Hye-Young Heo^1,2, Casey P. Johnson¹, Daniel R. Thedens¹, John A. Wemmie^3,4, and Vincent A. Magnotta^1,3
1Department of Radiology, University of Iowa, Iowa City, IA, United States, ²Department of Biomedical Engineering, University of Iowa, Iowa City, IA, United States, ³Department of Psychiatry, University of Iowa, Iowa City, IA, United States, ⁴Department of Neurosurgery, University of Iowa, Iowa City, IA, United States

The purpose of this study is to assess the temporal dynamics of BOLD, ASL, and T1ρ using phase-encoded visual stimulation. Phase-encoded activation maps were calculated from the measurements obtained using an expanding ring stimulus that induce traveling waves of neural activity in the visual cortex. The results show that pH-sensitive T1ρincreased slightly first, followed by ASL and BOLD signal in the primary visual cortex, V1. This study support that local acidosis induced by the neural activity might be a principle factor in neurovascular coupling.

Louis O. Gagnon¹, Sava Sakadzic¹, Anna Devor², Qianqian Fang¹, Frederic Lesage³, Emiri T. Mandeville¹, Vivek J. Srinivasan¹, Mohammad A. Yaseen¹, Emmanuel Roussakis⁴, Eng H. Lo¹, Sergei Vinogradov⁴, Richard B. Buxton², Anders M. Dale⁵, and David A. Boas^1
1Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, Massachusetts, United States,²Department of Radiology and Neuroscience, University of California San Diego, La Jolla, California, United States, ³Department of Electrical Engineering, Ecole Polytechnique Montreal, Montreal, Quebec, Canada, ⁴Department of Biochemistry and Biophysics, University of Pensylvania, Philadelphia, Pensylvania, United States, ⁵Department of Radiology and Neuroscience, University of California, San Diego, La Jolla, California, United States

We propose a new methodology for modeling the fMRI signals at the microscopic level from quantitative optical microscopy. Two-photon microscopy O2 saturation measurements and Optical Coherence Tomography cerebral blood flow data were acquired in layers 1-3 of the mouse cortex during forepaw stimulation. The gradient echo and spin echo fMRI signals were then computed by simulating the diffusion of millions of proton over the tri-dimensional volume. This detailed model will serve as a gold standard to test the accuracy of more simplified models and new quantitative fMRI sequences to recover clinically relevant physiological parameters from fMRI measurements.

Sun Mi Park¹, Francesca Branzoli², Misun Kim¹, Hyerin Lim¹, Matthias J.P. van Osch², Itamar Ronen², and Dae-Shik Kim^1
1Department of Electrical Engineering, Korea Advanced Institute of Science and Technology, Daejeon, Korea, ²C. J. Gorter Center for High-field MRI, Department of Radiology, Leiden University Medical Center, Leiden, Netherlands

Although Multi-voxel pattern analysis (MVPA) has been widely studied to classify recognizing object categories, the signal characteristic of distributed patterns on the human ventral stream remains elusive. Ultra-high field MRI enables imaging at higher spatial resolution and improved localization, thereby potentially leading to improved discrimination between different neuronal patterns. Our results showed that the distributed patterns of 7T BOLD fMRI were better to distinguish each object category than 3T despite equal spatial resolution to provide fair comparison. It has been hypothesized that improving SNR and signal localization at 7T BOLD fMRI resulted in improved discrimination in MVPA methods.

Simon Schwab¹, Thomas Koenig¹, Andrea Federspiel¹, Thomas Dierks¹, and Kay Jann^1
1Dept. of Psychiatric Neurophysiology, University Hospital of Psychiatry / University of Bern, Bern, Switzerland

In the present work, we aimed to define the frequency specific thalamic areas generating distinct synchronized cortical networks, as indexed by EEG scalp fields oscillating with a common phase. Fourteen healthy subjects underwent simultaneous EEG/fMRI. The EEG was subjected to a Topographic Time-Frequency Decomposition that formed 6 classes of transient states of synchronized oscillation which were used to fit the thalamic BOLD response. For each of the EEG scalp fields, subregions corresponding to thalamic nuclei were identified. Therefore, our study provides a novel way to elucidate the systematics of subcortical effects on the formation of large scale cortical networks.

Nurhan Erbil¹ and Gopikrishna Deshpande^1,2
1Department of Electrical and Computer Engineering, Auburn University, Auburn, Alabama, United States, ²Department of Psychology, Auburn University, Auburn, Alabama, United States

Resting state networks, as coherent low frequency fluctuations of fMRI, are correlated with smoothed and downsampled EEG microstates. This coupling between EEG and fMRI can be attributed to scale free fractal properties of EEG microstates. By using fractal analysis of simultaneously acquired EEG and fMRI, we examined the existence of scale free dynamics in fMRI in addition to EEG. Simultaneously acquired resting state EEG and fMRI data from 4 subjects were analyzed by Cartool software and subjected to microstate segmentation for each subject and modality separately. Random walk embedding was applied to microstate segmentation sequences and corresponding Hurst exponents were calculated. We have shown both EEG and fMRI have scale free dynamics indicating a relation between resting state fMRI fluctuations and electrical oscillations underlying fast neuronal processes.

Siamak Salari Sharif¹, Michael D. Noseworthy², James P. Reilly³, and Daniel J. Bosnyak^4
1MR Innovations Inc., Detroit, MI, United States, ²Electrical and computer engineering, school of biomedical engineering, McMaster University, Hamilton, ON, Canada, ³Electrical and computer engineering, McMaster University, Hamilton, ON, Canada, ⁴Department of Psychology, Neuroscience and Behaviour, McMaster University, Hamilton, ON, Canada

Here we present a novel approach to removal of ballistocardiographic noise that contaminates EEG during combined EEG/fMRI. By using a soft constrained PLS algorithm (SC-PLS) we were able to extract EEG components that are essentially noise free. Our method is effective at removing BCG especially when ECG data is corrupt or not available. Furthermore, our method is advantageous as it does not require the finding of ECG QRS peaks, nor requires collection of ECG data.

Hu Cheng¹ and Aina Puce^1
1Indiana University, Bloomington, IN, United States

Motion correction is critical for data analysis of fMRI time series. Most motion correction algorithms treat the head as a rigid body. Respiration of the subject, however, can alter the static magnetic field in the head and result in slice shifts. To characterize the effect of respiration on Motion Correction, we acquired highly sampled fMRI data using multi-band EPI and then simulated different acquisition schemes. Our results show that respiration does affect the motion correction results and interleaved acquisition leads to larger between volume variations than ascending acquisition, suggesting a hybrid acquisition scheme is preferred.

Ekaterina Tchistiakova^1,2, David E. Crane², Jeremy Gilbert^3,4, Sandra E. Black^2,5, and Bradley J. MacIntosh^1,2
1Medical Biophysics, University of Toronto, Toronto, Ontario, Canada, ²Heart and Stroke Foundation Centre for Stroke Recovery, Sunnybrook Research Institute, Toronto, Ontario, Canada, ³Department of Medicine, Division of Endocrinology, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada,⁴University of Toronto, Toronto, Ontario, Canada, ⁵Department of Medicine, Division of Neurology, University of Toronto, Toronto, Ontario, Canada

Cerebrovascular reactivity (CVR) can provide valuable information on hymodynamic impairment in a wide range of neurological conditions. In this study we compared the spatial distribution of CVR maps obtained using CO₂ inhalation and breath hold (BH) tasks. The two hypercapnic challenges showed a good spatial overlap of CVR maps. Discrepancies were confined to inferior temporal and frontal pole regions with higher sensitivity during BH and CO₂ respectively. CO₂ CVR presented with increase in regions related to visual processing and sensorimotor areas compared to BH.

Ian D. Driver¹, Samuel J. Wharton¹, Paula L. Croal¹, Richard W. Bowtell¹, Susan T. Francis¹, and Penelope A. Gowland^1
1Sir Peter Mansfield Magnetic Resonance Centre, University of Nottingham, Nottingham, United Kingdom

The measurement of venous blood oxygenation (Yv) is important in cases where oxygen extraction fraction (OEF) may be perturbed. We assess phase-based methods for measuring Yv, either using intravascular phase of a large vein (the superior sagittal sinus in our case), or by comparing the phase distributions local to a vessel, as perturbed by hyperoxia. We measured similar Yv values using both approaches, although the hyperoxia-based method is suitable for small veins and independent of their orientation, providing localised Yv measurements. Some evidence for spatial heterogeneity in Yv was found using this method.

Yongquan Ye¹ and Ewart Mark Haacke^1
1Radiology, Wayne State University, Detroit, MI, United States

It is well known that the intravascular signals contribute about half of the functional contrast at 3T, and physiological effects also contribute to the BOLD signal. Although extensively studied in task/event related fMRI, the role of vascular and physiological effects in resting state (RS) functional connectivity is yet to be fully understood. By implementing flow dephasing with low VENC value, this study aims at estimating the contribution of the intravascular (IV) blood signal and physiological effects to local and long range resting state signal characteristics.

Paula L. Croal¹, Emma Louise Hall¹, Ian D. Driver¹, Susan T. Francis¹, and Penelope A. Gowland^1
1SPMMRC, School of Physics & Astronomy, University of Nottingham, Nottingham, Nottinghamshire, United Kingdom

The power law () relating blood oxygenation and tissue R2* is yet to be tested experimentally in humans. This is important as models and simulations do not accurately depict intravascular effects. Cross-field studies which use neuronal activation or hypercapnia are confounded by changes in CBV (ΔCBV) which also modulate R2*. Here, isocapnic-hyperoxia is used to measure the dependence of R2* on oxygenation, and the effect of ΔCBV simulated. Experimental and simulation results support a reduction of  with field strength. For a given field, β will not have a significantly impact on CMRO2 estimations within a normal physiological range.

Marta Bianciardi¹, Karleyton C. Evans², Jonathan R. Polimeni¹, Tian Y. Song², Boris Keil¹, Thomas Witzel¹, Bruce R. Rosen¹, David A. Boas¹, and Lawrence L. Wald^1
1Department of Radiology, A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States, ²Department of Psychiatry, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States

The velocity of cardiac and respiratory pressure-waves traveling in the brain and the pressure-change are indicators of vessel elasticity and of cerebrospinal-fluid response to vascular compliance. Recent work showed the feasibility of mapping the propagation of the cardiac pressure-wave in the brain by echo-planar-imaging (EPI). We further investigated the cardiac and also the respiratory pressure-wave in the brain by high spatio-temporal-resolution EPI at 7Tesla. Our results show the feasibility of mapping within <30-45s/slice the arrival time and the bulk flow velocity change of each pressure-wave, which are respectively related to the velocity of the traveling wave and to the pressure-change.

Hannah Hare¹, Michael Germuska¹, and Daniel P. Bulte^1
1FMRIB, University of Oxford, Oxford, Oxon, United Kingdom

This study consisted of a direct comparison between cerebrovascular reactivity (CVR) to 5% CO₂ in air and 5% CO₂ in oxygen (carbogen). We found that CVR as measured by BOLD and ASL response are correlated for CO₂ in air but not under a carbogen stimulus. In addition flow CVR as measured by ASL is strongly affected by inspired oxygen fraction and is not consistent between these two stimuli. We conclude that BOLD imaging should not be used in conjunction with a carbogen stimulus to measure CVR, and that flow CVR values acquired with carbogen and CO₂/air are not directly comparable.

Peng Wang¹, Bogdan Wilamowski², and Gopikrishna Deshpande^3
1AU MRI Research Center, Department of Electrical and Computer Engineering, Auburn University, Auburn, Alabama, United States, ²Department of Electrical and Computer Engineering, Auburn University, Auburn, Alabama, United States, ³AU MRI Research Center, Department of Electrical and Computer Engineering; Department of Psychology, Auburn University, Auburn, Alabama, United States

Brain disorder classification is traditionally done by Support Vector Machines (SVMs) due to SVM’s capability of handling data of high dimensionality and superior training speed. SVMs are effective in correctly identifying non-ADHD subjects. However SVMs are ineffective in correctly identifying ADHD subjects. Two-stage Fully Connected Cascade Deep neural network architecture has been designed and modified experimentally. This FCC Deep NN architecture significantly excels traditional NN architecture, overcomes data unbalance issue, is capable of handling data of high dimensionality and easy to train, generates better results, and therefore outperforms SVMs in total.

Feng Xu^1,2, Suresh E. Joel^1,2, Jun Hua^1,2, Craig K. Jones^1,2, Brian S. Caffo³, Martin A. Lindquist⁴, Ciprian M. Crainiceanu⁴, Peter C.M. van Zijl^1,2, and James J. Pekar^1,2
1Radiology, Johns Hopkins University, School of Medicine, Baltimore, Maryland, United States, ²F. M. Kirby Research Center, Kennedy Krieger Institute, Baltimore, Maryland, United States, ³Biostatistics, Johns Hopkins University, Baltimore, Maryland, United States, ⁴Biostatistics, Johns Hopkins University, School of Public Health, Baltimore, Maryland, United States

Multi-echo acquisitions can improve the sensitivity and specificity of resting-state functional connectivity MRI compared with conventional EPI. Analysis of multi-echo decays to estimate transverse relaxation in each voxel is a case of parallel estimation; statistical theory states that popular maximum likelihood (least-squares) methods are in some ways inferior to empirical Bayesian approaches. In this study we show that the James-Stein estimator yields modest increases in the spatial extent and inter-subject concordance of functional networks estimated from multi-echo resting BOLD data acquired at 7 Tesla.

Ali-mohammad Golestani¹ and Mariana Lazar^1
1Radiology, Center for Biomedical Imaging, NYU Medical Center, New York, NY, United States

Head motion during Resting-State fMRI artificially alters functional connectivity maps, with its effects persisting even after typical correction including frame realignment and regression of motion parameters. Adequate methods for correcting motion artifacts are currently a topic of debate. In this study we compared the ability of basic correction, data scrubbing (excluding volumes with excessive motion from the dataset), and independent component analysis (identifying and excluding motion components with ICA) to correct motion-induced connectivity alterations. Our result shows that ICA outperforms basic correction and data scrubbing and can suppress motion-induced connectivity changes.

Yash S. Shah¹, Ashish Farmer¹, Luis Hernandez-Garcia¹, Douglas C. Noll¹, Mark Greenwald², Jon-Kar Zubieta¹, and Scott J. Peltier^1
1University of Michigan, Ann Arbor, MI, United States, ²Wayne State University, Detroit, MI, United States

Multitask learning formulation presents a novel way of accommodating information from other subjects' data and building a generalized classifier. In our study, we use multitask learning to classify the temporal crave-state of a nicotine dependent subject and compare results to standard single subject SVM. We demonstrate that multitask learning is a promising novel analysis technique for fMRI data analysis.

Camille Maumet¹, Pierre Maurel¹, Jean-Christophe Ferré^1,2, and Christian Barillot^1
1Visages Project-team / U746, University of Rennes 1, INSERM, CNRS, Inria, RENNES, Brittany, France, ²Department of Neuroradiology, CHU Rennes, Rennes, Brittany, France

In this paper we propose an a contrario approach for the detection of activated brain areas in BOLD functional MRI data. The method focuses on subject-specific activations. Comparisons with the standard massively univariate general linear model are provided using Receiver-Operating-Characteristics curves for two different ground truths. The data used for validation include 12 subjects, scanned 3 times for a hand-motor paradigm. Overall, the proposed a contrario approach displayed better spatially defined activation with a more interesting trade-off between sensitivity and specificity by comparison to the standard massively univariate GLM.

Karthik Ramakrishnan Sreenivasan¹, Martin Havlicek², and Gopikrishna Deshpande^1,3
1AU MRI Research Center, Department of Electrical and Computer Engineering, Auburn University, Auburn, AL, United States, ²Department of Cognitive Neuroscience, Maastricht University, Maastricht, Limburg, Netherlands, ³Department of Psychology, Auburn University, Auburn, AL, United States

In this study we present a method which uses non-parametric blind deconvolution based on homomorphic filtering to investigate the over fitting problem of existing parametric methods. We compare our method to the performance of cubature Kalman filter (CKF)-based parametric approach. Simulations were performed with both methods and estimated neuronal responses were obtained. Correlation between the simulated and estimated neuronal responses indicated that in both cases (CKF-based and homomorphic deconvolution) the temporal neuronal events were correctly estimated, indicating that parametric methods such as CKF-based approaches are not susceptible to over fitting.

Masaki Sekino^1,2, Tomoaki Takewa¹, Dongmin Kim^1,2, Yusuke Inoue^1,2, and Hiroyuki Ohsaki^1
1Department of Electrical Engineering and Information Systems, Graduate School of Engineering, The University of Tokyo, Bunkyo-ku, Tokyo, Japan,²Exploratory Research for Advanced Technology, Japan Science and Technology Agency, Bunkyo-ku, Tokyo, Japan

Detection of weak magnetic fields arising from neuronal electrical activities using MRI is an attractive but challenging topic in functional MRI. We measured three-dimensional distribution of local field potentials in the rat somatosensory cortex using a 16-channel microelectrode array with a spatial resolution of 0.3 mm. Local distributions of electric current and magnetic field were calculated from the measured potentials. The resulting field magnitude of 2.8x10-12 T was comparable to the theoretical sensitivity limit of MRI.

Geon-Ho Jahng¹, Min-Ji Kim², Eo-Jin Hwang², Hyug-Gi Kim³, Kyung-Mi Lee⁴, Chang-Woo Ryu¹, Soo-Yeol Lee³, Wook Jin¹, Dal-Mo Yang¹, and Ji Seon Park^5
1Radiology, Kyung Hee University Hospital at Gangdong, Kyung Hee University, Seoul, Seoul, Korea, ²Radiology, Kyung Hee University Hospital at Gangdong, Seoul, Seoul, Korea, ³Biomedical Engineering, Graduate College of Electronics and Information, Kyung Hee University, Youngin, Gyeonggi-do, Korea, ⁴Radiology, Graduate College of of Medicine, Kyung Hee University, Seoul, Seoul, Korea, ⁵Radiology, Kyung Hee University Hospital, Kyung Hee University, Seoul, Seoul, Korea

To investigate an fMRI technique with an ultra-short TE (UTE) sequence, the UTE-based fMRI signal was obtained with a three-dimensional UTE sequence in brains of 18 young healthy volunteers during visual stimulations with TE=0.15 ms. The rationales of the use of UTE-based fMRI were to design a novel methodology for fMRI to more directly access cellular activation in addition to much lower sensitivity to field inhomogeneities compared to BOLD. A free induction decay (FID) signal has a potential to detect neuronal events other than T2 or T2* alternations. The UTE-based fMRI may provide a novel mechanism to investigate neuronal activations.

Jaroslav Marek¹, Jan Rieger^1,2, Darius Lysiak¹, Tomasz Kaczmarczyk³, Martin Kanowski⁴, Claus Tempelmann⁴, and Thoralf Niendorf^1,5
1Berlin Ultrahigh Field Facility (B.U.F.F.), Max-Delbrueck Center for Molecular Medicine, Berlin, Germany, ²MRI.TOOLS GmbH, Berlin, Germany, ³Blue Brick Ltd., Lodz, Poland, ⁴Department of Neurology, Otto-von-Guericke-University Magdeburg, Magdeburg, Germany, ⁵Joint cooperation - Charité Medical Faculty and the Max-Delbrueck Center for Molecular Medicine, Experimental and Clinical Research Center, Berlin, Germany

Temporal correlation of the stimulus paradigm with physiological noise induced by cardiac activity, blood flow, blood pulsation and respiration affects the BOLD signal. Realizing the unmet needs of today’s fMRI studies this abstract focuses on the monitoring and tracking of multiple physiological signals including ECG, phonocardiogram, magneto-hydrodynamic effect, pulse oximetry, respiration etc. to examine physiological fluctuations. For this purpose a dedicated physiological logging device that facilitates simultaneous and synchronized recording of multiple physiological signals as well as of the scanner trigger in-/output is proposed.

Swati Rane¹, John T. Spear¹, Zhongliang Zu¹, Manus J. Donahue^1,2, and John C. Gore^1,3
1VUIIS, Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, ²Psychiatry, Vanderbilt University, Nashville, TN, United States, ³Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

This study describes a novel approach to assess activity dependent diffusion changes using spin locking preparation and R₁ measurement

Casey P. Johnson¹, Hye-Young Heo², Daniel R. Thedens¹, John A. Wemmie³, and Vincent A. Magnotta^1,3
1Radiology, University of Iowa, Iowa City, IA, United States, ²Biomedical Engineering, University of Iowa, Iowa City, IA, United States, ³Psychiatry, University of Iowa, Iowa City, IA, United States

A spin-lock-prepared SE-EPI sequence for dynamic T1rho imaging of the brain with 4.0sec temporal resolution is characterized in simulation, phantom, and in vivo experiments. It is demonstrated in a flashing checkerboard experiment that the method can simultaneously measure both T1rho and BOLD activation. Additionally, it is shown in phantoms that the sequence is sensitive to pH changes on the order 0.1 units or less. This evidence supports that dynamic T1rho imaging is a viable tool to study brain activation that is unique from BOLD and is sensitive to activity-evoked acidosis. A potentially significant application is the study of psychiatric diseases.

Russell W. Chan^1,2, Iris Y. Zhou^1,2, Y. X. Liang³, Yong Hu⁴, K. F. So³, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong SAR, China, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong SAR, China, ³Department of Anatomy, The University of Hong Kong, Hong Kong SAR, China,⁴Department of Orthopaedics & Traumatology, The University of Hong Kong, Hong Kong SAR, China

Previously, a case study reported that complete transection of the corpus callosum induced loss of interhemispheric correlations in resting-state functional connectivity MRI (RSfcMRI). However, the results were limited by the lack of any electrophysiological recordings. Therefore, it could not be excluded that the loss of interhemispheric correlations in RSfcMRI might arise due to non-neuronal physiological modulations. In this study, intra-cortical electroencephalography (EEG) signals were recorded in the complete corpus callosotomy rats to understand the neural origin of the loss of interhemispheric correlations in RSfcMRI. The results clearly supported that the loss of interhemispheric correlations in RSfcMRI reflects the changes in spontaneous brain activity and its coherence. More importantly, the results strongly suggested that resting-state spontaneous fluctuations have strongest correlation with delta oscillations.

Iris Y. Zhou^1,2, Y. X. Liang³, Russell W. Chan^1,2, Joe S. Cheng^1,2, Patrick P. Gao^1,2, K. F. So³, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong SAR, China, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong SAR, China, ³Department of Anatomy, The University of Hong Kong, Hong Kong SAR, China

Changes in connectivity strength or spatial map of the low-frequency fluctuations in resting-state fMRI are the common focus when comparing the normal and diseased brains. However, limited work has studied the underlying spectral alterations. In this study, we investigated the spectral changes of resting-state connectivity in a rat model of complete corpus callosum (CC) transection. Interhemispheric connectivity was disappeared in cortical regions where the corresponding callosal connections were severed. Spectral analysis of these regions showed stronger power at relative high frequency around 0.1 Hz in transection group than in sham group. These spectral changes were observed in brain regions showing predominately intrahemispheric connectivity and therefore may arise from the faster intrahemispheric communication. Our experimental findings indicate that spectral characteristics of rsfMRI connectivity can be modulated by neural disruption and spectral analysis of rsfMRI data may provide a new dimension of information regarding the brain organization and connectivity.

Edmund W. Wong¹, David Matthew Carpenter¹, Johnny Ng¹, Jessica Roman¹, Ying Wei Wu², Xiaofeng Tao², and Cheuk Ying Tang^1
1Radiology, Mount Sinai School of Medicine, New York, New York, United States, ²Radiology, Shanghai East Hospital, Shanghai Tongji University, Shanghai, China

Functional connectivity has previously been shown to correlate with structural (anatomical) connectivity. Structural connectivity of brain white matter is quantified using diffusion tensor imaging to calculate fractional anisotropy (FA). In this study, we looked specifically at the superior longitudinal fasciculus (SLF) due to its anatomical relevance to the attention network. The correlation between functional coactivation of the attention network and SLF FA was investigated using independent component analysis (ICA) and dual regression. We hypothesize that this coactivation within the nodes of the network are positively correlated to SLF FA because the structural integrity of this neural pathway has importance to the neural communication between the nodes.

Joe S. Cheng^1,2, Iris Y. Zhou^1,2, Patrick P. Gao^1,2, Russell W. Chan^1,2, Queenie Chan³, Henry Ka Fung Mak⁴, Pek Lan Khong⁴, and Ed X. Wu^5
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, Hong Kong, ²Department of Electrical and Electronic Engineering, Hong Kong, Hong Kong, ³Philips Healthcare, Hong Kong, Hong Kong, ⁴Diagnostic Radiology, The University of Hong Kong, Hong Kong, Hong Kong, ⁵The University of Hong Kong, Hong Kong, SAR, China, Hong Kong

Resting-state functional MRI (rsfMRI) using blood-oxygen-level-dependent (BOLD) contrast has emerged as a valuable become an increasingly important tool for mapping inter and intra-hemispheric connectivity in normal and diseased brains. To date, most rsfMRI studies have exploited blood oxygenation level-dependent (BOLD) contrast using T2*- weighted gradient-echo (GE) EPI, which suffers from signal drops and image distortion due to magnetic susceptibility and inherent long TE. Alternatively, T2/T1- weighted passband balanced SSFP (bSSFP) has been promoted for distortion-free, high spatial resolution task-based functional imaging. In this study, we studied the capability of passband b-SSFP for rsfMRI. Resting-state networks (RSNs) of similar but different spatial and temporal pattern were identified, especially into areas near air sinus.

Patrick P. Gao^1,2, Russell W. Chan^1,2, Joe S. Cheng^1,2, Iris Y. Zhou^1,2, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, Hong Kong SAR, China, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, Hong Kong SAR, China

Resting-state fMRI (rsfMRI) has been increasingly used to study brain function on humans and animals. Currently, most rsfMRI studies adopt the GE-EPI sequence which has a blood oxygenation level dependent (BOLD) contrast. However, GE-EPI is affected by signal dropout and image distortion due to nature of GE and long echo time. These problems become more severe at high field and in regions close to air-tissue interface. Balanced steady-state free procession (bSSFP) imaging is free from these problems. Recently it has been demonstrated to detect BOLD-like activation signals for fMRI. In this study, we investigated the feasibility of using bSSFP to detect resting-state networks.

A. Alhamud¹, F. Robertson¹, D. Donaldson¹, André J. W. van der Kouwe², and Ernesta M. Meintjes^1
1Human Biology, MRC/UCT Medical Imaging Research Unit, University of Cape Town, Cape Town, Western Cape, South Africa, ²Department of Radiology, Massachusetts General Hospital, Charlestown, Massachusetts, United States

Diffusion tensor imaging (DTI) is a promising technique to map white matter tracts, while fMRI is a technique to measure brain activity. Recent studies have combined data from separate fMRI and DTI acquisitions to improve accuracy of seed regions for fiber tracking. It remains unclear whether brain activation during DTI acquisition itself may affect DTI data. The present study exploits the navigated diffusion sequence, which has recently been introduced for real time motion correction, to perform simultaneous fMRI-DTI by using the navigators in the navigated diffusion sequence to acquire BOLD data.

Joe S. Cheng^1,2, Iris Y. Zhou^1,2, Hua Guo³, Patrick P. Gao^1,2, Russell W. Chan^1,2, Queenie Chan⁴, Henry Ka Fung Mak⁵, Pek Lan Khong⁵, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, Hong Kong, ²Department of Electrical and Electronic Engineering, Hong Kong, Hong Kong, ³Biomedical Engineering & Center for Biomedical Imaging Research, Tsinghua University, Beijing, China, ⁴Philips Healthcare, Hong Kong, Hong Kong, ⁵Diagnostic Radiology, The University of Hong Kong, Hong Kong, Hong Kong

Resting-state functional MRI (rsfMRI) has emerged as a valuable become an increasingly important tool for mapping inter and intra-hemispheric connectivity in normal and diseased brains. To date, most rsfMRI studies have used T2*-weighted gradient-echo (GE) for the benefit of sensitivity. Theoretically, spin-echo (SE) provides better spatial specificity due to the elimination of static dephasing around macro-vessels. At 3T, the benefit however is limited because of a nearly equal share of undesired intravascular (IV) and well-localized extravascular (EV) contribution from microvasculature. Diffusion gradients induce velocity-dependent phase shifts and thus reduce signal from blood due to inhomogeneous velocities within vessel and presence of vessels with different orientations in a pixel. In this study, we aimed to investigate rsfMRI connectivity at 3T using diffusion-weighted (DW) SE to eliminate IV effect from large vessels. By using interleaving non-DW and DW acquisition, we identified default mode networks with similar but different pattern, spatially and spectrally.

Mark Chiew¹, Stephen M. Smith¹, Peter J. Koopmans², Thomas Blumensath³, and Karla L. Miller^1
1FMRIB Centre, University of Oxford, Oxford, United Kingdom, ²Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands, ³ISVR, University of Southampton, Southampton, Hampshire, United Kingdom

In FMRI, measurements of resting state functional connectivity are often preceded by a principal component analysis to reduce data dimensionality. We propose a new method for the acceleration of FMRI acquisitions that exploits the decrease of information in a dimensionality reduction to facilitate the undersampling of k-t space. We call this approach k-t FASTER: FMRI Acceleration in Space-time via Truncation of Effective Rank. This technique is demonstrated on 4x retrospectively undersampled FMRI data to reproduce resting state networks with high spatial fidelity.

Michael E. Kelly¹, Eugene P. Duff¹, Janine D. Bijsterbosch¹, Natalie L. Voets¹, Nicola Filippini¹, Steen Moeller², Junqian Xu², Essa S. Yacoub², Edward J. Auerbach², Kamil Ugurbil², Stephen M. Smith¹, and Karla L. Miller^1
1FMRIB Centre, University of Oxford, Oxford, United Kingdom, ²Center for Magnetic Resonance Research, University of Minnesota Medical School, Minneapolis, Minnesota, United States

Multiband (MB) EPI is a recent MRI technique that offers increased temporal and/or spatial resolution as well as increased temporal SNR due to increased temporal degrees-of-freedom (DoF). However, MB-EPI may exhibit increased motion sensitivity due to the combination of short TR with parallel imaging. In this study, the performance of MB-EPI with different acceleration factors was compared to that of standard EPI, with respect to subject motion. Although MB-EPI with 4 and 8 times acceleration exhibited some motion sensitivity, retrospective clean-up of the data using independent component analysis was successful at removing artefacts. By increasing temporal DoF, accelerated MB-EPI supports higher spatial resolution, with no loss in statistical significance compared to standard EPI. MB-EPI is therefore an important new technique capable of providing high resolution, temporally rich FMRI datasets for more interpretable mapping of the brain's functional networks.

Yongquan Ye¹ and Ewart Mark Haacke^1
1Radiology, Wayne State University, Detroit, MI, United States

Resting state fMRI reflects the BOLD signal induced by spontaneous neuronal activities at rest. rs-fMRI signal is especially sensitive to physiological noise, which often is the dominant source of noise. Recently it has been demonstrated that using a flip angle (FA) as low as 9¡ã can reduce the relative weighting of physiological noise in BOLD signal while still sustaining sufficient BOLD contrast [1]. In this study, we will study how different flip angles affect rs-fMRI signal characteristics and the calculation of functional connectivity, and analyze the underlying mechanism of the origin of resting state signals.

Hsu-Lei Lee¹, Jakob Assländer¹, Pierre Levan¹, and Jürgen Hennig^1
1University Medical Center Freiburg, Freiburg, Germany

In this study we used MREG sequence with stack-of-spirals trajectory to obtain resting-state fMRI signal at 10 Hz sampling rate. Data were filtered into three different frequency bands and put through the same post-processing steps. Results of spatial ICA were checked with ICASSO and the reliable components were used in the following connectivity analysis to generate functional modules in the brain and calculate the modularity coefficients. We found that as some modules can be seen at all frequencies, and some only appears in one or two of the bands.

Waqas Majeed¹, Matthew E. Magnuson², and Shella Keilholz^2
1Department of Electrical Engineering, Lahore University of Management Sciences, School of Science and Engineering, Lahore, Punjab, Pakistan,²Biomedical Engineering, Georgia Institute of Technology/Emory University, Atlanta, GA, United States

This work demonstrates the use of spatiotemporal Fourier analysis for studying dynamic characteristics of low frequency fluctuations (LFFs) in resting state fMRI data. LFFs in rats contain propagating waves moving in lateral to medial direction. Our work demonstrates that this propagation pattern primarily consists of fast moving spectral components, and spatiotemporal velocity-selective filtering can be used improve the contrast-to-noise ratio of this pattern, when detected automatically. This type of analysis can be useful for detecting new patterns in LFFs, as well as analyzing dynamic imaging data in general.

Thomas Allan¹, Matthew J. Brookes¹, Susan T. Francis¹, and Penelope A. Gowland^1
1SPMMRC, University of Nottingham, Nottingham, Nottinghamshire, United Kingdom

The brain is known to be segregated into relatively few large scale networks but it has been shown using fMRI that the brain has highly focal sub regions that respond to a specific task. During a task, whether it is externally or internally driven, multiple nodes of a network can be recruited to successfully complete the task. We investigate how spontaneous BOLD events, detected using paradigm free mapping, synchronise transiently and have specific plausible substructures within networks similar to fMRI studies by using temporal ICA.

Pan Lin¹, De Pisapia Nicola², and Jorge Jovicich^2
1Institute of Biomedical Engineering,Xi'an Jiaotong University, xi'an, shaanxi, China, ²Center for Mind/Brain Sciences, University of Trento, Trento, Trentino, Italy

Functional connectivity within the default network dynamic changes is associated with internal mental state across rest or task state. Characterization of the dynamic resting sate network organization as a complex network of dynamic temporal evolving networks has received great interest for the study of its neural mechanisms. To better understand temporal dynamic topology structure of DMN, we used weighted complex network and clustering method to investigate default mode network across different brain mental state. Our results suggest the important role of the dynamic temporal-topological structure of DMN link to underling brain adaptive processing function.

Suresh Emmanuel Joel¹, Anand Narasimha Murthy¹, Ek T. Tan², Dattesh D. Shanbhag¹, John F. Schenck², and Rakesh Mullick^1
1Diagnostics and Biomedical Technologies, General Electric Global Research, Bangalore, Karnataka, India, ²Diagnostics and Biomedical Technologies, General Electric Global Research, Niskayuna, New York, United States

Differences in functional networks have been used to understand brain in health and disease. Variability in functional networks can be due to disease, state (of mind) or trait. Here we study intra-subject and inter-subject reproducibility and classify in to a) networks that are highly reproducible within and across subjects making them good candidates for studying changes associated with disease, b) networks that are variable within subjects making them good candidates for studying state and c) networks that are variable between subjects making them good candidates for studying traits.

Jiancheng Zhuang¹ and Savio Wong^1
1University of Southern California, Los Angeles, California, United States

This study proposes an approach to estimate the functional connectivity networks from CBF and BOLD signals simultaneously measured by ASL (arterial spin labeling) MRI using exploratory Structural Equation Modeling analysis. Two structural equation models were estimated at each voxel regarding to the sensory-motor network and default-mode network. The resulting connectivity maps indicate that supplementary motor area has significant connections to left/right primary motor areas, and inferior parietal lobules link significantly with posterior cingulate cortex and medial prefrontal cortex. The model fitting results imply that BOLD signal is more directly linked to the underlying cause of functional connectivity than CBF signal.

Yunzhi Wang¹, Santosh Katwal^2,3, Baxter Rogers^2,4, John C . Gore^4,5, and Gopikrishna Deshpande^1,6
1AU MRI Research Center, Department of Electrical and Computer Engineering, Auburn University, Auburn, AL, United States, ²Vanderbilt University Institute of Imaging Science (VUIIS), Nashville, TN, United States, ³Department of Electrical Engineering and Computer Science, Vanderbilt University, Nashville, TN, United States, ⁴Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, ⁵Vanderbilt University, Nashville, TN, United States, ⁶Department of Psychology, Auburn University, Auburn, AL, United States

Dynamic Granger causality (DGC) was performed to infer sub-100ms neuronal timing differences in BOLD responses from the visual cortex. DGC connectivities were calculated from experimental fMRI data which introduced a known delay between right and left hemified stimuli and a GLM was fit between the experimental paradigm and dynamic Granger causality difference series. The results come out that dGCD significantly covaried with the paradigm for all delays except zero. Our experimental validation of DGC to detect sub-100 ms neuronal timing differences without the confounding effect of hemodynamic variability provide a reliable data-driven method for effective connectivity analysis of fMRI data.

Kaiming LI¹ and Xiaoping P. Hu^1
1Emory University, Atlanta, GA, United States

This paper describes a new framework to characterize the connectomic profiles for distinct functional regions on the cortical surface. Unlike existing group ICA approaches that heavily rely on spatial smoothing and registration techniques, this framework employs two measures, cortical parcellation by BOLD signal homogeneity and over complete dictionary learning, to account for the well-known anatomical variability across individuals. Our results show that the resultant connectomic profiles are robust and can be used for the identification of both distinct functional regions and functional networks, facilitating building statistical models for these profiles and pinpointing disrupted regions in pathological/psychiatric brain disorder datasets.

Hang Joon Jo¹, Richard C. Reynolds¹, Stephen J. Gotts², Alex Martin², Robert W. Cox¹, and Ziad S. Saad^1
1Scientific and Statistical Computing Core, National Institutes of Mental Health, Bethesda, MD, United States, ²Section on Cognitive Neuropsychology, Laboratory of Brain and Cognition, National Institutes of Mental Health, Bethesda, MD, United States

Brain function in the “resting” state has been extensively studied with MRI since its discovery of correlated temporal signals in distant parts of the cortex. Recent studies have reported major corrupting effects of subject head motion on the results of resting state FMRI analyses. The first study suggested that motion effects induce bias between short- and long-range correlations, casting suspicion on inferences about brain networks (Power et al., 2012). The second study showed how differences in average subject movement between two subjects groups could produce significant differences in apparent brain connectivity (Van Dijk et al., 2012). Herein, we replicate these results, and qualify as follows the conclusions that ascribed them solely to the presence of motion difference between groups.

Swati Rane¹, Brandon Ally^2,3, Tracy Wilson¹, Tricia Thornton-Wells^4,5, John C. Gore^1,6, Erin Hussey³, and Manus J. Donahue^1,2
1VUIIS, Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, ²Psychiatry, Vanderbilt University, Nashville, TN, United States, ³Neurology, Vanderbilt University, Nashville, TN, United States, ⁴Center for Human Genetics Research, Vanderbilt University, Nashville, TN, United States, ⁵Molecular Physics and Biology, Vanderbilt University, Nashville, TN, United States, ⁶Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

This study examines baseline BOLD connectivity using R2* maps obtained from multi-echo FMRI, in relation to multi-subject comparison.

Ying-Hui Chou¹, Pooja Gaur², Carol P. Weingarten¹, Mei-Lan Chu¹, David Madden¹, Allen W. Song¹, and Nan-Kuei Chen^3
1Duke University Medical Center, Durham, NC, United States, ²Vanderbilt University, Nashville, TN, United States, ³Duke University, Durham, NC, United States

In this study, we demonstrated that the behavior-based connectivity analysis and support vector machine methods can be used to decode the whole-brain resting-state functional connectivity patterns and classify individuals who reported inner language as the dominant mental activity during resting-state fMRI scan from those who did not with a sensitivity of 0.88 and a specificity of 0.9. Our findings can lead to a better understanding of variations in resting-state fMRI signals and their dependence on the spontaneous cognition/mind wandering.

Li Min Chen¹, Alex V. Maier², Arabinda Mishra¹, Feng Wang¹, Daniel Colvin¹, Allen T. Newton¹, M. Young², John C. Gore¹, and Jeffrey D. Schall^2
1Vanderbilt University Institute of Imaging Science, Vanderbilt University, Nashville, Tennessee, United States, ²Department of Psychological Sciences, Vanderbilt University, Nashville, Tennessee, United States

Being off-limits to the invasive anatomical techniques, knowledge of human brain connectivity lags behind nonhuman primate. The proposed study is intended to investigate the homology of the primate frontal lobe using a novel imaging approach that bridges the evolutionary division between new world monkeys (Ceboidea) and old world primates (Cercopithecidae). We implemented an unsupervised self-organization mapping1,2 (SOM) technique to study the functional connectivity network of the frontal eye field (FEF) of macaques and squirrel monkeys. Comparison of these connectivity maps across monkey species is a promising initial step to establish the degree of homology of cortical sub-regions in primates.

Yinan Liu¹, Karl Young², Duygu Tosun², Yu Zhang², and Norbert Schuff^2
1San Francisco VA Medical Center, San Francisco, CA, United States, ²Department of Radiology and Biomedical Imaging, University of California at San Francisco, San Francisco, CA, United States

Resting-state fMRI has been an important tool to understand brain networks. However, the intrinsic properties of BOLD fluctuations have been rarely studied. In this preliminary study, we used transient information to compare the degree of complexity of the BOLD fluctuations between normal and Alzheimer’s disease subjects. Results exhibited significantly smaller degrees of complexity in AD than CN subjects in brain regions typically affected by disease. In contrast, AD and CN subjects had similar degrees of complexity in regions usually spared by disease. In conclusion, transient information could provide a new metric to study brain functions in healthy conditions and disease.

Unal Sakoglu¹ and Kushal Bohra^1
1Computer Science Department, Texas A&M University - Commerce, Commerce, Texas, United States

In this work, we study the effects of task-related signal variance differences to functional connectivity analysis using simulated fMRI signals. Considerable correlation was found between the functional connectivity and the standard deviations of the fMRI signals. This shows that functional connectivity differences are accompanied by systematic differences in the variance of the fMRI signal under different experimental tasks/conditions. This result highlights the need for caution in interpreting condition-dependent differences in functional connectivity analyses; one cannot rule out the possibility that the observed differences were merely a consequence of differences in the variance of the respective signal time-courses.

Zhiye Chen¹, Lin Ma¹, Xin Lou¹, and Jinfeng Li^1
1Department of Radiology, Chinese PLA General Hospital, Beijing, Beijing, China

This study mainly focused on the age effects on the marginal division of the neostriatum in health adults revealed by a resting functional connectivity

Zhuo Fang^1,2, Senhua Zhu^1,2, Hui Shi², Siyuan Hu², Huajian Cai³, Marc Korczykowski², and Hengyi Rao^1,2
1Department of psychology, Sun Yat-sen University, GuangZhou, GuangDong, China, ²Center for functional Neuroimaging, University of Pennsylvania, Philadelphia, PA, United States, ³Institute of psychology, Chinese Academy of Sciences, Beijing, Beijing, China

Recent neuroimaging research has demonstrated great interest in resting brain function. However, the exact function of brain default mode network (DMN) remains largely unknown. The present study used arterial spin labeling (ASL) perfusion MRI to non-invasively measure resting cerebral blood flow (CBF) in a large cohort of normal subjects (n=81) and examined the relationship between default brain function and self-esteem, a core component of the self. The results showed that self-esteem is positively correlated with regional CBF in the posterior DMN regions, supporting the view that self-introspective processes is associated with default brain function.

Hsin-Long Hsieh¹, Pin-Yu Chen², Fu-Shan Jaw¹, and Wen-Yih Isaac Tseng^2
1Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan, ²Center for Optoelectronic Biomedicine, National Taiwan University College of Medicine, Taipei, Taiwan

The structural and functional connectivity are correlated. However, mismatch of functional and structural connectivity occurs due to indirect connections and limitations of current neuroimaging approaches and analysis. To address the limitation, we focus on interhemispheric connection because it is direct connection and the most clearly defined white matter structure. We investigated the interhemispheric structure-function relationship combining DSI and resting-state fMRI. Our results showed that the degree of structure-function relationship varied in different lobes and also differed in homotopic and heterotopic interlobular regions as well. We suggest that topology may be a contributing factor of the structure-function relationship.

Asimina Lazaridou^1,2, Loukas Astrakas^1,2, Dionyssios Mintzopoulos^1,2, Azadeh Khanicheh³, Aneesh Singhal⁴, Michael Moskowitz², Bruce Rosen², and Aria A. Tzika^1,2
1NMR Surgical Laboratory, Massachusetts General Hospital and Shriners Burn Institute, Harvard Medical School, Boston, MA, United States, ²Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Boston, MA, United States, ³Mechanical Engineering, Northeastern University, Boston, MA, United States,⁴Department of Neurology, Stroke Research Center, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States

Stroke may alter functional connectivity of motor execution networks. Our purpose was to demonstrate functional reorganization of motor systems by exploring connectivity alterations within the motor related areas using fMRI with a novel MR-compatible hand-induced robotic device (MR_CHIROD). Our findings suggest that enhancement of SMA activity benefits M1 dysfunction in stroke survivors. Assessing changes in connectivity by means of fMRI in conjunction with MR_CHIROD might be used in the future to further illustrate the neural network plasticity that underlies functional recovery in chronic stroke patients.

Amir Abduljalil¹, Chima Oluigbo², Xiangyu Yang¹, Andrew Kalnin³, Michael V. Knopp⁴, and Ali Rezai^2
1Wright Center of Innovation, The Ohio State University, Columbus, Ohio, United States, ²Neurosurgery, The Ohio State University, Columbus, Ohio, United States, ³Radiology, The Ohio State University, Columbus, Ohio, United States, ⁴The Ohio State University, Columbus, Ohio, United States

Spinal cord stimulation is an established treatment option for chronic neuropathic pain. However, the cortical processing of sensory information regarding the perception of pain and the stimulation induced pain relief has not yet been well understood. Resting state fMRI and graph theory methodology can provide an insight to the cortical response. In this study, the functional response to four different settings of the spinal cord stimulation was analyzed and significant correlation to the functional brain network measures has been detected.

Yongquan Ye¹, Jie Yang¹, Ewart Mark Haacke¹, and Shuang Xia^2,3
1Radiology, Wayne State University, Detroit, MI, United States, ²Radiology, Nanjing Jinling Hospital, Nanjing University, Nanjing, Jiangsu, China, ³Radiology, Tianjin First Central Hospital, Tianjin, Tianjin, China

The development and elaboration of auditory cortical areas and its interactive connections with other brain functions is yet to be understood, as it is of primary importance to development of speech, language, cognition (memory and attention), literacy, etc. However, few studies, if any, have compared the functional connectivity between children (2-5 y/o) with profound bilateral hearing loss and normal hearing. In this study, we will explore the variations in resting state functional connectivity related to hearing capability, and to obtain an insight to what can be at stake if hearing loss is not properly treated in a timely manner.

Siamak P. Nejad-Davarani^1,2, Valentina Gumenyuk³, Hassan Bagher-Ebadian^4,5, Scott J. Peltier², John Budaj¹, Christopher L. Drake³, Douglas C. Noll², Quan Jiang^1,5, and Michael Chopp^1
1Neurology, Henry Ford Hospital, Detroit, MI, United States, ²Biomedical Engineering, University of Michigan, Ann Arbor, MI, United States, ³Sleep Disorders and Research Center, Henry Ford Hospital, Detroit, MI, United States, ⁴Radiology, Henry Ford Hospital, Detroit, MI, United States, ⁵Physics, Oakland University, Rochester, MI, United States

We have used resting state fMRI to study the functional connectivity network in individuals with Shift Work Sleep Disorder (SWSD) and compared it to the connectivity networks of asymptomatic night shift workers (SWC) and daytime workers (NC). We used a framework in our analysis for segmenting the brain and finding the correlation between the fMRI signals in each segment pair. Our results show differences in the functional connectivity network between the NC and SWC groups however, the connectivity pattern of the SWC group is more correlated to the NC group than the SWSD group pattern which confirms previously reported results.

Edmund W. Wong¹, Roxana Teodorescu², Lazar Fleysher², Cheuk Ying Tang¹, Matilde Inglese², and David Matthew Carpenter^1
1Radiology, Mount Sinai School of Medicine, New York, New York, United States, ²Neurology, Mount Sinai School of Medicine, New York, New York, United States

We combined diffusion tensor imaging and resting state fMRI to assess the relationship between white matter integrity and functional coactivation during resting state in patients with relapsing-remitting MS (RR-MS) and healthy controls. The analysis was performed using independent component analysis to identify the default mode network, followed by dual-regression analyses used to find voxels where coactivation of the default mode network was correlated with global and, in a separate analysis, cingulum bundle FA. An inverse relationship was observed in both global and cingulum bundle FA with coactivations within the default mode network.

Alexandre Coimbra¹, Farshid Faraji², Alex De Crespigny¹, and David Clayton^2
1Genentech Inc, South San Francisco, California, United States, ²Genentech, Inc., South San Francisco, California, United States

In single-site studies, brain functional connectivity metrics (FCMs) measured by resting-state functional MRI (rs-fMRI) have been reported to be sensitive to Alzheimer Disease (AD, [1]), beta-amyloid burden [2], and therapeutic effect [3]. Exploratory RS-fMRI was included in two global Phase II studies of a novel AD amyloid drug. We studied different types of data processing pipelines and FCMs make recommendations based on ensuing FCMs test-retest performance and ability to differentiate cohorts scanned at 1.5T vs. 3T magnets. A discussion on FCMs and statistical power to detect changes due to disease progression and/or therapeutic effect is also provided.

Guangyu Chen¹, Chunming Xie¹, Guangyu Chen¹, Wenjun Li¹, B. Douglas Ward¹, Jennifer L. Jones², Malgorzata Franczak², Piero G. Antuono², and Shi-Jiang Li^1
1Department of Biophysics, Medical College of Wisconsin, Milwaukee, WI, United States, ²Department of Neurology, Medical College of Wisconsin, Milwaukee, WI, United States

Understanding the reliability of resting-state functional magnetic resonance imaging is essential to the successful application of the technique. This study will demonstrate the key factors that affect the reliability of detecting functional connectivity changes in studying brain disorders.

Dongyi Liao¹, Lena Palaniyappan², Karthik Ramakrishnan Sreenivasan¹, Peter Liddle³, Molly Simmonite², and Gopikrishna Deshpande^1,4
1AU MRI Research Center, Department of Electrical and Computer Engineering, Auburn University, Auburn, AL, United States, ²Translational Neuroimaging, Institute of Mental Health, University of Nottingham, Nottingham, United Kingdom, ³Department of Psychiatry, Institute of Mental Health, University of Nottingham, Nottingham, United Kingdom, ⁴Department of Psychology, Auburn University, Auburn, AL, United States

In this study, we demonstrate decreased effective connectivity in resting state networks of Schizophrenics as compared to healthy individuals. Resting state fMRI data were acquired from both groups and subjected to multivariate Granger causality analysis coupled with blind deconvolution of the hemodynamic response using a cubature Kalman filter. Casual interactions were significantly impaired for subjects with schizophrenia, especially the outputs from Insula and posterior parietal cortex. This disconnectivity is likely to be an important factor in the core pathophysiology of schizophrenia.

Trey Shenk¹, Kausar Abbas¹, Evan Breedlove², Victoria N. Poole^3,4, Katherine Breedlove⁵, Larry Leverenz⁵, Eric Nauman^2,4, Thomas Talavage^1,3, and Meghan Robinson^3,6
1Electrical Engineering, Purdue University, West Lafayette, IN, United States, ²Mechanical Engineering, Purdue University, West Lafayette, IN, United States,³Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, United States, ⁴Basic Medical Sciences, Purdue University, West Lafayette, IN, United States, ⁵Health and Kinesiology, Purdue University, West Lafayette, IN, United States, ⁶Translational Research Center for TBI and Stress Disorders (TRACTS), VA Boston Healthcare System, Boston, MA, United States

We have explored resting state functional connectivity in asymptomatic high school American football players and a cohort of non-contact athletes peers. Seed-based default mode network (DMN) analysis reveals that while the total number of voxels included in the DMN did not change significantly with hits, the location of the DMN, as defined by number of voxels in MarsBaR regions, did. Independent component analysis (ICA) on all sessions was used to find stable networks. Pairwise correlations between networks were significantly different in players than controls in 8 of the 15 pairs, and further were correlated with hits in some cases.

Jeiran Choupan¹, Zhengyi Yang², Luca Cocchi³, Julia Hocking¹, and David C. Reutens^1
1Centre for Advanced Imaging, Brisbane, QLD, Australia, ²School of Information Technology and Electrical Engineering, Brisbane, QLD, Australia, ³Queensland Brain Institute, Brisbane, QLD, Australia

This abstract focus on finding comprehensive set of measures of functional connectivity in order to improve the localization of the seizure focus in temporal lobe epilepsy patients. The measurement includes resting state functional connectivity, regional homogeneity, amplitude of low frequency fluctuation, fractional amplitude of low frequency fluctuation and voxel-mirrored homotopic connectivity. We have found that, not only the whole brain hippocampal functional connectivity, but also other high level features of resting state fMRI signal, such as regional homogeneity, amplitude of low frequency fluctuation differ between left and right temporal lobe epilepsy groups. These results can be incorporated in pattern classification methods as prior knowledge to increase the classification accuracy.

Steven M. Stufflebeam¹, Naoro Tanaka¹, Boris Keil¹, Linda Douw¹, Lawrence L. Wald¹, and Christina Triantafyllou^1
1Radiology, MGH-Martinos Center, Charlestown, MA, United States

We acquired resting-state BOLD fMRI at 7T MRI in six patients with medically refractory partial epilepsy and identified 3 clinically relevant resting-state networks: default mode, motor, and language networks. The laterality of these networks was calculated by computing a laterality index based on the left and right hemisphere correlation maps. We localized all networks and found that the language network was left lateralized, as was the default mode network in a subject with temporal lobe epilepsy. This suggests that, even at an individual subject level, 7T task-free BOLD fMRI localizes and lateralizes eloquent cortex that is critical for presurgical mapping. including motor, language and memory tasks performed.

Svyatoslav Vergun¹, Veena A. Nair², Matthew Jensen³, Marcus Chacon³, Justin Sattin³, and Vivek Prabhakaran^2
1Medical Physics, UW-Madison, Madison, WI, United States, ²Radiology, UW-Madison, Madison, WI, United States, ³Neurology, UW-Madison, Madison, WI, United States

Multivariate pattern analysis methods have been shown successful in extracting significant information and classifying individual scans. In this work, a support vector machine classifier accurately discriminated between stroke and normal aging subjects based on their resting state functional connectivity. 50 resting state fMRI scans from 24 normal and 26 stroke subjects were preprocessed and time series from 160 functional ROIs were correlated to produce a functional connectivity matrix for each subject. Each subject’s correlations were input as features into the classifier, which predicted subjects with 80% accuracy using leave-one-out cross validation. Sensorimotor network connectivity was most influential for classification.

Benito de Celis Alonso¹, Silvia Hidalgo Tobón², Pilar Dies Suarez², Manuel Obregon Espejel², Porfirio Ibañez Fernández², Heidi de la Rosa Guzmán², Eduardo Castro Sierra², and Eduardo Barragán Pérez^2
1Faculty of Medicine, BUAP, Puebla, Puebla, Mexico, ²Hospital Infantil de México, Mexico City, México FD, Mexico

ADHD is the most common neurological disorder in intants with a prevalence circa 5%. This illness is considered to be in an 80% originated buy genetic factors and in a 20% by environmental factors. Almost all of the literature containing MR work in this field is based in studies on subjects from European, North American or Asian origin. No studies of this kind exist on Latin American subjects (different genetic pool). In this abstract we compared resting state scans of healthy and non-healthy ADHD patients of Latin origin and compared these results to the literature findings with a special focus on the role of the posterior cingulate cortex.

Ariane Orosz¹, Leila Maria Soravia¹, Andrea Federspiel¹, Roland Wiest², Thomas Dierks¹, and Kay Jann^1
1Dept. of Psychiatric Neurophysiology, University Hospital of Psychiatry / University of Bern, Bern, Bern, Switzerland, ²Institut of Diagnostic and Interventional Neuroradiology, University of Bern / Inselspital, Bern, Bern, Switzerland

The effect of the stress-hormone cortisol on cerebral blood flow (CBF) within functionally connected networks (FCN) was investigated using arterial spin labeling (ASL). The cortisol level fluctuates in a diurnal rhythm and might alter network connectivity and metabolism in FCNs representing higher cognitive functions or emotional processing. Three groups with different cortisol levels (high, medium, low) were measured. The FCNs displaying the spatial pattern of the medial-temporal network and the default mode network were identified by independent component analysis. Network CBF quantification revealed different average CBF in distinct FCNs, but no particular impact of cortisol at resting state.

Hazim Omar¹, Theodore Kypraios², Dorothee P. Auer¹, and Naranjargal Dashdorj^1
1Radiological and Imaging Sciences, University of Nottingham, Nottingham, Nottinghamshire, United Kingdom, ²School of Mathematical Sciences, University of Nottingham, Nottingham, Nottinghamshire, United Kingdom

Resting state fMRI had been reported to have small-world architecture in the brain. We investigate the effect of hypnotic and sedative drug, midazolam administration on the functional brain network by using graph analysis and found a small-world network characteristic in both condition pre and post midazolam administration with a significant increase in clustering coefficient which indicate synchronized cliquishness functional network. There is also significant increase in local efficiency which reflects the increase in ability of a network in transmitting the information. This evidence suggests graph theory analysis can provide an insight in exploring the effect in functional brain connectivity.

Kang Wang¹, Zachary M. Smith¹, Chi Wah Wong¹, and David J. Dubowitz^1
1University of California San Diego, La Jolla, CA, United States

High altitude hypoxia increases CMRO2, CBF, and OEF. We investigated how this might impact BOLD functional connectivity. 5 subjects were studied during normoxia, and following 2-days sustained hypoxia with and without acetazolamide prophylaxis (a treatment for high altitude hypoxia symptoms). Hypoxia increased overall correlations with posterior cingulated cortex, and the global signal amplitude. Acetazolamide mitigated these global effects, while increasing correlations in the default mode network. CO2 has been implicated as a modulator of BOLD functional connectivity, but during hypoxia this does not seem to be the sole contributor. Alterations in CBF, CMRO2 and neuronal activity must also be considered.

Johnny Ng¹, Ying Wei Wu², Xiaofen Tao², Edmund Wong¹, David Carpenter¹, and Cheuk Tang^1,3
1Dept. of Radiology, Mount Sinai School of Medicine, New York, NY, United States, ²Dept. of Radiology, Shanghai East Hospital, Shanghai Tongji University, Shanghai 200120, China, ³Dept. of Psychiatry, Mount Sinai School of Medicine, New York, NY, United States

Using pure oxygen as an MR contrast agent (Oxygen-enhanced magnetic resonance imaging) has been used in recent years. It is being used to investigate its effect on brain function and epilepsy in normal controls. However, there is no previous work has studied the effect of oxygen on resting-state fMRI in human. The aim of this study to investigate the difference in activation in the resting-state networks of healthy controls when performing functional scans with and without oxygen enhancement. 26 healthy subjects participated in this study. Significant differences were detected when comparing oxygen enhanced scans versus ambient air scans.

Joe S. Cheng^1,2, Iris Y. Zhou^1,2, Patrick P. Gao^1,2, Russell W. Chan^1,2, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, Hong Kong, ²Department of Electrical and Electronic Engineering, Hong Kong, Hong Kong

Resting-state functional MRI (rsfMRI) utilizing blood-oxygen-level-dependent (BOLD) contrast has increasingly used to map inter and intra-hemispheric connectivity in normal and diseased brains. However BOLD is a complex function of changes in oxygen metabolism, cerebral blood flow (CBF), and cerebral blood volume (CBV). While the effect of various gas inspirations including hypoxia on baseline and stimulus-evoked BOLD signal have been documented, how BOLD synchronization or rsfMRI can be affected by hypoxia has not been investigated. This study aims to assess the effect of graded hypoxia on BOLD rsfMRI connectivity in an animal model and shows that both mild and severe hypoxia suppressed BOLD rsfMRI connectivity significantly.

Elisabeth Jonckers¹, Rafael Delgado y Palacios¹, Disha Shah¹, Caroline Guglielmetti¹, Marleen Verhoye¹, and Annemie Van der Linden^1
1Biomedical Sciences, Bio-Imaging Lab, University of Antwerp, Wilrijk, Belgium

The use of resting state functional MRI (rsfMRI) in preclinical research is expanding progressively, but the majority of resting state imaging is obtained in anesthetized animals. Since it is known that anesthesia may influence the functional connectivity (FC) outcome, we compared rsfMRI data from awake mice with results obtained from mice anesthetized with á-chloralose, urethane or isoflurane (1%). Overall, we can conclude that to study interhemispheric FC using ICA in mice, a 15 component analysis should be used in urethane or á-chloralose anesthetized animals. Moreover awake mouse rsfMRI is proven to be possible and useful but needs to be optimized.

Robert James Anderson¹, Benedikt A. Poser¹, and Victor Andrew Stenger^1
1Dept. of Medicine, University of Hawaii, Honolulu, HI, United States

Simultaneous Multi-Slice (SMS) imaging methods can increase BOLD fMRI acquisition speeds by factors of four and greater. However, fMRI in the inferior brain still remains difficult due to susceptibility induced signal loss in gradient-echo EPI. Recently, spectral spatial (SPSP) pulses have been shown to correct signal loss in a single excitation. Here we extend the SPSP method to SMS excitations and demonstrate Multi-Band (MB) and Power Independent Number of Slices (PINS) SPSP pulses that are capable of exciting multiple signal-loss corrected slices simultaneously. Using breath-hold fMRI at 3T, we demonstrate that the pulses successfully recover lost signal and activation.

Kawin Setsompop^1,2, Stephen F. Cauley³, Himanshu Bhat⁴, Jonathan R. Polimeni^1,2, and Lawrence L. Wald^1,2
1A.A. Martinos Center for Biomedical Imaging, MGH, Charlestown, MA, United States, ²Harvard Medical School, Boston, MA, United States, ³Massachusetts General Hospital, Charlestown, MA, United States, ⁴Siemens Medical Solutions, Charlestown, MA, United States

Simultaneous Multi-Slice (SMS) acquisition with blipped-CAIPI scheme has enabled dramatic reduction in imaging time for fMRI and Diffusion imaging. The signal leakage is an important metric that characterizes signal corruption (due to leakage of signal from one slice to another) for such an acquisition. . In this work, we demonstrate a technique that can be used to rapidly compute signal leakage metrics, and demonstrate two techniques to modifying the slice-GRAPPA (SG) reconstruction to significantly reduce leakage artifact without affecting the g-factor penalty.

David Feinberg^1,2, Liyong Chen^2,3, and An T. Vu^4
1Advanced MRI Technologies, Sebastopol, CA, United States, ²Helen Wills Neuroscience Institute, University of California, Berkeley, CA, United States,³Advanced MRI Technologies, LLC, Sebastopol, CA, United States, ⁴CMRR, University of Minnesota, Minneapolis, MN, United States

Recently the simultaneous multi-slice (SMS) EPI technique which reduces the scan time has been combined with in-plane parallel imaging (e.g. GRAPPA) to achieve high spatial resolution at the cost of increased g-factors, phase drift, motion sensitivity and temporal instability. We introduce Zoomed Multi-Band Imaging (ZOMBI) which combines SMS EPI with outer volume suppression (OVS) instead of GRAPPA and compare against SMS EPI with GRAPPA. We find that ZOMBI achieves high isotropic spatial resolution while maintaining faster TR and improving temporal SNR. This obtainable high resolution is immediately useful for vision science research at 3T and 7T.

Alan Chu¹ and Douglas C. Noll^1
1Biomedical Engineering, University of Michigan, Ann Arbor, MI, United States

We propose a simultaneous multislice spiral imaging method that uses z-gradient modulation during readout for better slice separation. The method uses a multiple-coil receive array and sensitivity maps for the iterative reconstruction. Although the method can be used in a variety of applications, it has particular potential in fMRI because of the inherent advantages of spiral readout and increased efficiency from multislice acquisition. We demonstrate the method in an fMRI scan with promising results.

Valur Olafsson¹, Prantik Kundu², Chi Wah Wong¹, Jia Guo³, Peter A. Bandettini^4,5, Eric Wong¹, and Thomas Liu^1
1Center for Functional MRI, UCSD, La Jolla, CA, United States, ²Section on Functional Imaging Methods, NIMH, Bethesda, MD, United States, ³Center for Functional MRI, University of California San Diego, La Jolla, CA, United States, ⁴Section on Functional Imaging Methods, National Institute of Mental Health, Bethesda, MD, United States, ⁵Functional MRI Facility, NIMH, Bethesda, MD, United States

Resting-state functional connectivity of human subcortical structures is of a particular interest due to its role in common cognitive functions. In this study we investigated the detection performance of subcortical resting-state functional connectivity networks, when using a multi-echo simultaneous multi-slice (MESMS) acquisition to increase the number of volumes collected per scan time, as compared to a multi-echo single slice acquisition. Our results showed that higher acquisition efficiency provided by MESMS provided an increase in the robust detection of subcortical networks.

Zhan Xu¹, Guangyu Chen², Andrew S. Nencka¹, and Shi-Jiang Li^1
1department of biophysics, medical college of wisconsin, milwaukee, wi, United States, ²department of biophysics, Medical College of Wisconsin, milwaukee, wi, United States

By applying multi-slice acquisition technique in resting state fMRI, we find out brain functional connectivity is stronger in this modality compare to standard slice by slice acquisition, in addition to less scan time. By using open source data from U- Minn, in which both Multi-slice acquired data and standard acquired data have same resolution, we did test with under same scan time and same slice number. The multi-slice data displays higher brain connectivity. A reasonable explanation is multi-slice method has higher sample frequency which leads to stronger temporal SNR. We then did a simulation to test our hypothesis

Hing-Chiu Chang¹, Pooja Gaur², Ying-Hui Chou³, and Nan-Kuei Chen^1
1Brain Imaging and Analysis Center, Duke University, Durham, NC, United States, ²Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, United States, ³Brain Imaging and Analysis Center, Duke University Medical Center, Durham, NC, United States

The spatial resolution of fMRI data can be significantly improved with segmented EPI acquisition, which is highly susceptible to undesirable aliasing artifacts originating from subtle subject motion, B0 drifting, and other types of signal inconsistencies among multiple EPI segments of each fMRI volume. These aliasing artifacts lead to increased time-domain signal fluctuation in dynamic segmented EPI scans. A novel algorithm, termed multiplexed parallel reconstruction, to effectively remove undesirable aliasing artifacts in segmented EPI based fMRI was reported to produce high-quality, high-resolution and high-SNR fMRI without relying on any external navigator signal.

Liyong Chen^1,2, An T. Vu^1,2, Junqian Xu³, Kamil Ugurbil³, Essa S. Yacoub³, and David Feinberg^2,4
1Advanced MRI Technologies, LLC, Sebastopol, CA, United States, ²Helen Wills Neuroscience Institute, University of California, Berkeley, CA, United States,³CMRR, University of Minnesota, Minneapolis, MN, United States, ⁴Advanced MRI Technologies, Sebastopol, CA, United States

Multiplexed EPI utilizes multiband (MB) rf pulses and simultaneous image refocusing (SIR) to reduce the sampling TR of fMRI experiments. Here we evaluate different combinations of MB and SIR for accelerations up to N=48 images per EPI echo train. We found that for high accelerations, N>=8, multiplexed EPI with SIR 2 gave improvement over accelerations with MB alone in terms of tSNR, BOLD CNR, and BOLD information content. TRs around 600 – 300 ms gave the best single trial classification accuracy, suggesting significant BOLD information exists at very high temporal frequencies and demonstrating the importance of highly accelerated EPI for fMRI.

Jürgen Finsterbusch^1,2
1Department of Systems Neuroscience, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, ²Neuroimage Nord, Hamburg-Kiel-Lübeck, Germany

A single T2*-weighted echo-planar image of the human brain can be acquired within 50-80ms on standard whole-body MR system. In most fMRI experiments, distributed regions-of-interest are targeted and several slices must usually be acquired. Thus, the experiment’'s temporal resolution is reduced accordingly and the different regions-of-interest are not acquired at identical time points. With a tailored 2D-selective RF excitation, several small target volumes at dedicated locations can be excited and acquired in a single projection image for an appropriate image orientation. Thus, the different regions-of-interest can be acquired simultaneously with the high temporal resolution of a single-slice acquisition.

Zhongwei Chen¹, Rong Xue¹, and Danny J.J. Wang^2
1State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, Chaoyang District, Beijing, China,²Department of Neurology, University of California Los Angeles, Los Angeles, CA, United States

In this study, we systematically evaluated the sensitivity of 4 emerging fMRI methods including fast gradient echo (GRE), passband and transition band balanced steady-state free precession (SSFP), and multiband EPI to visual cortex stimulation at 7 Tesla using both block and event-related designs. Our results showed that GRE and passband SSFP provide the highest functional sensitivity (peak and mean t-value) and can reliably detect initial dip in hemodynamic response function curves. We also found that transition band SSFP has the most sensitivity to physiological noise.

Steve Patterson^1,2 and Chris Bowen^1,2
1Dalhousie University, Halifax, NS, Canada, ²Institute for Biodiagnostics (Atlantic), NRC, Halifax, NS, Canada

Most pass-band SSFP fMRI studies employ the signal-optimizing flip angle (_s), which maximizes on-resonant signal and produces a flat pass-band signal vs. off-resonance profile. In this work we derive an analytical expression for the BOLD contrast-optimizing flip angle (_c). Using literature-reported relaxation times for grey matter, we found that _c ≈ _s + 20^o for B₀ from 1.5-7T. We tested our result against Monte Carlo simulations of SSFP fMRI BOLD contrast at 3T and found that the BOLD contrast-optimizing flip angle maximizes on-resonant BOLD contrast (23% increase from _s) and produces a flat pass-band BOLD contrast vs. off-resonance profile.

Manus J. Donahue¹ and William A. Grissom^2
1Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States, ²Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

The purpose of this study is to develop and implement tailored radiofrequency (RF) excitation and refocusing pulses at 7T to allow for robust detection of R2-weighted spin echo (SE) blood oxygenation level-dependent (BOLD) activity in multiple brain regions. A 150% improvement in spatial specificity and 52% improvement in CNR was demonstrated for 7T SE BOLD performed using tailored spokes and kT point refocusing pulses compared with a standard sequence using sinc-excitation and composite refocusing.

Stephen James Wastling¹, David J. Lythgoe¹, and Gareth John Barker^1
1Department of Neuroimaging, King's College London, London, United Kingdom

Functional MRI data acquired using gradient-echo echo-planar imaging (GE-EPI) suffers from signal-dropout in the orbitofrontal cortex (OFC) and temporal lobes (TL) caused by susceptibility gradients in the slice-selection, Gz,s, phase-encoding, Gy,s, and readout, Gx,s ,directions. For the first time we combine the use of full-passage scaled-down Hyperbolic Secant (HS) excitation pulses and gradient compensation in the readout direction to recover BOLD signal in regions with signal dropout caused by Gz,s and Gx,s. We demonstrate improvements in temporal signal-to-noise ratio and BOLD sensitivity in the OFC and TL of six healthy male volunteers compared to conventional GE-EPI.

Benedikt A. Poser¹, Benjamin Zahneisen¹, Robert James Anderson¹, Markus Barth^2,3, and Victor Andrew Stenger^1
1John A Burns School of Medicine, University of Hawaii, Honolulu, Hawaii, United States, ²Donders Institute for Brain, Cognition and Behaviour, Centre for Cognitive Neuroimaging, Radboud University, Nijmegen, Netherlands, ³Erwin L Hahn Institute for Magnetic Resonance Imaging, University Essen-Duisburg, Essen, Germany

Signal loss in GE-EPI is a problem that compromises the detection of BOLD activation near the air cavities, e.g. in frontal cortex and around the ear canals. The ‘Interleaved Dual-Echo Acquisition”(IDEA EPI) sequence that was recently proposed as a solution to Nyquist ghosting, is modified by incorporating a z-shim. With this we present a novel and effective method that (a) allows through-plane gradient compensation in a single shot, (b) is compatible with high-resolution fMRI and (c) remains insensitive to N/2 ghosting. The approach is evaluated using in vivo scans and breathold fMRI at 1.5x.1.5mm2 resolution; signal voids are drastically reduced.

Brice Fernandez¹, Gavin Houston², Victor Spoormaker³, Marion Smits⁴, Michael Czisch³, and Patrick Le Roux^5
1EMEA Research and Collaboration, Applied Science Laboratory, GE Healthcare, Munich, Germany, ²EMEA Research and Collaboration, Applied Science Laboratory, GE Healthcare, Rotterdam, Netherlands, ³RG Neuroimaging, Max Planck Institute of Psychiatry, Munich, Germany, ⁴Departement of Radiology, Erasmus MC - University Medical Centre Rotterdam, Rotterdam, Netherlands, ⁵EMEA Research and Collaboration, Applied Science Laboratory, GE Healthcare, Palaiseau, France

Hybrid gradient (GRE) and spin-echo (SE) pulse sequences are an interesting alternative to GRE functional MRI (fMRI). In this preliminary work, to minimize the difference between the GRE and the SE profile, two pairs of RF pulses (excitation and refocusing pulses) were designed, implemented and evaluated in a basic visual fMRI experiment. The results shows that a typical profile difference (<6%) between the SE and the GRE profile is not problematic in the context of fMRI. Our results give also some insight for further improvement of the RF pulse pair used in this hybrid pulse sequence.

W. Scott Hoge¹, Hong Pan¹, Ravi T. Seethamraju², Keith A. Heberlein³, and Emily Stern^1
1Radiology, Brigham and Women's Hospital, Boston, MA, United States, ²Siemens Healthcare, Boston, MA, United States, ³Siemens Healthcare USA, Boston, MA, United States

We present a method combining a reduced field of view (rFOV) EPI acquisition sequence with z-shim compensation. Echo planar imaging of the human brain suffers from geometric distortion and signal loss near regions of high magnetic susceptibility, such as near the nasal sinuses. rFOV is effective in reducing in-plane distortion, but requires z-shim compensation to mitigate signal loss from local through-plane gradients. Z-shim compensation combined with rFOV-EPI enables a comprehensive solution to susceptibility artifacts in EPI, and is an important step towards achieving high temporal- and spatial-resolution functional imaging of deep brain structures.

Ryan K. Robison^1,2 and Allen T. Newton^2,3
1Philips Healthcare, Nashville, Tennessee, United States, ²Vanderbilt University Institute of Imaging Science, Vanderbilt University, Nashville, Tennessee, United States, ³Monroe Carell Jr. Children's Hospital at Vanderbilt, Nashville, Tennessee, United States

Readout-segmented EPI is applied herein to functional MRI to reduce the effects of shot-to-shot signal variations. At ultra high field, geometric distortion may compromise the use of single-shot EPI for functional studies, particularly for high spatial resolution. Interleaved multi-shot EPI has been investigated as an alternative but is strongly affected by shot-to-shot signal variations. Improvement in temporal SNR is shown for readout-segmented EPI as compared to interleaved multi-shot EPI. Improved geometric distortion is also shown for readout-segmented EPI as compared to single-shot EPI.

Mario Zeller¹, Andreas J. Bartsch^2,3, Alexander Müller¹, Marcel Gutberlet⁴, Dietbert Hahn¹, and Herbert Köstler^1
1Institute of Radiology, University of Würzburg, Würzburg, Germany, ²Department of Neuroradiology, University of Heidelberg, Heidelberg, Germany,³FMRIB Centre, Oxford University, Oxford, United Kingdom, ⁴Institute for Interventional and Diagnostic Radiology, Hannover Medical School, Hannover, Germany

Echo planar images of conventional Cartesian fMRI are often spatially smoothed by a Gaussian filter prior to statistical analysis. K-space density weighting, on the other hand, allows to already record the data with a Gaussian shaped point spread function. This is achieved by sampling the k-space with a non-Cartesian trajectory. Such density weighting can, at the same time, be optimized for SNR efficiency by applying a matched filter. In this study, fMRI finger-tapping experiments were recorded with both approaches. Density weighting demonstrates significantly increased sensorimotor activations compared to conventional Cartesian imaging. Potential benefits of density weighted fMRI are discussed.

Lars Kasper¹, Maximilian Haeberlin¹, Benjamin E. Dietrich², Simon Gross¹, Christoph Barmet^2,3, Christian C. Ruff⁴, Klaas E. Stephan^1,5, and Klaas P. Pruessmann^1
1Institute for Biomedical Engineering, University & ETH Zurich, Zurich, Switzerland, ²Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland, ³Skope Magnetic Resonance Technologies, Zurich, Switzerland, ⁴Department of Economics, University of Zurich, Zurich, Switzerland,⁵Wellcome Trust Centre for Neuroimaging, University College of London, London, United Kingdom

Matched-filter fMRI utilizes 2D-density weighted EPI to increase BOLD sensitivity. It matches acquisition density in k-space to the desired spatial response – typically a Gaussian kernel. Theory predicts significant SNR benefits from such acquisitions, which are maximal when thermal noise dominates and expected to decrease with higher physiological noise contributions. We explore the validity of this argument for different regimes of physiological noise in the brain and report consistent and replicable SNR increases of 20-40 % compared to uniform EPI acquisitions. For task-based fMRI, we observe a consistent increase in BOLD-sensitivity of 30 % (average t-value), and show reproducibility both within and between subjects (N=4).

Jon-Fredrik Nielsen¹, Hao Sun², Jeffrey A. Fessler², and Douglas C. Noll^1
1Biomedical Engineering, University of Michigan, Ann Arbor, Michigan, United States, ²Electrical Engineering and Computer Science, University of Michigan, Ann Arbor, Michigan, United States

"Small-tip fast recovery" (STFR) is a recently-proposed steady-state imaging sequence that is a potential alternative to balanced SSFP. STFR relies on a tailored "tip-up" pulse to produce similar image contrast as bSSFP but with reduced banding artifacts. We evaluate the use of this new sequence for functional MRI using Monte Carlo Bloch simulations, and fMRI observations in visual cortex. Our results indicate that STFR can produce enhanced functional contrast compared to passband bSSFP.

Jingwei Zhang¹, Mengchao Pei¹, Tian Liu^1,2, Ajay Gupta¹, Cynthia Wisnieff¹, Pina C. Sanelli,¹, Pascal Spincemaille¹, and Yi Wang^1
1Department of Radiology, Weill Medical College of Cornell University, New York, NY, United States, ²MedImageMetric LLC, New York, NY, United States

Accurate measurement of cerebral metabolic rate of oxygen (CMRO2) is highly desired for the assessment of brain cell function in health and in stroke.MRI offers the potential to map CMRO2 by estimating paragmagnetic deoxyhemoglobin concentration ([dHb]) from detected signal, which requires MRI signal modeling. We propose to use a rigorous deconvolution technique called quantitative susceptibility mapping (QSM) to map [dHb] and hence CMRO2.

Bryan Haddock¹, Henrik B.W. Larsson¹, Adam E. Hansen¹, and Egill Rostrup^1,2
1Department of Diagnostics, Glostrup Hospital, Glostrup, Denmark, ²Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark

We evaluated changes in intra- and extravascular brain oxygenation during oxygen challenge on 11 subjects with a new MRI technique (TOLD). A model is presented to quantify extravascular changes in oxygen tension from the TOLD signal. We Found an increase in extravascular tissue oxygenation for white and grey matter. Our Model calculates a change in oxygen tension close to literature values from data. TOLD and BOLD data sets for the same breathing challenge are compared revealing intravascular changes in oxygen tension is more rapid than extravascular responses.

Christian Federau¹, Kieran O'Brien², Adrien Birbaumer¹, Reto Meuli¹, Patric Hagmann¹, and Philippe M. Maeder^1
1Radiology, CHUV, Lausanne, VD, Switzerland, ²CIBM, Université de Genève, Lausanne, VD, Switzerland

Intravoxel Incoherent Motion MRI allows quantitative measurement of microvascular blood flow, and could therefore potentially have a higher spatial correlation with neuronal activation than the currently used BOLD technique. We demonstrate functional imaging with IVIM in the visual human brain, and present quantitative analysis showing increase in IVIM perfusion parameters during visual stimulation, in both the visual cortex, as well as, to a lesser extend, in the subcortical white matter.

Tae Kim¹, Seong-Gi Kim¹, and Kyongtae Ty Bae^1
1University of Pittsburgh, Pittsburgh, PA, United States

Cerebral blood flow (CBF) and arterial blood volume (CBVa) in baseline and their changes associated with visual stimulation in humans were successfully measured using multi-slice FAIR with bipolar gradients at 3T. CBF and CBVa values were dynamically synchronous and highly overlapping during visual stimulation, suggesting that the dilation of CBVa is the major contributor to an increase in CBF in normal physiology. Simultaneous quantification of CBV and CBVa without the use of exogenous contrast agents is highly useful and clinical significant for the assessment of cerebrovascular viability and pathological conditions.

Valerie E M Griffeth¹, Nicholas P. Blockley², Aaron Simon³, and Richard B. Buxton^4
1Bioengineering, University of California San Diego, La Jolla, CA, United States, ²FMRIB, University of Oxford, Oxford, United Kingdom, ³Bioengineering, University of California, San Diego, La Jolla, CA, United States, ⁴Radiology, University of California San Diego, La Jolla, CA, United States

The simple and straightforward “ratio method” can be used test whether the CBF-CMRO₂ coupling ratio is the same for two different stimulus responses using only combined CBF and BOLD measurements, independent of model parameters, and without an additional calibration experiment. We tested the ratio method using a previously published detailed BOLD model demonstrating that it is an effective approach for comparing stimulus responses in order to differentiate between values of CBF-CMRO₂ coupling. We also include a reanalysis of recently published data on varying visual stimulus contrast showing consistency of these results in comparison to the traditional Davis model approach.

Laurentius Huber¹, Dimo Ivanov^2,3, Steffen N. Krieger², Claudine Joëlle Gauthier², Elisabeth Roggenhofer⁴, Ilona Henseler⁴, Robert Turner², and Harald E. Moeller^1
1NMR-Unit, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, ²Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, ³Cognitive and Clinical Neuroscience, Maastricht University, Maastricht, Netherlands, ⁴Neurology, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

VAscular Space Occupancy (VASO) is the most widely used method to investigate human functional cerebral blood volume (CBV) changes. These measurements are limited at high field strengths by low signal-to-noise ratio (SNR) or inflow effects. A high-SNR VASO variant is introduced that can avoid inflow effects of fresh blood by using a newly-designed partial inversion pulse at 7T. The high SNR of the developed method was utilized to obtain voxel-wise estimates of the calibration factor M and CMRO2 changes on a single-subject basis.

Felipe B. Tancredi^1,2 and Richard D. Hoge^1,2
1Institut de génie biomédical, Université de Montréal, Montreal, Quebec, Canada, ²Unité de neuroimagerie fonctionnelle, CRIUGM, Montreal, Quebec, Canada

ASL measurements of the cerebrovascular reactivity to CO₂ (CVR) provide an index of the cerebral vasculature’s health status and are also an important input for calibrated MRI. CVR is expressed as the increase in ASL-flow signal that is induced per unit increase in respiratory levels of CO₂. We examined whether different respiratory manipulations lead to consistent measures of CVR. We found that because of the non-linearity of CO₂-CBF dose-response curve, CVR values differ according to the manipulated levels of CO₂. However, estimates tend to converge when CVR is expressed as the percent increase in flow signal.

Manus J. Donahue¹ and Swati Rane^1
1Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, United States

The purpose of this work is to outline a quantitative framework for interpreting the primarily correlative relationships that have very recently been reported between blood oxygenation level-dependent (BOLD) contrast and ongoing inhibitory neuronal activity measured by baseline γ-aminobutyric acid (GABA) concentration.

Ashley D. Harris¹, Kevin Murphy¹, and Richard G. Wise^1
1CUBRIC - School of Psychology, Cardiff University, Cardiff, United Kingdom

Using a working memory task, the modulation the BOLD response at different resting CBF is examined in regions associated with the task activation and deactivation. Task activation is reduced with increased baseline CBF as expected from modelling. Task deactivation differs regionally; the default mode network has increased task deactivation while other task-negative regions show less deactivation.

Kevin Murphy¹, Ashley D. Harris¹, and Richard G. Wise^1
1CUBRIC, School of Psychology, Cardiff University, Cardiff, Wales, United Kingdom

Hypercapnia is often used in calibrated BOLD techniques to determine the maximum possible BOLD response. The assumption is that an increase in arterial CO₂ tension will not affect CMRO₂, a measure of brain metabolism. This assumption has been called into question with studies showing conflicting results. A framework is presented that models changes in CMRO₂ caused by hypercapnia at two different CO₂ levels: +4mmHg and +8mmHg. Decreases in CMRO₂during hypercapnia were observed but only for the +8mmHg condition. This suggests that a low level of hypercapnia (~4mmHg) must be used for the assumption of calibrated BOLD to hold.

Alan J. Stone¹, Kevin Murphy¹, Ashley D. Harris¹, and Richard G. Wise^1
1Cardiff University Brain Research Imaging Centre, School of Psychology, Cardiff University, Cardiff, United Kingdom

We present a refined method for hyperoxic and hypercapnic calibration of BOLD signal that yields measurements of the absolute rate of cerebral oxygen consumption (CMRO2). We demonstrate that the measurements can be obtained in 9 mins, are repeatable within a scan session and are sensitive to sustained (18 min) task-related increases in absolute CMRO2. The sensitivity to sustained changes in CMRO2 indicates that the method proposed is promising for clinical or pharmacological applications in which CMRO2 is altered.

Molly G. Bright¹ and Kevin Murphy^1
1CUBRIC, School of Psychology, Cardiff University, Cardiff, Wales, United Kingdom

Resting state BOLD fMRI is a powerful technique for probing brain function. However, it is difficult to characterize these fluctuations using arterial spin labeling. We present a new method of mapping changes in CBF that are coupled to BOLD fluctuations, rather than linked to a modeled stimulus, within the framework of a general linear model and utilizing a dual-echo sequence to capture ASL and BOLD data simultaneously. This is a more sensitive method for mapping coupled CBF and BOLD changes throughout the brain than conventional correlation approaches, and also improves the robustness of CBF statistical maps during a known stimulus.

Sungho Tak¹, Danny J.J. Wang², Lirong Yan², and J. Jean Chen^3
1Rotman Research Institute at Baycrest Centre, University of Toronto, Toronto, ON, Canada, ²Neurology, University of California, Los Angeles (UCLA), Los Angeles, CA, United States, ³University of Toronto, Toronto, ON, Canada

In this study, we investigated the contribution of cerebral blood flow (CBF) fluctuations to the resting-state BOLD signals. A dual-echo pseudo-continuous spin labeling (pCASL) method was used to simultaneously measure the BOLD and CBF responses, and a multivariate general linear model was applied to statistically analyze the relationship. We accounted for the time lag between BOLD and CBF, and modeled systemic physiological noise separately. Results show that the low-frequency fluctuations of the CBF and BOLD signals are significantly correlated but spatially variable, with particularly strong CBF contributions across major resting-state networks, including the default mode and anti-correlated networks.

Maryam Falahpour^1,2, Hazem Refai², and Jerzy Bodurka^1,3
1Laureate Institute for Brain Research, Tulsa, Oklahoma, United States, ²Electrical and Computer Engineering, University of Oklahoma, Tulsa, Oklahoma, United States, ³College of Engineering, University of Oklahoma, Tulsa, Oklahoma, United States

Slow changes in the heart rate and breathing pattern alter BOLD fMRI and can be misinterpreted as neural activation. It is common to employ respiratory and cardiac response functions (RRF, CRF) to map and remove the low-frequency physiological fluctuations from BOLD. However, both RRF and CRF have been analytically defined based on the average of multi-subject data and don’t account for intra-subject differences. We propose a novel approach that employs subject-specific, individualized RRFi and CRFi obtained from the whole brain global signal (GS). Our subject-specific approach explains more variance in data without creating the problems caused by GS regression.

Marta Bianciardi¹, Karleyton C. Evans², Jonathan R. Polimeni¹, Tian Y. Song², Boris Keil¹, Christina Triantafyllou¹, Bruce R. Rosen¹, David A. Boas¹, and Lawrence L. Wald^1
1Department of Radiology, A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States, ²Department of Psychiatry, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States

Respiration generates fMRI signal instability in the brain by different mechanisms, including chest motion, and pulsatile motion of blood/CSF due to respiratory-pressure-wave. We investigated the relative contribution of these effects, their spatial distribution, and the performance of different correction methods by acquiring whole-brain high-spatial-resolution magnitude and phase fMRI resting-state data at 7Tesla. Our results indicate that the major sources of respiratory-related signal instability are: chest motion in phase-data, and pulsatile motion in magnitude-data. Optimized correction of respiratory effects in both magnitude- and phase-data was achieved by combining RETROICOR with a method for estimating off-resonance changes in phase-data.

Andrew T. Curtis¹ and Ravi S. Menon^1
1Medical Biophysics, University of Western Ontario, London, Ontario, Canada

With the popular development of multi-channel, form fitting receive coils used in fMRI acquisitions, compromises in the EPI image reconstruction process result in a non-zero residual zero-order phase of a complex valued fMRI time-series. This phase is sensitive to residual motion and artifacts. It is proposed to use this signal as a reference to identity and remove artifact components in ICA based filtering.

Vadim Zotev¹, Han Yuan¹, Masaya Misaki¹, Raquel Phillips¹, Kymberly D. Young¹, and Jerzy Bodurka^1,2
1Laureate Institute for Brain Research, Tulsa, OK, United States, ²College of Engineering, University of Oklahoma, Tulsa, OK, United States

We have applied EEG-assisted retrospective motion correction (E-REMCOR) to evaluate effects of rapid head movements on fMRI functional connectivity data. Twenty MDD patients underwent a resting EEG-fMRI scan. Functional connectivity patterns for seed ROI in the medial prefrontal cortex obtained with and without E-REMCOR were compared. The comparison shows that rapid head movements increase the apparent connectivity strength with the seed ROI for anterior brain regions and reduce it for posterior regions. Our results confirm that effects of head motions are observed in group-level fMRI analysis. They further suggest that E-REMCOR is an efficient motion correction technique for fMRI.

Sharmishtaa Seshamani¹, Chris Gatenby², Mads Fogtmann¹, Xi Cheng¹, Manjiri Dighe², and Colin Studholme^1
1Pediatrics, Bioengineering, Radiology, University of Washington, Seattle, WA, United States, ²Radiology, University of Washington, Seattle, WA, United States

This abstract presents a method for combining R2* mapping with volume and slice correction to address motion and slice spin history artifacts in fMRI data of moving subjects. Experiments are presented on an adult with and without controlled motion, using a finger tapping paradigm. Results show that signal changes in the R2* map of the moving subject do not suffer from the severe motion artifacts present in the T2* weighted images. This is a very promising approach for both task based and resting state analysis of moving subjects.

Daniel J. Sheltraw¹ and Ben Inglis^1
1Henry H. Wheeler Jr. Brain Imaging Center, University of California Berkeley, Berkeley, CA, United States

The combination of receive field contrast and motion correction can introduce temporal modulations in fMRI data. In this work we report on simulations which estimate the size of this effect and its potential consequences to fMRI data acquired using a 32-channel head array.

Arno Solin¹, Simo Särkkä¹, Aapo Nummenmaa², Aki Vehtari¹, Toni Auranen³, Simo Vanni^3,4, and Fa-Hsuan Lin^1,5
1Department of Biomedical Engineering and Computational Science, Aalto University, Espoo, Finland, ²Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Boston, MA, United States, ³Advanced Magnetic Imaging Centre, Aalto University, Espoo, Finland, ⁴Brain Research Unit, O.V. Lounasmaa Laboratory, Aalto University, Espoo, Finland, ⁵Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan

In this work we form volumetric temporal phase and amplitude maps of physiological oscillatory signals using fMRI data. We study the spatio-temporal structure of respiration and heart beat induced noises in the brain. The separation of the fMRI data into components was done with a Kalman filter and smoother (the DRIFTER method) using physiological reference signals for identifying the frequencies. We present proof-of-concept maps for one test subject, and discuss how this prior information can be applied to removal or estimation of quickly varying noise components in slow fMRI, and utilized in future fast-imaging techniques.

Christina Triantafyllou¹, Jonathan R. Polimeni¹, Boris Keil¹, and Lawrence L. Wald^1,2
1A.A. Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States, ²Health Science and Technology, MIT, Cambridge, MA, United States

Sensitivity in BOLD fMRI is characterized by the time-series SNR, which contains fluctuations from thermal and physiological noise sources. In this work we investigate the need for a modified model describing the behavior of physiological noise in the fMRI time-series in multi-channel acquisitions and assess the new model’s dependence on the physiological noise correlations, tissue class and coil combination methods across field strengths of 3T and 7T. Our findings demonstrate that the proposed model could be used to characterize multi-channel array acquisitions at high and ultra-high field strengths and ultimately to optimize fMRI protocols towards maximizing tSNR.

Jonathan R. Polimeni¹, Kawin Setsompop¹, Christina Triantafyllou¹, and Lawrence L. Wald^1,2
1Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Harvard Medical School, Massachusetts General Hospital, Charlestown, MA, United States, ²Harvard-MIT Division of Health Sciences and Technology, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States

Here we introduce a coil combination method that accounts for the spatially-varying noise covariance matrix—present in both accelerated imaging and in fMRI time-series—to boost SNR. Because this matrix must be inverted and its rank varies spatially, a per-voxel regularization is required to yield SNR gains. We demonstrate an improvement in image SNR for GRAPPA-reconstructed phantom data and a tissue-specific improvement in time-series SNR for resting-state fMRI data. These results can also provide insight into the non-stationarity of physiological noise covariance across coil channels.

Christine Law^1
1University of Oxford, Oxford, Oxfordshire, United Kingdom

The conventional technique for physiological noise reduction in fMRI uses Fourier analysis over long duration usually spanning the entire record. We propose, instead, a sliding window technique (cardioX) that more closely approximates the cardiac component variation over the short term. CardioX removes more cardiac energy in its frequency band than the coventional method.

Deqiang Qiu¹, Greg Zaharchuk¹, Thomas Christen¹, Wendy W. Ni¹, and Michael E. Moseley^1
1Radiology, Stanford University, Stanford, CA, United States

In this study, we characterize the noise behaviour of post contrast fMRI (termed ICE-BVI) using the ultrasmall superparamagnetic iron particle (USPIO) under different echo times and static SNR. The result is useful in guiding study designs in task based and resting state ICE-BVI. It suggests that higher resolution task based ICE-BVI can be achieved without significantly sacrificing sensitivity, and that effective methods for reducing physiological noises are important for resting state ICE-BVI for studying functional connectivity.

Helen Erica D'Arceuil¹, Alexandre Coimbra², Pamela Triano³, Margaret Dougherty³, Julie Mello³, Michael E. Moseley⁴, Gary H. Glover⁵, Maarten Lansberg⁶, and Francis Blankenberg^7
1Diagnostic Radiology, Stanford, Stanford, CA, United States, ²Genentech Inc, South San Francisco, CA, United States, ³Department of Rehabilitation Services, Stanford Hospital and Clinics, Stanford, CA, United States, ⁴Diagnostic Radiology, Stanford University, Stanford, CA, United States, ⁵Radiology, Stanford University, Stanford, CA, United States, ⁶Neurology and Neurological Sciences, Stanford Stroke Center, Stanford Hospital and Clinics, Stanford, CA, United States, ⁷Pediatric Radiology, Stanford Hospital and Clinics, Palo Alto, CA, United States

Enhanced task-based fMRI signals in primate and human brains, compared to the BOLD effect, have been realized using Ferumoxytol. We explored the use of ferumoxytol as a susceptibility contrast agent for CBV weighted resting state fMRI in 6 volunteers at 3T. While the amplitude of low frequency fluctuations was greater post ferumoxytol, the mean z-score and volume of visual and default mode networks were similar for BOLD and CBV weighted scans. CBV-RS-fMRI is robust and may be suitable for use in neurologically impaired or head trauma patients where regional tissue loss renders task FMRI problematic or even untenable.

Jie Huang^1
1Department of Radiology, Michigan State University, East Lansing, MI, United States

Neural activity associated local neuronal currents produce weak transient magnetic fields that would attenuate local MR signal intensity. There is not yet consensus as to whether the size of this attenuation is detectable with the current MRI techniques. Some groups have reported a successful detection of neuronal currents using MRI, but other groups failed to detect the activity. This study investigates the magnitude of the neuronal current-induced signal attenuation using a temporally well-controlled visual stimulation paradigm. The results show that the visual stimulus-evoked neuronal currents in the visual cortex do not induce detectable signal attenuation with the current MRI technique.

Carlos Ernesto Garrido Salmon¹ and Danilo Maziero^1
1Department of Physics, University of Sao Paulo, FFCLRP, Ribeirao Preto, Brazil

In this study we model the magnitude and phase effect of one metallic wire with/without current inside an infinite medium during a single EPI sequence. We carry out an experimental comparison of the MR signal obtained in two solutions with the same magnetic permeability but different viscosities. The wire position inside the voxel and the mechanical properties of the medium have a strong influence in the magnitude and phase of the MRI signal in an in vitro ncMRI experiment. We conclude that mechanical effect could be dominant in very low current and low viscosity phantoms.

Mukund Balasubramanian¹, Padma Sundaram¹, William M. Wells, III², Robert V. Mulkern¹, and Darren B. Orbach^1
1Radiology, Boston Children's Hospital, Boston, Massachusetts, United States, ²Radiology, Brigham and Women's Hospital, Boston, Massachusetts, United States

Understanding the behavior of ionic volume currents in the presence of strong magnetic fields could assist efforts to use functional MRI to detect epileptiform discharges, especially when these discharges occur near cerebrospinal fluid spaces such as the lateral ventricles. However, studies on this topic are controversial: two competing mechanisms have been proposed to account for the behavior of these currents—the Lorentz effect model and the magnetohydrodynamic (MHD) model. Here, we address this controversy by directly testing the MHD hypothesis using an ionic current phantom and currents of varying duration. Our results provide experimental support for the MHD hypothesis.