Imaging and Analysis Center, Duke University Medical Center, Durham, NC, United States

Mapping of intrinsic connectivity networks is widely applied to imaging different patient populations. However, conventional fMRI protocols used in ICN mapping only provide data at a relatively low spatial-resolution. Although low-resolution ICN mapping may be sufficient to detect disruptions of major large-scale networks, it cannot measure the networks comprising nodes that are smaller in size, such as the midbrain nuclei. To address this limitation, here we report a novel time-domain phase-regularized parallel (T-PREP) reconstruction algorithm, retaining the fidelity of BOLD time-course profiles derived from under-sampled fMRI data, so that midbrain connectivity ICNs can be reliably measured from high-resolution resting-state fMRI.

14:00 3585.   High Resolution Functional Connectivity Mapping at 7T 
Christina Triantafyllou1,2, Boris Keil2, Sheeba Arnold1, Susan Whitfield-Gabrieli1, and Lawrence L Wald2,3
1A.A. Martinos Imaging Center, McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, MA, United States, 2A.A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States, 3Harvard-MIT Division of Health Sciences and Technology, Cambridge, MA, United States

Increases in BOLD CNR by improvements in array coils and higher field strength potentially allow sub-structure in connectivity networks to be examined at higher spatial resolutions. In this work we assess the potential of sub-millimeter functional connectivity mapping at 7T by characterizing the effect of voxel size across a range of isotropic resolutions. The results indicate that robust connectivity can be detected at all resolutions including the isotropic 500µm data at acquisition time of 5min, on a single subject level; suggesting that fcMRI studies of small-scale networks are only limited by the biological point spread function of the underlying BOLD mechanism.

14:30 3586.   High Resolution Functional Mapping of Primary Motor Cortex and Primary Somatosensory Cortex in Humans at 7 T 
Robert Trampel1, Andreas Schäfer1, Robin Martin Heidemann1, Dimo Ivanov1, Gabriele Lohmann1, Stefan Geyer1, and Robert Turner1
1Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

The functional properties of the sensorimotor system have been intensively studied by functional MRI over the last decades. Typical fMRI resolution at field strengths at 3 Tesla and below enables a rough somatotopic mapping, but may not unambiguously differentiate between primary motor cortex (M1) and primary somatosensory cortex (S1). Here, we performed isotropic sub-millimeter fMRI at 7 Tesla, using a paradigm comprising finger tapping, finger movement without touching the finger tips and imagined finger movement. Spatial differences in activation in S1 and M1 could be distinguished, and also signal time courses dependent on the laminar position across M1 cortex were obtained.

15:00 3587.   High Resolution BOLD fMRI of the Human Retina of Oxygen and Carbogen Inhalation 
Yi Zhang1,2, Qi Peng1,2, Oscar San Emeterio Nateras2, and Timothy Q. Duong1,2
1Research Imaging Institute, University of Texas Health Science Center at San Antonio, San Antonio, TX, United States, 2Radiology, University of Texas Health Science Center at San Antonio, San Antonio, TX, United States

BOLD fMRI of the human retina is challenging mainly due to thin retina thickness, large magnetic susceptibility variation in the orbital region and potential eye movements. This study demonstrated an innovative fMRI application to image normal human retina associated with oxygen and carbogen challenges. fMRI utilized an inversion-recovery of Echo Shifting using a Train of Observations (IR-PRESTO) acquisition to suppress vitreous signal and to achieve high spatiotemporal resolution free of susceptibility artifacts. This approach has the potential to open up new avenues for retinal research and complement existing optical imaging techniques.

Thursday May 12th
  13:30 - 15:30 Computer 53

13:30 3588.   Comparison of Acceleration Techniques Applied to Multi-shot 3D EPI for fMRI Studies 
Onur Afacan1,2, Dana H. Brooks1, W. Scott Hoge2, and Istvan A. Morocz2
1ECE Dept., Northeastern University, Boston, MA, United States, 2Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, United States

In this work we evaluated the temporal performance of three acceleration techniques applied to multi-shot 3D EPI for fMRI. In particular we implemented a multi-shot sequence accelerated by GRAPPA, UNFOLD, and a combination of UNFOLD and GRAPPA, all along the slice encoding direction. We show results from a phantom experiment and an in-vivo fMRI study that compares the accelerated sequences to a full 3DEPI acquisition in terms of temporal SNR, temporal artifacts, and functional sensitivity.

14:00 3589.   Functional MRI using Super-Resolved Spatiotemporally-Encoded Imaging Techniques 
Noam Ben-Eliezer1, Ute Goerke2, and Lucio Frydman1
1Chemical Physics, Weizmann Institute of Science, Rehovot, Israel, 2Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, United States

Recently, new single-shot imaging sequences such as RASER and Hybrid SPEN have been proposed, relying on spatiotemporal-encoding (SPEN) and offering significantly higher immunity to B0-inhomogeneities and chemical-shift offsets than conventional EPI. fMRI experiments in the orbitofrontal cortex have been successfully performed using RASER – a highly challenging task using EPI due to strong distortions near the air-filled sinuses. Despite these advantages systematic analyses have shown that SPEN sensitivity is lower and its RF power deposition higher, than in EPI. We hereby explore a new reconstruction algorithm, capable of alleviating these shortcomings and analyze its performance in the context of fMRI applications.

14:30 3590.   Multi-Banded T2-Weighted fMRI with a z-Encoding RF Coil Array for Whole Brain Coverage at 7 T 
johannes Ritter1, Pierre-Francois Van de Moortele1, Steen Moeller1, Eddie Auerbach1, Kamil Ugurbil1, and Gregor Adriany1
1CMRR, University of Minnesota, Minneapolis, MN, United States

Due to significant B1 inhomogeneities at high fields it is challenging to achieve effective whole brain fMRI. However, it has recently been demonstrated at 7 T that B1-shim homogeneity, T2-weighted contrast and hence activation in different target volumes of the human brain can be improved significantly with a multi-region B1-shim Here we built upon this and present a new integrated approach, consisting of a novel multi banded sequence, a 3D B1 shim T/R coil array and large volume B1-shim. Only the combination of these components allow to effectively address SAR and B1-inhomogeneity challenges and enable whole head T2 weighted fMRI.

15:00 3591.   Highly Sparse Spiral fMRI Reconstructed with Compressed Sensing: Trajectory Optimization for BOLD Contrast 
Daniel Holland1, Careesa Liu2, Chris V. Bowen2,3, Andy Sederman1, Lynn Gladden1, and Steven D. Beyea2,3
1Department of Chemical Engineering and Biotechnology, University of Cambridge, Cambridge, United Kingdom, 2Institute for Biodiagnostics (Atlantic), National Research Council Canada, Halifax, Nova Scotia, Canada, 3Departments of Physics & Radiology, Dalhousie University, Halifax, Nova Scotia, Canada

The use of compressed sensing (CS) introduces the possibility of increased k-space sparsity without the increase in artifact, which widens the range of variable density (VD) spiral trajectories for improving fMRI sensitivity. We have explored a variety of heavily undersampled VD spiral trajectories in order to optimize the fMRI. By combining CS with VD spiral, we demonstrated that optimal fMRI acquisitions are achieved using significantly sparser data than was previously reported for non-CS reconstructed VD spiral data. A representative CS-VD acquisition with 35% undersampling exhibited significantly improved fMRI sensitivity (e.g. 60% more active voxels and 13% increase in CNR).

Electronic Posters : Functional MRI
Click on to view the abstract pdf and click on to view the video presentation.
fMRI Signal Contributions

Monday May 9th
Exhibition Hall  14:00 - 16:00 Computer 54

14:00 3592.   Dynamical statistical modeling of physiological noise for fast BOLD fMRI 
Simo Sarkka1, Aapo Nummenmaa1,2, Arno Solin1, Aki Vehtari1, Thomas Witzel3, Toni Auranen4, Simo Vanni4, Matti S. Hamalainen2, and Fa-Hsuan Lin1,5
1Department of Biomedical Engineering and Computational Science, Aalto University, Espoo, Finland, 2Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, Massachusetts, United States, 3Harvard-MIT Division of Health Sciences and Technology, Harvard University, Cambridge, Massachusetts, United States, 4Advanced Magnetic Imaging Centre, Low Temperature Laboratory, Aalto University, Espoo, Finland, 5Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan

In this work, we propose a statistical model based method for removal and analysis of physiological noise in fast BOLD fMRI acquisition methods. The proposed stochastic state space model allows for accurate dynamic tracking of time-varying physiological signal frequencies and the estimation method is based on the Interacting Multiple Models (IMM) Kalman filter (KF) algorithm, which is widely used in real time target tracking applications. The method forms statistically the best possible separation of the spatiotemporal BOLD and physiological signals into separate components, which allows for further processing of the de-noised BOLD signal or analysis of the spatial characteristics of the physiological signals. The proposed method was applied to a three-slice EPI data and the results indicate that the method is able to accurately separate the cardiac and respiration signals from the BOLD signal.

14:30 3593.   Improved Model for Physiological Fluctuations in fMRI 
Christina Triantafyllou1,2, Boris Keil2, Jonathan R Polimeni2, and Lawrence L Wald2,3
1MIT, Athinoula A. Martinos Imaging Center, McGovern Institute for Brain Research, Cambridge, MA, United States, 2A.A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States, 3Harvard-MIT Division of Health Sciences and Technology, Cambridge, MA, United States

Sensitivity in BOLD fMRI is characterized by the time-series SNR (tSNR), which contains fluctuations from thermal (σ0) and physiological (σp) noise sources. The relationship between tSNR and the image SNR (SNR0)often fit to a model where σp is proportional to signal. Our findings demonstrate that the relationship between tSNR and SNR0 is well described by this for small count of channels, however for higher field strengths and multiple channel arrays, a modified model with an additional constant term in σp(which does not scale with signal strength) produces a significantly better fit to both SE and GrE data.

15:00 3594.   Sources of signal fluctuations in single-shot 2D EPI and segmented 3D EVI acquisitions for fMRI at 7T 
João P. F. Jorge1,2, Patrícia Figueiredo1,2, Wietske van der Zwaag3,4, Mayur Narsude3, and José P. Marques3,4
1Instituto Superior Técnico, Lisbon, Portugal, 2Institute for Systems and Robotics, Lisbon, Portugal, 3Laboratory for Functional and Metabolic Imaging, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 4Department of Radiology, Université de Lausanne, Lausanne, Switzerland

Recently, there has been an increased interest in segmented EVI acquisitions in fMRI studies due to their improved image SNR compared to traditional EPI. However, the segmented EVI is more susceptible to noise arising from temporally-correlated, signal-dependent noise sources. In this work, the ability of regressors based on simultaneously acquired physiological signals to explain the signal variance of fMRI data acquired with these two sequences was evaluated. The use of a physiological signal-based correction method improved the temporal SNR and BOLD sensitivity of the segmented EVI data to levels superior to those of conventional EPI.

15:30 3595.   The Effect of Cardiac Synchronization on the Temporal Characteristics of 3D SSFP and 3D SGPR 
Rob Hendrikus Tijssen1, and Karla Loreen Miller1
1FMRIB Centre, Oxford University, Oxford, Oxon, United Kingdom

Recently, several groups have re-visited the potential of low-distortion sequences like 3D SSFP and SPGR sequences for high-field FMRI. This is particularly relevant for sub-cortical structures, which require high resolution to resolve small nuclei. 3D acquisitions however are susceptible to flow-induced instabilities and show reduced temporal SNR in inferior regions of the brain. Our results suggest that cardiac synchronization could significantly improve the tSNR of both SSFP and SPGR in these regions.

Tuesday May 10th
  13:30 - 15:30 Computer 54

13:30 3596.   Feasibility of quantitative measurements for regional cerebral metabolic rate of oxygen (CMRO2) during functional change with visual stimulus using MRI 
Audrey Peiwen Fan1, Jonathan R Polimeni2, Bruce R Rosen2,3, and Elfar Adalsteinsson1,3
1Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States, 2Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States, 3Health Sciences and Technology, Harvard-MIT, Cambridge, MA, United States

Phase-based regional oxygen metabolism (PROM) is an MRI method to quantify regional cerebral metabolic rate of oxygen (CMRO2). The technique combines measurements of oxygen saturation (Yv) from MR susceptometry in cerebral veins and local cerebral blood flow (CBF) from arterial spin labeling to estimate regional CMRO2. Using this method at rest and during visual stimulation, baseline CMRO2 in the visual cortex was estimated 158±23 μmol/100g/min and increased by 13.1% to 178±18 μmol/100g/min with functional activation (p=0.02).

14:00 3597.   A new approach for venous blood oxygenation and calibrated BOLD using hyperoxia 
Ian Driver1, Emma Hall1, Susan Pritchard1, Susan Francis1, and Penny Gowland1
1Sir Peter Mansfield Magnetic Resonance Centre, University of Nottingham, Nottingham, United Kingdom

Current methods for BOLD calibration assume Grubb’s coupling constant between cerebral blood flow and blood volume, which is poorly characterised, measured for total blood volume and likely to vary between brain regions and subjects. Here we present a new method based on hyperoxia for the measurement of venous blood volume (vCBV), blood oxygenation and oxygen metabolism (CMRO2) changes on neuronal activation without the need to assume any flow/volume coupling relationship. Data is presented for a motor task where changes were of 32±2% for vCBV, 38±3% for blood oxygenation, and 30±6% for CMRO2 which is in close agreement to literature.

14:30 3598.   Breathing gas calibration for MR CMRO2 measurements: comparative effects on functional brain networks 
Dimo Ivanov1, Gabriele Lohmann1, Stefan Kabisch1,2, Ilona Henseler1, Haiko Schloegl1,2, Wolfgang Heinke3, Chloe Hutton4, and Robert Turner1
1Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2Department of Medicine, University Hospital Leipzig, Leipzig, Germany, 3Department of Anestesiology and Intensive Care Therapy, University Hospital Leipzig, Leipzig, Germany, 4Wellcome Trust Centre for Neuroimaging, University College London, London, United Kingdom

Regional changes in the cerebral metabolic rate of oxygen (CMRO2) associated with neural activation can be estimated with MRI using the deoxyhaemoglobin dilution model, whose application requires the calibration of resting-state BOLD signal with a gaseous challenge. In this study we used eigenvector centrality mapping (ECM) to assess whether the brain network connectivity is influenced by commonly used calibration gases. The results suggest that these functional brain networks are only slightly modulated, confirming the applicability of the CMRO2 mapping methodology across most of the brain.

15:00 3599.   High resolution CMRO2 in visual cortex of macaca mulatta 
Yvette Bohraus1, Nikos K Logothetis1,2, and Jozien Goense1
1Dept. Physiology of Cognitive Processes, Max Planck Institute for Biological Cybernetics, Tübingen, Germany, 2University of Manchester, Manchester, United Kingdom

Current fMRI-methods are based on changes in cerebral blood flow and/or oxygenation. Since these methods measure hemodynamic signals, changes in BOLD or CBF may not always accurately reflect changes in the actual energy use of the brain. We determined CMRO2 in macaques during visual stimulation at high resolution. The CMRO2values and the ratio of fractional CBF and CMRO2 changes were consistent with those reported in the literature.

Wednesday May 11th
  13:30 - 15:30 Computer 54

13:30 3600.   Differences in neurovascular coupling in areas with positive and negative BOLD signal 
Jozien Goense1, and Nikos K Logothetis1,2
1Dept. of Physiology of Cognitive Processes, Max-Planck Institute for Biological Cybernetics, Tuebingen, Germany, 2Division of Imaging Science and Biomedical Engineering, University of Manchester, United Kingdom

While the dynamics and mechanism of the positive BOLD response have been well studied, much less is known about the mechanism of the negative BOLD response. We studied the properties of the negative BOLD response by comparing the BOLD-, functional CBV- and CBF-responses in regions exhibiting positive and negative BOLD in primary visual cortex (V1) of anesthetized monkeys. Both positive and negative BOLD signals were accompanied by increases in CBV, although CBF was decreased in the unstimulated region.

14:00 3601.   Neurovascular coupling and uncoupling in negative fMRI response 
Chiao-Chi V Chen1,2, Yen-Yu I. Shih3, Yi-Hua Hsu1,2, Bai-Chuang Shyu1, and Chen Chang1,2
1Institute of Biomedical Sciences, Academic Sinica, Taipei, Taiwan, 2Functional and Micro-magnetic Resonance Imaging Center, Academic Sinica, Taipei, Taiwan,3Research Imaging Institute, University of Texas Health Science Center, San Antonio, TX, United States

Negative fMRI response is less well understood in fMRI studies. Elucidating its coupling and uncoupling to brain activity is important for referring the negative fMRI signals to a correct source. Afferent activity and local processing is recognized as the activity source of fMRI signals, but their role in negative fMRI signals with respect to neurovascular coupling and uncoupling remains unclear. The present study characterized a negative fMRI response occurring in the striatum, triggered by nociceptive electrical stimulation. We found that the negative fMRI response was uncoupled to the local neuronal activity yet coupled to afferents via neurotransmission.

14:30 3602.   The source of the early-negative blood oxygenation signal 
Hiro Fukuda1, Alberto Vazquez1, and Seong-Gi Kim1
1Radiology, University of Pittsburgh, Pittsburgh, PA, United States

Relationships between changes in oxygen saturation level in emerging veins and the presence of early dip in blood oxygenation signal was examined. When oxygen consumption increases in active parenchyma, it causes oxygen saturation levels to decrease and increases deoxyhemoglobin in blood. These changes are expected to be observed in emerging pial veins because of blood draining. We found that an increase in CBV contributes dominantly to the early dip when CBF response is fast while a decrease in blood oxygen saturation level due to an increase in tissue oxygen consumption also contributes to the dip when CBF response is slow.

15:00 3603.   The BOLD fMRI post-stimulation undershoot in human primary motor cortex is not caused by elevated CBV 
Peter Dechent1, Gunther Helms1, Dietmar Merboldt2, and Jens Frahm2
1MR-Research in Neurology and Psychiatry, Universitymedicine Göttingen, Göttingen, Germany, 2Biomedizinische NMR Forschungs GmbH am MPI für biophysikalische Chemie, Göttingen, Germany

The post-stimulation undershoot in BOLD fMRI may be due to a delayed recovery of elevated CBV or CMRO2. To exclude the possibility that contradicting results in humans and animals originate from studies of different cortical systems, we performed contrast-enhanced T1-weighted fMRI in humans to evaluate CBV during a motor task. The results confirm a CBV increase during finger movements and reveal CBV baseline levels in the post-stimulation phase. This finding is in line with previous results from human visual cortex. It renders discrepant findings from human and animal studies unlikely to be caused by the specific cortical system investigated.

Thursday May 12th
  13:30 - 15:30 Computer 54

13:30 3604.   Vascular-space-occupancy (VASO) MRI in human brain at 7T 
Jun Hua1,2, Craig K Jones1,2, and Peter C.M. van Zijl1,2
1Neurosection, Div. of MRI Research, Dept. of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, 2F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

VASO-MRI exploits the difference between blood and tissue T1 to null blood signal and measure CBV changes using residual tissue signal. At 7T, VASO-MRI faces two challenges: converging blood and tissue T1, and a large positive BOLD effect counteracting the negative VASO change. Off-resonance (>20ppm) MT effects are large in tissue but minimal in blood, which was used to magnify tissue signal while nulling blood, and boost SNR by 112+/-33% at 7T. A 3D turbo-field-echo (TFE) sequence with short TE (1.77ms) is employed to minimize the BOLD effect, and provide whole brain coverage (2mm isotropic) for VASO fMRI at 7T.

14:00 3605.   Non-invasive Quantification of Absolute Cerebral Blood Volume 
Pelin Aksit Ciris1, Maolin Qiu1, and Robert Todd Constable1
1Yale University, New Haven, CT, United States

Non-invasive absolute CBV quantification could foster improved understanding of fMRI signal mechanisms, and potentially find clinical utility in evaluating vascular state, damage, and monitoring treatments. A non-invasive method based on a biophysical model and multi-slice acquisition with varying contrast weightings, with efficient multi-slice imaging using a rotating slice acquisition, is introduced. The method was evaluated by fitting to a slight change in data and model parameters with activation. Consistent and physiologically expected CBV estimates were obtained in mL of blood / 100mL of parenchyma in normal human volunteers. Results indicate feasibility of non-invasive quantification applicable to the whole human brain.

14:30 3606.   Cerebral arterial and venous blood volume changes during the post-stimulus BOLD undershoot period 
Tae Kim1, and Soeng-Gi Kim1
1Neuroimaging Laboratory, Radiology, University of Pittsburgh, Pittsburgh, PA, United States

Sources of post-stimulus BOLD undershoots were investigated by measuring cerebral arterial and venous blood volume changes in isoflurane-anesthetized cats. After the cessation of 40-s visual stimulation, arterial CBV showed post-stimulus undershoots, similar to BOLD, while venous CBV returned to baseline slowly. Both CBF undershoot (due to arterial CBV undershoot) and slow venous CBV return contribute to the post-stimulus BOLD undershoot

15:00 3607.   CBV Measurements-Gd_DTPA vs. VASO- and Their Relationship with CBF in Activated Human Visual Cortex 
Ai-Ling Lin1, Hanzhang Lu2, Peter T Fox1, and Timothy Q Duong1
1Research Imaging Institute, University of Texas Health Science Center, San Antonio, TX, United States, 2Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, United States

The purpose of the study was to compare the VAcular Space Occupancy (VASO) techniques and a contrast agent-based (Gd-DTPA) method in determining CBV changes during multi-frequency visual stimulation (4 and 8 Hz). Specifically, we aimed to indentify the impact of repetition time (TR) on CBV changes determination using VASO. With additional measurements in cerebral blood flow (CBF), the flow-volume coupling relationship (£\ value) and cerebral metabolic rate of oxygen (CMRO2) were further determined. The results showed that with short TR (2s), the CBV changes determined by VASO were overestimated and thus caused the underestimated of CMRO2 changes with a fMRI biophysical model. With Long TR (6s), CBV changes determined by VASO were consistent with those obtained with the Gd-DTPA method. In addition, the results showed that tasked-induced CBF-CBV coupling was stimulus frequency-dependent, i.e., £\ = 0.35-0.38 at 4 Hz and £\ = 0.51-0.53 at 8 Hz.

Electronic Posters : Functional MRI
Click on to view the abstract pdf and click on to view the video presentation.
Improving fMRI Acquisition

Monday May 9th
Exhibition Hall  14:00 - 16:00 Computer 55

14:00 3608.   Multi-echo EPI with parallel transmission z-shimming for increased sensitivity in BOLD fMRI 
Benedikt A Poser1, Cungeng Yang1, Weiran Deng1, Vijayanand Alagappan2,3, Lawrence L Wald2,4, and V Andrew Stenger1
1University of Hawaii, John A. Burns School of Medicine, Honolulu, Hawaii, United States, 2A.A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Boston, Massachusetts, United States, 3Department of Biomedical Engineering, Tufts University, Medford, Massachusetts, United States,4Harvard-MIT Division of Health Sciences and Technology, Cambridge, Massachusetts, United States

Signal loss remains the most obdurate problem in gradient-echo BOLD fMRI. Here we address the issue by merging two highly promising techniques, parallel-transmission z-shimming and multi-echo EPI acquisition. ME-EPI has previously been shown to increase BOLD sensitivity over the entire brain; however it does not actively compensate for through-plane dephasing. This makes multi-channel transmission for the application of local corrective z-shims is highly complementary. The combined method is found to be more effective in reducing susceptibility artifacts than each individual approach; In vivo experiments reveal improved sensitivity in the frontal lobe, demonstrating the method’s potential for fMRI application.

14:30 3609.   fMRI with concurrent magnetic field monitoring 
Christoph Barmet1, Bertram Jakob Wilm1, Lars Kasper1, Christian C Ruff2, Klaas Enno Stephan2,3, and Klaas Paul Pruessmann1
1Institute for Biomedical Engineering, University and ETH Zürich, Zurich, Zurich, Switzerland, 2Laboratory for Social and Neural Systems Research, University of Zurich, Zurich, Zurich, Switzerland, 3Wellcome Trust Centre for Neuroimaging, University College of London, London, London, United Kingdom

In fMRI it is attractive to precisely know the spatio-temporal magnetic field evolution in the imaging volume during the scan. On the one hand, such knowledge can be taken into account for image reconstruction; on the other hand, it could be a useful means of subject surveillance and of monitoring the course of a study for retrospective analysis. Recent advances in field monitoring hardware and methods actually promise to offer such capability. The present work aims to explore this proposition, using recent concurrent magnetic monitoring technology.

15:00 3610.   Slice-Specific Gradient Compensation of Magnetic Field Inhomogeneities to Improve T2*-Weighted Imaging of the Human Spinal Cord 
Jürgen Finsterbusch1,2, and Falk Eippert1,2
1Department of Systems Neuroscience, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 2Neuroimage Nord, University Medical Centers Hamburg-Kiel-Lübeck, Hamburg-Kiel-Lübeck, Germany

BOLD-based spinal cord imaging suffers from the magnetic field inhomogeneities induced by the different susceptibilities of the vertebrae and the surrounding tissue. In T2*-weighted acquisitions, these inhomogeneities cause a signal dephasing that varies between different transverse sections. In this study, it is shown that applying slice-specific gradient compensation moments reduce the signal variations and drop outs which could help to improve the detectability and reliability of BOLD-based functional neuroimaging of the spinal cord.

15:30 3611.   Parallel Imaging with Asymmetric Acceleration (ASYA) to Reduce Susceptibility Artifacts in BOLD fMRI 
Kwan-Jin Jung1, and Tiejun Zhao2
1Scientific Imaging Brain Research (SIBR), Department of Psychology, Carnegie Mellon University, Pittsburgh, PA, United States, 2MR R&D Collaborations, Siemens Medical Solutionsn USA, Siemens Healthcare, Pittsburgh, PA, United States

Parallel imaging with acceleration was noted to pronounce the ripple artifact near the susceptibility-affected region in the gradient echo EPI for BOLD fMRI. Using the extended EPI sequence, which collected extended readouts outside the regular data acquisition window, the pronounced ripple artifact was analyzed and found to be caused by an increased echo shift in the pre-TE period in accelerated parallel imaging. This was also confirmed by theoretical derivation of the echo shift due to the magnetic field susceptibility. A new EPI sequence was developed to reduce the ripple artifact as well as to restore the signal level to the level of un-accelerated parallel imaging by applying the acceleration asymmetrically only to the post-TE period. The un-accelerated portion in the pre-TE period utilized the delay for the optimum BOLD sensitivity at 3T, maintaining the same slice coverage as the conventional acceleration in both pre- and post-TE periods.

Tuesday May 10th
  13:30 - 15:30 Computer 55

13:30 3612.   Comparison of fMRI with Accelerated Variable Density Spiral and EPI 
Wei Lin1, Enrico Simonotto1, Feng Huang1, Charles Saylor1, George R Duensing1, and Arne Reykowski1
1Invivo Corporation, Philips Healthcare, Gainesville, FL, United States

A recently proposed rapid k-space-based parallel imaging method for variable-density (VD) spiral, Generalized GRAPPA for wider spiral bands (GROWL) is applied to fMRI. When compared with standard EPI images acquired with identical scan parameters, VD spiral-in images shows significantly less signal dropout in the frontal orbital region. A second potential advantage is to achieve higher temporal/spatial resolution and larger volume coverage with highly undersampled VD spiral trajectories.

14:00 3613.   Effects of a slice-dependent template-based gradient compensation method on the BOLD sensitivity 
Jochen Rick1, Oliver Speck2, Jürgen Hennig1, and Maxim Zaitsev1
1Dept. of Radiology, Medical Physics, University Medical Center Freiburg, Freiburg, Germany, 2Biomedical Magnetic Resonance, Otto-von-Guericke University, Magdeburg, Germany

Anatomy-related susceptibility gradients in the human head lead to artefacts in echo planar imaging (EPI). The use of a slice-dependent template-based gradient compensation method improves BOLD sensitivity (BS) in areas affected by strong susceptibility-induced field gradients. Here, an evaluation of BS changes is performed for a compensated measurement in relation to an uncompensated measurement. It is shown that the ratio of signal loss to signal gain is well balanced, but the method allows the sensitivity to be optimized in target areas. These results are affirmed in a functional experiment and shows that the method can be used for event-related functional experiments.

14:30 3614.   Optimizing EPI for Functional MRI using Multi-directional Shimming in a Single Shot Acquisition 
Jaemin Shin1, Sinyeob Ahn1, and Xiaoping P Hu1
1Biomedical Engineering, Georgia Tech/Emory University, Atlanta, GA, United States

Susceptibility-induced field inhomogeneity is a major cause of signal loss in functional MRI. Z-shimming has been widely used in the slice selection (SS) direction. Shimming technique has been extended to the phase encoding (PE) direction. However, the gradient of a fixed amplitude in PE direction may not achieve the optimal signal recovery of whole brain. In this work, we describe a single shot EPI sequence that can optimally compensate for the inhomogeneity in both SS and PE directions. Signal recovery with only Z- shimming was only 38% of the optimal shimming in the two directions. In conclusion, multi directional shimming is better than Z-shimming alone for signal recovery in fMRI.

15:00 3615.   Dependence of acquisition trajectory on BOLD sensitivity changes due to magnetic susceptibility differences in the brain 
Thomas Le Paine1,2, and Brad P Sutton1,2
1Bioengineering, University of Illinois Urbana-Champaign, Urbana, IL, United States, 2Beckman Institute, University of Illinois Urbana-Champaign, Urbana, IL, United States

Magnetic susceptibility differences between air and tissues lead to magnetic field inhomogeneity in the brain that can cause artifacts during functional neuroimaging using the BOLD mechanism. When a gradient echo acquisition is used, gradients in the magnetic field inhomogeneity can cause effective echo time shifts that result in spatial variations in the sensitivity of BOLD. With a measured field map, these variations can be calculated and used to calibrate BOLD maps. We show BOLD maps from a breath hold task for four different acquisition trajectories on the same subject and compare them to the expected BOLD sensitivity maps.

Wednesday May 11th
  13:30 - 15:30 Computer 55

13:30 3616.   Implementation of navigator phase correction in multi-echo non-balanced SSFP at 7T 
Pål Erik Goa1,2, Benedikt Andreas Poser2,3, and Markus Barth2,3
1Department of Medical Imaging, St.Olav University Hospital, Trondheim, Norway, 2Erwin L. Hahn Institute for Magnetic Resonance Imaging, University Duisburg-Essen, Essen, Germany, 3Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands

Navigator phase correction is used in GRE sequences to remove respiration induced global phase variations. Non-balanced Steady-state free precession might also benefit from phase correction. Here we investigate the phase variation in S1 and S2 signals individually and evaluate two different navigator methods. Results show that the S2 phase is different from S1 phase and cannot be corrected using a single S1 navigator. An interleaved scheme where navigator echoes for both S1 and S2 are sampled every m TR will successfully correct both signals individually. BOLD-fMRI experiments on 7T shows significant increase in number of activated voxels using the interleaved scheme.

14:00 3617.   Impact of TE on Short-TR Pass-band b-SSFP BOLD Contrast at 3T 
Qi Peng1,2, Yi Zhang1,2, Oscar San Emeterio Nateras1,2, and Timothy Q Duong1,2
1Radiology, UT Health Science Center at San Antonio, San Antonio, TX, United States, 2Research Imaging Institute, UT Health Science Center at San Antonio, San Antonio, TX, United States

Pass-band balanced steady-state free precession (pbSSFP) fMRI offers high BOLD sensitivity and is less susceptible to physiological noise compared to traditional gradient echo planar imaging sequences. However, the contrast mechanism is still under investigation, and the impact of TE on commonly used short-TR (<10ms) pbSSFP on BOLD sensitivity has not been experimentally investigated at 3.0T. In this study, we demonstrated that a TE/TR ratio approaching unity had much higher BOLD compared to the same pbSSFP with TE/TR¡Ü0.5. Therefore, k-space trajectories leading to larger TE have advantages in short-TR pbSSFP fMRI studies to obtain high BOLD sensitivity.

14:30 3618.   A Real-Time Cardiac Synchronization Method for Reducing Flow-Induced Instabilities in SSFP FMRI of the Brainstem 
Rob Hendrikus Tijssen1, Thomas William Okell1, and Karla Loreen Miller1
1FMRIB Centre, Oxford University, Oxford, Oxon, United Kingdom

Low distortion and high signal-to-noise ration make Steady-state free precession (SSFP) FMRI an attractive pulse sequence for high resolution brainstem FMRI. Unfortunately the multi-shot character of the readout makes the sequence highly susceptible to flow-induced instabilities that correlate with the cardiac cycle. We present a readout method that achieves real-time cardiac synchronization without varying frame rate, based on GRAPPA and partial-Fourier reconstruction. This method significantly improves temporal stability in the brainstem.

15:00 3619.   Balanced Steady State Free Precession fMRI Using Intravascular Susceptibility Contrast Agent 
Iris Yuwen Zhou1,2, Matthew M. Cheung1,2, Kevin C. Chan1,2, Condon Lau1,2, and Ed X Wu1,2
1Laboratory of Biomedical and Signal Processing, The University of Hong Kong, Hong Kong SAR, China, People's Republic of, 2Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong SAR, China, People's Republic of

The major challenges in EPI-fMRI are the image distortions and constraint in spatial resolution, thus limiting accurate and high resolution mapping of brain functions especially in small animals at high fields. To overcome these limitations and also to achieve high CNR and robust signal changes, we investigated the feasibility of fMRI using bSSFP together with intravascular susceptibility contrast agent MION in rodent brains. We also compared with the conventional SE-EPI and GE-EPI based fMRI. The results demonstrated that bSSFP in combination with intravascular contrast agent provides a robust CBV based fMRI approach. Such brain functional mapping is distortion free and can be of high resolution, and it is particularly suited for high field fMRI study of animal models.

Thursday May 12th
  13:30 - 15:30 Computer 55

13:30 3620.   Dynamic changes in the tissue microenvironment induced by hypercapnia and hyperoxia: a T1rho dispersion study at 9.4 T 
Tao Jin1, and Seong-Gi Kim1
1Neuroimaging laboratory, Department of Radiology, University of Pittsburgh, Pittsburgh, PA, United States

The spin-lattice relaxation time in the rotating frame (T1rho) has been applied in many pathological studies, including cartilage degradation, cerebral ischemia, and neurodegeneration diseases. Recently, it has also been reported that the T1rho contrast can detect dynamic changes in the tissue microenvironment induced by hypercapnia, hyperoxia challenges, or neuronal activation. T1rho is most sensitive to molecular fluctuations with correlation times close to the inverse of Rabi frequency of the applied spin-locking (SL) pulse. Thus, the T1rho relaxation time, measured with different SL frequencies, which is termed T1rho dispersion, provides valuable information about the underlying physiological mechanisms. Previous studies have demonstrated that the chemical exchange between labile protons of proteins and the bulk water may be an important contributor to T1rho dispersion in biological tissues in the low-frequency range of below several kHz. In order to gain more insight about the underlying mechanisms of dynamic T1rho changes, we investigated the T1rho response during hypercapnia and hyperoxia for two different SL frequencies.

14:00 3621.   Direction-dependent diffusion fMRI signals during hypercapnia and hyperoxia 
Tao Jin1, and Seong-Gi Kim1
1Neuroimaging laboratory, Department of Radiology, University of Pittsburgh, Pittsburgh, PA, United States

Recently, diffusion-weighted fMRI (DfMRI) signals of brain water were reported to increase with the degree of diffusion sensitization during human visual stimulation, indicating a decrease of the apparent water diffusivity; however, the interpretation of the signal origin was controversial. Le Bihan et al. attributed the activation-induced change of the apparent water mobility to a functional expansion of neuronal cell membrane. Miller et al. found similar DfMRI signal change during a hypercapnia challenge, which was also dependent on the direction of diffusion-sensitizing gradients, thus suggesting that the DfMRI signal change might be due to residue intravascular signals. In this preliminary study, we measured the direction-dependent change of DfMRI signal i) during intravascular susceptibility change without changes in vascular physiology by the intravascular injection of a small amount of iron oxide nanoparticle, and ii) during global hypercapnia and hyperoxia stimulations in anesthetized rats after the suppression of the intravascular signals.

14:30 3622.   Assessment of hemodynamic effects in functional diffusion-weighted MRI 
Umesh Suryanarayana Rudrapatna1, Maurits P A van Meer1, Annette Van der Toorn1, and Rick M Dijkhuizen1
1Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands

As diffusion-based functional imaging schemes are being introduced as alternatives to BOLD fMRI, it is imperative to assess their resilience to hemodynamic effects. We report our findings from a cerebrovascular challenge study in rats, where DTI was applied with relatively high diffusion-weighting. Data reveal that hemodynamic changes significantly influence signals in diffusion-weighted data even at high b-values, in gray as well as white matter. The changes were similar in different directions, suggesting a common vascular origin. This highlights the difficulty of discriminating hemodynamic and cellular responses with diffusion-weighted fMRI during stimulus-induced brain activation.

15:00 3623.   Magnetization transfer fMRI in humans at 7T 
Sung-Yeon Park1, Dae-Hoon Kang1, Se-Hong Oh1, Myoung-Kyun Woo1, Joshua H. Park1, Jun-Young Chung1, Young-Bo Kim1, Zang-Hee Cho1, and Seong-Gi Kim2
1Neuroscience Research Institute, Gachon University of Medicine and Science, Incheon, Korea, Republic of, 2Radiology, University of Pittsburgh, United States

Recently a magnetization transfer (MT)-varied fMRI technique was proposed to simultaneously measure stimulus-induced arterial CBV (CBVa) change and BOLD response. In previous 9.4-T animal MT fMRI contradicts to previous 1.5-T human MT studies, in which MT reduced the percentage signal change (i.e., increased MTR). This discrepancy can be due to different magnetic field (9.4 T vs. 1.5 T), different spatial resolution (0.3 mm vs. 3.7 mm), different MT pulse scheme (long CW vs. one short pulse), or different species. Thus, in order to investigate the source of discrepancy, we performed MT fMRI with high spatial resolution in humans at 7 T.