Preclinical fMRI

Tuesday 2 June 2015

Arabinda Mishra¹, Feng Wang¹, John C Gore¹, and Chen Min Li^1
1Radiology, Vanderbilt University, Nashville, TN, United States

Spinal cord is primarily responsible for transmitting sensory and motor information to and from the brain. Disruptions following injury on the spinal cord affect the functional integrity of somatic sensory networks. This study aims to determine whether the disruption of afferent inputs from the cervical spinal cord alter the fine - scale functional connectivity between sub-regions of SI cortex in the resting state. We performed an ROI based analysis of resting state BOLD MRI signals obtained in pre- versus post- lesion conditions in anesthetized squirrel monkeys at 9.4T to quantify the inter-regional functional connectivity differences between the sub-regions of the SI cortex.

Peter Herman¹, Basavaraju G. Sanganahalli¹, Daniel Coman¹, Hal Blumenfeld², Lihong Jiang¹, Douglas L. Rothman^1,3, and Fahmeed Hyder^1,3
1Diagnostic Radiology, Yale University, New Haven, Connecticut, United States, ²Neurobiology, Yale University, New Haven, Connecticut, United States,³Biomedical Engineering, Yale University, New Haven, Connecticut, United States

Resting-state fMRI (R-fMRI) is powerful for mapping networks. Since oscillations in metabolic/neural events are linked to R-fMRI networks and studies suggest that absolute metabolic/neural baseline interacts with evoked/spontaneous signals, we measured blood flow, neural activity, glucose oxidation, glutamatergic neurotransmission, and BOLD in relation to two states. Fluctuations in metabolic/neural activities underlying connectivity maps, regardless of the state, represented at most 5% of the total baseline metabolic/neural level. Functional correlation density (FCD) maps, not seed-based correlation maps, showed significant alteration between the states in accord with other absolute measures of states, signifying the importance of FCD and absolute baseline in R-fMRI.

Sheng-Min Huang¹, Wei-Cheng Lee¹, Kung-Chu Ho², and Fu-Nien Wang^1
1Department of Biomedical Engineering & Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan, ²Nuclear Medicine, Chang Gung Memorial Hospital, Taoyuan, Taiwan

Resting state functional MRI is an emerging neuroimaging method, examining the connectivity of brain neural circuits. In this work a seed-based analysis was carried out to compare the resting state network between normal (WKY) and ADHD rat model (SHR). Results show that ADHD rat presented a network contains caudate putamen region but WKY didn't. This difference may be tied to the symptoms of ADHD since putamen is involved in the regulation of motor behavior and caudate nucleus is related to social behavior.

Zhentao Zuo¹, Xudong Zhao¹, Yifan Miao², Shuo Shen¹, Zuxiang Liu¹, and Yuanye Ma^1
1Chinese Academy of Sciences, Institute of Biophysics, Beijing, Beijing, China, ²State Key lab of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, Beijing, Beijing, China

Resting state-fMRI (rs-fMRI) allows studying aesthesis status through functional connectivity (FC) and amplitude low-frequency fluctuation (ALFF), as well as the mechanisms of different neurological diseases. Several monkeys scanner with different anesthesia depths underlying rs-fMRI were scanned to explore the variation of the monkey brain FC patterns and brain region neuron fluctuation. We found different drip concentrations will affect the monkey anesthesia depth. Higher concentration suppresses more neurons activity and also decreases the connections between different cerebral hemispheres and surround regions. It also opened a new method to explore how the intrinsic network created, and whether the connectome dependent on the neuron activity.

Sheng-Min Huang¹, Wei-Cheng Lee¹, Kung-Chu Ho², and Fu-Nien Wang^1
1Department of Biomedical Engineering & Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan, ²Nuclear Medicine, Chang Gung Memorial Hospital, Taoyuan, Taiwan

In this abstract the resistance to isoflurane anesthesia between ADHD model (SHR) and normal control (WKY) was compared by using resting state functional MRI. A different distribution of remaining network under high dose anesthesia was found. WKY group tends to preserve the frontal region of brain, while SHR group keep the central region. Since we can speculate that the connections in high anesthesia level are very fundamental networks, these basic differences between SHR and WKY may be one of the origins of functional abnormality in ADHD rat.

Tuo Shi¹, Oleg Pustovyy², Yun Wang¹, Paul Waggoner³, Ronald Beyers¹, Jessica Fleming⁴, Paul Hammond⁴, Edward Morrison², Thomas S. Denney Jr.^1,5, Vitaly Vodyanoy², and Gopikrishna Deshpande^1,5
1AU MRI Research Center, Dept. of Electrical & Computer Engineering, Auburn University, Auburn, AL, United States, ²Dept. of Anatomy, Physiology & Pharmacology, Auburn University, Auburn, AL, United States, ³Canine Detection Research Institute, Auburn University, Auburn, AL, United States, ⁴iK9 LLC, Auburn, AL, United States, ⁵Dept. of Psychology, Auburn University, Auburn, AL, United States

Humans have long made use of olfactory detection capabilities of dogs. Training dogs to use those capabilities can be expensive due to the manpower and time involved. Therefore, predicting the training ease and behavioral of working dogs before their recruitment will be beneficial. We explored the canine olfactory system using functional MRI and correlated the canine behavior with imaging metrics derived from activation to discriminative odors, to predict dogs’ training ease and behavior. We found the discriminative odorant stimulus lead to higher activity in olfaction-related and higher order brain areas, also were significantly correlated with integrated behavior and training ease.

Jevin W. Zhang^1,2, Patrick P. Gao^1,2, Shu-Juan Fan^1,2, Dan H. Sanes³, 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, ³Center for Neural Science, New York University, New York, NY, United States

The auditory cortex (AC) is the source of one of the largest inputs to the inferior colliculus (IC). Normal IC exhibits stronger BOLD response to forward vocalizations than to the temporally inverted one. But this responses selectivity to behaviorally relevant sound requires AC inputs. After bilateral AC ablation, the BOLD response difference between the two vocalizations in the IC was diminished. In the unilateral (right side) AC ablation animals, the BOLD response difference between the two vocalizations in the ablation (right) side was negligible while the contralateral (left) ECIC still showed a stronger response to the forward vocalization.

Daniel Albaugh^1,2, Garret Stuber³, and Yen-Yu Ian Shih^4
1Curriculum in Neurobiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States, ²Biomedical Imaging Research Center, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States, ³Department of Psychiatry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States, ⁴BRIC, Department of Neurology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, United States

Deep brain stimulation (DBS) of the nucleus accumbens is a promising therapy for a wide variety of intractable neuropsychiatric disorders. Here, we have employed an animal model of DBS, with simultaneous fMRI, to decipher the neural circuitry modulated by accumbens DBS.

Patrick P. Gao^1,2, Jevin W. Zhang^1,2, Shu-Juan Fan^1,2, Dan H. Sanes³, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, HKSAR, China, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, HKSAR, China, ³Center for Neural Science, New York University, New York, NY, United States

The cortex is commonly thought of as the site at which ascending projections from all sensory modalities are integrated, yet cortical feedback to subcortical nuclei modulates early information processing. Here, we demonstrate that descending inputs from both auditory and visual cortex are integrated in the auditory midbrain. Using BOLD fMRI to measure sound-evoked responses throughout the auditory midbrain, we show that auditory cortical input normally suppresses the gain of midbrain response, while visual cortical input increases the gain. Our results demonstrate the large-scale influence of cortical projections from more than one sensory modality, demonstrating that while ascending integration occurs in cortex, descending integration occurs in the brainstem.

Condon Lau¹, Jevin W Zhang², and Ed X Wu^2
1Department of Physics and Materials Science, City University of Hong Kong, Kowloon, Hong Kong SAR, Hong Kong, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, Hong Kong SAR, Hong Kong

Long-term acoustic noise exposure at moderate sound pressure levels (SPLs) (within occupational limits) can be detrimental to speech intelligibility by altering its spectrotemporal representation in the auditory system. To test this, we performed fMRI on adult rats exposed to 65dB SPL noise for two months. The results show noise exposure reduces fMRI signals in the auditory thalamus and cortex. The reduction is greater during 10 than 5Hz pulse rate acoustic stimulation. These findings are important for speech processing, which depends on accurate processing of sounds with a wide spectrum of rates.

Tomokazu Tsurugizawa^1,2, Yukari Takahashi¹, Akihiko Kitamura^1,3, and Fusao Kato^1
1Jikei University School of Medicine, Tokyo, Japan, ²NeuroSpin/CEA, Gif-sur-Yvette, Essone, France, ³Ajinomoto Co. Inc., Kawasaki, Japan

We aimed to investigate the dose-dependence of sevoflurane on BOLD responses in the upper layer (layer 1-3) and in the lower layer (layer 4-6) of the somatosensory cortex to electrical stimulation in the hindpaw. The BOLD responses in the upper layer were greater than those in the lower layer. The estimation of hemodynamic response indicates that time-courses of BOLD responses were dependent on the concentration of sevoflurane, indicating the disturbance of neurovascular coupling by sevoflurane dos-dependently.

Dávid Z Balla¹, Rolf Pohmann¹, Shajan G¹, Philipp Ehses¹, Arno Nauerth², Thomas Steudel¹, Yusuke Murayama¹, Axel Oeltermann¹, Matthias H Munk¹, Hellmut Merkle¹, Michael Beyerlein¹, Henry C Evrard¹, Nikos K Logothetis¹, and Klaus Scheffler^1
1Max Planck Insitute for Biological Cybernetics, Tübingen, Germany, ²Bruker Biospin GmbH, Ettlingen, Germany

We developed fast MRI methods facilitating concurrent electrophysiological recordings and fMRI with high spatio-temporal resolution (500ms / 750µm) and full brain coverage in macaques. The significant improvements in MR signal sampling efficiency without and with the use of parallel imaging are demonstrated by presenting activation maps and time-courses of fMRI experiments using a simple visual stimulation paradigm.

fMRI Methods

Tuesday 2 June 2015

Anthony N. Price^1,2, Lucilio Cordero-Grande^1,2, Shaihan J. Malik^1,2, Maryam Abaei¹, Tomoki Arichi¹, Emer J. Hughes¹, Daniel Rueckert³, A. David Edwards¹, and Joseph V. Hajnal^1,2
1Centre for the Developing Brain, King's College London, London, United Kingdom, ²Division of Imaging Sciences & Biomedical Engineering, King's College London, London, United Kingdom, ³Biomedical Image Analysis Group, Imperial College London, London, United Kingdom

Simultaneous multi-slice EPI acquisitions using multiband (MB) excitation have recently been demonstrated to provide high resolution fMRI data with significantly reduced acquisition time, sufficient enough to directly determine cardiac and respiratory influences from spontaneous activation in the brain. These have mainly been applied to adult subjects, but here we demonstrate MB EPI applied to preterm infants with cardiac rates exceeding 160 bpm. Resting state fMRI data collected with MB factors of 9 demonstrate it is possible to distinguish the neonatal cardiac physiology from activations in resting state networks.

Vincent Jansen¹, Rasim Boyacioglu¹, Jenni Schulz¹, and David G Norris^1,2
1Radboud University, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, Netherlands, ²University Duisburg-Essen, Erwin L. Hahn Institute for Magnetic Resonance Imaging, Essen, Germany

Following a recent implementation1 of a Multi-Band2 Multi-Echo3 (MBME) sequence at 7T for resting state fMRI that showed improved sensitivity, the potential benefits of MBME with respect to a single band multi-echo (SBME) were investigated at 3T using an event related design. : Implementation of Multi-Band in a Multi-Echo sequence shows improved sensitivity in several GM areas and benefits more from automatic non-BOLD related signal removal than a standard Multi-Echo sequence in a standard resolution, event related design at 3T.

Russell W. Chan^1,2, Alex T.L. Leong^1,2, Joe S. Cheng^1,2, Victor B. Xie^1,2, Partick P. Gao^1,2, Aaron Mok², Kevin K. Tsia², and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, China, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, China

Previously, it has been reported that blue light delivery in the naïve brain resulted in pseudo positive and negative fMRI responses, which were attributed to NMR frequency shifts, and T1 and T2* changes. EPI is known to be susceptible to NMR frequency shifting, field inhomogeneity and T2* changes. However, the effects of blue light delivery in the naïve brain and EPI distortion have not been examined. The aim of this study was to investigate the local EPI distortion induced by laser stimulation in the naïve posterior thalamus. Our results showed that the pseudo negative fMRI response was at the fiber tip, while the pseudo positive fMRI response was adjacent to the pseudo negative fMRI response along the phase encoding direction. This study brings attention to EPI distortion as a possible confounder which must be taken into account when optogenetic fMRI experiments are designed.

Rita G. Nunes^1,2, Giulio Ferrazzi¹, Anthony Price¹, Matthew Fox¹, Christina Malamateniou¹, Mary Rutherford¹, and Joseph Hajnal^1,3
1Centre for the Developing Brain, King's College London, London, United Kingdom, ²Instituto de Biofisica e Engenharia Biomedica, Universidade de Lisboa, Lisbon, Portugal, ³Division of Imaging and Sciences and Biomedical Engineerin, King's College London, London, United Kingdom

Initial fMRI studies of the fetal brain have used standard multi-slice 2D EPI sequences and either required intra-volume motion correction or led to a high data rejection rate. We propose combining an EVI readout with localized excitation of the fetal brain to achieve a TR in the order of 150ms therefore significantly reducing sensitivity to motion. This methodology was shown to be feasible in three adult volunteers and a first acquisition test was performed on the fetal brain in utero.

Yuankui Wu^1,2, David Woods³, Moshe T. Stern⁴, Nicholas I.S. Blair⁵, Raag D. Airan⁶, James J. Pekar^1,7, Peter C. M. van Zijl^1,7, and Jun Hua^1,7
1Neurosection, Div. of MRI Research, Dept. of Radiology, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States, ²Department of Medical Imaging, Nanfang Hospital, Southern Medical University, Guangzhou, Guangdong, China, ³Department of Orthodontics and Pediatric Dentistry, University of Maryland School of Dentistry, Baltimore, Maryland, United States, ⁴Department of Orthodontics and Pediatric Dentistry, University of Maryland, Baltimore, Maryland, United States, ⁵Department of Biomedical Engineering, Johns Hopkins University, Baltimore, Maryland, United States, ⁶Div. of Neuroradiology, Dept. of Radiology, Johns Hopkins University School of Medicine, Baltimore, Maryland, United States, ⁷F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, Maryland, United States

Gradient echo (GRE) EPI BOLD sequences are sensitive to susceptibility effects and render signal dropouts in the brain in the presence of metallic objects such as dental braces. This impedes the application of fMRI especially for studies involving teenage participants. T2prep-BOLD sequences use a readout similar to MPRAGE, which is much less sensitive to susceptibility effects. Here, we compare GRE-EPI and T2prep-BOLD in whole brain resting state fMRI scans in the presence of metallic orthodontic braces. T2prep-BOLD scans showed minimal dropout in the whole brain, and greater temporal SNR and BOLD sensitivity in the EPI dropout regions than GRE-EPI BOLD.

Mark Chiew¹, Nadine N Graedel¹, Stephen M Smith¹, and Karla L Miller^1
1FMRIB Centre, University of Oxford, Oxford, Oxfordshire, United Kingdom

We demonstrate application of k-t FASTER, a low-rank acceleration technique designed for FMRI data, on visual task FMRI data acquisition. Here, we used a 3D hybrid radial-Cartesian sampling strategy with combined parallel imaging to acquire 2 mm isotropic whole brain BOLD-weighted images at 0.6 s temporal resolution. Our results show that k-t FASTER at high accelerations is capable of capturing task-related FMRI activation with high sensitivity, and that an independent components analysis of the principal component subspace directly estimated by k-t FASTER reproduces the expected temporal and spatial patterns elicited by the visual task with high fidelity.

Catarina Rua¹, Stephen James Wastling², Mauro Costagli³, Laura Biagi⁴, Mark Roger Symms⁵, Alberto del Guerra¹, Mirco Cosottini^1,3, Michela Tosetti^3,4, and Gareth John Barker^2
1University of Pisa, Pisa, Italy, ²Neuroimaging, King's College London, London, United Kingdom, ³IMAGO7 Foundation, Pisa, Italy, ⁴IRCCS Stella Maris, Pisa, Italy,⁵GE Healthcare, Pisa, Italy

We show that signal losses associated with slice selective excitation in Gradient Echo (GE) EPI at 7T can be recovered by the use of specially designed Tailored-RF (TRF) pulses. Three subjects were tested with both a standard GE sequence and a TRF modified GE-EPI pulse sequence at three different resolutions. Also, a spin-echo EPI was taken in one of the subjects for comparison. It was possible to observe signal recovery in lower temporal lobe regions in the TRF sequence even with an expected global SNR drop. This work shows a potential applicability for functional brain imaging in lower brain regions.

Ying-Hua Chu¹, Alexandre Vignaud², Ruo-Ning Sun¹, Christophe Pallier³, Wen-Jui Kuo⁴, Denis Le Bihan², and Fa-Hsuan Lin^1
1Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan, ²CEA\DSV\I2BM\Neurospin\UNIRS, Gif sur Yvette, France,³CEA\DSV\I2BM\Neurospin\UNICOG, Gif sur Yvette, France, ⁴Institute of Neuroscience, National Yang Ming University, Taipei, Taiwan

The spatial resolution of MR inverse imaging (InI) was empirically tested at 3T and 7T. By using a coil array of the same number of channel and a similar geometry at a higher field, we found that the coil sensitivity becomes more disparate and improves the condition of the spatial encoding. Compared to results at 3T, the InI spatial resolution quantified by the average point-spread function at 7T improved by about 65% and 90% at SNR = 0.1 and 1, respectively.

Ruo-Ning Sun¹, Yi-Cheng Hsu¹, Ying-Hua Chu¹, Shang-Yueh Tsai², Wen-Jui Kuo³, and Fa-Hsuan Lin^1
1Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan, ²Institute of Applied Physic, National Chengchi University, Taipei, Taiwan,³Institute of Neuroscience, National Yang Ming University, Taipei, Taiwan

We use the “dynamic off-resonance in k-space” (DORK) method to correct the phase drift in magnetic resonance inverse imaging (InI), Empirical results show that DORK can reduce the InI fluctuation in the respiratory frequencies, improve the stability of the fMRI time series, and increase the peak value of hemodynamic response estimates.

Rebecca Ramb¹, Pierre Levan¹, and Jürgen Hennig^1
1Department of Radiology, Medical Physics, University Medical Center, Freiburg, Germany

PEAK-EPI, i.e. permuted interleaved EPI acquisition together with PEAK-GRAPPA reconstruction, facilitates high spatial resolution (2mm at 3T) with whole brain coverage (60-63 slices) in task-related fMRI while maintaining high SNR and tSNR. Due to the shortening of the acquisition times per slice, distortion effects are mitigated. The feasibility PEAK-GRAPPA in task-related fMRI is presented with two different acquisition and reconstruction schemes and magnitude images as well as activation maps are compared to a reference EPI acquisition.

Mahdi Khajehim¹, Abbas Nasiraei Moghaddam^1,2, Gholam-Ali Hossein-Zadeh^2,3, Thomas Martin⁴, and Danny JJ Wang^4
1BME, Amirkabir University of Technology (Tehran Polytechnic), Tehran, Tehran, Iran, ²School of Cognitive Sciences, Institute for Research in Fundamental Sciences (IPM), Tehran, Iran, ³ECE, University of Tehran, Tehran, Tehran, Iran, ⁴Neurology, UCLA, Los Angeles, CA, United States

Generally magnitude data is used in BOLD fMRI for detecting activation. By using a recently introduced steady state fMRI sequence called A-BOSS, phase changes can be reliably detected and quantified. A-BOSS has several advantages over BOSS: Its spatial coverage is not limited and the contribution of phase is not dependent on the center frequency position. Our preliminary data showed that 1) a considerable number of voxels showed phase activation (42% of total activated voxels) which could not be detected just based on magnitude data; 2) on average 0.07 radian phase change due to task activation which is consistent with Bloch equation simulation.

Vânia Tavares¹, André Santos Ribeiro^1,2, Carlos Capela³, Luís Cerqueira⁴, and Hugo Alexandre Ferreira^1
1Institute of Biophysics and Biomedical Engineering, Faculy of Sciences of the University of Lisbon, Lisboa, Portugal, ²Centre for Neuropsychopharmacology, Division of Brain Sciences, Department of Medicine, Imperial College London, London, United Kingdom, ³Department of Neurology, Centro Hospitalar Lisboa Central, Lisbon, Portugal, ⁴Department of Neuroradiology, Centro Hospitalar Lisboa Central, Lisbon, Portugal

A method for localizing the epileptogenic focus from functional magnetic resonance imaging data was developed. For that purpose a bi-dimensional temporal clustering analysis was used to identify potential epileptogenic foci. Then, the complexity of the corresponding BOLD signals was analyzed to identify the most probable focus. Complexity analysis comprised the use of a modified multiscale entropy analysis algorithm and the detrended fluctuation analysis algorithm. Results showed that this method is able to localize epileptogenic tissue in agreement with clinical knowledge. Results also showed that the epileptogenic foci influence the dynamics of related brain regions within epileptic networks.

Alejandro Veloz^1,2, Luis Hernandez-Garcia³, Hector Allende², Claudio Moraga⁴, Rodrigo Salas¹, and Steren Chabert^1
1Biomedical Engineering School, Universidad de Valparaiso, Valparaiso, Chile, ²Department of Informatics, Universidad Tecnica Federico Santa Maria, Valparaiso, Chile, ³Functional Magnetic Resonance Imaging Laboratory, University of Michigan, Ann Arbor, Michigan, United States, ⁴European Centre for Soft-Computing, Mieres, Spain

Since the introduction of fMRI, accurate delineation of brain activity is a relevant topic. This is a difficult task, among other reasons, due to the fact that the Haemodynamic Response varies over time, and across individuals or brain regions. This work focuses on developing a tool more adequate to represent a broader range of possible shapes of the HRF, based on the framework of fuzzy variables. Promising results are obtained in both simulation and healthy volunteer data, where the activated region obtained with the fuzzy GLM completely intersects the canonical GLM, in addition to obtaining a broader activated region.

Frank Riemer^1,2, Bhavana S. Solanky¹, Xavier Golay², Egidio U. D'Angelo³, and Claudia A. M. Wheeler-Kingshott^1
1NMR Research Unit, Department of Neuroinflammation, Queen Square MS Centre, Institute of Neurology, University College London, London, United Kingdom,²Department of Brain Repair and Rehabilitation, Queen Square MS Centre, Institute of Neurology, University College London, London, United Kingdom, ³Brain Connectivity Center, C. Mondino National Neurological Institute, Pavia, Italy

Sodium (23Na) fMRI is a novel approach to study brain function. Its physiological underpinning may come from the temporal increase in intracellular sodium during neuronal firing, but also from an increase in blood volume. During finger-tapping, contralateral motor, premotor and ipsilateral somatosensory, insula and cerebellum grey matter areas are activated. Interestingly, 23Na-fMRI reveals unique activations in grey matter/white matter bordering regions where we expect the highest presence of sodium channels and in areas of associative white matter fibers such as the superior longitudinal fasciculus and the inferior cerebellar peduncle, supporting the hypothesis that 23Na-fMRI is sensitive to intracellular sodium accumulation.

Rosa Sanchez Panchuelo¹, Keren Yang¹, Martin Buehrer², Richard Bowtell¹, and Susan Francis^1
1University of Nottingham, Nottingham, United Kingdom, ²Gyrotools, Zurich, Switzerland

By using multiband acquisition to extend slice coverage, we investigate body somatotopy in SI and SII with high spatial resolution at 7T. fMRI was performed during both fingertip and face-hand-foot mapping paradigms in three subjects. The fingertip mapping paradigm revealed clear digit somatotopic representation within SI, but not in SII for all subjects, suggesting an overlapping representations of the fingers in SII. The face-hand-foot mapping showed a somatotopic arrangement in SII, with the representation of the face, hand and foot ordered from lateral to medial, in agreement with previous fMRI experiments and electrophysiology in primates.

Sachiko Kiyama¹, Mitsunobu Kunimi¹, Katsuo Tamaoka², Rinus Verdonschot³, and Toshiharu Nakai^1
1National Center for Geriatrics and Gerontology, Ohbu, Aichi, Japan, ²Nagoya University, Nagoya, Aichi, Japan, ³Osaka University, Osaka, Japan

The present study aims to assess the age-related neural changes particularly concerning the role memory plays during emotional sentence processing, by employing functional magnetic resonance imaging (fMRI). Although emotional valence did not explicitly moderate behavior within sentence recognition task both in young and elderly groups, neural activity during the task showed particular interaction effects between emotional valence and age group.�@ Among the enhanced regions for recognizing positive emotional sentences, the insula is known to be involved in emotional awareness. Our results suggest that the insula might reflect the elderlies�f greater awareness to positive emotional sentences.

Rosa Sanchez Panchuelo¹, Keren Yang¹, Martin Buehrer², Miles Humberstone³, and Susan Francis^1
1University of Nottingham, Nottingham, United Kingdom, ²Gyrotools, Zurich, Switzerland, ³Nottingham University Hospitals Trust, Nottingham, United Kingdom

We investigate digit representation and tactile attention in SI and SII with high spatial resolution at 7T. fMRI was performed during a novel task to modulate spatial attention between digits 2, 3 and 4 during stimulation of the digits of the left hand, each with a different periodicity. This novel paradigm revealed a clear digit somatotopy within SI, but no modulation to attention. In contrast, no somatotopic map was seen in SII, with digit responses largely overlapping, whilst SII and other cortical areas, such as SMA and pre-SMA, were strongly modulated by tactile attention in an overlapping manner.

Christopher Thomas Sica¹, Prasanna Karunanayaka¹, Jeff Vesek², Jianli Wang¹, and Qing X Yang^1,3
1Radiology, Penn State College of Medicine, Hershey, PA, United States, ²Molecular Biology, Penn State College of Medicine, Hershey, PA, United States,³Neurosurgery, Penn State College of Medicine, Hershey, PA, United States

Susceptibility field gradients (SFG) cause undesired through-plane dephasing and associated signal loss near the sinus region, especially for fMRI scans with substantial echo times. One recently developed approach to mitigating the signal loss is the use of spectral-spatial pulses that apply a frequency dependent through-plane phase precompensation. This pre-phasing of the magnetization will cancel out the dephasing due to SFG’s, leading to signal recovery at TE. We evaluated and demonstrated the ability of spectral-spatial pulses to improve detection of BOLD activation in the olfactory regions near the base of the brain.

Nicholas G Dowell¹ and Eleanor Denny^1
1Brighton and Sussex Medical School, Falmer, Brighton, United Kingdom

In this study, we establish a sham MRI session that replicates the primary conditions of a genuine MRI scan, but without the need of a costly MRI simulator. We demonstrate that such a session can reduce the fMRI session effects that are currently observed with multi-session fMRI studies and will improve the power of these studies, increasing the likelihood of identifying subtle differences between groups. Currently, the only solution is to increase the sample size, although the costs increase as the square of the standard deviation of the unwanted effects. We introduce a more cost-effective and academically-sound approach to measure, understand and then reduce unwanted sources of variation.

Joe S. Cheng^1,2, Iris Y. Zhou^1,2, Mengye Lyu^1,2, and Ed X. Wu^1,2
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong, Hong Kong, China, ²Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, China

Modern neuroimaging techniques, especially fMRI, have advanced our understanding of the neuroanatomical basis of cognitive processes such as mental arithmetic. Most studies so far only attempted to relate individual activated regions with specific processes. However, the neural activation within a given region may reflect the summation of multiple distinct networks that carry different functional purposes. By combining general linear model (GLM) and independent components analysis (ICA) into contributive source analysis (CSA), this study aimed to test the hypothesis that multiple functional networks engage in simple mental task of single-digit multiplication and contribute collectively to the activations in prefrontal and parietal cortices.

fMRI: Bold Physiology & Multimodal Imaging

Tuesday 2 June 2015

PINAR SENAY ÖZBAY^1,2, Cristina Rossi¹, Geoffrey Warnock³, Felix Kuhn³, Klaas Paul Prüssmann², and Daniel Nanz^1
1Department of Radiology, University Hospital Zürich, Zürich, Switzerland, ²Institute of Biomedical Engineering, ETH Zürich, Zürich, Switzerland, ³Department of Nuclear Medicine, University Hospital Zürich, Zürich, Switzerland

This study compared BOLD- and QSM-based neuronal activation patterns and changes of venous magnetic susceptibility under visual-stimulation as observed in data acquired with (i) traditional repeated fast whole-brain covering 2D GRE-EPI scans versus with (ii) single 3D-GRE scans. T-score-maps from statistical parametric mapping of the EPI-acquired fQSM time-series-data and difference maps (ON-OFF) for both EPI-fQSM and 3D-GRE-fQSM are presented. Image quality of 3D-GRE-fQSM and its difference-map were superior to corresponding EPI-QSM data. 3D-GRE-fQSM detected larger reduction of venous magnetic susceptibility upon visual stimulation than EPI-fQSM. 3D-GRE-fQSM might have a potential for studies with a focus on the venous vessel tree.

PINAR SENAY ÖZBAY^1,2, Cristina Rossi¹, Geoffrey Warnock³, Felix Kuhn³, Burak Akin⁴, Klaas Paul Prüssmann², and Daniel Nanz^1
1Department of Radiology, University Hospital Zürich, Zürich, Switzerland, ²Institute of Biomedical Engineering, ETH Zürich, Zürich, Switzerland, ³Department of Nuclear Medicine, University Hospital Zürich, Zürich, Switzerland, ⁴Medical Physics, University Medical Center, Freiburg, Germany

ICA has been widely used in task-based-fMRI in order to separate independent signal components, without supplying à-priori knowledge of the paradigm. The aim of this work was to identify and characterize signal components that capture neuronal activation in quantitative susceptibility data (QSM) acquired under visual-stimulation. The effect of temporal-filtering on activation maps, signal time-course and corresponding power-spectra were investigated and results compared with those from traditional BOLD analysis. There was a strong correlation between BOLD and filtered QSM data. ICA of QSM data seems promising for an accurate localization of neuronal activation and a better understanding of the underlying mechanisms.

Junseok Kim^1,2, Jackie Leung², Indra Narang², Paula Louise Croal², and Andrea Kassner^1,2
1University of Toronto, Toronto, ON, Canada, ²Hospital for Sick Children, Toronto, ON, Canada

Obesity is the most common cause of obstructive sleep apnea (OSA) in children. With childhood obesity at epidemic levels, the incidence of OSA is rising. OSA is characterized by intermittent episodes of nocturnal hypoxia, hypercapnia and sleep disruption. Moreover, intermittent hypoxia leads to oxidative damage of the endothelial cells, resulting in endothelial dysfunction which compromises vasodilatory capacity. Reduction in vasodilatory capacity can be quantified experimentally using MR-based cerebrovascular reactivity (CVR). We found that obese children with OSA have reduced CVR, both globally and regionally, compared to controls with no OSA.

Sebastian Domsch¹, Sebastian Weingärtner¹, Jascha Zapp¹, and Lothar R. Schad^1
1Computer Assisted Clinical Medicine, Heidelberg University, Mannheim, Baden-Württemberg, Germany

The oxygen extraction fraction is of great clinical interest providing a biomarker for brain tissue viability and a parameter for the evaluation of diseases such as tumor, stroke, and Alzheimer’s disease. The analytical tissue model in the static dephasing regime has facilitated promising in-vivo results mapping hemodynamic parameters such as the OEF and the venous blood volume fraction separately using MRI. This work shows a comparison between the fit results using the full analytical model and the more frequently used asymptotic approximation in terms of accuracy and precision based on Monte-Carlo simulations and robustness in-vivo.

David E Crane¹, Anoop Ganda¹, David J Mikulis², Sandra E Black¹, and Bradley J MacIntosh^1
1Sunnybrook Research Institute, Toronto, ON, Canada, ²Toronto Western Hospital, Toronto, ON, Canada

Hypercapnia-induced cerebrovascular reactivity is a powerful tool to study the ability of the brain’s blood vessels to vasodilate. The clinical utility of CVR has been demonstrated in multiple applications. There are however continued efforts to improve the CVR time-series analysis. In this study we introduce a deconvolution approach that produces a vascular transfer function, which can be calculated by tissue type or at a voxel-wise level. We found this approach had higher sensitivity to patient group differences in white matter, compared to conventional CVR, and therefore may be useful as an adjunct approach.

Christina Y. Shu¹, Daniel Coman², Basavaraju G. Sanganahalli², Helen Wang², Christoph Juchem², Peter Herman², and Fahmeed Hyder^1,2
1Biomedical Engineering, Yale University, New Haven, CT, United States, ²Diagnostic Radiology, Yale University, CT, United States

In calibrated fMRI the BOLD signal range from deoxyhemoglobin to oxyhemoglobin is reflected by M. Since blood oxygenation changes are captured with relaxation rates by gradient echo (R2*) and spin echo (R2), M is echo time multiplied by R2’ (i.e., R2*- R2). Because it is unclear if the M range reported in calibrated fMRI studies is due to regional/state differences, we tested M estimated from R2’ in rat brain. Homogenous M values were found in the cortex regardless of the brain state. These results suggest that a direct R2’ mapping method may provide more accurate M values for calibrated fMRI.

Christina Y. Shu¹, Douglas Rothman^1,2, Basavaraju G. Sanganahalli², Daniel Coman², Peter Herman², and Fahmeed Hyder^1,2
1Biomedical Engineering, Yale University, New Haven, CT, United States, ²Diagnostic Radiology, Yale University, New Haven, CT, United States

In calibrated fMRI the power-law relationship between the BOLD signal with blood flow and oxygen consumption is quantified by the parameter β. The frequently used β value of 1.5 comes from simulation results. To make calibrated fMRI independent of assumptions and hence applicable for clinical settings, we developed a method for β mapping by measuring R2’, the most sensitive relaxation component of BOLD signal, as a function of intravascular susceptibility weighting. Since we measured lower β values in cortex than previously assumed and observed β heterogeneities across other regions, there is need for reevaluation of physiological parameters for calibrated fMRI.

Deqiang Qiu¹, Junjie Wu¹, Fadi Nahab², and Seena Dehkharghani^1
1Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States, ²Neurology, Emory University, GA, United States

In this paper, we propose to image cerebrovascular reserve (CVR) by combining repeated CBF measurements using arterial spin labeling (ASL) and BOLD. Complementary information can be provided by using both ASL and BOLD, and this is demonstrated in patients with cerebrovascular diseases.

Hannah Hare¹ and Daniel Bulte^1
1FMRIB, University of Oxford, Oxford, Oxfordshire, United Kingdom

A dual-gas hypercapnia/hyperoxia paradigm as used in quantitative physiological MRI was performed while subjects were monitored with a NIRS cerebral oximeter, a finger-clip pulseox, and respiratory gas analyser to obtain arterial and tissue oxygen saturation levels. Increases in arterial and venous saturations were found by combining the group data that could not be measured in individuals. This has implications for the use of the Davis model in calibrated fMRI.

Jingwei Zhang^1,2, Thanh D. Nguyen², Pascal Spincemaille², Tian Liu³, Dong Zhou², Ajay Gupta², and Yi Wang^1,2
1Biomedical Engineering, Cornell University, New York, New York, United States, ²Radiology, Weill Cornell Medical College, New York, New York, United States,³Medimagemetric, LLC, New York, United States

Quantitative susceptibility mapping (QSM) is advantageous over R2* for determining deoxyhemoglobin concentration from gradient echo MRI data. QSM can be used for quantitative CMRO2 mapping using pre- and post- caffeine challenge, a procedure much simpler than that required by R2*. In general, CMRO2 is sensitive to the poor SNR inherent in MRI based deoxyhemoglobin measurements. One of the noise consequence is to cause the estimated OEF exceed its physiological range. To overcome this limitation, we propose to solve for CMRO2 by constraining OEF within its physiological limit.

João Jorge^1,2, Frédéric Grouiller³, Wietske van der Zwaag⁴, Rolf Gruetter¹, and Patrícia Figueiredo^2
1Laboratory for Functional and Metabolic Imaging, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ²Department of Bioengineering, Instituto Superior Técnico, Lisbon, Portugal, ³Biomedical Imaging Research Center, University of Geneva, Geneva, Switzerland, ⁴Biomedical Imaging Research Center, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland

During simultaneous EEG-fMRI, the artifacts induced on EEG recordings by the strong magnetic fields employed can largely overwhelm the signals of interest. Unlike gradient and cardiac pulse-related artifacts, contributions due to spontaneous head motion in B0 have so far received little attention, but can become considerably problematic at higher field strengths. Here, an approach is presented for detection and reduction of EEG artifacts due to head motion in B0. This approach produced important reductions in EEG noise power on EEG-fMRI recordings at 7T, and largely improved sensitivity to visual responses at both a trial-average and a single-trial level.

Lars Michels^1,2, Steffen Bollmann², Diego Manuel Baur², Anthony Schläpfer³, Maya Schneebeli³, Carmen Ghisleni², Peter Klaver^2,4, Daniel Brandeis^3,5, and Ruth O'Gorman^2
1Institute of Neuroradiology, University Hospital Zurich, Zurich, Zurich, Switzerland, ²Center for MR-Research, University Children’s Hospital, Zurich, Zurich, Switzerland, ³Department of Child and Adolescent Psychiatry University of Zürich, University of Zurich, Zurich, Switzerland, ⁴Institute of Psychology, University of Zurich, Zurich, Switzerland, ⁵Department of Child & Adolescent Psychiatry and Psychotherapy, Central Institute of Mental Health Ma, Medical Faculty Mannheim / Heidelberg University, Mannheim, Mannheim, Germany

Attention-deficit/hyperactivity disorder (ADHD) has a high prevalence but the neurobiological mechanisms underlying ADHD remain unclear. We characterize resting EEG-fMRI coupling differences between ADHD and controls by EEG-fMRI. Based on the literature, we expected EEG-fMRI alterations in central resting and cognitive control networks. In 33 adults (17 ADHD), frequency-band specific EEG power was correlated with fMRI signal alterations during alternating eyes open and closed blocks (p < 0.05 corrected). We found bidirectional group differences in low (theta) to high (alpha) frequency bands in regions of the described networks. We conclude that sensory- and cognitive processing is compromised in adults with ADHD.

Muhammad E H Chowdhury¹, Karen J Mullinger^1,2, Glyn S Spencer¹, and Richard Bowtell^1
1SPMIC, School of Physics and Astronomy, University of Nottingham, Nottingham, United Kingdom, ²BUIC, School of Psychology, University of Birmingham, Birmingham, United Kingdom

EEG data recorded during fMRI are compromised by large gradient artefact (GA) voltages. The GA is usually corrected using average artefact correction; requiring the amplifier to have a large enough dynamic range to characterise the artefact voltages. Here we re-designed the EEG cap-cable configuration so that the GA induced in the 1 m ribbon cable by an AP gradient partially cancels that induced in the EEG cap and head. We demonstrate that the range and amplitude of the GA can be significantly reduced by cap-cable re-wiring, allowing recording at higher EEG bandwidths or increased achievable image resolution without saturation.

Yun Wang¹ and Gopikrishna Deshpande^1,2
1AU MRI Research Center, Department of Electrical & Computer Engineering, Auburn University, Auburn, Alabama, United States, ²Department of Psychology, Auburn University, Auburn, Alabama, United States

Previous studies have showed that the envelope of EEG gamma band power is correlated with slow fMRI fluctuations in resting state networks. However, simultaneous EEG/fMRI studies have not been able to assess whether fMRI RSNs have a true neural basis in millisecond-scale fast neuronal dynamics. In order to address this issue, we acquired simultaneous EEG/fMRI data multiband EPI with TR =1000 and 200 ms and fused the modalities using parallel independent component analysis (pICA) such that native resolution of either modality is not compromised. Our results suggest high frequency electrical correlates for visual resting state network

Rodolfo Abreu¹, Marco Leite^1,2, Alberto Leal³, and Patrícia Figueiredo^1
1Institute for Systems and Robotics and Department of Bioengineering, Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Portugal, ²Department of Clinical and Experimental Epilepsy and The Wellcome Trust Centre for Neuroimaging, University College London Institute of Neurology, Queen Square, London WC1N 3BG, United Kingdom, ³Centro de Investigação e Intervenção Social and Department of Neurophysiology, Centro Hospitalar Psiquiátrico de Lisboa, Lisbon, Portugal

Ballistocardiogram artefact (BCG) seriously compromises EEG data quality when simultaneously-acquired with fMRI. Here, we propose a novel ICA-based approach (PROJIC) for BCG artefact correction as well as a novel evaluation pipeline that assesses both artefact and background signal removal. The proposed evaluation pipeline for BCG artefact correction allowed different weightings of the importance of removing the artefact against preserving the signal background, showing that different methods may be preferred in different situations. The proposed ICA-based approaches outperformed both previous ICA-based methods as well as the Optimal Basis Sets (OBS) approach, in all conditions.

Maxime Chamberland¹, Michaël Bernier¹, David Fortin², Kevin Whittingstall³, and Maxime Descoteaux^4
1Nuclear Medecine and Radiobiology, Université de Sherbrooke, Sherbrooke, Québec, Canada, ²Neurosurgery, Université de Sherbrooke, Sherbrooke, Québec, Canada, ³Diagnostic Radiology, Université de Sherbrooke, Sherbrooke, Québec, Canada, ⁴Computer science, Université de Sherbrooke, Sherbrooke, Quebec, Canada

The human brain can be viewed as a collection of networks. Those highly specialized networks can be referred to as a set of nodes (gray matter functional areas) linked together by edges (for example white matter axonal structure). In this work, we propose an interactive tool for the exploration of functional connectivity in a fully 3D fashion, which can be coupled with our existing real-time fiber tracking module.

Richard Baumgartner¹, Arie Struyk², Jeff Evelhoch², Cynthia Gargano², Francheska Colon Gonzalez², Haiying Liu¹, Ruben Declercq³, Hans Verheyden³, Ingeborg Heirman³, Hans De Pla⁴, Griet Van Lancker⁴, Sofie Van den Abeele⁴, Adelheid Hollebosch⁴, Brant Delafontaine⁴, Luc Van Bortel⁴, Rik Achten⁴, Patricia Clement⁴, Pieter Vandemaele⁴, Dai Feng¹, and Sofia Apreleva^1
1Merck and Co, Rahway, New Jersey, United States, ²Merck and Co, Pennsylvania, United States, ³Merck and Co, Belgium, ⁴Ghent University, Belgium

We present results from a clinical trial of pharmaco-MRI (phMRI) employing cerebral blood volume (CBV) imaging using ferumoxytol (Rienso/Feraheme, AMAG) as a blood pool contrast agent. The study examined the pharmacodynamic effects of two single doses of buprenorphine (0.2mg/70kg and 0.1mg/70kg administered intravenously). We found that contrast-enhanced CBV phMRI signals are more sensitive reporters of pharmacodynamic effects than conventional blood oxygen level dependent (BOLD) phMRI. In particular, higher sensitivity of CBV phMRI compared to BOLD allows for elucidation of PD responses at lower doses of buprenorphine, which has practical implications for similar phMRI studies with centrally acting drugs.

Functional Connectivity Materials & Applications

Tuesday 2 June 2015

Bumhee Park¹, Jose A Palomares¹, Mary A Woo², Daniel W Kang³, Paul M Macey², Frisca L Yan-Go⁴, Ronald M Harper⁵, and Rajesh Kumar^1,6
1Anesthesiology, University of California at Los Angeles, Los Angeles, CA, United States, ²School of Nursing, University of California at Los Angeles, Los Angeles, CA, United States, ³Medicine, University of California at Los Angeles, Los Angeles, CA, United States, ⁴Neurology, University of California at Los Angeles, Los Angeles, CA, United States, ⁵Neurobiology, University of California at Los Angeles, Los Angeles, CA, United States, ⁶Radiological Sciences, University of California at Los Angeles, Los Angeles, CA, United States

OSA subjects show impaired autonomic, affective, executive, sensori-motor, and cognitive functions. Brain injury, assessed by various MRI procedures, appears in multiple sites regulating these functions; however, the integrity of functional networks remains unclear. We examined resting functional interactions and complex network organization across the whole-brain in OSA over controls and found aberrant functional connections and altered brain network organization in those regions. The findings suggest that impaired functions in OSA may stem from altered functional connectivity and brain network organization. The outcomes likely result from prominent structural brain changes in both axons and nuclear structures reported-earlier.

Elisa Marchetta^1,2, Sara Cirillo¹, Pasquale Della Rosa³, Silvia Pontesilli¹, Andrea Falini^1,4, Graziano Barera⁵, Cristina Baldoli^1,4, and Paola Scifo^6,7
1Neuroradiology Dept., San Raffaele Hospital, Milan, -, Italy, ²University of Milano-Bicocca, Milan, -, Italy, ³Institute of Molecular Bioimaging and Physiology, CNR, Segrate, -, Italy, ⁴Vita-Salute San Raffaele University, Milan, -, Italy, ⁵Division of Neonatology, Pediatrics Dept, San Raffaele Hospital, Milan, -, Italy, ⁶Nuclear Medicine Dept., San Raffaele Hospital, Milan, -, Italy, ⁷CERMAC, San Raffaele Hospital, Milan, -, Italy

The brain functional architecture grows fast during prenatal period. This study aims to investigate the development of DMN and its correlation with frontal task-positive areas by means of Resting State fMRI in a group of healthy preterm newborns. Interestingly, we found significant correlations between mPFC and the frontopolar region. This functional association was also found to be significantly correlated with age. Our results suggest that the recruitment of regions involved in high order functions might start since the very early age, although frontal areas are characterized by late myelination and expected late connectivity.

Giusy Olivito^1,2, Maria Leggio^1,2, Fiorenzo Laghi³, Roberto Baiocco³, Anna Maria Tedesco¹, Silvia Clausi¹, Chiara Mastropasqua⁴, Marco Molinari⁵, Mara Cercignani^4,6, and Marco Bozzali^4
1Ataxia Research Laboratory, IRCCS Santa Lucia Foundation, Rome, Italy, Italy, ²Department of Psychology, Faculty of Medicine and Psychology, University of Rome "Sapienza", Rome, Italy, Italy, ³Department of Developmental and Social Psychology, Faculty of Medicine and Psychology, University of Rome "Sapienza", Rome, Italy, Italy, ⁴Neuroimaging Laboratory, IRCCS Santa Lucia Foundation, Rome, Italy, Italy, ⁵Neurological and Spinal Cord Injury Rehabilitation, Department  A, IRCCS Santa Lucia Foundation, Rome, Italy, Italy, ⁶Clinical Imaging Science Center (CISC), Brighton and Sussex Medical School, Brighton, Sussex, United Kingdom

The cerebellum has emerged as one of the regions affected in Autism spectrum disorders (ASDs). The cerebro-cerebellar disconnection could in part underlie autistic symptoms, mainly involving “Theory of Mind” (ToM) processes. In the present study we demonstrate the usefulness of resting-state fMRI in detecting changes in functional connectivity (FC) between the cerebellar dentate nucleus, the sole cerebellar output channel, and cerebral cortex regions by comparing ASD patients and tipically developing subjects. Using a seed-based approach, we found altered FC in ASDs patients between DN and regions of the Default Mode Network, known to be related to social deficits seen in ASD.

Mario Torso¹, Chiara Mastropasqua¹, Giovanni Giulietti¹, Laura Serra¹, Giusy Olivito^2,3, Elisa Tuzzi¹, Barbara Spanò¹, Carlo Caltagirone^4,5, Mara Cercignani⁶, and Marco Bozzali^1
1Neuroimaging Laboratory, Santa Lucia Foundation, IRCCS, Rome, Italy, ²Ataxia research Laboratory, Santa Lucia Foundation, IRCCS, Rome, Italy, ³Department of psychology, University of Rome Sapienza, Rome, Italy, ⁴Department of Clinical and Behavioural Neurology, Santa Lucia Foundation, IRCCS, Rome, Italy,⁵Department of Neuroscience, University of Rome ‘Tor Vergata’, Rome, Italy, ⁶CISC, Brighton & Sussex Medical School, Brighton, Sussex, United Kingdom

In this study we used resting state f-MRI to investigate the contribution of White Matter Hyperintensities (WMHs) in determining functional connectivity (FC) changes in amnestic Mild Cognitive Impairment patients, the prodromal stage of Alzheimer disease. WMHs volumes were quantified by semiautomatic method on TSE images and used to distinguish patients in two different groups (high WMHs volume –H and low WMHs volume –L). The main finding of the study was that a higher WMHs volume can determine changes in FC and that these alterations are associated with the presence of cognitive deficits in the patients.

Mingfei Ni¹, Weiwei Wang¹, Ziheng Zhang², Qingwei Song¹, Ailian Liu¹, and Yanwei Miao^1
1Radiology Department, the First Affiliated Hospital of Dalian Medical University, Dalian, Liaoning, China, ²GE Healthcare China, Beijing, China

Objective:To investigate the activity changes of the baseline brain in patients with Benign Essential Blepharospasm (BEB) by resting-state fMRI fractional amplitude of low-frequency fluctuation (fALFF) method.Methods:28 patients with BEB and 28 healthy controls group were scanned on a 3.0T MRIscanner. The image data were analyzed with software SPM8, DPARSF and REST. Results The whole brain analysis indicated that in comparison with the normal control group, there was a ncreased fALFF in the right major sensorimotor area£¬right caudate head£¬left thalamus. the fALFF in the left thalamus were positively correlated with the JRS total score.

Subash Khushu¹, Sadhana Singh¹, Mukesh Kumar¹, Shilpi Modi¹, Prabhjot Kaur¹, and L Ravi Shankar^2
1NMR Research Centre, INMAS, DRDO, Delhi, Delhi, India, ²Thyroid Research Centre, INMAS, DRDO, Delhi, Delhi, India

Hypothyroidism affects brain functioning as suggested by various neuroimaging studies. The primary focus of this study was to examine whether hypothyroidism would impact connectivity among resting state networks with the use of resting-state fMRI. Our results showed significantly decreased functional connectivity in the region of right fronto-parietal network, medial visual network and motor network in hypothyroid patients as compared with healthy controls. These findings suggest dysfunction of motor and cognitive functions in hypothyroidism.

Keith M Vogt¹ and James W Ibinson^2
1Anesthesiology, University of Pittsburgh Medical Center, Pittsburgh, PA, United States, ²Center for Pain Research, Dept of Anesthesiology, University of Pittsburgh, Pittsburgh, PA, United States

The portion of the insula chosen as the seed region for functional connectivity MRI (fcMRI) analysis can dramatically affect the resulting maps. This is particularly important when comparing maps obtained during rest to those during experimental pain. In this fcMRI study, seed region analyses revealed that the difference maps for posterior insula Pain vs Rest connectivity was obviously different than those seen with an anterior insular seed region. Thus, the posterior insula is a putative brain area for differentiating the experience of acute pain from rest, while the anterior insula does not exhibit this specificity.

Zhen-Hui Li^1,2, Vincent Chin-Hung Chen³, Ming-Chou Ho⁴, and Jun-Cheng Weng^1,2
1Department of Biomedical Sciences, Chung Shan Medical University Hospital, Taichung, Taiwan, ²School of Medical Imaging and Radiological Sciences, Chung Shan Medical University, Taichung, Taiwan, ³Department of Psychiatry, Chung Shan Medical University Hospital, Taichung, Taiwan, ⁴Department of Psychology, Chung Shan Medical University, Taichung, Taiwan

Major depressive disorder (MDD) is a public health problem in recent years. MDD is characterized by emotional imbalance with extremely in emotional processing. MDD patient with self-harm may eventually result in the death. Previous studies showed abnormal functional connectivity between specific brain regions, and few studies demonstrated the functional network can be observed by the large-scale structural pathways interconnecting. Graph theory is capable of evaluating the topological organization of the human brain. Therefore, in this study we tried to find out the functional connectomic difference between MDD patients and healthy subjects based on resting-state functional magnetic resonance imaging (rs-fMRI) using graph theoretical and network-based statistic (NBS) analyses. Our results revealed that MDD patients exhibit a disruption in the topological organization of functional brain networks.

Matthew C. Murphy¹, Amy C. Nau¹, Christopher Fisher¹, Seong-Gi Kim^2,3, Joel S. Schuman^1,4, and Kevin C. Chan^1,4
1Department of Ophthalmology, University of Pittsburgh, Pittsburgh, PA, United States, ²Departments of Biological Sciences and Global Biomedical Engineering, Sungkyunkwan University, Suwon, Korea, ³Department of Radiology, University of Pittsburgh, Pittsburgh, PA, United States, ⁴Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA, United States

Visual deprivation is known to induce plasticity of the visual system, which can be observed through alterations in functional connectivity (FC). How these FC changes accrue over time and how this relationship between FC and duration of blindness differs between early and late acquired blindness are unknown. In this work, we modeled the effects of prior visual experience on visual FC in blind subjects. The results suggest that alterations in FC due to visual deprivation progress over time, however the direction of this progression in congenitally blind subjects is opposite to that in subjects with acquired blindness.

Gloria Castellazzi^1,2, Fulvia Palesi^2,3, Ahmed T. Toosy⁴, Stefania Bruno⁵, Egidio D'Angelo^2,6, and Claudia A.M. Wheeler-Kingshott^7
1Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, PV, Italy, ²Brain Connectivity Center, C. Mondino National Neurological Institute, Pavia, PV, Italy, ³Department of Physics, University of Pavia, Pavia, PV, Italy, ⁴Department of Brain Repair and Rehabilitation, Queen Square MS Centre, UCL Institute of Neurology, London, England, United Kingdom, ⁵Overdale Hospital, Jersey, England, United Kingdom, ⁶Department of Brain and Behavioral Sciences, University of Pavia, Pavia, PV, Italy, ⁷NMR Research Unit, Department of Neuroinflammation, Queen Square MS Centre, UCL Institute of Neurology, London, England, United Kingdom

During complex “continuous” cognitive tasks, the brain elaborates information over multiple domains and time scales. These operations contribute to dynamically shape, over space and time, the whole-brain functional connectivity. We used rs-fMRI to investigate the changes in brain functional connectivity occurring in subjects listening to a narrated story. Results show that the listening task dynamically alters the shape of the functional connectome in a non-random way, affecting sensory prior that cognitive brain networks. A possible interpretation looks at the brain as a “prediction engine” that constantly generates predictions about the optimal configuration of the networks to process the impending input.

Hesamoddin Jahanian¹, Samantha Holdsworth¹, Thomas Christen¹, Hua Wu², Kangrong Zhu³, Adam Kerr³, Mathew J Middione⁴, Robert F Dougherty², Michael Moseley¹, and Greg Zaharchuk^1
1Department of Radiology, Stanford University, Stanford, CA, United States, ²Center for Cognitive and Neurobiological Imaging, Stanford University, Stanford, CA, United States, ³Department of Electrical Engineering, Stanford University, Stanford, CA, United States, ⁴Applied Sciences Laboratory West, GE Healthcare, Menlo Park, CA, United States

In this work we studied the dynamics of functional connectivity within the default mode network using high temporal sampling rate (TR=350 ms) Simultaneous MultiSlice (SMS) Resting-state fMRI. We performed wavelet coherence analysis in two frequency bands: 0.01-0.1 Hz and above 0.1 Hz. Our results indicate that functional connectivity occurs at multiple frequency bands – including those above 0.1 Hz that are commonly ignored in rsfMRI studies – and exhibits dynamic changes within time scales of seconds to minutes.

Hsu-Lei Lee¹, Jakob Assländer¹, Pierre LeVan¹, and Jürgen Hennig^1
1Medical Physics, University Medical Center Freiburg, Freiburg, BW, Germany

We used wavelet analysis to demonstrate dynamic coherence maps with respect to different brain regions. A frequency-dependent connectivity behaviour of default mode network was observed, where idividual parts of the network become less synchronized as signal frequency increases.

Francisca Marie Tan^1,2, Karen Mullinger¹, Yaping Zhang², David Siu-Yeung Cho², Susan Francis¹, and Penny Gowland^1
1Sir Peter Mansfield Imaging Centre, The University of Nottingham, Nottingham, Nottinghamshire, United Kingdom, ²Department of Electrical and Electronic Engineering, The University of Nottingham Ningbo China, Ningbo, Zhejiang, China

The Default Mode Network (DMN) is associated with anti-correlation with task positive networks. In this study, we investigate the DMN’s response to spontaneous motor events that are detected in the Somatosensory Network. Sparse paradigm free mapping is used to detect three movement types: long and short motor tasks, and spontaneous movements without prior timing information. DMN activity is compared for these three types. Results indicate that the DMN has higher average of activation for spontaneous events compared to task events. This analysis can be further extended to explore the DMN’s behaviour for spontaneous events in other networks.

Yan Bai¹, Carlos Torres², Peng Liu³, Xuejuan Yang³, Dapeng Shi¹, Jie Tian⁴, and Meiyun Wang^1
1Department of Radiology, Henan Provincial People¡¯s Hospital, Zhengzhou, Henan, China, ²Division of Neuroradiology, Department of Radiology, University of Ottawa, Ottawa, ON, Canada, ³School of Life Sciences and Technology, Xidian University, Shaanxi, China, ⁴Institute of Automation, Chinese Academy of Sciences, Beijing, China

The prevalence rate of Chronic prostatitis/Chronic pelvic pain syndrome (CP/CPPS) is approximate 9-16% in the world.However, the pathogenesis of CP/CPPS has been largely unknown so far. Our findings may be helpful for further study on the central mechanism of CP/CPPS.

Yong Zhang¹, Kang-An Li², Yun-Cheng Wu², Qian Jiang¹, Zhenyu Zhou³, and Gui-Xiang Zhang^2
1GE Healthcare China, Shanghai, Shanghai, China, ²Shanghai First People’s Hospital, Shanghai, Shanghai, China, ³GE Healthcare China, Beijing, Beijing, China

This preliminary study used regional homogeneity (ReHo), a novel resting-state fMRI parameter to investigate local brain activity changes in restless legs syndrome (RLS), a common neurological disorder characterized by an urge to move the legs and paraesthesias deep in the legs. Ten RLS patients and ten age- and gender-matched healthy controls were recruited for comparison. The RLS group showed increased ReHo regions in bilateral middle frontal gyrus, anterior cingulate cortex, caudate nucleus, insula, left putamen and thalamus as compared to normal controls, which might provide interesting insight into the mechanism of the disease.

Giulio Ferrazzi¹, Rita G. Nunes^1,2, Tomoki Arichi¹, Maryam Abaei¹, Emer Hughes¹, Anthony Price¹, and Joseph Hajnal^1,3
1Centre for the Developing Brain, King's College London, London, United Kingdom, ²Instituto de Biofisica e Engenharia Biomedica, Universidade de Lisboa, Lisbon, Portugal, ³Division of Imaging and Sciences and Biomedical Engineering, King's College London, London, United Kingdom

There is growing interest in fMRI of neonates. However, T2* is longer in immature brains, suggesting that optimal fMRI requires different parameter settings. We explore the use of echo-shifted EPI to detect task activation in a motor paradigm in neonates using longer echo-times to enhance sensitivity to the BOLD effect, first testing for a suitable echo time and then running dual echo and echo shifted sequences on a small group of infants. The pilot data obtained confirmed that longer echo times improve detection of motor activation and that signal recovery combined with increased efficiency can be achieved with echo-shifting.

Gregory Kirk¹, Rasmus Birn², and Andrew Alexander^3
1Waisman Laboratory for Brain Imaging and Behavior, University of Wisconsin,Madison, Madison, Wi, United States, ²Psychiatry, University of Wisconsin,Madison, Madison, Wi, United States, ³University of Wisconsin,Madison, Madison, Wisconsin, United States

We present a method of characterizing the state of functional connectivity of every vertex over the entire cerebral cortex on an individual subject level. The method is based on a discovered relation between the synchrony of bold fmri time series in a small neighborhood of a vertex and the scale of global functional connectivity of the time series at the vertex. The scale is defined by the number of vertices with epi time series correlation at or above a reference pearson linear correlation level r. The relation is demonstrated by a large scale computation of 120 resting fmri scans.

Venkata Veerendra Nadh Chebrolu¹, Suresh Joel¹, Brice Fernandez², Ek Tsoon Tan³, Luca Marinelli³, Dattesh Shanbhag¹, Radhika Madhavan¹, Rachel Connett⁴, Ajit Shankaranarayanan⁴, and John Schenck^5
1Medical Image Analysis Lab, GE Global Research, Bangalore, Karnataka, India, ²GE Healthcare, Muenchen, Germany, ³MRI Laboratory, GE Global Research, NY, United States, ⁴GE Healthcare, CA, United States, ⁵MRI Technologies & Systems, GE Global Research, NY, United States

In recent years, resting state functional connectivity (rs-FC) has been widely used to image brain function. rs-FC is measured by correlation of blood oxygen level dependent (BOLD) MRI signal time-courses. BOLD signal has contributions from cerebral blood flow (CBF), cerebral blood volume (CBV) and cerebral metabolic oxygen rate (CMRO2). The initial magnetization (M) is associated with CBF and CBV. Transverse relaxation rate (R2*) is associated with CMRO2, which is related more specifically with neuronal activity than CBF and CBV. In this work, we estimate functional connectivity using R2* time-courses and compare the results with conventional BOLD rs-fMRI.

Mark Chiew¹, Nadine N Graedel¹, Jennifer A McNab², Stephen M Smith¹, and Karla L Miller^1
1FMRIB Centre, University of Oxford, Oxford, Oxfordshire, United Kingdom, ²Radiology, Stanford University, California, United States

In this work, the k-t FASTER method for accelerating resting state FMRI data acquisition is demonstrated using a 3D hybrid radial-Cartesian acquisition, which facilitates accelerated reconstruction at multiple temporal resolutions. We combine k-t FASTER acceleration with parallel imaging, using radial acceleration factors ranging from R=1.67 to R=12.5x to produce 2 mm isotropic whole brain images at TRs ranging from 3 s down to 0.4 s. Resting state network expression is found to be optimised when the benefit of increased temporal degrees of freedom (provided by higher acceleration), is offset by diminishing reconstruction quality.

Hesamoddin Jahanian¹, Samantha Holdsworth¹, Thomas Christen¹, Hua Wu², Kangrong Zhu³, Adam Kerr³, Matthew J Middione⁴, Robert F Dougherty², Michael Moseley¹, and Greg Zaharchuk^1
1Department of Radiology, Stanford University, Stanford, CA, United States, ²Center for Cognitive and Neurobiological Imaging, Stanford University, Stanford, CA, United States, ³Department of Electrical Engineering, Stanford University, Stanford, CA, United States, ⁴Applied Sciences Laboratory West, GE Healthcare, Menlo Park, CA, United States

In an effort to distinguish cognitive states of the brain from rsfMRI data, we studied the dynamics of the whole-brain functional connectivity using high temporal sampling rate (TR=350 ms) Simultaneous MultiSlice (SMS) Resting-state fMRI. We probed the whole-brain functional connectivity in a wide frequency spectrum over a sliding window (duration:17.5 s, steps:7 s) and further characterized its dynamic changes into distinct connectivity states using k-means clustering.

Afonso Dias¹, Marta Bianciardi², Sandro Nunes¹, Rodolfo Abreu¹, Juliana Rodrigues¹, L. Miguel Silveira³, Lawrence L. Wald², and Patricia Figueiredo^1
1Institute for Systems and Robotics and Department of Bioengineering, Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Lisbon, Portugal, ²Department of Radiology, A.A. Martinos Center for Biomedical Imaging, MGH and Harvard Medical School, Boston, MA, United States, ³INESC-ID and Department of Electrical and Computer Engineering, Instituto Superior Técnico, Universidade de Lisboa, Lisbon, Lisbon, Portugal

Parcellation of the brain into functionally meaningful regions is a crucial step in studies of brain connectivity using complex network analysis methods based on resting-state fMRI (rs-fMRI). With the recent development of fast acquisition sequences at ultra-high-field (7T), high-spatial-resolution rs-fMRI can now be collected from the whole-brain with sufficient temporal resolution to capture the slow haemodynamic fluctuations underlying functional brain connectivity. We present a modification based on discrete Morse theory of a previously proposed method of hierarchical brain parcellation with rs-fMRI. We show that the method is able to produce parcellations at high levels of detail, with good intra-subject reproducibility.