Molecular Imaging Agents & Techniques
Hall D Tuesday 13:30-15:30
13:30
3202.
A Novel Targeted Iron Oxide Nanocolloid Agent for T1 and T2* Imaging of
Fibrin Using Conventional MR Techniques
Computer 25
Shelton
D. Caruthers1,2, Angana Senpan1, Dipanjan Pan1,
Mike J. Scott1, Patrick J. Gaffney3, Christian Stehning4,
Jochen Keupp4, Patrick M. Winter1, Samuel A. Wickline1,
Gregory M. Lanza1
1Washington
University, St. Louis, Missouri, USA; 2Philips Medical Systems,
Andover, Massachusetts, USA; 3St. Thomas' Hospital, London, UK; 4Philips
Research Europe, Hamburg, Germany
A
novel nanocolloid contrast agent with multiple iron-oxide crystals per
nanoparticle (1240 &[mu]g Fe per g of emulsion) has been targeted to fibrin
clot phantoms and human endarterectomy specimens in vitro. The agent can be
visualized as the typical signal dearth on T2* imaging, but also as bright
signal on conventional T1-weighted turbo spin echo imaging. The agent is constrained by size to the
vasculature and is predicted to allow imaging to be performed within minutes
post-injection.
Computer 25
Richard
Southworth1, Junji Chen2, Lei Zhang2, Megan
Kaneda2, Huiying Zhang2, Samuel Wickline2
1King's
College London, London, UK; 2Washington University School of
Medicine, St. Louis, Missouri, USA
Vascular
Cell Adhesion Molecule-1 (VCAM-1) is responsible for the tethering of
leukocytes to the vascular lumen in early inflammation. It has been implicated in the pathogenesis of
numerous inflammatory diseases, and therefore represents a potentially useful
molecular imaging target. We have
developed unique liquid perfluorocarbon nanoparticles which can be
functionalised with homing ligands in their outer lipid layer, allowing us to
target them to intravascular biomarkers of disease. These nanoparticles are capable of delivering
a targeted payload of over 50,000 Gd atoms, or by virtue of their high 19F
content (98% by volume), providing a quantifiable 19F MR signal. Here, we
describe their use in specific visualisation and quantification of VCAM-1
expression in the kidneys of atherosclerotic ApoE-/- mice.
14:30
3204.
First Results of an Ex-Vivo Experiment on Human Plaques Using a Contrast
Agent Targeting Activated Platelets
Computer 25
Dominik
Paul1, Constatin von zur Mühlen1, Julia Möller1,
Timo Spehl1, Christoph Bode1, Karlheinz Peter2,
Dominik von Elverfeldt1
1University
Hospital Freiburg, Freiburg, Germany; 2Baker Heart Research
Institute, Melbourne, Australia
Targeted
MRI contrast agents are gaining importance in clinical diagnostic. Here we
present an experimental ex-vivo environment for the evaluation of target
specific MRI contrast agents on human plaque tissue. The contrast agent
consists of microparticles of iron oxide and single-chain antibodies targeting
ligand-induced binding sites on activated glykoprotein IIb/IIIa-receptors.
Specific target binding resulted in a clear signal drop in high resolution,
high field T2 and T2* imaging and was verified by immunohistochemistry. The
experimental setup proves to be a promising tool for pre-clinical research on
target specific contrast agents.
15:00
3205.
Towards Dual-Mode Imaging of Vulnerable Plaques
Computer 25
Ben
Jarrett1, Bjorn Gustafsson1, Angelique Louie1
1UC
Davis, Davis, California , USA
We
have developed dual-mode imaging agents detectable by both MRI and PET for
detection and diagnosis of plaque vulnerability. The probes are targeted to macrophages, whose
density correlates with plaque vulnerability.
Computer 26
Sharon
Portnoy1, Jonathan Bishop1, Jun Dazai1,
Shoshana Spring1, R.M. Henkelman1,2
1Toronto
Centre for Phenogenomics, Toronto, Canada; 2University of Toronto,
Toronto, Canada
The
objective of this investigation is to determine the enhancement time-course and
dosage requirements in mice for intraperitoneally (IP) administered Gadolinium
contrast agents. Although higher doses
are required (~2.5 mmol/kg) and enhancement is slightly delayed, results
suggest that IP injection may be an effective method for Gadolinium
administration. The relative simplicity
of IP contrast administration compared to traditional tail-vein injection makes
this method a convenient alternative, particularly in longitudinal studies and
high-throughput imaging of large numbers of mice.
14:00
3207.
Fluorine-19 MRI of the Lung: First Human Experiment
Computer 26
Ursula
Wolf1, Alexander Scholz1, Maxim Terekhov1,
Kerstin Muennemann1, Karl Kreitner1, Christian Werner1,
Christoph Dueber1, Wolfgang Guenter Schreiber1
1Johannes
Gutenberg University, Mainz, Germany
As
fluorine-19 MRI of the lung appears to be a promising tool for the diagnosis of
obstructive lung diseases such as COPD, efforts have been made to improve image
quality and to develop a technique which can be safely and effectively used in
humans. Since 1984, a lot of animal experiments have been performed using SF6,
C2F6, C4F8, C3HF7, and CF4 as contrast gases. Here, we present the data of the
first human experiment using up to 78% SF6. This experiment is an important
milestone towards the use of fluorine-19 MRI in patients.
14:30
3208.
In-Vivo Ultra-High Resolution Imaging of Small Vessels Using Improved
Sensitivity and Long Circulation Time of FeCo-Graphitic Carbon Shell
Nanocrystals
Computer 26
Jin
Hyung Lee1, Sarah Sherlock1, Masahiro Terashima1,
Hisanori Kosuge1, Won Seok Seo1,2, Yoriyasu
Suzuki1, Michael V. McConnell1, Dwight G. Nishimura1,
Hongjie Dai1
1Stanford
University, Stanford, California , USA; 2Sogang University, Seoul,
Republic of Korea
FeCo-Graphitic
Carbon Shell Nanocrystals have been recently reported to have an unprecedented
high relaxivity with multi-functional capabilities. In this paper, we
demonstrate how the improved sensitivity provided by the high relaxivitiy
combined with the long circulation time can be used to generate high-resolution
in-vivo vessel images.
Computer 26
Jesus
Pacheco1,2, Paloma Ballesteros2, Sebastian
Cerdan1, Pilar Lopez-Larrubia1
1Instituto
de Investigaciones Biomedicas "Alberto Sols"- CSIC, Madrid, Spain; 2Instituto
Universitario de Investigacion - UNED, Madrid, Spain
Hypoxia
is known to be an important physiological parameter determining tumour
progression and malignancy. In this study, we report the synthesis and in vitro
evaluation of a new series of nitroimidazolyl derivatives as quantitative pO8 markers. The
NADPH:cytochrome P450 reductase enzymatic systems was found to reduced the
probes only under anoxic conditions. Incubation of one of this derivative with
C6 cells under different oxygen concentrations depicted clearly visible changes
in the 1H-NMR spectrum at the probe, which depended on the degree of
hypoxia.
13:30
3210.
Targeted MR Imaging of CD44-Positive Breast Cancer Stem-Like Cell
Phenotype
Computer 27
Dmitri
Artemov1, Wenlian Zhu1, Yoshinori Kato1,
Marie-France Penet1, Farhad Vesuna1, Zaver Bhujwalla1,
Venu Raman1
1Johns
Hopkins University, Baltimore, Maryland, USA
Recently,
a subpopulation of highly tumorigenic and drug resistant cancer cells was
identified in multiple solid tumors, which is currently referred to as cancer
stem-like cells. In breast cancer, these
cells are characterized by high level expression of the cell surface receptor
CD44 and decreased expression of another marker, CD24.. Here we have developed a CD44-targeted system
for MR imaging of breast cancer stem-like cells and demonstrated the MR imaging
of CD44 positive MDA-MB-231 human breast cancer cell in vitro.
14:00
3211.
Nanoglobular Gd-DO3A Conjugates as Highly Effective MRI Contrast Agents
Computer 27
Todd
Kaneshiro1, Eun-Kee Jeong, Dennis Parker, Zheng-Rong Lu1
1University
of Utah, Salt Lake City, Utah, USA
Novel
macromolecular Gd(III) chelates with well-defined nanosizes and globular
morphology were synthesized as highly effective MRI contrast agents. The nanoglobular MRI contrast agents resulted
in significant blood and tumor enhancement at a substantially reduced dose,
e.g. 10 µmol-Gd/kg.
14:30
3212.
Manganese-Alginate Gels for Controlled-Release of Mn2+
Computer 27
Christian
Brekken1, Ioanna Sandvig1, Olav Haraldseth1,
Gudmund Skjåk-Bræk1, Yrr Mørch1
1NTNU,
Trondheim, Norway
In
the present study we aimed at designing alginate beads for controlled release
of Mn2+. Both elemental analysis and swelling studies of the alginate gel beads
showed great differences in the ion binding properties of alginate to
manganese. Dynamic T1-weighted MRI of single alginate beads immersed into
NaCl-solution showed that the Mn2+ release rate differed by a factor of up to
~100% between the 4 differently designed beads imaged. The results indicate
that nano-fabrication of Mn-alginates can tailor-make biocompatible manganese
delivery systems and hopefully help in introducing MEMRI in targeted
contrast-enhanced MRI of otherwise toxicity-limited organs, such as the brain.
15:00
3213.
Novel Receptor-Targeted Nanoparticles for MR Imaging and Specific
Delivery of Gene Therapy
Computer 27
Panagiotis
Kyrtatos*1, Michele Writer*, Anthony N. Price1, Stephen
Hart, Mark F. Lythgoe1
1Institute
of Child Health and Department of Medicine, University College London, London,
UK
Multimodal
nanoparticles offer a promising application as MRI contrast agents with
therapeutic capabilities. Our group is developing a novel gene therapy vector,
which utilises an anti-cancer targeting peptide for specific transfection of
tumour cells with therapeutic genes. In addition, a Gd3+ moiety has been
incorporated for non-invasive real-time monitoring of the delivery. Here we
present a pilot in-vitro tumour cell study investigating the MRI properties of
the vector, confirming its potential for both specificity of delivery of DNA to
the target cells for gene therapy and also providing evidence of the MR
contrast enhancement in targeted cells.
13:30
3214.
MR Imaging of Breast Cancer Using the Folate-Receptor Targeted Contrast
Agent P1133
Computer 28
Reinhard
Meier1,2, Tobias D. Henning1, Sophie
Boddington1, Sidhartha Tavri1, Sandeep Arora1,
Claire Corot3, Ernst J. Rummeny2, Heike E. Daldrup-Link1
1University
of California San Francisco, San Francisco, California , USA; 2Technical
University Munich, Klinikum Rechts der Isar, Munich, Germany; 3Guerbet,
Paris, France
The
purpose of this study was to assess the uptake of the new FR-targeted USPIO
P1133 in breast cancers. In vivo studies demonstrated a progressive enhancement
of central tumor areas with the FR-targeted USPIO P1133 in FR-positive
MDA-MB-231 breast cancers. Corresponding SNR data were significantly higher for
P1133 compared to P904B, indicating at least a component of FR-specific
enhancement with P1133. The P1133 tumor uptake was not significantly inhibited
by FFA in vivo, most likely due to the rapid FFA metabolism in the liver. Thus,
the FR-targeted USPIO P1133 provides a significant and specific enhancement of
FR-positive MDA-MB-231 breast cancers.
14:00
3215.
Synthesis and Characterization of a Redox- And Light-Responsive MRI
Probe
Computer 28
Chuqiao
(Tom) Tu1, Ryan Ngao1, Angelique Louie1
1UC
Davis, Davis, California , USA
We
have previously developed a gadolinium contrast agent that reversibly changes
relaxivity properties in response to irradiation by different wavelengths of
light. We here demonstrate that the
contrast agent can also be modulated by reduction/oxidation. The mechanism for the relaxivity effects in
response to redox appear to differ from the light induced response.
14:30
3216.
Quantitative Molecular Imaging with a Combined Fluorescence Diffuse
Optical Tomography and MRI System
Computer 28
yuting
lin1, Orhan Nalcioglu2, Gultekin Gulsen2
1University
of California, Irvine, California , USA; 2University of California,
Irvine, California , USA
An
ideal molecular imaging technique should have both high sensitivity for
molecular probes and also provide high-resolution images. Our solution to this
demanding requirement is to employ a multimodality imaging strategy. In this
study, we show the feasibility of using a combined MRI and optical fluorescence
imaging approach to quantitatively resolve the fluorescence contrast agent
concentration. The true fluorophore concentration was recovered only if the MRI
anatomical information was employed.
Computer 28
Thomas
Sheung Chee Ng1,2, Daniel Procissi1, Andrey
Demyanenko1, Yibao Wu3, Ciprian Catana4, Simon
R. Cherry3, Andrew A. Raubitschek5, Russell E. Jacobs1
1California
Institute of Technology, Pasadena, California, USA; 2University of
Southern California, Los Angeles, California , USA; 3University of
California, Davis, Davis, California ,
USA; 4Massachusetts Gener
Positron
Emission Tomography (PET) and Magnetic Resonance (MR) offer complementary
functional and anatomic information that provide unique windows into biological
processes. As well as integrating images in space and time, the combined PET/MR
scanner offers the potential to perform real time analysis of multi-modal data
that can feedback to direct further studies in a single imaging session. We
demonstrate the feasibility of this paradigm. Mice implanted with tumor cells
were imaged simultaneously with PET/MR. Functional data derived from the PET
was used as a basis for 1H-MR spectroscopy studies, demonstrating metabolic
heterogeneity within a tumor cell mass.
Hall D Wednesday 13:30-15:30
Computer 25
Sven
Gottschalk1, Michaela Hohnholt2, Dieter Leibfritz2,
Marc Bilodeau1, Claudia Zwingmann1
1University
of Montréal, Montréal, Canada; 2University of Bremen, Bremen,
Germany
Application
of stable isotope labeling in vitro is a powerful method to study metabolic
pathways and fluxes. We applied [3-13C]pyruvate on primary mouse
hepatocyte cultures to establish labeled pyruvate for metabolic studies and
flux analysis in this cellular model. Our results show: [3-13C]pyruvate
was metabolized by lactate dehydrogenase, alanine-/aspartate-aminotransferase,
PC, PDH and subsequent metabolic pathways through the TCA-cycle. Considering
that almost no 13C-NMR studies in isolated hepatocytes have been
performed so far, labeled pyruvate will provide an important physiological
substrate to assess hepatocellular pathways and the de novo synthesis of metabolites in these cells under normal and
pathological conditions.
14:00
3219.
A Versatile NMR-Compatible Bioreactor for Alginate-Encapsulated Liver
Cells
Computer 25
Rex
Errol Jeffries1, Kayvan R. Keshari, Chris Seagle, Peter
Pediaditakis, Michael P. Gamcsik, John Kurhanewicz, Jeffrey M. Macdonald
1University
of North Carolina, Chapel Hill, USA
Synopsis:
A remodeled NMR-compatible bioreactor was used with electrostatic alginate
encapsulation of hepatocytes to create a bioartificial liver ultimately
designed for metabolomic studies. The 500 μm diameter spherical
encapsulates are well perfused and permit the use of 20% oxygen rather than the
standard 95% oxygen. The 31P NMR spectra from a rat liver cell line, JM1, and
primary rat hepatocytes were compared to oxygen uptake rates for bioenergetics.
14:30
3220.
Metabolism of Colonic Mucosa in Patients with Ulcerative Colitis (UC)
and Crohn’s Disease (CD): An NMR Study
Computer 25
B
Krithika1, S Kumar1, R R. Singh1, Uma Sharma1,
V Ahuja1, G K. Makharia1, N R. Jagannathan1
1All
India Institute of Medical Sciences, New Delhi, India
In
vitro NMR spectroscopy demonstrated significant differences in the
concentration of amino acids (isoleucine/leucine/valine, glutamic acid +
glutamine, alanine), membrane components (choline, glycerophosphorylcholine,
myoinositol), glycolytic product (lactate) in patients with active state of
ulcerative colitis (UC) and Crohn’s disease (CD) compared to controls
indicating a decreased protein and carbohydrate metabolism. In the remission state, levels of most of the
metabolites were similar to controls. A significant difference in the
concentration of formate was observed between patients with active states of UC
and CD suggesting its potential as a biomarker for distinguishing UC and CD.
Computer 25
Sven
Gottschalk1, Michaela Hohnholt2, Dieter Leibfritz2,
Claudia Zwingmann1, Marc Bilodeau1
1University
of Montréal, Montréal, Canada; 2University of Bremen, Bremen,
Germany
We
have recently shown that the initial phase of the apoptotic process is
associated with alterations in specific glucose metabolic pathways. The
synthetic glucocorticoid Dexamethasone is known for its anti-apoptotic effects
and impact on glucose metabolism (in particular, anaplerosis and
gluconeogenesis). We therefore characterized the effects of Dexamethasone on
hepatocellular metabolism with multinuclear NMR-measurements. Our results
further support the strong relationship noted between changes in cellular
metabolism and apoptosis. This paves the way toward the therapeutic modulation
of cell metabolism in order to influence upon cell survival/death particularly
in the context of liver injury.
Hall D Wednesday 13:30-15:30
13:30
3222.
In Vivo Imaging of Endogenous Neural Stem Cells Labelled with a
Ferumoxide-Polycation Complex Using a Low Field System
Computer 26
Rachael
Dobson1,2, Patrizia Ferretti1, Mark Lythgoe1
1University
College London, London, UK
Adult
neural stem cells (NSC) migrate towards the olfactory bulb and differentiate
into neurons. Ferumoxide-based contrast agents have been used to track the
migration of exogenously-labelled NSC, there are limited methods that allow for
longitudinal tracing of endogenous neural stem cells using MRI. In this study,
two MRI contrast agents were used to label the endogenous population of NSC and
neuroblasts in situ in the subventricular zone.
Cell labelling and T2* contrast were assessed up to 28 days
post-labelling using a low-field system.
Cell migration was observed in vivo and ex vivo, and histology confirmed
labelling of migrating neuroblasts.
14:00
3223.
Quantification of Cell Trafficking in Vivo Using Magnetically Sensitive
Histograms
Computer 26
Christopher
M. Long1, Hyam I. Levitsky1, Jeff W.M. Bulte
1Johns
Hopkins University School of Medicine, Baltimore, Maryland, USA
Quantification
of dendritic cell migration to lymph nodes in vivo is critical for evaluating
the efficacy of tumor vaccines. We
developed a new method based on magnetically sensitive histograms and cell-cell
transfer of SPIO particles in vivo. It
was found to correlate exceptionally well with ex vivo cell counts after
magnetically activated cell sorting The
method may be used to monitor the biological variability associated with cancer
vaccination and to evaluate the efficacy of immunoadjuvants.
14:30
3224.
In Vivo Stem Cell Tracking Using Magnetic Resonance Imaging in the Rat
Genitalia of a Radical Prostatectomy Model
Computer 26
Young
Taik Oh1, Jang Hwan Kim1, Yong-Min Huh1,
Myeong-Jin Kim1, Jin-Suck Suh1
1Yonsei
University College of Medicine, Seoul, Republic of Korea
Erectile
dysfunction is a major complication after radical prostatectomy for prostate
cancer. There have been vigorous trials for improving erectile dysfunction but
no satisfying method until now. Previous reports have shown that stem cell
injection improved erectile function in a rat model erection dysfunction. However,
they used the immunohistochemical stain for the evaluation of stem cell. A
reliable in vivo imaging method to localize transplanted cells and monitor
their restorative effects will enable a systematic investigation of cell
therapy. Our results showed that in vivo stem cell tracking using MR imaging in
a rat model of radical prostatectomy was feasible.
15:00
3225.
Comparison of Different Iron-Oxide Agent Detection Methods Using a
Single Dataset
Computer 26
Gopal
Varma1, Richard Tavare1, Hannes Dahnke2,
Stephen Keevil1, Tobias Schaeffter1
1King's
College London, London, UK; 2Philips Research Laboratories, Hamburg,
Germany
Detection
of super-paramagnetic iron oxide can be achieved at varying sensitivities
through its inherent negative contrast or by positive contrast methods. A
single multi-echo dataset is used to compare T8* weighted imaging, R8*
mapping, and positive contrast methods. In our study this includes inversion
recovery on-resonance (IRON), gradient dephasing (“white marker”), and
susceptibility gradient mapping (SGM). T8* images are found to be
most sensitive to the smaller concentrations at short echo times (TEs), but
high sensitivity of SGM and T8* imaging suggests a combination of
these two methods is ideal. Data acquired for T8* images and SGM at
2 different TEs to address different concentrations would provide a more
sensitive differentiation to the background.
13:30
3226.
Stem Cell Treatment for Peripheral Arterial Disease - Comparison of
Plain and Encapsulated X-Ray Visible Mesenchymal Stem Cell Transplants
Computer 27
Dorota
A. Kedziorek1, Wesley D. Gilson2, Kenyatta Cosby1,
Matthias Stuber1, Bradley P. Barnett1, Royston C. Boston3,
Grigorios Korosoglou4, Bernard E. Kohl1, Gary Huang1,
Brady Sieber1, Aravind Arepally1, Jeff W.M. Bulte1,
Lawrence V. Hofmann5, Dara L. Kraitchman1
1Johns
Hopkins University, Baltimore, USA; 2Siemens, Baltimore, USA; 3University
of Pennsylvania, Kennet Square, USA; 4University of Heidelberg,
Heildelberg, Germany; 5Stanford University, USA
The
resented work comapares plain and encapsulated Mesenchymal Stem Cell therapy in
Peripheral Arterial Disease model. Non-invasive imaging demonstrated a more
robust angiogenic response with encapsulated MSCs, while "naked" MSCs
can incorporate into the vessel walls.
Computer 27
Yidong
Yang1,2, Ben Waghorn1,2, Yuhui Yang1,
Brianna Klein1, Nathan Yanasak1, Tom C-C. Hu1,2
1Medical
College of Georgia, Augusta, Georgia, USA; 2Georgia Institute of
Technology, Atlanta, Georgia, USA
Inflammatory
process plays an important role in myocardial injury and the subsequent
recovery. In this study, we examined cell mobilization due to myocardial
infarction post injection of micrometer-sized particles of iron oxide (MPIOs).
MRI was performed to potentially visualize the migration and trafficking of
inflammatory cells as well as to provide quantitative information. This
technique indicates potential to track disease progression in a preclinical
model of myocardial injury.
14:30
3228.
In Vitro MR Thermometry on Magnetically Heated Iron Oxide Labeled Stem
Cells
Computer 27
Daniel
Haddad1, Jochen Lorenscheit2, Markus Hildenbrand1,
Meike Weber2, Regina Ebert2, Peter Michael Jakob1,2
1MRB
Research Center Magnetic-Resonance-Bavaria, Würzburg, Germany; 2University
of Würzburg, Germany
Iron-oxide
nanoparticles are commonly used as MR markers to label and track (stem) cells.
Magnetic particle heating (magnetic field hyperthermia) can be used to heat the
iron-oxide particles and thus the iron labeled stem cells. A proof-of-principle
in vitro study shows that MR thermometry can be used to visualize the spatial
and temporal heat distribution with high accuracy in the vicinity of iron-oxide
labeled stem cells that were heated outside the MR spectrometer via magnetic
particle heating. Since the heating capacity depends on the composition and
structure of the material of interest, different types of iron-oxide particles
yield different heating rates. The heating rates can also be controlled by
varying the amount of iron-oxide incorporated in the cells.
15:00
3229.
Tracking of Spio Labeled Natural Killer Cells to Epcam Positive Prostate
Cancer with Mr Imaging
Computer 27
Reinhard
Meier1,2, Sidhartha Tavri1, Tobias D. Henning1,
Winfried S. Wels3, Rick Baehner1, Ernst J. Rummeny2,
Heike E. Daldrup-Link1
1University
of California San Francisco, San Francisco, California , USA; 2Technical
University Munich, Klinikum Rechts der Isar, Munich, Germany; 3Chemotherapeutisches
Forschungsinstitut, Frankfurt, Germany
Development
and optimization of a technique for labeling of human natural killer (NK) NK-92
cells and genetically engineered NK-92-scFv(MOC31)-zeta cells, targeted to the
Ep-CAM (epithelial cell adhesion molecule) on prostate cancer cells, with
superparamagnetic iron oxides (SPIO). SPIO labeled NK cells were injected
intravenously into rats with implanted Ep-CAM positive prostate cancers and the
accumulation of the SPIO labeled NK cells in prostate tumors could be monitored
with a non-invasive magnetic resonance (MR) imaging technique.
13:30
3230.
Study on the Magnetic Relaxation of Superparamagnetic Nanotubes as
Magnetic Resonance Imaging Contrast Agent
Computer 28
xia
Bai1, sang jun Son1,2, shixiong Zhang1,
wei Liu3, elaine k. Jordan4, joseph a. Frank4,
Thirumalai Venkatesan1, Sang Bok Lee1
1university
of maryland college park, college park, Maryland, USA; 2Kyungwon
University, Republic of Korea; 3Philips Research North America,
Briarcliff Manor, New York, USA; 4National Institutes of Health,
Bethe
This
work describes the synthesis of magnetic nanotubes (MNTs) which is silica
nanotube (SNT)/magnetic iron oxide nanoparticle (MION) composite with MIONs
loaded in SNTs. This unique structure combines the easy chemistry for
differential functionalization of the inner and outer surfaces of SNTs with the
superparamagnetic characteristics of MIONs. The MNTs had well-controlled
dimensions and retained a high saturation magnetization. It was proved that
MNTs worked as an effective T2 magnetic resonance (MR) contrasting agent. The
in vitro cell labeling was effective without showing significant cytotoxicity.
Our results indicate that MNTs could be an ideal candidate for image-guided
drug delivery.
Computer 28
Jian
Yang1,2, Jian Xin Liu3, Gang Niu2,
Yong Liu3, Ed X. Wu4
1Department
of Electrical and Electronic Engineering, The University of Hong Kong , Hong
Kong, People's Republic of China; 2The First Hospital of Xi'an
Jiaotong University, Xi'an, People's Republic of China; 3The School
of
In
this study,10-day-old normal rats (n=6) and hypoxic-ischemic (H-I) neonatal
rats (n=6) were injected with the micronsized iron oxide particles (MPIOs) into
the anterior lateral ventricle. 2D and 3D gradient echo MRI was performed with
a 7T animal scanner in hour 3, day 3, day 7 and day 14 after the MPIOs
injection. Then animals were sacrificed for double staining with iron and
mature neurons. In normal neonatal rat brain, the migrating pathway of the
endogenous neural progenitor cells (NPCs) with MPIO is mainly along the rostral
migratory stream to the olfactory bulb. In H-I neonatal rat brain, the
migration of endogenous NPCs with MPIO is mainly toward injured boundary. MRI
can facilely detect the above migrations in 2 weeks. Therefore, in vivo
magnetic cell labeling of endogenous NPCs with MPIO and subsequently
non-invasive, serial MRI monitoring should open up a new approach to probe into
the mechanism of cell migration in the developmental brain under physiological
and pathologic states.
14:30
3232.
Quantitative Assessment of Magnetically Labeled Luciferase Positive
Cells Using Multimodality Imaging
Computer 28
April
M. Chow1,2, Kwan Man1, Jerry S. Cheung1,
Tracy Y. Chow1, Ed X. Wu1
1The
University of Hong Kong, Pokfulam, Hong Kong
Cellular
imaging using magnetically labeled luciferase positive cells with multimodality
imaging has found numerous biological applications. However, studies only
involve imaging dual-labeled cells, tracking their migrations without
quantification of local density of the dual-labeled cells. Quantitative
analysis of cells after transplantation may allow more accurate assessment of
cell delivery and subsequent distribution and migration, which can lead to more
effective monitoring and optimization of therapeutic paradigms. In this study,
we demonstrated that dual-labeled cells exhibited excellent linear correlations
between cell concentration with photon flux and &[Delta]R2* in vitro,
illustrating that quantitative assessment of dual-labeled cells can be
achieved.
Computer 28
Olaf
Saborowski1, Francesco Santini2, Melpomeni Fani3,
Philippe Robert4, Sebastien Ballet4, Jean Sebastien
Raynaud4, Robin Santus4, Johannes Froehlich3,
Klaus Scheffler2, Georg Bongartz3, Helmut R. Maecke3
1University
Hospital Basel, Basel , Switzerland; 2University Basel/University
Hospital, Basel, Switzerland; 3University Hospital Basel, Basel,
Switzerland; 4Guerbet Group, Aulnay-Sous-Bois, France
Iron
oxide particles (SPIO and USPIO) are widely used for MRI-based cell labeling
and tracking because they can be loaded on cells by simple protocols and
provide high image contrast due to their large susceptibility effect. We
compared 2 different cell labeling models (AGAR and Ficoll) in human KB cells.
Both presented cell labeling methods (AGAR and Ficoll model) are feasible for
in vitro MR characterization of KB tumor cells at 1.5, 2.35 and 3T. Ficoll
method is easier to perform because it does not need a heating procedure for
preparation without air bubbles as compared to AGAR method.
13:30
3234.
Longitudinal Tracking and Quantification of T Cells Using in Vivo 19F
MRI
Computer 29
Mangala
Srinivas1, Michael S. Turner2, Penelope A. Morel2,
Jelena M. Janjic1, David H. Laidlaw3, Eric T. Ahrens1,4
1Carnegie
Mellon University, Pittsburgh, Pennsylvania, USA; 2University of
Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA; 3Brown
University, Providence, USA; 4Pittsburgh NMR Center for Biological
19F
MRI was used to track T cells in a murine model of acute inflammation for up to
21 days. Quantification of total 19F signal in the DLN, and hence
apparent cell numbers, from the in vivo
data showed 2.1x106 ± 9x105 apparent cells at day 2 and
3.6x106 ± 9x105 at day 7 post-transfer in the draining
lymph node. The murine inflammation model can be applied to study the effect of
therapeutics on T cell trafficking. The in
vivo 19F platform developed can readily be extended to other
cells types including, immunotherapeutics or stem cells.
14:00
3235.
A Novel Bimodal Fluorescent and Paramagnetic Lipid for Cell Labelling
and Tumour Magnetic Resonance Imaging
Computer 29
Nazila
Kamaly1, Tammy Louise Kalber1, Jimmy D. Bell1,
Michael R. Jorgensen1, Andrew David Miller1
1Imperial
College London, London, UK
A
novel bimodal fluorescent and paramagnetic lipid, Gd.DOTA.Rhoda.DSA was
synthesised. Cationic and neutral PEGylated bimodal liposomes were formulated
for cell labelling and tumour imaging respectively. Effective IGROV-1 (human
ovarian carcinoma) cell labelling was demonstrated in vitro post incubation of
cells with the cationic liposome formulations containing Gd.DOTA.Rhoda.DSA. The
enhanced permeation and retention (EPR) effect of tumour tissue was exploited
post-injection of the neutral PEGylated liposomes in mice to image IGROV-1
xenografts. Tumour T1 values were reduced 14-24 h post-injection of the bimodal
liposomes by a substantial 65%. These MRI findings were supported by
fluorescence findings co-validating the presence of the bimodal liposomes/lipid
within tumour tissue.
14:30
3236.
Bimodal Intracellular Nanoparticles Based on Quantum Dot Technology for
High Field MR Microscopy at 21.1 T
Computer 29
Jens
Thorvald Rosenberg1,2, Joshua M. Kogot1,
Goeffery F. Strouse1, Samuel Colles Grant1
1The
Florida State University, Tallahassee, USA
A
bimodal contrast agent for the intracellular transfection of mammalian cells
has been optimized for high magnetic field MR imaging at 21.1 T. With a
fluorescent InP/ZnS quantum dot as a substrate, a peptide-bound lanthanide (Dy3+)
component has been added to generate MR contrast. The CAAKA–DOTA-Ln3+-Qdot
agent displays strong MR contrast enhancement for all relaxation mechanisms,
but the strongest contrast for intracellular loading is seen with T2/T2*
weighting. Utilizing this agent, Chinese Hamster Ovary (CHO) cells were
transfected, with MR and confocal images displaying contrast enhancement and
the intracellular localization of nanoparticles.
15:00 3237.
19F NMR Molecular
Imaging: Cell Labeling With Emulsified Perfluoro-15-Crown-5-Ether
Computer 29
Samir
Mulla-Osman1, Henrike Goetze2, Ute Bommerich3,
Johannes Bernarding
1University
of Magdeburg, Magdeburg, Germany; 2University of Magdeburg, Germany;
3Leibniz Institute for Neurobiology, Germany
Monitoring
19F-labelled cells is an important new technique to evaluate in vivo
the fate of implanted cells such as in stem cell therapy. The use of 19F-marker
substances such as perfluorocarbons (PFC) has been explored by 19F
spectroscopy and magnetic resonance imaging (MRI) for cell tracking. We present
first results of single volume selective spectroscopy of fibroblasts, which
have been labeled with perfluoro-15crown-5-ether (PFCE) emulsion.
Hall D Thursday 13:30-15:30
13:30
3238.
MEGA-Editing of
Spermine/Spermidine in Healthy Human Prostates Using External
Phased-Array Coil Assembly
Computer 25
Michael
Albert Thomas1, S.Sendhil Velan2, Saadalah Ramadan3,4,
Nagarajan Rajakumar1, Daniel J. Margolis1, Maria Ana Gomez1,
Steve S. Raman1, Carolyn E. Mountford3,4
1UCLA,
Los Angeles, California , USA; 2West Virginia University,
Morgantown, West Virginia, USA; 3Brigham and Women's Hospital,
Boston, Massachusetts, USA; 4Harvard Medical School, Boston,
Massachusetts, USA
Polyamines
such as spermidine and spermine play a critical role in cell proliferation and
differentiation in cancer. Overlap of metabolite resonances is a major concern
in conventional one-dimensional (1D) MR Spectroscopy (MRS), hence
spectral-editing techniques either based on signal subtraction or
multiple-quantum filter have been proposed and tested in human brain in vivo. A
goal of this work was to implement a MEGA-based spectral editing sequence on a
3T MRI scanner, to optimize the technique using prostate phantom, and to
evaluate the feasibility of detecting spermine/ spemidine in healthy human
prostates using an external phased-array matrix coil assembly.
14:00
3239.
Optimized MRS of Neurotransmitters: How Far Do You Need to Go?
Computer 25
Paul
Mullins1,2, Hongji Chen2, Jing Xu2,
Arvind Caprihan2, Charles Gasparovic2,3
1Bangor
University, Bangor, UK; 2The MIND Research Network, Albuquerque, New
Mexico, USA; 3University of New Mexico, Albuquerque, New Mexico, USA
Standard
and optimized MRS sequences for the detection of Glutamate and other
neurotransmitters are compared for test re-test reliability.
14:30
3240.
High Speed Multi-Voxel Thermography with Free Induction Decay Echo
Planar Chemical Shift Imaging
Computer 25
Matthew
A. Neimark1, Scott Henneman2, Yingli Yang2,
Jae Choi2, Angelos Aristeidis Konstas3, Melvyn B. Ooi,
Hamed Mojahed2, Andrew F. Laine2, John Pile-Spellman2,
Truman R. Brown2
1Columbia
University, New York, New York, USA; 2Columbia University, New York,
USA; 3Massachusetts General Hospital, Boston, USA
In
this study, we present a fast method of performing spectroscopic imaging using
free induction decay echo planar chemical shift imaging (FID EP-CSI). This technique was performed on a phantom
made up of Ringer’s lactate and 10 mM NAA.
The phantom was cooled from 38 to 29°C while temperature was
continuously monitored, and 13 FID EP-CSI acquisitions were performed. Each acquisition was 28 seconds. This is 12 times faster than conventional
CSI. For voxels in a central region, the overall regression for NAA-H8O difference (δNAA-H8O) vs. temperature (±95% confidence intervals)
was T=(100.67±1.07)δNAA-H8O+300.36±2.86 (R2=0.9967;
rms=0.16°C).
15:00
3241.
Fully Automated Shimming for High Lipid Regions Using Phased Arrays at
3T
Computer 25
Gamaliel
Isaac1, Jeremy jeremy.magland@upenn.edu Magland, Hoby Patrick
Hetherington2
1University
of Pennsylvania, Philadelphia, Pennsylvania, USA; 2Yale University,
New Haven, Connecticut, USA
We
have developed a robust fully automated method for shimming regions of the body
with high lipid content which is compatible with phased array detection. The
method utilizes multi evolution delay B0 maps to generate high accuracy while
retaining a large bandwidth for poor starting homogeneity. The method includes
an embedded Dixon image to identify fat and water content such that the
chemical shift due to lipid is corrected for allowing for the use of arbitrary
evolution times. To demonstrate the method we have applied it to shim the human
calf at 3T on a Siemens Trio system.
13:30
3242.
SVD 8-Channel Coil Combine for 2D Chemical Shift Imaging (CSI) MRI
Computer 26
Mithun
Diwakar1, Mingxiong Huang1,2, Roland R. Lee1,2,
Rebecca J. Theilmann1
1UCSD,
La Jolla, California , USA; 2VA San Diego Healthcare System, San
Diego, California , USA
This
work describes a novel method for combining 2D Chemical Shift Imaging (CSI)
data recorded by 8 independent phased array coils (8-channel GE head
coil). Singular Value Decomposition
(SVD) is a form of dominant mode analysis that can be used to combine time
decay signals. The work was carried out
on a phantom containing physiologic concentrations of key metabolites found in
the brain (NAA, Ch, Cr). The advantages of SVD over conventional methods of
combining signals from multiple coils include preservation of phase-information,
automatic zero-order phasing, and no need of a priori information.
14:00
3243.
Improved Spectral Quality Through Enhanced Shimming on a Clinical
Platform at 3T and 7T
Computer 26
Ralf
Mekle1,2, Giulio Gambarota1, Vladimir Mlynarik1,
Rolf Gruetter1,3
1Ecole
Polytechnique Federale de Lausanne, Lausanne, Switzerland; 2University
of Lausanne, Lausanne, Switzerland; 3Universities of Lausanne and
Geneva, Lausanne and Geneva, Switzerland
Reliable
localized shimming is an essential prerequisite for obtaining high-quality MR
spectra, especially at higher fields (≥3T). For the first time, the
fast automatic shimming technique by mapping along projections (FASTMAP) and
its descendant FASTESTMAP were implemented on a clinical platform of a 3T and a
7T system. In vivo proton spectra from various brain regions were acquired and
showed excellent quality due to the enhanced local field homogeneity resulting
from improved shimming with FASTMAP/FASTESTMAP that translated into reduced
water and metabolite linewidths at both field strengths.
14:30
3244.
Application of Forward Linear Prediction Method to High-Resolution NMR
Spectra in Inhomogeneous Fields
Computer 26
Hai
Feng1,2, Zhiwei Chen1, Shuhui Cai1,
Xiaohong Wang1, Ji Feng3, Zhong Chen1
1Xiamen
University, Xiamen, People's Republic of China; 2Institute of
Physics, Chinese Academy of Sciences, Beijing , People's Republic of China; 3Institute
of Physics, Chinese Academy of Sciences, Beijing, People's Republ
The
forward linear prediction (LP) method was applied to deal with 2D
intermolecular double quantum coherence spectra from inhomogeneous fields. The
results show that compared to normal discrete Fourier transform, the use of
forward LP extrapolation can shorten sampling time by a factor of eight or more
at the same level of sensitivity and resolution. It can effectively extend the
data sets acquired from inhomogeneous fields even for shorter data records and
lower signal-to-noise ratio.
Computer 26
Ralf
Mekle1,2, Vladimir Mlynarik1, Giulio Gambarota1,
Martin Hergt3, Gunnar Krueger3, Rolf Gruetter1,4
1Ecole
Polytechnique Federale de Lausanne, Lausanne, Switzerland; 2University
of Lausanne, Lausanne, Switzerland; 3Siemens Medical Solutions-CIBM,
Lausanne, Switzerland; 4Universities of Lausanne and Geneva, Lau
The
implementation of the spin echo full intensity acquired localized (SPECIAL)
spectroscopy technique on a clinical platform was taken to the next level by
combining interleaved water suppression (WS) and outer volume saturation (OVS),
optimized sequence timing, and large B1 fields producing coils in addition to
improved shimming. High-quality single voxel spectroscopy (SVS) data of the
human brain were acquired at TEs below 6 ms on 3T and 7T systems. The high SNR
of the spectra enabled reliable metabolite quantitation at both field
strengths. Moreover, the enhanced sensitivity at the higher B0 field allowed a twofold
reduction in scan time.
13:30
3246.
Automating Brain Tumour Classification Using High Resolution Magic Angle
Spinning Data
Computer 27
Jean-Baptiste
Poullet1, Daniel Monleon2, M. Carmen Martinez-Bisbal3,4,
Bernardo Celda3,4, Sabine Van Huffel5
1Katholieke
Universiteit Leuven , Leuven, Belgium; 2Fundación Investigación
Hospital Clínico Universitario Valencia, Valencia, Spain; 3University
of Valencia, Spain; 4Networking Research Center on Bioengineering
The
limited success of in vivo MRS to classify some brain tumor types tends to
encourage the use of other data types. Thanks to their narrow line widths and
their large signal-to-noise ratios, high resolution magic angle spinning
(HR-MAS) data provides much more information than in vivo MRS. Since biopsy
extraction is a routine procedure in the clinical practice, incorporating HRMAS
in the decision system is a reasonable solution to improve brain tumor
diagnosis. This study propose a fully automated procedure to classify HR-MAS
spectra.
14:00
3247.
Proton Decoupled 31P MRS of Head and Neck Tumors in Vivo at
1.5T and 3T
Computer 27
Sungheon
Kim1, Sanjeev Chawla1, Laurie A. Loevner1,
Harry Quon1, Eric J. Sherman1, Gregory S. Weinstein1,
Harish Poptani1
1University
of Pennsylvania, Philadelphia, USA
The
purpose of this study was to implement proton decoupled 31P MRS at
3T for assessing smaller tumors and to investigate its feasibility in
predicting treatment response in head and neck tumors. Proton decoupled 31P
MRS data were acquired from eleven patients who were treatment naïve.
Substantial increase in SNR with smaller voxel sizes was achieved at 3T as
compared to 1.5T. In a small cohort of
patients, studied at 1.5T, 31P MRS studies partial responders
exhibited higher PME/βNTP and PDE/βNTP ratio
than complete responders indicating the potential of 31P MRS in
prediction of treatment response.
14:30
3248.
In Vitro Metabolites Limit of Detection by Localised NMR Spectroscopy
Using Micro Coils
Computer 27
Nicoleta
Baxan1, Aziz Kadjo1, Guillaume Pasquet1, Jean-François
Chateaux1, André Briguet1, Pierre Morin1,
Danielle Graveron-Demilly1, Latifa Fakri-Bouchet1
1Université
Lyon1, Villeurbanne, France
We
present a new concept of receiver planar micro coil fabricated by
electroplating technique. The innovative part of this work concerns:
-
Implemented MR instrumentation at micro-scale,
-
Restricted volumes detection by MR spectroscopy and imaging.
The
spectrum of a solution containing eleven MR-observable 1H metabolites in human
brain (50 mM) is presented, the concentration limit of detection (LODc) for the
Choline and NAA case are estimated at 3.3 mM and 10.4 mM respectively. The
measurements reproducibility was established using six micro coil prototypes.
The
micro coils performance improvement open the way to highly spatially resolved
explorations on animal model.
15:00
3249.
Diffusion-Weighted 1H NMR Single Voxel Spectroscopy at 3T and 7T
Computer 27
Daniela
Wagegg1, Ralf Lützkendorf2, Wolfgang Dreher3,
Claus Tempelmann4, Jörg Stadler5, Johannes Bernarding2
1Institute
for Biometry and Medical Informatics, Medical
Faculty, University of Magdeburg, Magdeburg, Germany; 2Institute
for Biometry and Medical Informatics, Medical Faculty, University of Magdeburg,
Magdeburg, Germany
We
implemented Diffusion-weighted Single Voxel Spectroscopy on a 3T- and
7T-MR-Scanner. The Sequence was based on a STEAM-Sequence. The DW-gradients
where applied simultaneously in each direction. In phantom studies, this
allowed the calculation of the ADC values of several brain metabolites
successfully.
Hall D Thursday 13:30-15:30
13:30
3250.
Sodium Mapping in Focal Cerebral Ischemia in the Rat by Quantitative 23Na
MRI
Computer 28
Victor
E. Yushmanov1, Boris Yanovski1, Alexander Kharlamov1,
George LaVerde2, Fernando E. Boada2, Stephen C. Jones1
1Allegheny-Singer
Research Institute, Pittsburgh, Pennsylvania, USA; 2University of
Pittsburgh, Pittsburgh, Pennsylvania, USA
23Na
twisted projection MRI was validated as a quantitative technique to assess
local brain sodium concentration ([Na+]br) every 5.3 min
after MCAO. The MRI protocol included a
0.4-ms echo-time, a B1 mapping, and 0-154 mM NaCl calibration
standards. [Na+]br
values were obtained by MRI and flame photometry in exactly the same ROIs of
~0.5 mm3 size. The Bland-Altman
comparison of the two techniques in normal and ischemic cortex and caudate
putamen of seven rats yielded a mean bias and limits of agreement of -4%±42% of
average. A linear increase in [Na+]br
was observed at 1-6 hours after MCAO.
14:00
3251.
Sodium MRI with Triple Quantum Filter and Inversion Recovery at 7T
Computer 28
Guillaume
Madelin1, Niels Oesingmann2, Glyn Johnson1,
Alexej Jerschow1, Matilde Inglese1
1New
York University, New York, New York, USA; 2Siemens Medical Solutions
USA, New York, New York, USA
This
study demonstrates the feasability of Sodium MRI at 7T with Triple Quantum
Filter (TQF) and Inversion Recovery (IR) in differentiating mobile vs. less
mobile 23Na nuclei in phantoms. The combination of TQF and IR can give
complementary informations about the sodium environment and allow a better
differentiation of the intracellular and extracellular compartments in
biological tissues. Future studies will focus on TQF and IR 23Na MRI on human
brain in vivo at 7T in order to assess the intracellular sodium concentration,
with the goal of localizing pathologies.
14:30
3252.
Clean Separation of Bound vs. Free Sodium by 23Na Inversion
Recovery
Computer 28
Peng
Rong1, Ravinder R. Regatte2, Alexej Jerschow1
1New
York University, New York , New York, USA; 2New York University, New
York, New York, USA
Monitoring
the bound sodium pool often allows one to characterize tissue disorders, such
as osteoarthritis, or cell dysfunction. Separation methods have been used based
on residual quadrupolar interactions, or slow molecular tumbling. In the
current study, we demonstrate the feasibility of employing the inversion
recovery (IR) sequence to selectively detect the bound or the free sodium pools
in cartilage tissue. The advantages of this method are its simplicity, the
ability to selectively detect either the ordered or free sodium signal, the use
of only a small phase cycle, and the independence of residual quadrupolar
couplings.
15:00 3253.
23Na DQF Signal
Induced by Paramagnetic Shift Reagents: Dependence on the Pseudo-Contact Shift
Computer 28
Peng
Rong1, Gil Navon2, Alexej Jerschow1
1New
York University, New York , New York, USA; 2Tel Aviv University, Tel
Aviv, Israel
23Na
NMR/MRI in combination with paramagnetic shift agents is often used to separate
the intra- from the extracellular sodium components. In addition to shifts, we
recently reported relaxation effects which lead to the appearance of
double-quantum coherences in isotropic phases and without slow motion. Here we
show that these effects correlate with the pseudo-contact shift induced by the
agents by examining the effects of the LnDOTP series.
13:30
3254.
Sodium MRI Using a Density Adapted 3D Radial Acquisition
Computer 29
Armin
Michael Nagel1, Frederik Bernd Laun1, Marc-André Weber1,
Lothar Rudi Schad
1German
Cancer Research Center, Heidelberg, Germany
A
density adapted 3D Radial sampling scheme (DA-3D-RAD) was implemented. The
readout gradients of the DA-3D-RAD are switched constantly up to a k-space
radius k0. Beyond this radius the gradient’s amplitude is reduced
such that the sampling density in all spherical shells is kept constant. The
benefits for sodium MRI compared to conventional 3D Radial sampling are
demonstrated. In vivo sodium brain and phantom imaging was performed. This
technique leads to a higher SNR compared to conventional radial sequences and
improved resolution.
14:00
3255.
Sodium Imaging on a Whole-Body 7T Scanner: SNR and Resolution Benefits
Computer 29
Yongxian
QIAN1, Yik-Kiong Hue1, Tiejun Zhao2, Tamir S.
Ibrahim1, Fernando E. Boada1
1University
of Pittsburgh, Pittsburgh, Pennsylvania, USA; 2Siemens Medical
Solutions USA, Pittsburgh, Pennsylvania, USA
This
study demonstrates sodium (23Na) images of a healthy human brain acquired on a
whole-body 7T scanner. The signal-to-noise ratio (SNR) of the 7T images was
increased by a factor of 3.5 in comparison with the images acquired on a 3T
scanner with the same acquisition parameters. The SNR gain immediately resulted
in an increase in spatial resolution of the 7T image from 3.44mm to 1.72mm.
14:30
3256.
31P Imaging of the Human Brain with Balanced SSFP - Preliminary Results
Computer 29
Jochen
Leupold1, Roza Umarova2, Jürgen Hennig1,
Cornelius Weiller2, Holger Kaube2
1University
Hospital Freiburg, Dept. of Diagnostic Radiology, Medical Physics, Freiburg,
Germany; 2University Hospital Freiburg, Dept. of Neuroradiology,
Freiburg, Germany
Magnetic
Resonance Spectroscopy of 31P is a promising tool for the examination of
several pathologies and metabolic processes. A beneficial reduction of total
scan time could result from utilisation of fast pulse sequences, which can
acquire spatial and spectroscopic information at the same time. We show
preliminary results of imaging 31P in the brain with a balanced SSFP sequence,
as this sequence has potential to be the basic signal generating sequence in
several fast spatial-spectroscopic methods due to its high steady state signal.
15:00
3257.
Assessment of SNR and Detection Sensitivity of F-UTSI
Computer 29
Muhammed
Yildirim1,2, Jochen Keupp3, Klaas Nicolay2,
Rolf Lamerichs1
1Philips
Research, Eindhoven, Netherlands; 2Eindhoven University of
Technology, Eindhoven, Netherlands; 3Philips Research, Hamburg,
Germany
F-uTSI
(Fluorine ultrafast Tutbospectroscopic Imaging) is a Turbo Spin Echo (TSE) based
method developed for imaging and detection of 19F based contrast agents by
avoiding the well known chemical shift. Here, the preliminary results of
performance assessment of the two basic variants of the method employing
cartesian and pseudo-radial sampling of the k-space are presented. A small part
of the potentially large parameter space was explored for two different
perfluorocarbon compounds; perfluoroctyl bromide (PFOB) and perfluoro crown
teher (PFCE), by altering the field the field of view and TR and the
corresponding detection limits are investigated.
Measurement of Perfusion & Permeability Using Contrast Agents
Hall D Monday 14:00-16:00
14:00
3258.
Sampling the Arterial Input Function in T1 Weighted Dynamic Contrast
Enhanced Perfusion
Computer 32
Adam
Espe Hansen1,2, Henrik Bodil Wiberg Larsson1
1Glostrup
Hospital, University of Copenhagen, Denmark
When
using dynamic contrast enhanced (DCE) MRI to assess cerebral blood flow (CBF),
correct determination of the arterial input function (AIF) is critical to CBF
quantification. With T1-weighted DCE-MRI, we demonstrate the result of the
partial volume effect (PVE) on CBF quantification in vivo. The larger
vasculature can easily be identified and a PVE on the AIF can be introduced by
displacing the AIF from the vessel centre. We further propose and test a method
we denote venous normalization to compensate for PVE. This method might in
particular be utilized to correct local arterial input functions.
14:30
3259.
Correcting Susceptibility Artifacts in Arterial Input Function (AIF) for
Dynamic Contrast Enhancement Magnetic Resonance Imaging (DCE-MRI) at 3T
Computer 32
Poe-Jou
Chen1, Wei-Ting Zhang2, Rakesh K. Jain3, T T.
Batchelor3, A Gregory Sorensen2
1Massachusetts
Institute of Technology, Cambridge, Massachusetts, USA; 2Athinoula
A. Martinos Center for Biomedical Imaging, Charlestown, Massachusetts, USA; 3Massachusetts
General Hospital, Charlestown, Massachusetts, USA
Dynamic
contrast-enhanced magnetic resonance imaging (DCE-MRI) is a non-invasive imaging
technique that has the ability to study tumor vascular functions. It has been
widely used for a range of clinical oncologic applications including cancer
detection, grading, and evaluation of therapeutic response. Unfortunately, in
high field MRI (3T), the image signal is prone to susceptibility effects due to
the increased relaxivity of contrast agent. In this abstract, we propose a
novel method that could fully compensate the susceptibility effect and restore
the full range of plasma concentration dynamic curve. The results demonstrated
that, despite severe T2* effects embedded in the DCE-MRI data, the technique is
able to reconstruct the plasma concentration time course and provide us a more
accurate vascular input function
15:00
3260.
A Blood Circulatory Model to Estimate the Arterial Input Function in MR
Brain Perfusion Studies
Computer 32
Hassan
Bagher-Ebadian1,2, Kourosh Jafari-Khouzani1,
Hamid Soltanian-Zadeh1,3, James R. Ewing1,4
1Henry
Ford Hospital, Detroit, Michigan, USA; 2Amir-Kabir University of
Technology, Tehran, Iran; 3University of Tehran, Tehran, Iran; 4Oakland
University, Rochester, Michigan, USA
A
mathematical model of Contrast Agent(CA) concentration in the circulatory
system as a function of time is of interest since it may allow the description
of the AIF based on the Intravenous (IV) injection function In this study, as a building block, the
concentration-time profile in the circulatory system for each compartment was
modeled by simple physical and pharmacokinetic assumptions using the Fick
equation and Kety’s model. A complete model of the CA concentration as a
function of time in the circulatory system was then constructed by combining
those building blocks using typical flow and volume parameters for the various
compartments. Using a model of the IV bolus injection, its results were
compared to MR perfusion (T2*) signal, indicate that this model is a good
candidate to be used to define a cost function for detection of the AIF in MR
brain perfusion studies.
15:30
3261.
The Use of Phase to Measure the Arterial Input Function for Quantitative
T1-Weighted DCE-MRI in Human Brain Tumors
Computer 32
Claire
Foottit1, Greg O. Cron2, Matthew Hogan2, Thanh
Nguyen2, Ian Cameron,12
1Carleton
University, Ottawa, Canada; 2Ottawa Health Research Institute,
Ottawa, Canada
When
measuring perfusion quantitatively in tumors with dynamic contrast-enhanced
T1-weighted MRI, it can be challenging to measure the arterial input function
(AIF) in large vessels. Blood flow and
saturation effects can lead to incorrect conversion of signal magnitude to
contrast agent concentration. The signal
phase, however, is largely immune to such effects, potentially providing
improved AIF measurement with no added imaging time or pulse sequence
modification. Here, we used this
technique to measure quantitative perfusion (Ktrans) in human brain tumors.
Computer 33
Junyu
Guo1, Mark A. Rosen1, Hee Kwon Song1
1University
of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA
We
investigate the feasibility and accuracy of principal component analysis (PCA)
for radial DCE-MRI. It is shown that with judicious selection of principal components,
PCA can effectively remove image streaking and reduce noise in a highly
undersampled radial DCE-MRI data set, permitting accurate, pixel-wise
measurement of the perfusion parameters.
14:30
3263.
Assessment of Image Reconstruction Methods for Subsampled DCE-MRI
Computer 33
Florian
Knoll1, Franz Ebner2, Stephen Keeling3, Rudolf
Stollberger1
1TU
Graz, Graz, Austria; 2Medical University Graz, Graz, Austria; 3University
of Graz, Graz, Austria
Undersampled
imaging techniques, in combination with advanced image reconstruction methods
like HYPR or parallel imaging, have the potential to deliver high resolution
DCE-MRI data sets in spatial and temporal domain, which can be used for the
identification of pharmacokinetic parameters. It is, however, not clear to
which extent these techniques change the time course of contrast enhancement,
in particular for the arterial input function. The performance of four state of
the art reconstruction methods is evaluated in this work. It is shown that the
estimation of pharmacokinetic parameters can be improved significantly using
these undersampled imaging techniques.
15:00
3264.
Accelerating Dynamic MRI with Efficient Multiple Acquisitions by SPEED
Using Shared Information
Computer 33
Zheng
Chang1, Jim Ji2, Qing-San Xiang3, Fang-Fang
Yin1
1Duke
University, Durham, USA; 2Texas A&M University, College Station,
Texas, USA; 3University of British Columbia, Vancouver, Canada
The
efficient multiple acquisition method using Skipped Phase Encoding and Edge
Deghosting (SPEED) has been successfully demonstrated in water-fat imaging with
double acquisitions. In this work, it is further developed to accelerate
dynamic MRI. In dynamic MRI, multiple
acquisitions of different frames are spatially correlated, and thus share
similar spatial information. By taking
advantage of the shared information, dynamic MRI is accelerated by SPEED with
factors greater than that achievable with a single acquisition. In this work, a dynamic contrast enhanced
mice tumor study is accelerated by a factor of 2, with results comparable to
the images without acceleration.
Computer 33
Adriana
Teodora Perles-Barbacaru1, Laurent Lamalle1, Emmanuel
Barbier1, Christoph Segebarth1, Hana Lahrech1
1UMR-S
836 INSERM-UJF-CEA Functional & Metabolic Neuroimaging, La Tronche Cedex,
France
The
USPIO SINEREM is used for cerebral blood volume fraction (CBVf) mapping with
the Rapid Steady State T1 (RSST1) method. A steady state
signal for at least one hour is obtained in rats after an intravenous injection
of 0.2 mmol/kg SINEREM and was exploited for CBVf mapping in healthy and RG2
glioma bearing rats with the RSST1 method in a 3D projection
reconstruction acquisition mode enabling short echo time and with a
susceptibility based steady state method for comparison. In opposite to results
reported for the C6 tumor model, SINEREM seems to leak from the
microvasculature in the RG2 tumor model.
Hall D Monday 14:00-16:00
Computer 34
Mazhar
Husain1, Raj Kumar1, Rakesh K. Gupta2, Khader
M. Hasan3, Mohammad Haris2, Atul Agarwal1,
Ponnada A. Narayana3
1Chhatrapati
Shahuji Maharaj Medical University, Lucknow, India; 2Sanjay Gandhi
Postgraduate Institute of Medical Sciences, Lucknow, India; 3University
of Texas Medical School at Houston, Houston, USA
Serial
DTI within 2 weeks and at 6 months after TBI was performed on 38 patients with
TBI [frontal (n=12), temporal (n=6), fronto-temporal (n=9) and multifocal (n=11)].
Neuropsychological tests (NPT) were performed at follow-up study and 30 age/sex
matched controls. DTI was also performed in controls. The fractional anisotropy
(FA) and mean diffusivity (MD) were quantified from different region of the
corpus callosum (CC) in patients and controls. Decreased FA in callosal regions
on follow-up study compared to controls was observed in different groups. We
conclude that widespread primary abnormalities in acute stage and secondary
damage after 6 months in the CC can be demonstrated on DTI. These DTI
abnormalities significantly correlate with some of the NPT.
Computer 34
Rakesh
K. Gupta1, Sona Saksena1, Vijan Rai1, Vivek
Anand Saraswat1, Ram KS Rathore2, Ankur Purwar2,
Manoj Kumar1, M A. Thomas3
1Sanjay
Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India; 2Indian
Institute of Technology, Kanpur, India; 3David Geffen School of
Medicine at UCLA, Los Angeles, USA
In
vivo PMRS and DTI data were acquired from FHF patients (n=10) and controls
(n=10). Five of the 10 patients, had a repeat imaging after three weeks.
N-acetylaspartate, choline (Cho), glutamine (Gln), glutamine/glutamate (Glx),
and myoinositol ratios were calculated with respect to creatine (Cr).
Fractional anisotropy (FA) and mean diffusivity (MD) were quantified on
different white and gray matter regions. Patients exhibited significantly
increased Gln/Cr and Glx/Cr ratios compared to controls. Significantly
decreased Cho/Cr was observed in deceased patients compared to controls. In
patients, significantly decreased MD and FA values were observed in different
brain regions compared to controls. In follow-up study, MD and FA values showed
an insignificant increase.
15:00
3268.
Longitudinal Diffusion Tensor MR Imaging Study of Radiation Induced
White Matter Damage in a Rat Model
Computer 34
Silun
Wang1, Chung Nga Tam1, Ed X Wu1, Ho-Fai Lau1,
Deqiang Qiu1, Lucullus H.T. Leung2, Pek-Lan Khong1
1The
University of Hong Kong, Hong Kong, Hong Kong; 2Princess Margaret
Hospital, Hong Kong, Hong Kong
We
evaluate the longitudinal changes in DTI indices and its histological
correlates in a radiation induced white matter (WM) injury rat model.
The
results show that radiation induced WM injury is reflected by significant
reduction in FA and ¦Ë// as early as 4 weeks after injury, prior to T2WI signal
changes. These changes correlate with astrocytic hypertrophy, suggesting that the
relative abundance of astrocytes influence ¦Ë// in the early changes of
radiation induced WM injury, and this occurs before necrosis, axonal loss and
demyelination. Therefore, FA and ¦Ë//
may be biomarkers for early detection of radiation induced WM injury.
Computer 34
Rakesh
K. Gupta1, Vivek A. Saraswat1, Pooja Upreti1,
Rajendra Singh1, Jitesh K. Singh2, Richa Trivedi1,
Sunil Kumar1, Ram KS Rathore2
1Sanjay
Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India; 2Indian
Institute of Technology, Kanpur, India
Diffusion
tensor imaging (DTI) was performed in 28 patients with end stage chronic liver
disease and 25 age/sex matched controls. Mamillary body volume and DTI metrics
in mamillary body were quantified by using in-house JAVA based software.
Significantly decreased fractional anisotropy (FA) and linear anisotropy (CL)
with increased spherical isotropy (CS) and mean diffusivity (MD) values were
observed in patients compared to controls. No change in volume of mamillary
body was observed between patients and controls. Absence of the changes in the
volume of the mamillary bodies compared to controls suggests that DTI is more
sensitive for the early detection of changes due to thiamine deficiency in
these patients.
Hall D Monday 14:00-16:00
14:00
3270.
Diffusion Tensor Imaging in Glioblastoma Multiforme and Brain
Metastases: The Role of P, Q, L and FA
Computer 35
Chris
Steward1,2, Wayland Wang1, Patricia Desmond1,2
1University
of Melbourne, Melbourne, Australia; 2Royal Melbourne Hospital,
Melbourne, Australia
Micro-invasive
tumour cells, which are not detected on conventional imaging, contribute to
poor prognoses for patients diagnosed with glioblastoma multiforme (GBM; WHO
grade IV). Diffusion tensor imaging (DTI) shows promise in being able to detect
this infiltration. This study aims to detect a difference in diffusion properties
between GBM (infiltrative) and brain metastases (non-infiltrative). In particular, to compare the diffusion
tensor metrics p, q, L, and FA from tumoural and peritumoural regions of
glioblastoma multiforme and metastatic brain tumours 2/ To determine whether
these parameters can be correlated with the type of tumour and the extent of
infiltration into surrounding white matter 3/ To determine whether these
parameters, and more broadly DTI, has a significant clinical meaning, and thus
whether it should be included in imaging protocols for suspected brain tumour
and for treatment planning.
14:30
3271.
About "Axial" and "Radial" Diffusivities
Computer 35
Claudia
Angela Michela Wheeler-Kingshott1, Mara Cercignani1,2
1Institute
of Neurology, UCL, London, UK; 2Fondazione Santa Lucia, Rome, Italy
DTI
allows the quantitative assessment of diffusion anisotropy in tissues. It is
well known that the DT can be diagonalised, determining three eigenvectors and
their eigenvalues. We have questioned
the rationale that underpins the relationship between axial and radial
diffusivities and specific biophysical substrates, based on the already
established knowledge of the uncertainty of the eigenvectors/values measures.
Using two healthy-controls and two people-with-MS we have demonstrated what
kind of serious interpretation errors can arise when comparing one or more
eigenvalues of the DT across datasets. Our results are strongly supporting the
not-yet-recognised need for including the eigenvectors when comparing eigenvalues.
15:00
3272.
Quantitative Relations of Axial and Radial Diffusivities to Anisotropy
Indices in Diffusion Tensor Imaging
Computer 35
Ling-chih
Lin1, Jianhui Zhong1
1University
of Rochester, Rochester, New York, USA
Sensitivities
of radial and axial diffusivities with respect to other diffusion anisotropy
indices (DAIs) were evaluated numerically. Monte Carlo Simulation was performed
to evaluate the effect of the signal to noise ratio (SNR) to their accuracy.
Correlations between the axial and radial diffusivities and other DAIs were
examined to facilitate the optimal selection of diffusion anisotropy
measurement parameters.
15:30
3273.
Three Novel Methods for Studying Complicated Diffusion Behavior Using
Simple Capillary Structures
Computer 35
Nathan
Yanasak1, Qun Zhao2, Tom C.-C. Hu1, Jerry
Allison1
1Medical
College of Georgia, Augusta, USA; 2University of Georgia, Athens,
USA
Although
DTI has shown great clinical potential for studying a variety of neurological
conditions, procedures and equipment for quality assurance of clinical data are
still in development. In particular,
noise can bias anisotropy measurements, yet phantom designs are generally
limited to simple anisotropic structures less characteristic of tissue. This study introduces three new techniques to
assist in the characterization of data quality for complicated tissue-like
structures. Images acting as input are
acquired using a phantom with capillary structures. In this manner, data quality from images of
complex structures can be inferred from a QA phantom.
14:00
3274.
Constrained Single Step Diffusion Tensor Reconstruction Using Cholesky
Decomposition
Computer 36
Murat
Aksoy1, Roland Bammer1
1Stanford
University, Stanford, California , USA
In
DTI, the diffusion model requires the tensor to be positive definite. Several
methods have been proposed to ensure positive definiteness, such as replacing
negative eigenvalues or constraining tensors to be positive definite using
matrix decomposition. In this study, we propose a single step non-linear tensor
estimation scheme that utilizes Cholesky decomposition to guarantee the
positive definiteness of the reconstructed diffusion tensors. The presented
algorithm estimates the diffusion tensor directly from the unity of all k-space
data in a non-linear fashion and it is especially effective in correcting gross
patient motion in the case of multi-shot DTI scans.
14:30
3275.
Using Fractional Anisotropy Neighbourhood Information in a Bayesian
Based Regularisation Technique for DTI
Computer 36
Marta
Morgado Correia1, Tim Hosey1, Sally G. Harding1,
Thomas Adrian Carpenter1, Guy B. Williams1
1University
of Cambridge, Cambridge, UK
A
new regularisation method for DT-MRI, which is capable of improving the
confidence in the FA maps obtained from diffusion weighted MRI datasets, is
introduced. The idea behind this prior is that the variation of FA along a
fibre tract should be smooth, and therefore if there is a tract connecting two
neighbour voxels, the FA variation between these two voxels should be small. The
regularisation is applied to the priors of a MCMC model rather than the data
itself, meaning that the fit is biased towards spatially consistent solutions
but this only when supported by the data.
Computer 36
Jennifer
Andrea McNab1, Karla L. Miller1
1University
of Oxford, Oxford, UK
Diffusion-weighted
steady-state free precession (DW-SSFP) accumulates signal from multiple echoes
over several TRs yielding a strong sensitivity to diffusion with short gradient
durations and imaging times. DW-SSFP is thus of great interest as a potential
method for high angular and spatial resolution diffusion imaging. The DW-SSFP
signal is well characterized for isotropic, Gaussian diffusion, however, it is
unclear how the multi-echo signal propagates for inhomogenous media. In this
study, the DW-SSFP signal equation is presented for the first time in terms of
the reciprocal spatial wave vector (<B>q</B>) and an arbitrary
diffusion probability density function pdf (P(r, δ)).
15:30
3277.
Toroid-Based Characterization of Cardiac DT-MRI
Computer 36
Choukri
Mekkaoui1, Marcel P. Jackowski2, Albert J. Sinusas1
1Yale
University School of Medicine, New Haven, USA; 2Institute of
Mathematics and Statistics, São Paulo, Brazil
The
aim of the present study is to introduce a novel toroid-based representation of
diffusion tensor fields to improve myofiber visualization and macrostructure
analysis. A new diffusivity index, the toroidal volume (TV), and a new
coefficient of anisotropy, the toroidal volume ratio (TVR), are defined.
Toroidal glyph renderings from a normal canine heart were compared to
ellipsoidal and superquadric glyph fields. TV and {1-TVR} maps were calculated
for a DT-MRI dataset of an infarcted canine heart and compared to contemporary
indices. Results demonstrate that the proposed methodology improves diffusion
tensor visualization and may complement existing diffusion indices.
MRI & MRS of Cerebral Ischemia
Hall D Tuesday 13:30-15:30
13:30
3278.
Susceptibility-Weighted Imaging in Hyperacute Cerebral Ischemia
Computer 30
Masahiro
Ida1, Kennichi Motoyoshi1, Hiroyuki Fukuyama1,
Hisashi Yoshizawa1, Naoya Yorozu1, Keiko Hino1
1Tokyo
Metroplolitan Ebara Hospital, Oota-ku, Japan
Increased
vessel contrast (IVC) suggests a relative increase of intravenous paramagnetic
deoxyHb and a relative reduction of oxyHb due to the impaired oxygenation
(misery perfusion state). SWI can detect not only an area of perfusion
impairment but also occluded artery. SWI provides important adjunct information
for HIS. IVC is a useful finding to assess acute misery perfusion state without
contrast media and a reliable indicator for determining whether a patient
should undergo Gd-perfusion study.
Computer 30
Juan
Chen1,2, Daniel J. Licht3, Yong Fan1,
Sabrina E. Smith3, Shannon C. Agner3, Stefanie Mason3,
Robert A. Zimmerman3, Rebecca N. Ichord3, Jiongjiong Wang1
1University
of Pennsylvania, Philadelphia, Pennsylvania, USA; 2Beijing Hospital,
Beijing, People's Republic of China; 3the Children's Hospital of
Philadelphia, Philadelphia, Pennsylvania, USA
This
is a longitudinal study of cerebral perfusion and lesion volume changes in
three pediatric patients with acute stroke. The aim of the study was to
demonstrate feasibility of pulsed arterial spin-label (PASL) perfusion MRI to
follow perfusion deficits and their relation with changes in lesion volume over
time. Ischemic lesions showed a trend of decreasing volume over time
accompanied by decreasing perfusion within the shrinking lesion. Voxel-based
scatterplot analysis of the region of diffusion restriction indicate that
tissue within the lesion has relatively higher perfusion and may represent
viable and potentially salvageable tissue.
14:30
3280.
Altered Regional CO2 Vasoreactivity in Patients with Ischemic Stroke
Using CASL MRI
Computer 30
Peng
Zhao1, David Alsop2, Magdy Selim3, Amir
Abduljalil4, Peter Novak5, Lewis Lipsitz1, Kun
Hu1, Sarah LaRose1, Vera Novak1
1Beth
Israel Deaconess Medical Center, Boston, Massachusetts, USA; 2Radiology,
Beth Israel Deaconess Medical Center,, Boston, Massachusetts, USA; 3Neurology,
Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA
This
study investigated the regional differences in CO2 vasoreactivity (CO2VR) after
stroke and their relationship to clinical outcomes. 27 subjects with chronic
large vessel infarcts in the middle cerebral artery (MCA) territory were
compared with 43 controls by using continuous arterial spin labeling (CASL)
perfusion MRI. Impairment of CO2VR after stroke was observed which extended
into additional brain regions and vascular territories not within the ischemic
region.
15:00
3281.
Quantitative Assessment of Brain Mass Effect Using Mid-Brain Surface in
Stroke Patients
Computer 30
Yasheng
Chen1, Chung-Yi Yang2, Cheng-Hong Toh1, Hongyu
An1, Allyson Zazulia3, Thomas Vedeen3, Michael
Diringer3, William Powers1, Weili Lin1
1The
University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA; 2National
Taiwan University Hospital, Taiwan; 3Washington University in St.
Louis, St. Louis, Missouri, USA
The
current approach for quantitatively measuring brain mass effect employs manual
measurements of the 2D displacement between the altered brain midline and the
putative normal midline in specific anatomical locations including the pineal
body, septum pellucidum, and third ventricle.
One major limitation associated with this approach is the inability to
assess the true 3D extent of brain mass effect.
An automated approach reconstructing the mid-brain surface is developed,
allowing a direct assessment of mass effects through measuring the volume
enclosed by the deformed and the estimated of un-deformed brain mid-surface.
13:30
3282.
White Matter Integrity But Not BOLD Response Predicts Upper Limb Motor
Function in Patients After Stroke
Computer 31
mingguo
qiu1, Justin Rajendra2, Warren Darling3,
Robert Morecraft4, Chunchun Ni2, Andrew J. Butler2
1Southwest
hospital of Third military medical university, chongqing, People's Republic of
China; 2Emory University, School of Medicine, Atlanta, Georgia, USA;
3The University of Iowa, Iowa City, Iowa, USA; 4Univ
The
aim was to establish the relationship between upper-extremity motor functional
outcomes with the asymmetry of fractional anisotropy (FA) from DTI and with the
laterality index (LI) derived from functional MRI. Seventeen patients after
stroke were enrolled and measured the motor function by Wolf Motor Function
Test (WMFT) and Fugl-Meyer (FM) assessment. The data showed a significant
relationship between FA asymmetry of PLIC with FM and WMFT asymmetry, white
matter integrity can strongly predict current motor function. However, LI can’t predict the current
clinical outcomes; no relationship was found between the LI with the FA
asymmetry and clinical motor outcomes.
14:00
3283.
Assessing the Evolution of Sodium MRI with Time After Onset in Human
Stroke
Computer 31
Robert
Wayne Stobbe1, Muhammad Shazam Hussain2, Yusuf A. Bhagat2,
Ken S. Butcher1, Derek J. Emery1, N Rizvi1,
Perkash Maheshwari1, Ashfaq Schuaib1, Christian Beaulieu1
1University
of Alberta, Edmonton, Canada; 2University of Alberta, Edmonton ,
Canada
Sodium
MRI signal intensity within the ischemic lesion is shown to evolve over time
after symptom onset in human acute stroke. Given significant early changes it
may be possible to estimate time of onset in patients when this important
factor is unknown, and possibly increase options for treatment.
Computer 31
Metin
Bora Vardar1, Gur Akansel,12, Nagihan Inan1,1,
Hasan Tahsin Sarisoy1,1, Arzu Serpil Arslan1,1,
Ercument Ciftci1,1, Ali Demirci1,1
1;
2Kocaeli University School of Medicine, Kocaeli, Turkey
Although
the MRI diagnosis of cerebral infarction is usually straightforward with the
use of diffusion-weighted imaging, infarcts first seen in the late subacute
stage remain a challenge to differentiate from neoplasm. Since cytotoxic edema
subsides in the late subacute period, abnormalities of apparent diffusion
coefficient due to restricted water diffusion are of limited help. Hemorrhagic
transformation and abnormal contrast enhancement due to blood brain barrier
breakdown may further complicate the picture.
A
common clinical indication for proton MRS is the differentiation of neoplastic
from non-neoplastic central nervous system lesions. MRS is frequently performed
to differentiate subacute infarcts from neoplasm. However, elevated choline
content that may be encountered in subacute infarcts may impede the
contribution of MRS in this differentiation.
In
this study, we evaluated choline/Creatine ratios in cases of acute, subacute
and chronic central nervous system infarction to detect the frequency, timing
and the extent of the elevation in choline/Creatine ratio. Our aim was to
determine if the MRS pattern in subacute infarction was sufficiently different
from that which characterizes a neoplasm.
15:00
3285.
fMRI of Rehabilitation in Chronic Stroke Using MR-Compatible Robots
Computer 31
Dionyssios
Mintzopoulos1,2, Loukas G. Astrakas1, Walter
J. Koroshetz1, Michael A. Moskowitz1, Bruce R. Rosen1,
A Aria Tzika1
1Massachusetts
General Hospital, Boston, Massachusetts, USA
Using
a hand motor task, we investigated brain activation after chronic stroke by
combining fMRI at 3T with a novel MR-compatible hand-induced, robotic device
(MR_CHIROD). Patients trained at home using a gel ball; serial neuroimaging was
performed before, during, upon completion of training, and after a non-training
period, to assess permanence of effects. Training significantly increased the
number of activated voxels in the cortex as a function of effort level,
suggesting functional cortical plasticity in chronic stroke. The result¢s
persistence indicates permanence of rehabilitation, which is remarkable given
that training is generally effective during a narrow window after stroke.
13:30
3286.
MRI Metrics Detected Axonal Outgrowth and Plasticity in Rat Brain After
Embolic Stroke
Computer 32
Guangliang
Ding1, Quan Jiang1, Lian Li1, Li Zhang1,
Zhanggang Zhang1, Karyn A. Ledbetter1,2,
Swayamprava Panda1, Siamak P.N. Davarani1,3,
James R. Ewing1, Michael Chopp1,4
1Henry
Ford Health System, Detroit, USA; 2Michigan State University, East
Lansing, USA; 3University of Michigan, Ann Arbor, USA; 4Oakland
University, Rochester, USA
Both
angiogenesis and axonal outgrowth were detected as early as 1 week after stroke
in sildenafil treated rats. Angiogenesis and axonal plasticity were
co-localized with each other, measured histologically at 6 weeks and
dynamically in vivo by MRI up to 6 weeks after stroke. This indicates that
sildenafil treatment of stroke may simultaneously promote angiogenesis and
axonal plasticity after embolism in rats. Local CBF in the co-localized area
was increased after angiogenesis. Both elevated CBF and re-organized white
matter in ischemic areas may contribute to recovery of neuronal function after
stroke in rats.
Computer 32
Lian
Li1, Quan Jiang1, Guangliang Ding1, Li Zhang1,
Zhenggang Zhang1, James R. Ewing1, Michael Chopp1,2
1Henry
Ford Health System, Detroit, Michigan, USA; 2Oakland University,
Rochester, Michigan, USA
The
objective of the present study was to noninvasively identify and monitor the
progress of white matter reorganization within 6 weeks after the onset of
stroke and correlate this structural change with improved neurological function
using MRI. Our data in rats indicate that FA is a sensitive measure of white
matter changes after stroke and provides an important noninvasive means for
real-time evaluation of treatment efficacy and functional outcome. Treatment
with EPO significantly enhances white matter reorganization, which correlates
with local restoration of CBF and recovery of neurological function
Computer 32
Maurits
Pieter Adriaan van Meer1,2, Wim M. Otte1, Kajo
van der Marel1, J. W. Berkelbach van der Sprenkel1, Rick
M. Dijkhuizen1
1University
Medical Center Utrecht, Utrecht, Netherlands
Manganese-enhanced
MRI was applied after injection of MnCl2 in the contralesional primary
sensorimotor cortex of rats at 10 weeks after unilateral stroke. Subsequent ROI
analysis of the time-course of manganese-induced R1 increases revealed
increased tracer uptake and distribution in contralesional subcortical ROIs and
ipsilesional cortical ROIs as compared to controls. Our data support the concept
of structural remodeling of the contralesional cortex and increased
connectivity with other sensorimotor network areas, which may significantly
contribute to post-stroke functional recovery.
15:00
3289.
Magnetic Resonance Imaging of Delayed Administration of Manganese After
Transient Focal Cerebral Ischemia
Computer 32
Kevin
C. Chan1,2, Victor K. Hung1, Matthew M. Cheung1,
Ke Xia Cai1, Dave K. Cheung1, Chi-tat Chiu1,
Po-mak Chan1, Xiao-guang Zhao1, Sookja K. Chung1,
Ed Xuekui Wu1
1The
University of Hong Kong, Pokfulam, Hong Kong
This
study aims to employ in vivo magnetic resonance imaging to understand
longitudinally the effect of systemic MnCl2 administration on delayed secondary
changes after transient middle cerebral artery occlusion (MCAO). Results showed a significant increase in T1
signal intensities in the perilesional rim compared to the ischemic core in
dorsolateral striatum 24 hours after Mn2+ injection, whereas colocalization was
found between manganese-enhanced MRI pattern and immunostaining for GFAP, MnSOD
and GS, likely due to their Mn-dependent upregulation against oxidative stress
and glutamate excitotoxicity. Total ipsilateral infarcted volumes in the
posterior parts of the brain had also significantly reduced after Mn2+
injection, illustrating the potential neuroprotective effects of manganese upon
delayed postischemic administration.
Computer 33
Rachael
Dobson1,2, Patrizia Ferretti1, Mark F. Lythgoe1
1University
College London, London, UK
Transplantation
of stem cells following cerebral ischaemia has been shown to improve functional
outcome. MRI contrast agents have been used to track the migration of transplanted
cells towards lesions. The protamine sulphate-ferumoxide complex – FePro
–enhances cell labelling. In this study, migration of neural stem cells,
labelled with FePro, was monitored using MRI in a model of cerebral ischaemia.
Little migration was observed. Hypointense regions on T2-weighted images
developed within lesions, and iron was detected histologically, in cell
treatment and control groups. These results highlight issues for cell tracking
studies using MRI contrast agents in animal models of disease.
14:00
3291.
MRI Evaluation of the Effect of a COX-2 Inhibitor on BBB Permeability in
a Rat Stroke Model
Computer 33
Saeid
Taheri1, Eduardo Candellario-Jalil1, Eduardo Estrada1,
Gary Rosenberg1, Rohit Sood1
1University
of New Mexico, Albuquerque, New Mexico, USA
Recent
research has demonstrated an increase in expression of cyclooxygenase (COX-2)
mRNA within neurons and vascular cells and it has been postulated that COX-2
plays an important role in secondary events that amplify brain damage
post-ischemia. There is additional evidence that COX-2 is involved in the
increase in BBB permeability. The purpose of this study was to use MRI based
BBB permeability estimation technique to evaluate the effects of nimesulide, a
non-steroidal anti inflammatory drug on BBB permeability in a rat model with
focal cerebral ischemia at 48h post MCAO. Initial results suggest that
nimesulide is effective in reducing BBB damage 48h post ischemia and would be
potential candidate for therapeutically reducing late phase brain damage post
ischemic stroke.
14:30
3292.
in vivo 1H/13C MRSI of Changes in Neurotransmitter
Metabolism in Rats Recovering from Stroke
Computer 33
Pieter
van Eijsden1, Jet P. van der Zijden2, Robin A. de Graaf3,
Rick M. Dijkhuizen2
1University
Medical Center, Utrecht, Netherlands; 2University Medical Center
Utrecht, Netherlands; 3Yale University, New Haven, USA
Post-stroke
functional recovery has been associated with brain plasticity in the lesion
borderzone, but its metabolic basis is unknown. To characterize alterations in
oxidative glycolysis and neurotransmitter metabolism, we applied in vivo dynamic 1H/13C
MR spectroscopic imaging after infusion of 13C-labeled glucose in
rats after transient stroke. Semi-acutely, reductions in NAA levels and Glu
turnover implied severe neuronal dysfunction in the lesion borderzone. At 3
weeks post-stroke, NAA levels and Glu turnover had normalized, suggestive of
neuronal recovery. Our findings show that early metabolic impairment in the
morphologically intact lesion borderzone recovers over time, contributing to
post-stroke functional improvement.
15:00
3293.
Magnetic Resonance Spectroscopy and Imaging at 9.4T in P4 Rat Pup Brain
Following Cerebral Hypoxia-Ischemia
Computer 33
Yohan
van de Looij1,2, Alexandra Chatagner2, Nicolas
Kunz1, Petra S. Hüppi2, Stéphane V. Sizonenko2,
Rolf Gruetter1,3
1Laboratory
for Functional and Metabolic Imaging, Ecole Polytechnique Fédérale de Lausanne,
Lausanne, Switzerland; 2Division of Child Growth & Development,
Department of Pediatrics, University of Geneva, Geneva, Switzerland
To
define the nature of diffuse injuries of the very preterm infant, we use the
model of neonatal hypoxic-ischemic injury in the 3-day old (P3) rat pup and we
investigate acute and long term brain alterations using high-field MRI and MRS.
At P4, the ispilateral cortex indicated significant modifications for several
metabolites reflecting an energetic and functional slowing-down of the injured
cortex. At P11 no significant differences between both cortices were found, ascribed
to a low number of animals and possibly to intervening partial tissue repair.
Relation between metabolic changes and severity of the injury requires further
investigations.
13:30
3294.
In Vivo Early MRI and MRS at 9.4T of Combined Hypoxia-Hypotension and
Traumatic Brain Injury in the Rat
Computer 34
Bich-Thuy
Doan1,2, Fanny Noury3, Thomas Geeraerts4,
Arnaud Friggeri5, Philippe Meric1, Sandra Meme3,
Bernard Vigue4, Jean-Claude BeloeilL3
1CNRS,
Gif sur Yvette, France; 2Orléans, France; 3CNRS, Orléans,
France; 4AP-HP, University of Paris Sud, Kremlin Bicetre, France; 5AP-HP,
France
Following
traumatic brain injury (TBI), hypoxia and hypotension are frequent and damaging
secondary insults.
The
aim of this work is to characterize the traumatic injury by MRI (T2, diffusion
tensor, angiography) and MRS experiments in vivo on rat brain. The final aim is
to understand the role of secondary insults like hypoxia and hypotension (HH)
in the development of post-traumatic brain oedema.
Using
an animal model of diffuse TBI complicated by HH, we focused on post-traumatic
brain oedema development using MRI combined with MRS, as well as consequences
of post-traumatic perturbations on brain energy metabolism using cerebral
microdialysis.
14:00
3295.
Single Voxel MR Spectroscopy with Echo Times Below 2 Ms at 16.4 T in the
Rat Brain: First in vivo Results
Computer 34
David
Zsolt Balla1, Sung-Tak Hong1, Gunamony Shajan1,
Rolf Pohmann1, Kamil Ugurbil1,2
1Max
Planck Intitute for Biological Cybernetics, Tuebingen, Germany; 2Center
for Magnetic Resonance Research, Minneapolis, USA
Single
voxel MRS techniques employ at least three RF-pulses to generate an echo in the
selected volume and gradients to dephase magnetization in the outer region.
During the time between excitation and acquisition, relaxation effects and
phase modulation by J-couplings affect the NMR signal. The STEAM sequence can
be applied with very short echo times to obtain localized spectra containing
fast decaying components and peaks from all detectable coupled resonances. We
show in vivo spectra from the rat brain acquired at 16.4 T with localized
spectroscopy using echo times below 2 ms and compare them to results from the
literature.
14:30
3296.
Time Courses and Correlation of Diffusion-Weighted MR and CT Imaging in
Acute MCA Ischemia with Rat Model
Computer 34
Hyung
Suk Seo1, Dong Gyu Na2
1dongguk
university international hospital, goyangsi, Republic of Korea; 2Seoul
National University College of Medicine, Seoul, Republic of Korea
Unlike
ADC value, DWI signal in acute cerebral ischemia increases with time and
correlates with CT density, which decrease with time. It means that the signal
change of DWI after 3 hours is influenced by net water uptake in ischemic edema
rather than by restricted water diffusion and DWI will be used as the predictor
of ischemic injury severity like CT.
15:00
3297.
Brain Tissue Ischemic Transitions During Permanent Middle Cerebral
Artery Occlusion (PMCAO) in Rats
Computer 34
Lesley
May Foley1, T. Kevin Hitchens1, Brent Barbe2,
Joyce A. Horner1, Edwin M. Nemoto2
1Carnegie
Mellon University, Pittsburgh, Pennsylvania, USA; 2University of
Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
The
identification of the ischemic penumbra in acute stroke is important for
therapeutic intervention. Whereas
qualitative MRI perfusion and diffusion mismatch does not accurately identify
the ischemic penumbra, it may be identified by quantitative measurements using
ASL for perfusion and ADC with appropriate thresholds. Using these methods, we were able to describe
brain tissue ischemic transitions in the first four hours and up to three weeks
after pMCAO.
13:30
3298.
Mild Hypoxic-Ischemic Brain Injury in Neonatal Rats Using Diffusion
Tensor MR Imaging: A Longitudinal Study
Computer 35
Silun
Wang1, Ed X Wu1, Chung Nga Tam1, Edward S Hui1,
Pek-Lan Khong1
1The
University of Hong Kong, Hong Kong, Hong Kong
We
evaluate the longitudinal changes (D1 to D90 post-injury) in the white matter
(WM) of a mild hypoxic-ischemic (HI) rat brain injury model using DTI and
correlate the findings with histology. Significantly elevated ¦Ë¡Í with no
change in ¦Ë// in the ipsilateral WM was found, suggesting reduced myelination
in the ipsilateral WM as a consequence of mild HI injury and this was confirmed
by Luxol fast blue stain. Longitudinal changes of DTI indices on the
ipsilateral WM parallel changes of normal development in the contralateral WM
suggesting continual maturation processes after HI injury.
14:00
3299.
Comparison of Two Methods of Assessment of Perfusion-Diffusion Mismatch
in a Rodent Model of Ischemic Stroke
Computer 35
Feng
Chen1, Yicheng Ni1, Guy Marchal, Jie Yu1,
Yasohiro Suzuki, Nobuo Nagai
1Catholic
University of Leuven, Leuven, Belgium
The
present study demonstrated that the PWI-ADC pattern evolves from a mismatch
pattern through a match pattern to a reversed mismatch pattern after stroke
onset. The PWI-ADC mismatch pattern was observed up to 72h after MCA occlusion
in this animal model. There are linear correlations of volume and mismatch occurrence
between e(early)PWI-i(instant)ADC and iPWI-iADC models. The PWI-ADC pattern
could be defined either with ePWI or iPWI before 24h after stroke onset. The
iPWI appeared more adequate compared to ePWI to define the PWI-ADC pattern at
72h.
Diffusion: Artifacts, Phantoms, QA, Reproducibility
Hall D Tuesday 13:30-15:30
13:30
3300.
Quality Assessment Through Analysis of REsiduals of Diffusion Image
Fitting
Computer 36
Alexander
Leemans1, Christopher John Evans1,2, Derek K.
Jones1
1School
of Psychology, Cardiff University, Cardiff, UK; 2GE Healthcare,
Chalfont St. Giles, UK
In
order to obtain reliable quantitative results, it is important to assess the
quality of DTI data with respect to subject motion, distortions, signal
dropouts, etc. prior to performing further analyses. In this context, we
developed a comprehensive DTI quality assessment (QA) tool that provides a
‘direct feel’ and ‘global overview’ of the data in an automatic way. We show
how this exploratory tool can be used to efficiently identify subject motion,
signal dropouts, and image distortions, without the need to perform QA on all
the individual diffusion weigthed MR images.
14:00
3301.
Analysis of Noise Corrected Diffusion Decay of Human Brain
Computer 36
Elena
Olariu1, Arturo Cardenas-Blanco2, Ian Cameron,12
1Carleton
University, Ottawa, Canada; 2Ottawa Health Research Institute,
Ottawa, Canada
Several
studies have shown that, there is a significant deviation of the MR signal, for
both gray matter (GM) and white matter (WM), from the mono-exponential model;
however, there is no clear consensus as to whether the diffusion decay has one
or more exponential contributions. Our purpose was to carefully characterize
the diffusion attenuation in WM and GM over an extended range of b-values, and
to use a post-processing scheme to reduce the noise bias to see if a second
exponential component could be observed.
14:30
3302.
Quality Assurance of MR Scanner on Diffusion Tensor Imaging
Computer 36
Zili
Chu1,2, Jonathan Chia3, Zhiyue J. Wang1,2
1Baylor
College of Medicine, Houston, Texas, USA; 2Texas Children's
Hospital, Houston, Texas, USA; 3Philips Medical Systems, Cleveland,
Ohio, USA
Diffusion
Tensor Imaging (DTI) is prone to artifacts associated with EPI and strong
diffusion encoding gradients. Therefore, periodically assessing the performance
of the scanner on the DTI sequence is essential to ensure good image quality.
Here, we propose the use of a glycerin-gel spherical phantom constructed to be
similar to the human brain in terms of apparent diffusion coefficient (ADC) and
T2.The temperature of the phantom is controlled during the QA data acquisition.
The calculated ADC, FA, image center displacement and distortion on the DTI
data of the phantom can be analyzed to
serve as indicators of the scanner performance.
15:00
3303.
A Frame-Work for DTI Quantitation, Visualization & Analysis
Computer 36
Ankur
Purwar1, RKS Rathore1, RK Gupta2, D Rathore1,
G Bayu1, MK Sarma1, Anup Singh1, JK Singh1,
S Verma1
1IIT
Kanpur, Kanpur, India; 2SGPGIMS Lucknow, Lucknow, India
DTI
is an important tool to study brain white matter anatomy/abnormalities. The
data processing and analysis in DTI is quite elaborate and so far no commercial
and easily extendable/modifiable tool for the same is available. In view of the
same, we present a tool frame-work focusing on DTI technology combining all of
the above needs. It has 4 mutually independent modules namely DTI Processing
Module, DTI Visualization Module, ROI Analysis Module and Tracking and Fiber
data Visualization Module. The Tool can read various data formats from
different scanners including the raw binary format. An automatic de-scalping
procedure is also implemented.
Computer 37
Akira
Kunimatsu1,2, Masayuki Yamaguchi2, Yoshikazu
Okamoto1, Izumi Anno1, Hirofumi Fujii2,
Atsushi Nozaki3, Hiroyuki Kabasawa3, Manabu Minami1
1Institute
of Clinical Medicine, University of Tsukuba, Tsukuba, Japan; 2Research
Center for Innovative Oncology, National Cancer Center Hospital East, Kashiwa,
Japan; 3GE Yokogawa Medical Systems, Hino, Japan
We
aimed to develop small-diameter, flexible, ex-vivo phantoms for validation of
diffusion tensor (DT) imaging and tractography of the human peripheral nerve on
a clinical MR imager with standard coils. Our results suggest that DT imaging
and tractography are feasible even with a 4-mm-diameter phantom on a clinical
imager without any dedicated or special coil. Four or more numbers of
excitations and 15 or more motion probing gradient directions may provide
robust results of fractional anisotropy measurement and fiber-tracking of the
human peripheral nerve.
Computer 37
Xiaogeng
Feng1, Dae-Shik Kim1, Itamar Ronen1
1Boston
University School of Medicine, Boston, Massachusetts, USA
With
the proliferation of high field MRI scanners (B0 &[ge] 3T), it is now
possible to acquire DTI data of the human brain at a reasonably high spatial
resolution. Increased resolution, however, has implications on the assessment
of DTI parameters such as ADC and FA that are different for white matter and
cortical gray matter. This work aims to evaluate the tissue-specific effects of
increased resolution in carefully segmented DTI data sets and explain the
interplay between SNR, partial volume effects and macroscopic characteristics
(e.g. curvature effects), which is shown to be significantly different in brain
gray and white matter.
14:30
3306.
Simple, Reliable and Precise Quantitative Quality Assurance of In-Vivo
Brain ADC
Computer 37
Nicholas
G. Dowell1,2, Paul S. Tofts1,2
1University
of Sussex, Falmer, UK; 2UCL, London, UK
Measuring
diffusion with MRI is a useful and popular method of studying many pathological
conditions in vivo. The apparent diffusion coefficient (ADC) is commonly used
to quantify the extent of diffusion; however, there is a need for reliable
quantitative quality assurance of the technique. We use decane, dodecane and
tetradecane as test liquids that exhibit similar ADC values to those observed
in the human brain in vivo. We demonstrate that serial measurements of ADC can
be referenced to a standard temperature to avoid variation due to temperature
fluctuations. The materials used here are stable, readily available and safe to
handle.
15:00
3307.
Looking for the Optimal DTI Acquisition Scheme Given a Maximum Scan
Time: Are More B-Values a Waste of Time?
Computer 37
Marta
Morgado Correia1, Thomas Adrian Carpenter1, Guy B.
Williams1
1University
of Cambridge, Cambridge, UK
By
using simulated data, this study has shown that the use of more than 1 b-value
in DTI acquisition schemes can minimize the systematic bias of ADC and FA
estimates due to the Rician noise distribution, and also produce results closer
to the simulates ones and less variable due to noise and fibre rotation. In
addition, the use of more than 1 b-value accounts better for the great
diversity of diffusivities we find in the brain. For tractography studies,
however, we should use as many sampling directions as allowed by scan time
limitations.
Fiber Tracking & Connectivity Mapping
Hall D Wednesday 13:30-15:30 Chairs: Derek K. Jones and Mariana Lazar
13:30
3308.
In Vivo Localisation of Fibre Tracts: Optimisation of Fibre Tracking to
Reduce Voxel Misclassification
Computer 30
J-Donald
Tournier1,2, Fernando Calamante1,2,
Alan Connelly1,2
1Brain
Research Institute, Melbourne, Australia; 2University of Melbourne,
Melbourne, Australia
Diffusion
tractography is increasingly being used in clinical and neuroscientific
applications. However, reliable tracking requires methods to resolve crossing
fibres that have yet to be optimised. Here, we investigate the effects of
acquisition and reconstruction parameters on the reconstructed fibre tracts
estimated using sophisticated methods, by quantifying the volume of white
matter misclassified as belonging or not to the tract. We show that
misclassified volume is consistently reduced by using more diffusion-weighted
directions and higher spherical harmonic orders. Measures of false tract and
omitted tract volume also provide a means of tailoring the acquisition and
reconstruction to the application.
14:00
3309.
A New Fiber Tract Color-Encoding Scheme Based on Diffusion Tensor Model Residuals
Computer 30
Alexander
Leemans1, Derek K. Jones1
1School
of Psychology, Cardiff University, Cardiff, UK
Due
to the multi-component nature of DTI data, distinguishing between real and
artifactual observations is often non-trivial. To bridge this gap between the
acquired images and the understanding of the underlying WM microstructure,
specialized visualization methods have been developed that display various WM
properties along the fiber tracts. However, none of these visualization
approaches reflects the acquired data quality (artifacts) or Goodness-of-Fit
(GoF) to the model, which is crucial in interpreting analysis results. In this
context, we developed a new color-encoding scheme for visualizing fiber tracts,
based on the analysis of tensor model residuals, which allows the investigation
of data quality and GoF.
14:30
3310.
Combination of Distance Measures for Optimal Fiber Clustering in
Diffusion Tensor Imaging
Computer 30
Daniel
Güllmar1,2, Jana Langner1, Jens Haueisen2,
Jürgen R. Reichenbach1
1University
Hospital Jena, Jena, Germany; 2TU Ilmenau, Illmenau, Germany
From
Diffusion Tensor data of the human brain one can derive beautiful fiber tracts
employing deterministic algorithms in order to mimic the neuronal pathways.
However it is hard to distinguish between the different tracts, although the
tracts itself are easy to recognize by visual inspection. Several studies
showed that one can perform clustering based on distance measures between the
fiber path ways. Thus it is possible to objectively group the paths in order to
distinguish between at least the major tracts. The aim of the presented study
is to combine different distance measures to handle deficiencies of single
distance measures.
15:00
3311.
Tracking of CE-MR-Angiography Data Using Established Approaches in DTI
Computer 30
Daniel
Güllmar1,2, Andreas Deistung1, Stefan Richter1,
Jens Haueisen2, Jürgen R. Reichenbach1
1University
Hospital Jena, Jena, Germany; 2TU Ilmenau, Ilmenau, Germany
We
propose a method to generate tractograms from CE-MR-Angiography data using
existing tools designed to work on diffusion tensor data. The pseudo DT data
are generated by converting the Hessian matrix derived from a multi-scale
approach in order to obtain a highly anisotropic tensor representation along
the vascular system. The results derived from a high resolution 3D MR
angiographic extremity data set demonstrate that it is possible to reconstruct
the vascular system with diffusion tensor fiber tracking tools. Only
deterministic tracking was applied so far which leads to some extent to
fragmentary reconstruction, which may be overcome by probabilistic tracking.
13:30
3312.
Local DTI Connectivity Estimation Using Bayesian Probability Theory
Computer 31
Joshua
S. Shimony1, Adrian A. Epstein1, G Larry Bretthorst1
1Washington
University School of Medicine, St. Louis, Missouri, USA
The
diffusion tensor model has been used to analyze magnetic resonance diffusion
data and has been successful in both neuroscientific and clinical
applications. We propose an enhancement
of this model with a local connectivity parameter that better accords with the
known structure of white-matter. In
addition to providing diffusion tensor parameter estimation the calculation
provides the probability that a given pixel is connected to one of its nearest
neighbors. These probabilities can be
used in further calculations to determine the probability of connectivity
between different brain regions.
14:00
3313.
White Matter Tract Probability Atlas Derived from Diffusion Tensor
Tractography of a Large Population
Computer 31
Catherine
Lebel1, Luis Concha1, Gaolang Gong1, Christian
Beaulieu1
1University
of Alberta, Edmonton, Canada
Brain
white matter is usually crudely parcellated based on lobar location in existing
human brain atlases. Recent attempts to create a more detailed
white
matter atlas based on information derived from diffusion tractography have been
reported, but are limited by small sample sizes. We investigated the spatial
variability in standard space of thirteen white matter structures defined by
tractography in a group of 254 healthy volunteers. The core of the white matter
structures shows high spatial overlap amongst individuals, while the edges show
high variability. Our results could provide valuable information for the
interpretation of voxel-based analyses.
14:30
3314.
Improving the Efficiency of Tractography by Combining DTI with
Prospective Partial-Brain Q-Ball Imaging
Computer 31
Jiancheng
Zhuang1, Nicolas Lori1, Christine Vidal1,
Hanna Damasio1
1University
of Southern California, Los Angeles, USA
This
study proposed to develop a method which combines the conventional DTI and a
prospective partial-brain QBI to improve the efficiency of tractography. The
non-Gaussian ADC profile detected from DTI data was used to indicate which
brain region to be covered by QBI. The multiple wavevector fusion (MWF)
algorithm was performed in the combination of data, and multiple FACT algorithm
was applied in fiber tracking. The quality of connectivity detection in this
approach is sufficiently high, and the scan time is reduced by about 65.3% when
compared with a similar QBI with the whole brain coverage.
15:00
3315.
Connectivity-Based Probabilistic Parcellation of the Striatum in Human
Brain by Diffusion Tensor Imaging
Computer 31
mingguo
qiu1, qiyu li1, jian wang2, bing xie2,
shaoxiang zhang1
1Third
military medical university, chongqing, People's Republic of China; 2southwest
hospital of Third military medical university, People's Republic of China
The
aim is to investigate the corticostriatal connections in human by DTI. The
results showed the frontal projections were located to the head of the caudate
nucleus and the superior part of the putamen, the insular projections to the
lateral striatum, the projections from the parietal, temporal and occipital
lobes overlapped to the posterior part and the inferior part of the striatum.
M1 were connected to the posterosuperior part, SMA to the middle part and PMC
to the anterior and posteroinferior striatum. The striatum have specific
connections with the different cortex, connectivity-based probabilistic
tractography and parcellation is valuable in the functional anatomy of human
brain.
Computer 32
Stephane
Lehericy1, Essa Yacoub2, Eric Bardinet1,
Romain Valabregue1, Chantal Francois1, Geoff Ghose2,
Noam Harel2
1University
Pierre and Marie Curie, Paris, France; 2University of Minnesota,
Minneapolis, Minnesota, USA
In
this work, we used 7T whole body and diffusion tensor imaging (DTI) to
reconstruct the nigro-striatal fiber tract in the anesthetized macaque brain.
DTI was performed using a single shot SE acquisition. The nigro-striatal fiber
tract was reconstructed between the segmented substantia nigra and lenticular
nucleus. The tract coursed from the medial part of the substantia nigra
anteriorly toward the medial part of the globus pallidus. DTI reconstruction of
the nigro-striatal fiber tract provides a marker of the nigro-striatal pathway
that can be used to study basal ganglia pathology such as Parkinson’s disease.
Computer 32
Jennifer
A. McNab1, Saad Jbabdi1, Sean C.L. Deoni,12,
Gwenaelle Douaud1, Timothy E.J. Behrens1, Karla L. Miller1
1University
of Oxford, Oxford, UK; 2Institute of Psychiatry, King's College,
University of London, London, UK
This
study presents a robust method for estimating the principal diffusion direction
from multi-angle DW-SSFP data and demonstrates probabilistic tractography at
sub-millimeter resolution in a whole, fixed human brain.
14:30
3318.
Atlas-Based Reference Tracts Improve Automatic White Matter Segmentation
with Neighbourhood Tractography
Computer 32
Susana
Muñoz Maniega1, Mark E. Bastin1, Andrew M. McIntosh1,
Stephen M. Lawrie1, Jonathan D. Clayden2
1University
of Edinburgh, Edinburgh, UK; 2UCL , London, UK
Neighbourhood
tractography is a novel technique that aims to automatically segment the same
fasciculus in different subjects by scoring the similarity between a predefined
“reference tract” and a group of candidate tracts generated with different
initial seed points. In the current work we present a means of improving the
technique by generating the reference tract from an anatomical atlas. Using
diffusion data from 50 volunteers we found that atlas-generated reference tracts
improved the segmentation results in the four fasciculi examined, obtaining
lower variation of FA within the group and percentages of visually plausible
tracts of up to 82%.
15:00
3319.
Using BCP Index to Evaluate Coregistration Methods in Existing SPM2
Computer 32
Shiou-Ping
Lee1, Kao-Lun Wang1, Chung-Ming Chen2, Wen-Yih
Isaac Tseng3
1Far
Eastern Memorial Hospital, Taipei, Taiwan; 2National Taiwan
University Hospital, Taipei, Taiwan; 3National Taiwan University
Medical College, Taipei, Taiwan
A
novel indices, Bundle Curvature Profile (BCP), was proposed as the basis to
quantify the similarity of four fiber bundles to distinguish geometric
significant difference. In this study, we corrected diffusion tensor imaging
data using four different SPM2 built-in coregistration functions, i.e.,
normalized mutual information (NMI), mutual-information (MI), entropy
correlation coefficient (ECC) and normalized correlation coefficient (NCC). We
also applied coherence index (CI) to evaluate the accuracy of fiber tractography
in corpus callosum. In conclusion, the proposed BCP index is also capable in
revealing the difference between four coregistration functions which is
consistent with the results from CI index whereas the similarity of fiber
tracking using NCC correction method is higher than ECC, MI and NMI.
Crossing Fibers & Non-Tensor Approaches
Hall D Wednesday 13:30-15:30
13:30
3320.
Fibre Orientation Probability Maps from Q-Ball and the Model-Based Bootstrap – A Potential Segmentation
Tool
Computer 33
Hamied
Ahmad Haroon1, Karl V. Embleton1, Geoff J M Parker1
1The
University of Manchester, Manchester, UK
Model-based
residual (MBR) bootstrap enables quantification of the uncertainty in the
inferred fibre orientation for probabilistic fibre tracking using a single HARDI
dataset. Here we present probability maps of observing n fibre orientations estimated by MBR bootstrapping over i iterations in every voxel. These maps
provide information for the classification of tissues based on their
microstructural orientation complexity. The probability of finding any given
configuration reflects the underlying tissue microscopic complexity,
macroscopic partial volume, and data noise levels. This information will be of
use in probabilistic tractography and in monitoring changes in tissue
complexity due to disease or developmental processes.
14:00
3321.
4th Order Diffusion Tensor Estimation and Application
Computer 33
Aurobrata
Ghosh1, Maxime Descoteaux1, Rachid Deriche1
1INRIA,
Sophia Antipolis – Méditerranée, Sophia Antipolis, France
We
review and compare the existing methods for 4th order diffusion tensor imaging.
We also propose an extension to the Riemannian framework developed for
classical 2nd order DTI to the space of 4th order tensors. We test the methods
on synthetic HARDI data, and on real brain data. In conclusion our experiments
confirm that our Riemannian extension is capable of detecting fiber crossings,
is the only algorithm that guarantees positive definite diffusion and is also
computationally viable to be practicable.
14:30
3322.
Denoising HARDI Coefficients Using Spherical Wavelet Lifting
Computer 33
Sofia
Olhede1, Brandon Whitcher2
1University
College London, London, UK; 2GlaxoSmithKline, London, UK
We
propose to estimate the diffusion distribution at each voxel by using a
discrete wavelet transformation on the sphere, combined with an estimation
procedure that is adapted to the Rician distribution. To respect the underlying sampling scheme and
avoid smoothing out details in Q-space we construct a discrete wavelet
transform using lifting.
15:00
3323.
How Fast Can PAS Go?
Computer 33
Ken
Earl Sakaie1
1The
Cleveland Clinic, Cleveland, Ohio, USA
High
angular resolution diffusion imaging, combined with advanced postprocessing
methods, can resolve crossing fibers. Of the large number of postprocessing
methods, Persistent Angular Structure (PAS) has desirable properties in terms
of angular resolution, accuracy, and robustness against noise, but the publicly
available PAS calculation software can require weeks to months to analyze a
typical brain dataset on a single processor. We show that coding the PAS
calculation using widely available nonlinear optimization code can speed up the
computation considerably, making this powerful method an option for more
widespread use.
13:30
3324.
Bootstrap Methods for Estimating Uncertainty in Constrained Spherical
Deconvolution Fiber Orientations
Computer 34
Ben
Jeurissen1, Alexander Leemans2, J-Donald Tournier3,4,
Jan Sijbers1
1University
of Antwerp, Antwerp, Belgium; 2School of Psychology, Cardiff
University, Cardiff, UK; 3Brain Research Institute, Melbourne,
Australia; 4University of Melbourne, Melbourne, Australia
Monte
Carlo simulations were used to assess the accuracy and precision of bootstrap
estimates of the uncertainty associated with brain white matter fiber
orientations derived from diffusion weighted MRI using the new high angular
reconstruction technique called constrained spherical deconvolution (CSD). We
studied the accuracy and precision of these estimates as a function of repeated
acquisitions, bootstrap realizations and inter-fiber angle. We showed an
improvement of the accuracy of the uncertainty estimates of CSD fiber
orientations, using an adaptation of the bootstrap called bootknife. Estimation
of uncertainty is very important when fiber orientations are used in white
matter fiber tractography, since errors are known to propagate.
14:00
3325.
CSF Partial Volume Effect for Diffusional Kurtosis Imaging
Computer 34
Caixia
Hu1, Jens H. Jensen1, Maria Fatima Falangola1,
Joseph A. Helpern1
1Center
for Biomedical Imaging, New York, New York, USA
It
is well known that CSF partial volume effects affect the fractional anisotropy
(FA) and mean diffusivity (MD) values measured with diffusion tensor imaging
(DTI). Prior studies have shown how to
assess this quantitatively by comparing results from conventional DTI with
those from fluid-attenuated inversion recovery (FLAIR) DTI In this study, we
applied a recent generalization of DTI called diffusional kurtosis imaging
(DKI) to obtain measurements of FA, MD, and mean kurtosis (MK) in human brain
with and without FLAIR and found MK to be a more robust measurement with regard
to CSF contamination than either MD or FA.
14:30
3326.
Voxel Based Topometry of the ADC Profiles: Collapsing of the
Dimensionality
Computer 34
Oleg
Petrovich Posnansky1, Nadim Jon Shah1
1Institute
of Neurosciences and Biophysics (Medicine), Research Centre Juelich, Juelich,
Germany
We
explore the complex geometry of the apparent diffusion coefficient (ADC) profiles
and introduce a set of measures and their topological characteristics –
non-integer dimensions. Using the step-by-step scale blowing sphere method we
analyse the random spatial structure of the ADC profiles and present maps of
indices obtained by the proposed algorithm. The maps correlate with the
anatomical structure of the brain to different extents and provide more detail
non-Gaussian information about brain architecture.
15:00
3327.
A Simple Method for ODF Reorientation After Deformable Imaging Registraton
Computer 34
Jinsuh
Kim1, Madhura Ingalhalikar2, Vincent A. Magnotta2,
Andrew L. Alexander3
1University
of Iowa, Iowa City, Iowa, USA; 2University of Iowa, USA; 3University
of Wisconsin, Madison, Wisconsin, USA
We
propose a simple and efficient method of orientation distribution function
reorientation that allows correct estimation of fiber distribution by deforming
high angular resolution diffusion imaging data and applied diffusion gradient
direction based on the high dimensional transformation.
13:30
3328.
Comparison of Directional Diffusion Kurtoses and Diffusivities in EAE-
Induced Spinal Cord
Computer 35
Matthew
Man Hin Cheung1,2, Edward Sai Kam Hui1, Wutian
Wu1, Ed X. Wu1
1The
University of Hong Kong, Pokfulam, Hong Kong
Understanding
the complex pathology of MS is important in treatment strategies. In this
experiment, diffusion kurtosis imaging (DKI) was applied to ex-vivo EAE rat
spinal cord samples. The DK tensor and directional kurtoses were computed. The
differences between the EAE and the normal were evaluated. The directional
kurtosis analysis was shown to possess high sensitivity, thus potentially
capable of better differentiating MS pathology and understanding the disease.
14:00
3329.
Diffusional Kurtosis Approximation of the Orientation Distribution
Function in the Human Brain
Computer 35
Mariana
Lazar1, Jens H. Jensen1, Joseph A. Helpern1
1New
York University School of Medicine, New York, USA
The
Orientation Distribution Function (ODF) is used to describe the directionality
of multimodal diffusion in regions with complex fiber architecture. Here, we
present an approximation for the ODF of water diffusion from diffusional
kurtosis imaging (DKI). The DKI-based ODF approximation is decomposed into two
components representing the Gaussian and non-Gaussian (NG) diffusion
contributions, respectively. Orientation maps obtained for in-vivo brain
imaging data demonstrate multiple fiber components in brain regions with
complex anatomy, with the NG-ODF being the most sensitive to profiling the fibers’
directions. The results appear to be in agreement with known white matter
architecture.
14:30
3330.
Simulation Study of Kurtosis Measurements in MR Diffusion Kurtosis
Imaging
Computer 35
Kyle
Ho Yiu Cheng1,2, Ed X. Wu1
1The
University of Hong Kong, Pokfulam, Hong Kong
Monte
Carlo simulation was performed in 1D space to study the spin diffusion,
kurtosis characteristics, spin dephasing and MR signal attenuation in
compartment-restricted diffusion environment. The apparent kurtosis was
computed based on DKI formulation and compared to the true kurtosis and
calculated from the actual diffusion displacement profiles. The simple results
revealed that errors can be present in estimating true diffusion kurtosis by
the apparent kurtosis. Choices of diffusion weighting parameters affect the
outcomes. Both apparent diffusion kurtosis measurement and true diffusion
kurtosis are not entirely intrinsic to the tissue structures because they also
depend on the MR parameters. Therefore,
cautions must be taken in the quantitative interpretation of kurtosis
measurements in DKI experiments.
15:00
3331.
Diffusion Kurtosis Imaging (DKI) of in Vivo Human Brain at 7 T
Computer 35
Eric
Edward Sigmund1, Caixia Hu1, Mariana Lazar1,
Maria Fatima Falangola1, Jens H. Jensen1, Joseph A.
Helpern1
1New
York University, New York, New York, USA
Diffusion
kurtosis imaging (DKI) has been performed in healthy volunteer in vivo brain in
a 7 T scanner and 24-channel head coil.
DKI provides all of the
information in DTI in addition to higher order structural contrast determined
by non-Gaussian diffusion. High parallel
imaging factors were used to mitigate
high field EPI image artifacts. Mean
diffusivity MD, fractional anisotropy FA, and mean kurtosis MK were consistent
with results at 3 T. MD showed
comparable white matter / gray matter values, FA was elevated in WM and
negligible in GM, and MK was nonzero and distinct in both tissue types.
Hall D Wednesday 13:30-15:30
13:30
3332.
A Statistical Approach for Creating Anisotropy Maps of the Brain Using
Q-Space Diffusion Weighted Images
Computer 36
Siamak
Pourabdollah-Nejad1,2, Quan Jiang1,3,
Douglas C. Noll2, Guang Liang Ding1, Michael Chopp1,3
1Henry
Ford Hospital, Detroit, Michigan, USA; 2University of Michigan, Ann
Arbor, Michigan, USA; 3Oakland University, Rochester, Michigan, USA
In
this abstract, a new method for creating anisotropy maps of the brain in
introduced. In this method q-space data of the brain is acquired and as a
measure of anisotropy, the deviation of the diffusion vectors from a sphere is
calculated and is used for creating the new map which we have named
Standard-Deviation map. Comparison of our map with other conventional maps such
as the FA map shows that the new map can identify regions with crossing fibers
more realistically. We have validated our method with histology and q-space MR
images of a rat with traumatic brain injury.
14:00
3333.
Probing Restricted Microcompartments with Double PGSE
Computer 36
Noam
Shemesh1, Yoram Cohen1
1Tel
Aviv University, Tel Aviv, Israel
The
double PGSE (d-PGSE) sequence was used to study water diffusion in 20ìm
microtubes. Indeed we found, as predicted by a recent simulation, that a
negative diffraction pattern occurs. The results of non-collinear d-PGSE (i.e.
d-PGSExz or d-PGSEzx ) were found to be identical and independent of the mixing
time (tm). The signal decay and the diffraction patterns of the collinear
directions (i.e. d-PGSExx) depend on tm. Future work in micro scale phantoms
and neuronal tissues will demonstrate whether d-PGSE can overcome some of the
limitations of s-PGSE experiments.
14:30
3334.
The Dependency of the MR Indices, Observed by High B-Value Q-Space
Diffusion MRS on Fiber's Orientation
Computer 36
Amnon
Bar-Shir1, Yoram Cohen1
1Tel-Aviv
University, Tel-Aviv, Israel
The
present study examines the effect of the rotational angle, α, on the
apparent diffusion coefficient, extracted from low b-values and on the root
mean square displacement (rmsd) of the fast- and the slow-diffusing components
extracted from high b-value q-space MR diffusion experiments performed on optic
nerves. The rmsd of the slow-diffusing component exhibits the most significant
dependence on α. This phenomenon was found for both mature and newborn
optic nerves and for short and long diffusion times. Our findings imply that
the rmsd of the slow-diffusing component is the best predictor for restriction
and fiber's orientation.
15:00
3335. in vivo High Resolution Q-Space Imaging of the Spinal Cord Injury in Nonhuman
Primates
Computer 36
Keigo
Hikishima1,2, Masaya Nakamura, Kanehiro Fujiyoshi,
Masayuki Yamada,2, Kazuya Kitamura, Suketaka Momoshima, Kazuo Yagi,
Norikazu Tamaoki2, Yoshiaki Toyama, Hideyuki Okano
1Keio
University School of Medicine, Sinjuku, Japan; 2Central Institute
for Experimental Animals, Kawasaki, Japan
q-space
imaging(QSI) enables us to detect the size of microstructure quantitatively and
has been used gradually. However, there is no report about in vivo QSI of the
spinal cord (SC) in Nonhuman primates. To evaluate the structural changes of
the injured SC in the same animal longitudinally, we performed in vivo high
resolution QSI of both intact and injured spinal cords in common marmosets and
confirmed the accuracy of QSI through histology. As the results, high
Resolution QSI map reflected the histological changes of the injured SC and was
useful for its evaluation.
13:30
3336.
A Spectral Filtering View of Diffusion Gradient Encoding
Computer 37
Andrew
JM Kiruluta1
1Harvard
University, Cambridge, Massachusetts, USA
Diffusion
imaging gradients serve to spectrally filter the temporally evolving diffusion
tensor in the spectral domain. In this formulation, the design of diffusion
sensitizing gradients is reduced to the problem of adequately sampling q-space
in the spectral domain and hence one of designing a suitable filter. The practical limitations imposed by the
requirement for delta function type diffusion sensitizing gradients to
adequately sample q-space, can be relaxed if these impulse gradients are replaced
with chirped oscillatory gradients.
Computer 37
Henry
H. Ong1, Felix W. Wehrli1
1University
of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA
Knowledge
of white matter (WM) intra-cellular volume fraction and mean axon diameter (ICF
and MAD) would provide important insight into injury and pathology. Q-space
imaging (QSI) offers potential for indirect assessment of WM architecture but
is complicated by signal from both extra- and intra-cellular spaces (ECS and ICS). Here, we use a two-compartment
model to fit QSI molecular displacement distribution profiles from WM tracts in
healthy mouse spinal cords to characterize the ECS and ICS. The measured ICF
and MAD values showed excellent agreement with histology and demonstrate the
feasibility of this method to extract non-destructively accurate WM
architecture information.
Methods for Brain Arterial Spin Labelling
Hall D Thursday 13:30-15:30
Computer 30
Thomas
Liu1, Christina Wierenga2, Bryon Mueller, Jiong Jiong
Wang, Gary Glover, James Voyvodic, Doug Greve, Jessica Turner, Cynthia Wible,
Greg Brown, Function BIRN
1UCSD
Center for Functional MRI, La Jolla, California , USA; 2UCSD Dept.
of Psychiatry, La Jolla, California , USA
Arterial
spin labeling MRI was used to obtain whole brain cerebral blood flow measures
in a sample of 11 healthy volunteers who were scanned at three different sites
participating in the Function BIRN study.
There was not a significant effect of site on the cerebral blood flow
measures, but there was a significant effect of subject.
14:00
3339.
Pseudo-Continuous Arterial Spin Labeling at 7T
Computer 30
Wen-Ming
Luh1, Tie-Qiang Li1, Eric C. Wong2, Peter A.
Bandettini1
1National
Institutes of Health, Bethesda, Maryland, USA; 2University of
California, San Diego, La Jolla, California , USA
Arterial
spin labeling (ASL) techniques benefit from the increase in T1 at high field
strength. However, the SNR improvement at high field can only be realized with
an optimal tag duration which also increases with field strength. At 7T, the
optimal tag duration can reach 3 and 1.3 sec for continuous and pulsed ASL,
respectively. Pseudo-continuous ASL
allows for long tag duration without special hardware. However, both B1 and B0
inhomogeneity with the volume excitation coil toward the tagging location
presents challenges for PCASL. Here we implemented PCASL at 7T and demonstrated
these effects and a way to mitigate them.
14:30
3340.
Mapping PASL Arterial Transit Time in Normal Human Brain Using
[15O]water PET
Computer 30
Maolin
Qiu1, J Wang1, H Kim1, R E. Carson1,
R T. Constable1
1Yale
University School of Medicine, New Haven, Connecticut, USA
Arterial
transit time (ATT) in pulsed arterial spin labeling (PASL) refers to the time
it takes the arterial blood to travel from the labeling site to the
capillaries. It defines the earliest time data acquisition could start after IR
labeling and is required for CBF quantification. ATT measurements that use the
PASL sequence suffer from the low sensitivity of PASL and intravascular
contamination. ATT is usually determined and used on a per-slice basis and
within slice variations in ATT are not considered. The use of improper ATT
values can introduce errors in CBF quantification. In this study we used CBF
measured by PET to calculate the ATT on a per-voxel basis, which allows us not
only to examine the in-slice ATT variability, but also to optimize the imaging
parameters for PASL. More important, the ATT map, once estimated, can be used
for CBF mapping in a similar PASL imaging setup, providing voxel based ATT
values.
15:00
3341.
Tagging Efficiency Improvement Using Velocity-Matched Pseudo-Continuous
Arterial Spin Labeling and VERSE
Computer 30
Wen-Ming
Luh1, Eric C. Wong2, S Lalith Talagala1, Peter
A. Bandettini1
1National
Institutes of Health, Bethesda, Maryland, USA; 2University of
California, San Diego, La Jolla, California , USA
Pseudo-continuous
arterial spin labeling can be performed with standard commercial scanners
without special hardware. Moreover, several characteristics of the RF and
gradient pulses can be adjusted for optimal performance. However, changes in
velocity through the cardiac cycle can compromise tagging efficiency especially
during systolic phases. It is possible to improve tagging efficiency for high
velocity spins by increasing RF amplitude at the cost of SAR. Fortunately, the
increase in SAR can be mitigated with VERSE transformation if needed. Here we
dynamically raised PCASL RF amplitude according to the measured velocity
profile at labeling location with cardiac gating and VERSE modification.
13:30
3342.
Influence of Cardiac Cycle on Velocity Selective Arterial Spin Labeling
Computer 31
Wouter
M. Teeuwisse1,2, Mark A. van Buchem1, Matthias
J. van Osch1
1Leiden
University Medical Center, Leiden, Netherlands
In
velocity selective arterial spin labeling (VS-ASL), blood that flows faster
than a predetermined cut-off velocity is labeled. Because blood velocity is not
constant, labeling efficiency may vary over cardiac cycle. In this study,
dependency of perfusion signal on cardiac cycle was investigated. VS-ASL was
performed with cardiac triggering at various trigger delays. In all volunteers
mean slice perfusion showed variation up to ± 20% compared to mean perfusion
with trigger delay at diastole. Variation over cardiac cycle was different
between subjects. When cardiac triggering is applied, a trigger delay of 300 ms
gives highest label efficiency in most cases.
Computer 31
María
A. Fernández-Seara1, Jiongjiong Wang2, David A. Feinberg3,
John A. Detre2
1Center
for Applied Medical Research, University of Navarra, Pamplona, Spain; 2University
of Pennsylvania Medical Center, Philadelphia, Pennsylvania, USA; 3Advanced
MRI Technologies, Sebastopol, California , USA
A
modified version of the pseudo-CASL background suppressed 3D GRASE sequence is
presented. This sequence, which allows acquisition of perfusion maps with long
post-labeling delay, in a short scan time, will be useful in cases in which
arterial transit time is prolonged as occurs in cerebro-vascular disease or
acute stroke.
14:30
3344.
An Efficient Labeling Scheme for CASL for Use with Multiple,
Independently Switched Coils
Computer 31
Ralf
Berthold Loeffler1, Ruitian Song1, Yong Zhang1,
Adam Martin Winchell1,2, Josef Pfeuffer3,
Zoltan Patay1, Claudia Maria Hillenbrand1
1St.
Jude Children's Research Hospital, Memphis, Tennessee, USA; 2University
of Memphis and UTHSC Joint Biomedical Engineering Program, Memphis, Tennessee,
USA; 3Siemens Medical Solutions, Erlangen, Germany
A
new paradigm for efficient multi-coil continuous arterial spin labeling is
presented. The paradigm utilizes signal from every excitation to calculate the
contribution of a particular artery. The paradigm was implemented on a standard
clinical scanner for 3 label coils, which led to a paradigm length of 8. The
label coils could be switched independently by a custom built RF cabinet that
was controlled by the scanner. Volunteer measurements demonstrated good
separation of vascular territories in the brain.
15:00
3345.
Continuous Artery-Selective Spin Labeling (CASSL) Applied to Distal
Branches of Intracranial Arteries
Computer 31
Michael
Helle1,2, David Norris2, Karsten Alfke1,
Olav Jansen1
1Christian-Albrechts-Universität,
Kiel, Germany; 2FC Donders Centre for Cognitive Neuroimaging,
Nijmegen, Netherlands
We
demonstrate the feasibility of CASSL to selectively label blood of individual
branching intracranial arteries even in the immediate vicinity of other small
vessels in-vivo. A saturation of the magnetization in non-selected vessels
depends basically on the labeling gradient orientation, its rotation frequency
and the distance from the labeling focus. By empirical optimization of these
key parameters the labeling of small arteries with diameters of 3 mm and
less, that branch from the media and
anterior cerebral arteries, mainly to A2/A3 and M2/M3 segments respectively, is
demonstrated.
13:30
3346.
A Simple Model to Measure Arterial Cerebral Blood Volume by Arterial
Spin Labelling
Computer 32
Roman
Wesołowski1, Penny A. Gowland1, Susan T. Francis1
1University
of Nottingham, Nottingham, UK
A
method of quantifying LL-FAIR data for arterial cerebral blood volume (CBVa)
using simple correction factors is described. Monte-Carlo simulations are used
to compare the results of fitting CBVa using this simplified method with the
standard stepwise compartment model based on Bloch equations. We have used the
new method to quantify (CBVa) from experimental LL-FAIR data acquired at 3 T
during a 4.8 s visual stimulus.
14:00
3347.
A Probabilistic Approach to Model-Free Arterial Spin Labeling Perfusion
Quantification
Computer 32
Michael
A. Chappell1, Salima Makni1, Saad Jbabdi1,
Mark W. Woolrich1
1University
of Oxford, Oxford, UK
Traditional
'kinetic curve' model based methods for ASL analysis are not ideal for
perfusion quantification in pathological tissue, since assumptions about healty
tissue are made in the models. Recently the 'model-free' approach to ASL
analysis has been demonstrated, this treats the signal as the convolution of an
arterial input and redsidue function. Here we present an improved probabilistic
based method for model-free ASL analysis. The advantage of this approach is
that residue function shape can be determined under the constraint of temporal
smoothness, without neededing to set the functional form a priori and without
the regularisation aretefacts of a singular value decomposition approach. The
method also includes a procdure to determine arterial feeding regions, for
arterial input function assignment, based on blood arrival time information.
Ultimately the method is designed to provide robust perfusion measurments from
ASL data even in pathological tissue.
14:30
3348.
Post-Processing Correction for Extended Data Acquisition in Whole Brain
3D Quantitative PULSAR Imaging
Computer 32
Neville
Dali Gai1, John Anthony Butman1
1National
Institutes of Health, Bethesda, Maryland, USA
Here
we devise a simple first-order correction scheme for extended data acquisition
3D PULSAR brain perfusion imaging. Extended data acquisition (~600ms)
introduces some ambiguity in the transit time used in quantitative CBF
determination. By applying an exponential correction for tagged blood decay in
hybrid space, we attempt to reduce bolus dispersion effects during acquisition.
Corrected values were compared with corresponding single slice imaging. CBF
values in 3D slices were lower than corresponding single slice acquisition
values due to blurring. However, after correction the values showed better
agreement with the single slice acquisition case.
15:00
3349.
Comparison of Quantitative Perfusion 3D IR-PULSAR with Multi-Slice 2D and Single Slice Imaging
Computer 32
Neville
Dali Gai1, Sardha Lalith Talagala1, John Anthony Butman1
1National
Institutes of Health, Bethesda, Maryland, USA
In
this study, CBF values obtained with extended 3D data acquisition (for whole
brain coverage) QUIPSSII IR-PULSAR are compared with values obtained with
multi-slice 2D (MS-2D) and single slice acquisition. Excellent agreement in
gray matter (GM) global average CBF values was found between the 3D and 2D-MS
acquisition schemes. There was good agreement between the GM CBF values from 3D
slices and corresponding single slice acquisitions. Values were slightly lower
in the 3D case due to blurring. White matter CBF values were very low for all
acquisition schemes due to much longer transit delays (~1.6s). Whole brain
quantitative perfusion imaging with 3D PULSAR is feasible as long as the above
limitations are kept in mind.
Hall D Thursday 13:30-15:30 Chairs: Chunlei Liu and Claudia A. M. Wheeler-Kingshot
13:30
3350.
Simulations of SNR Efficiency of DTI Using Parallel Imaging and RFOV
Acquisition at 3 T and 7 T
Computer 33
Carolin
Reischauer1, Robert Stefan Vorburger1, Bertram Jakob Wilm1,
Thomas Jaermann1, Philipp Staempfli1, Peter Boesiger1
1ETH
and University of Zurich, Zurich, Switzerland
The
transition to ultra-high field strengths promises a boost in the SNR for
diffusion tensor imaging. Image acquisition typically relies on single-shot EPI
which is hampered by susceptibility-induced distortions and T2* related
blurring, especially at high field strengths. These challenges can be quite
effectively addressed by partial Fourier and reduced FOV acquisition, as well
as parallel imaging methods. In the present work simulations at 3 T and 7 T
were performed to investigate under which conditions the SNR gain can be
tapped, provided that higher reduction factors have to be applied at ultra-high
field strengths.
14:00
3351.
Apparent Diffusion Behaviors Modulated by Distant Dipolar Field in
Solution NMR
Computer 33
Shuhui
Cai1, Guiping Shen1, Congbo Cai1, Zhong Chen1
1Xiamen
University, Xiamen, People's Republic of China
A
modified CRAZED sequence was designed to observe and characterize apparent
diffusion behaviors of signals from intermolecular double-quantum coherences
during the mixing period. It is found that their apparent diffusion behaviors
are different from conventional single-quantum coherences, and different
orientation of diffusion weighting gradients relative to coherence selection
gradients results in different apparent diffusion behaviors. This indicates
that the apparent diffusion behavior is influenced by the distant dipolar
field.
14:30
3352.
Optimizing Diffusion Measurements for Large-Scale Multi-Centre Trials: A
Magnims DT MRI Sequence
Computer 33
Elisabetta
Pagani1, Jochen G. Hirsch2, Marco Rovaris1,
Achim Gass2, Petra Pouwels3, Stephan Roosendaal3,
Frederik Barkhof3, Federica Agosta1, Domenico Caputo4,
Antonio Giorgio5, Jacqueline Palace5, Silvia Marino6,
Nicola De Stefano6, Massimo Filippi1
1Scientific
Institute H San Raffaele, Milan, Italy; 2University Hospital Basel,
Basel, Switzerland; 3VU University Medical Centre, Amsterdam,
Netherlands; 4Scientific Institute Fondazione Don Gnocchi, Milan,
Ita
The
aims of the present study were: a) the development of an optimal acquisition
scheme of diffusion tensor (DT) magnetic resonance imaging (MRI) for
large-scale trials (considering the time issue in the context of
“multi-sequence scans”) and b) the evaluation of both the feasibility of the
sequence set-up on various scanners and the inter-centre reproducibility of
DT-derived metrics. Twenty-nine healthy subjects were studied in 7 MRI centres
using a standardized DT-MRI sequence. The inter-scanner variability of
DT-derived quantities was found to be lower for anisotropy than mean
diffusivity measures, especially when using 1.5 T magnets.
Computer 33
Donghui
Yin1, Jonathan C. Sharp2, Boguslaw Tomanek2
1National
Research Council Institute for Biodiagnostics, Winnipeg, Canada; 2National
Research Council Institute for Biodiagnostics (West), Calgary, Canada
Our
aim in this work is to extend a general-purpose integrated MR simulation / MR
console environment to include diffusion-weighting using the Monte Carlo stochastic
method. The general-purpose nature requires that any sequence written for the
MR console is also available for simulation without additional simulation
programming or expertise required of the user. We have demonstrated that the
Monte Carlo method can reliably produce diffusion weighting in many cases. Our
implementation has been optimized for multi-core processors so that the
performance can automatically improve as industry makes new CPUs available.
Computer 34
Hernan
Jara1
1Boston
University Medical Center, Boston, Massachusetts, USA
Purpose:
To develop a T1 relaxation theory incorporating the combined effects of
magnetization transfer and water micro-kinetics, for the purpose of computing
accurate correlation time diffusion coefficient (D(CT)) maps of structurally
complex biological tissues. Methods: Images obtained with the mixed-TSE sequence
were postprocessed for generating PD, T1, and T2 maps and used for computing
maps of the correlation time diffusion coefficient with the developed T1-MT
theory. Results: Excellent quantitative agreement is found in the brain
relative to standard pulsed-field-gradient MRI. Conclusion: Magnetization
transfer has a substantial effect on T1 relaxation for tissues containing a
restricted pool of 1H protons.
14:00
3355.
Single-Shot Multi-Echo Parallel EPI for DTI with Improved Efficiency and
Accuracy
Computer 34
Roger
Nana1, Tiejun Zhao2, Xiaoping Hu1
1Georgia
Institute of Technology / Emory University, Atlanta, USA; 2Siemens
Medical Solutions, Malvern, USA
We
introduce a multi-echo parallel EPI acquisition strategy to enhance SNR while
maintaining the advantages of parallel EPI. We show that an appropriate echo
combination strategy can provide significant gains in SNR as compared to using
only one echo. This SNR gain can be utilized to reduce the number of
measurements often required to ensure adequate SNR for accurate calculation of
various DTI measures or to improve spatial resolution. Furthermore, the
multiple echoes can be used to derive a T2 map, providing additional
information that might be useful in some applications.
14:30
3356.
Towards Quantitative Diffusion-Weighted Chemical Shift Imaging of Brain
Metabolites
Computer 34
Itamar
Ronen1, Dae-Shik Kim1
1Boston
University School of Medicine, Boston, Massachusetts, USA
Diffusion
tensor imaging (DTI) has become a ubiquitous MRI modality for the assessment of
microstructural properties of tissue. Its explanatory power, however, is
limited by the lack of compartmental specificity of the water MR signal. The
possibility of gaining compartment-specific diffusion information has been
explored using diffusion weighted spectroscopy of intracellular metabolites. In
this work we explore for the first time the feasibility of obtaining
quantitative diffusion information on intracellular metabolites using diffusion
weighted chemical shift imaging (CSI). The main hurdles for reliable diffusion
measurements, i.e. loss of phase consistency and signal across the acquisition
are addressed.
Computer 34
Geon-Ho
Jahng1, Michael W. Weiner2, Norbert Schuff2
1East-West
Neo Medical Center, Kyung Hee University, Seoul, Republic of Korea; 2University
of California,San Francisco, San Francisco, California , USA
To
investigate effects of EPI readout gradient modulation frequency on the
accuracy of diffusion tensor (DT)-MRI measurements, we studied the relationship
between the EPI bandwidth and the Nyquist ghost for a spin echo EPI acquisition
with DT preparation on a spherical water phantom and nine volunteers. In
result, there were significant effects on the bandwidths for the b=800 data and
for the b=0 data on the phantom study. On the human study, EPI bandwidth
variations substantially corrupted diffusion anisotropy indexes. The effect can
be minimized by tuning the modulation frequency of the EPI readout gradient.
13:30
3358.
Turboprop Diffusion Tensor Imaging for Computational Design of Drug
Transport in Brain
Computer 35
Xiaodong
Guo1, Mahadevabharath R. Somayaji2, Andreas A. Linninger2,
Jia-Hong Gao1, Richard D. Penn1
1University
of Chicago, Chicago, Illinois, USA; 2University of Illinois at
Chicago, Chicago, Illinois, USA
Three
dimensional subject-specific brain anatomy is reconstructed from Turboprop DTI
images. Rigorous first principles physical transport phenomena are applied to
predict the fate of a high molecular weight neurotropic factor infused into the
midbrain based on DT-MRI derived drug and tissue properties. For predicting
drug distribution in humans, comprehensive transport models considering
heterogeneous and anisotropic brain properties derived from patient-specific
images have not been adequately researched in open literature before.
14:00
3359.
Optimization of DTI Imaging Parameters Using Prior Information of Fiber
Orientation
Computer 35
Wei
Gao1, Hongtu Zhu, Hongyu An, Weili Lin
1University
of North Carolina-Chapel Hill, Chapel Hill, USA
The
aim of this study was to develop a new method for determining optimal imaging
parameters for diffusion tensor imaging. The accuracy and precision of DTI
experiments depend on both the choice of the imaging parameters and the noise
propagation in tensor estimation process. In this work, we propose to combine
both of these two processes during the optimization processes. Additionally, information on fiber
orientation distribution is also taken into account during the optimization
processes. As a result, the diffusion
encoding scheme could efficiently sample those “densely oriented cone areas” in
a shorter data acquisition. Thus, it leads to the efficient use of imaging
acquisition time, which is crucial for pediatric imaging.
14:30
3360.
Turbo Spin Echo Diffusion Tensor Imaging (TSE-DTI) in the Brain at 3 T
and 7 T
Computer 35
Eric
Edward Sigmund1, Daniel Kim1, Flavio Tulio Braga1,
Jian Xu2
1New
York University, New York, New York, USA; 2Siemens Medical Systems,
New York, New York, USA
This
study used a single-shot turbo spin echo sequence for diffusion tensor imaging
(TSE-DTI) in healthy volunteer brain scans in clinical scanners at 3 T and (for
the first time) at 7 T, in comparison with echo-planar imaging (EPI-DTI), which
suffers from known susceptibility artifacts at high field. T2 blurring was significantly reduced in TSE
images using a separate T2 map and k-space-based deconvolution procedure. At 3 T , EPI-DTI and TSE-DTI colormaps were
of comparable quality. At 7 T and at
higher resolution, EPI-DTI suffered significant image distortions, while
TSE-DTI showed correct anatomical proportions and uniform white matter
sensitivity.
15:00
3361.
Noise Bias Reduction and Parallel Imaging for the Measurement of
Diffusion Decay
Computer 35
Guillaume
Gilbert1, Georges A. Haddad1, Gilles Beaudoin1
1Centre
Hospitalier de l'Université de Montréal, Montreal, Canada
In
this abstract, we investigate the applicability of a noise bias reduction
method for the accurate measurement of diffusion decay. This method, based on
an improved combination of signals from array coils and complex averaging, is
used in conjunction with standard imaging and GRAPPA parallel imaging. While
some statistical noise bias correction methods developed for standard imaging
are no longer exact when parallel imaging is used, this approach is shown to
provide an accurate estimation of bi-exponential diffusion parameters, even
when combined with GRAPPA parallel imaging.
13:30
3362.
High-Resolution Diffusion Tensor Imaging Reveals Sub-Structure Whithin
Human Hippocampus in Vivo
Computer 36
David
Andrew Porter1, David Atkinson2, Rod Scott2,3,
Chris A. Clark2
1Siemens
Medical Solutions, Camberley, UK; 2University College, London, UK; 3Great
Ormond Street Hospital NHS Trust, London, UK
Diffusion
tensor imaging has the potential to provide detailed information about the
hippocampus that may be important in a number of diseases linked to hippocampal
abnormality. However, previous human DTI studies in vivo have had difficulty in
visualising hippocampal sub-structures due to the low resolution associated
with single-shot EPI. This study uses the readout-segmented EPI method for high
resolution DTI of the hippocampus and demonstrates some of this sub-structure
in fractional anisotropy maps.
Cerebral Perfusion & Diffusion Imaging of Animal Models
Hall D Thursday 13:30-15:30
13:30
3363.
Impact of Parallel Imaging Acceleration on Perfusion Measurements of the
Rodent Brain
Computer 36
David
Ratering1, Christof Baltes1, Ivana Kotevic1,2,
Markus Rudin1,2
1University
and ETH Zurich, Zurich, Switzerland; 2University Zurich, Zurich,
Switzerland
Dynamic
susceptibility contrast MRI in small rodents demands for high temporal
resolution to detect the fast change of the signal intensity at the first
passage of the contrast agent. The feasibility of acquiring a densely sampled
signal time curve using parallel imaging acceleration has been demonstrated in
humans. In this work, in vivo experiments were carried out to investigate the
influence of SENSE and GRAPPA acceleration on perfusion measurements of the rat
brain. In addition, computer simulations were performed to test the feasibility
of multi-slice imaging at reasonable temporal resolution.
14:00
3364.
Characterising the Origin of the Arterial Spin Labelling Signal in MRI
Using Multi-Echo Acquisitions
Computer 36
Jack
A. Wells1,2, Mark F. Lythgoe1, Mankin Choy1,
David G. Gadian1, Roger J. Ordidge1, David L. Thomas1
1University
College London, London, UK
We
estimate the relative signals deriving from labelled spins in the vasculature
and in the intra and extracellular space within the tissue, by calculating the
T2 of the ASL signal at a variety of post labelling delays and tagging
durations with and without bipolar crusher gradients, in the rat brain. The
results provide evidence for rapid exchange and suggest that the vascular
signal is comprised of both arterial and venous blood. The relative proportion
of extra- to intracellular signal is ~ 2 times greater for the labelled spins
in comparison to the control even at extended tagging duration (3s) and post
labelling delay (0.3s).
14:30
3365.
Novel Cardiac Spin Labeling (CSL) for CBF MRI in Mice
Computer 36
Eric
R. Muir1, Qiang Shen1, Timothy Q. Duong1
1Emory
University, Atlanta, Georgia, USA
Continuous
arterial-spin-labeling with a separate neck coil is generally more sensitive
relative to the single-coil technique. Cerebral-blood-flow (CBF) MRI
application in mice however has been limited by the close proximity between the
neck and the brain coil, which results in saturation of the brain signals by
the neck coil. We introduce an alternative approach to overcome this limitation
by placing the labeling coil at the heart position, which we termed the Cardiac
Spin Labeling (CSL) technique. This approach was applied to image quantitative
basal blood flow and physiologically evoked blood flow changes in normal mice
at high spatial resolution.
13:30
3366.
The Nuts and Bolts of Implementing a Two-Coil CASL Method for CBF
Measurement on a 9.4T/30cm Scanner
Computer 37
Rongyan
Zhang1, Yuan Ma1, Guangping Dai2, Kathleen Yin1,
Shi-Jiang Li1
1Medical
College of Wisconsin, Milwaukee, Wisconsin, USA; 2Massachusetts
General Hospital, Charlestown, Massachusetts, USA
We
recently implemented the two-coil CASL method on our Bruker Biospin 30cm bore
9.4T animal MRI system, and obtained excellent CBF images and time-series. The purpose of this abstract is to share our
experience, the challenges we met and present our solutions. We have found that
sometimes seemingly trivial details are key to the successful implementation.
It is our hope that this information can be helpful to more research and
application sites to utilize this powerful technique in their fMRI studies.
Computer 37
Kimmo
K. Lehtimäki1, Teemu P. Laitinen1, Alejandra Sierra1,
Jari Nissinen1, Asla Pitkänen1, Olli Gröhn1
1A.I.Virtanen
Institute for Molecular Sciences, University of Kuopio, Kuopio, Finland
We
implemented a whole brain voxelwise statistical analysis method, tract based
spatial statistics (TBSS), into the rat epilepsy model for the comparison of ex
vivo diffusion tensor data between control and kainic acid animals. Today, no
reports of TBSS in the rodent brain exist. With this approach we revealed both
brain regions that are known to be associated with epilepsy and those that have
not been earlier connected with epileptogenesis or epilepsy. TBSS combined with
animal models has great potential to serve as a robust screening method to
guide tedious histological analysis to novel target areas in the brain.
14:30
3368.
Toward a Comprehensive 3D DT-MRI Atlas for Marmoset Monkey Brain
Computer 37
Hao
Huang1, Xiaoqin Wang2, Madeleine Chollet2,
Susumu Mori2
1UT
Southwestern Medical Center, Dallas, Texas, USA; 2Johns Hopkins
University, Baltimore, Maryland, USA
Marmoset
has been more and more commonly used as an animal model in neurobiological
study. The digital 3D comprehensive atlas for marmoset brain appears to be an
urgent need. DTI can generate different contrasts of images which provide rich
information of not only cortical and subcortical gray matter but also
individual white matter fiber bundles. In this study, we acquired ex vivo DTI
data of adult marmoset head. Important neural structures were annotated,
segmented and reconstructed. The atlas is being built for the purposes of education,
anatomical reference, data registration and image guided invasive surgery.
Computer 37
Hamied
Ahmad Haroon1, David M. Morris1, Alexander Kaiser2,
Mark Augath2, Nikos K. Logothetis,12, Geoff J M Parker1
1The
University of Manchester, Manchester, UK; 2Max Planck Institute for
Biological Cybernetics, Tuebingen, Germany
We
present a study attempting to validate the corticocortical connection
information obtainable from diffusion-weighted MR data. We have implemented
probabilistic tractography in data acquired in a macaque model and compared
this with connection information in a database of invasive tracer studies in
the same model. The nature of the corticocortical interconnection information
gained from probabilistic tractography is different to that gained from
invasive studies, the latter also being sparse. Our results using the LVE00a
parcellation scheme indicate that probabilistic tractography is able to give
statistically comparable information on corticocortical interconnections to
invasive tracer studies.
13:30
3370.
Diffusion Tensor Tractography of Primate Visual Pathway
Computer 38
Masayuki
Yamada1,2, Suketaka Momoshima3, Yoshitaka
Masutani4, Kanehiro Fujiyoshi3, Keigo Hikishima,23,
Osamu Abe4, Masaya Nakamura3, Shigeki Aoki4,
Norikazu Tamaoki5, Hideyuki Okano3
1Fujita
Health University, Toyoake, Japan; 2Central Institute for Experimental Animals, Kawasaki, Japan; 3Keio
University, Shinjuku-ku, Japan; 4University of Tokyo Hospital,
Bunkyo-ku, Japan; 5Centra
We
evaluated the reliability of non-human primate visual pathway depicted by
diffusion tensor tractography (DTT) with the assistance of manganese-enhanced
magnetic resonance imaging (MEMRI) tract tracing. The DTT depicted the typical
primate retinogeniculate pathways branching bilaterally at the optic chiasm and
the MEMRI tract tracing also revealed this bilateral innervation. The
configurations visualized by the two modalities were nearly identical and these
morphological findings were corresponded with those obtained from previous
histopathological studies. Thus, the findings of our study suggested that the
DTT methods play a crucial role in the morphological analysis of the non-human
and/or human primate visual pathways.
14:00
3371.
Corpus Callosum Injury and Seizures in WAG/Rij Rats: Correlation Between
DTI and Disease Phenotype
Computer 38
Halima
Chahboune1,2, Asht Mangal Mishra1, Xenophon
Papademetris1, Fahmeed Hyder1, Hal Blumenfeld1
1Yale
University School of Medicine, New Haven, USA
In
vivo DTI has great potential for assessing morphological changes in the CNS.
Here we show alterations in morphology of neuronal pathways in a well-established
animal model for human absence epilepsy. Control and epileptic rats (28-30
weeks) were studied with DTI ex vivo. We report that the tissues integrity of
the corpus callosum was perturbed in epileptic rats. In conclusion, we have
shown that DTI is sensitive for the detection of white matter changes in the
WAG/Rij rat model of absence epilepsy.
Computer 38
Manisha
Aggarwal1, Jiangyang Zhang1, Susumu Mori1,2
1Johns
Hopkins University School of Medicine, Baltimore, USA; 2Kennedy
Krieger Institute, Baltimore, USA
Mouse
brain atlases have an important role in targeting brain structures during
surgical operations in experimental mice models. A three-dimensional
stereotactic surgical atlas of C57BL/6J mouse brains at six developmental
stages: postnatal day 7 (P7), P14, P21, P28, P63 and P140 was developed, using
high resolution diffusion tensor imaging (DTI) and micro-computed tomography
(MicroCT) of mouse head specimens. DTI generated sharp white and gray matter
contrasts throughout development, even in early postnatal mouse brains prior to
myelination. The atlas can be used to determine the exact stereotactic coordinates
of any location within the brain relative to reference skull landmarks.
Computer 38
Christine
Louise Mac Donald1, Krikor Dikranian1, Sheng-Kwei Song1,
Philip Bayly1, David Holtzman1, David Brody1
1Washington
University, St. Louis, USA
Diffusion
tensor imaging (DTI) was used to characterize the temporal changes occurring
following traumatic brain injury in an experimental mouse model. DTI signal changes correctly predicted the
presence of axonal injury and quite confidently predicted the approximate
timing of injury. The results suggest
that DTI could be used to assess axonal injury clinically if similar signal
changes were found in humans following traumatic brain injury.
13:30
3374.
Effects of Environment Enrichment on Hypoxia-Induced Injury to Corpus
Callosum and Cingulate in C57B/L6 Mice
Computer 39
Halima
Chahboune1,2, Laura Ment1, William Stewart1,
Douglas L. Rothman1, Fahmeed Hyder1, Micheal L. Schwartz1
1Yale
University School of Medicine, New Haven, USA
Using
in vivo DTI we examined morphological effects of environmental enrichment in a
clinically relevant neonatal model of chronic sublethal hypoxia (CSH) injury to
developing brain. Hypoxia (<10%) at the 3rd day (P3) after birth in C57B/L6
mice for a week provides an injury model which replicates neuropathologic
findings that accompany preterm birth. Normal and CSH mice raised in
non-enriched and enriched conditions were studied at P36 and P51. We found that
enriched environment enhances tissue anisotropy within the corpus callosum and
cingulate in both groups and may circumvent the delayed development observed in
CSH mice.
14:00
3375.
Manganese Enhanced MR Neuronal Tract Tracing: A Passive Stochastic
Process
Computer 39
Andrew
Sheridan Lowe1,2, Ian D. Thompson,23, Nicola
R. Sibson2
1University
College London, London, UK; 2Univerisity of Oxford, Oxford, UK; 3Kings
College London, London, UK
Manganese
enhanced MR tract tracing is based upon the intra-cellular uptake,
transportation and subsequent accumulation of manganese at terminal fields. A
recent study suggests neuronal activity is blocked, while axonal transport is
unaffected, by concentrations of Mn required for imaging. There remains
speculation, however, regarding the relative contribution of passive diffusion
as transportation rates were determined over the distal optic nerve. The
current study reports the consequences of pharmacological blockade of retinal
ganglion activity on remote terminal field enhancement. The paradox of blocked
activity and hence expected blocked intra-cellular uptake in the context of
evident terminal field enhancement is discussed.
14:30
3376.
Blood-Flow Magnetic Resonance Imaging of the Retina
Computer 39
Haiying
Cheng1, Yingxia Li1, Timothy Q. Duong1
1Emory
University, Atlanta, Georgia, USA
This
study describes a novel MRI application to image basal blood flow, and
physiologically induced blood-flow changes in the retina. Continuous
arterial-spin-labeling technique with a separate neck coil for spin labeling
was used to image blood flow of the rat retina at 90x90x1500-ƒÝm resolution.
The average blood flow was significantly higher (6.3¡Ó1.0mL/g/min under 1%
isoflurane) than the brain (1 mL/g/min). Breathing oxygen decreased blood flow
25¡Ó6% relative to baseline (air). Breathing 5%CO2 increased blood flow 16¡Ó6%.
Blood-flow MRI has the potential to provide unique insights into retinal
physiology and serve as an early biomarker for retinal diseases.
15:00
3377.
Cerebral Blood Flow MRI in Rats Using Cardiac Spin Labeling Technique
Computer 39
Qiang
Shen1, Eric Muir1, Timothy Q. Duong1
1Emory
Unviersity, Atlanta, Georgia, USA
Continuous
arterial-spin-labeling with a separate neck coil is generally more sensitive
relative to the single-coil technique, but the two coils need to be actively
decoupled to avoid coil-to-coil electromagnetic interaction. Additionally it is
difficult to measure cerebellum blood flow with the two coil technique, because
the neck coil does not label the vertebral arteries which supply the
cerebellum. We introduce a new approach to overcome these limitations by
placing the labeling coil at the heart position. We termed this approach the
<B>Cardiac Spin Labeling (CSL)</B> technique. We applied this
technique to image basal CBF and physiologically evoked CBF changes in normal
rats.
Fetus & Neonate: Normal, Injury, Animal, Human
Hall D Monday 14:00-16:00
14:00
3378.
Multi-Spectral Analysis of Relaxation Time Maps on Fetal Baboon Brains
Computer 38
Feng
Liu1,2, Marianne Garland1, Yunsuo Duan1,2,
Raymond Stark1, Bradley Peterson1,2, Alayar
Kangarlu1,2
1Columbia
University Medical Center, New York, New York, USA; 2New York State
Psychiatric Institute, New York, New York, USA
Relaxation
times T1 and T2 are sensitive to the rapid changes in the brain tissues during
early life. We acquired sequential measurements of T1 and T2 in fetal baboon
brains at different gestational ages. Either T1 maps or T2 maps provided better
contrast than the relaxation time weighted images. Nevertheless, we used both
relaxation time maps to extract multi-spectral maps as a quantitative
diagnostic tool for differentiation of immature brain tissues. Both prenatal
and postnatal data are presented to demonstrate the sensitivity of this tool to
the process of early brain development.
14:30
3379.
Distinguishing Primary from Non-Primary Visual Cortical Areas by DTI at
Early Stages of Brain Development
Computer 38
Christopher
D. Kroenke1, Erin N. Taber1, Andrew K. Knutsen2,
Philip V. Bayly2
1Oregon
Health & Science University, Portland, Oregon, USA; 2Washington
University, St. Louis, Missouri, USA
Water
diffusion anisotropy is observable within the developing cerebral cortex, and
becomes diminished with axonal/dendritic differentiation. Here we report cortical diffusion anisotropy
measurements in post-mortem, early postnatal ferret brain (developmentally
equivalent to mid/late gestational primate brain). Cortical surface modeling procedures are used
to define visual area boundaries in brains from several stages of
development. Differences in cortical
diffusion anisotropy are found between primary and non-primary areas,
reflecting differences in histological structure, or rate of development. These findings buttress prior observations in
non-human primate brain tissue, and suggest that DTI can be of utility for
delineating boundaries between functional cortical areas.
15:00
3380.
DTI Studies with Immunohistological Correlation in the Developing Human Frontal
Cerebrum
Computer 38
Rakesh
Kumar Gupta1, Richa Trivedi1, Nuzhat Husain2,
Sona Saksena2, Savita Srivastava2, Mandakini Pradhan1,
Vinita Das2, Gyanendra K. Malik2, Ram Kishore Singh
Rathore3
1Sanjay
Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India; 2CSMM
University, Lucknow, India; 3Indian Institute of Technology, Kanpur,
India
In-vivo
DTI was performed on freshly aborted 43 human fetuses with GA of 15–37 weeks.
Out of them immunohistochemical analysis by using glial fibrillary acid protein
(GFAP), and neuron specific enolase (NSE) antibodies was performed on 35
fetuses. Regions of interest were placed on cortical, subplate, intermediate
and germinal matrix zone of frontal lobe. Peak value of cortical fractional
anisotropy (FA) and maximal expression of GFAP was observed in frontal cortex
of 26 weeks of GA fetus. Significant correlation was observed between FA in
germinal metrics and number of NSE positive cells present in the germinal
matrix zone. The ability to noninvasively monitor neuronal migration and
maturation processes in vivo should greatly improve our understanding of the
normal developmental pattern of the cerebrum in human fetal brain.
Computer 38
Nuzhat
Husain1, Sona Saksena2, Savita Srivastava1,
Richa Trivedi2, Vinita Das1, Mandakini Pradhan2,
Ankur Purwar3, Gyanendra K. Malik1, Ram Kishore Singh
Rathore3, Rakesh K. Gupta2
1CSMM
University, Lucknow, India; 2Sanjay Gandhi Postgraduate Institute of
Medical Sciences, Lucknow, India; 3Indian Institute of Technology,
Kanpur, India
DTI
was performed on spontaneous abortion or intrauterine dead human fetuses (n=28)
with gestational age (GA) of 20-37 weeks. Fractional anisotropy (FA) values
were quantified on the cerebellar cortical regions. Immuohistochemistry using
Glial Fibrillary Acid Protein (GFAP) antibody was also performed on fetal
brains (n=22) ranging from 20-36 weeks GA. The observed increase in cortical FA
values during the early third trimester appears to be associated with the
radial organization of granule cell neurons seen on GFAP stained Bergmann glial
fibers. This study demonstrates the migrational and maturation changes in the
developing cerebellar cortex using DTI and its confirming on GFAP
immunostaining.
14:00
3382.
Diffusion Tensor Imaging and Immunohistochemical Studies in the
Developing Human Cerebellar Peduncles
Computer 39
Rakesh
K. Gupta1, Sona Saksena1, Nuzhat Husain2,
Savita Srivastava2, Richa Trivedi1, Vinita Das2,
Mandakini Pradhan1, Ankur Purwar3, Gyanendra K. Malik2,
RKS Rathore3
1Sanjay
Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India; 2CSMM
University, Lucknow, India; 3Indian Institute of Technology, Kanpur,
India
Diffusion
tensor imaging was performed on spontaneous abortion or intrauterine dead human
fetuses (n=23) with gestational age (GA) of 20-37 weeks. Fractional anisotropy
(FA) values were quantified on the middle cerebellar peduncles (MCP).
Immunohistochemistry using anti-Myelin Basic Protein (MBP) antibody was also
performed on fetal brains (n=9) ranging from 19-37 weeks GA. The observed
maximum increase in FA values during the late third trimester of gestation,
reaching a plateau by 37 weeks in MCP reflects myelination of axons during late
third trimester which is confirmed by the expression of myelinated fibers in
the MCP at 32 and 37 weeks GA. This study will provide a normative database of
developing cerebellar white matter using DTI and its confirming on MBP
immunostaining.
14:30
3383.
In-Vivo Subplate Development in Preterm Infants
Computer 39
Latha
Srinivasan1, Giuliana Durighel1, Serena J. Counsell1,
Joanna M. Allsop1, Julie A. Fitzpatrick1, A David Edwards1,
Mary A. Rutherford1
1Imperial
College London, London, UK
Subplate
is a transient neuronal layer that relays thalamic inputs to the developing
cortex. In-vivo quantification of 80 T2-weighted MR images between 25 and 45
weeks showed that the subplate was maximal in length at 30 weeks, after which
it decreased in length and was restricted to gyral crests. During development the depth remained
constant whereas there were regional variations in all subplate measurements.
Infants with lesions had decreased depth and T2-intensity. At term age preterm
infants had increased subplate depth, length and T2-intensity compared to term
controls suggesting a combination of delayed and abnormal subplate maturation
in preterm infants.
15:00
3384.
Corroboration of Disorganized Fetal Brain Lamination on Postmortem MR
and DTI with Autopsy Findings
Computer 39
Elysa
Widjaja1, Susan Blaser2, Patrick Shannon3
1Hospital
for Sick Children, Toronto, Canada; 2Hospital for Sick Children,
Canada; 3Mount Sinai Hospital, Canada
Postmortem
MR and diffusion tensor imaging in fetuses with brain malformations
15:30
3385.
Investigation of Neonate Brain Development Enabled by Tract-Oriented
Quantification
Computer 39
Mahnaz
Maddah1, Andrea U.J. Mewes2,3, Heidelise Als2,
Gloria McAnulty2, Eric L. Grimson1, Simon K. Warfield2
1Massachusetts
Institute of Technology, Cambridge, Massachusetts, USA; 2Harvard
Medical School and Children’s Hospital,, Boston, Massachusetts, USA; 3Charite,
Berlin, Germany
Tract-based
quantitative analysis reveals developmental differences that are not identified
by ROI-based methods. Spatial patterns of the tract development are clearly
observed once the parameters of interest are plotted along the tract arc
length. Comparison across different subjects or at different time points are
easily achieved by mapping the corresponding cluster centers. The proposed
approach opens new possibilities for more accurate analysis of neonate brain
development.
14:00
3386.
Thalamic Development in Preterm Infants During the Third Trimester
Computer 40
Latha
Srinivasan1, Ciara Abbott1, Serena J. Counsell1,
Joanna M. Allsop1, Julie A. Fitzpatrick1, Giuliana
Durighel1, A David Edwards1, Mary A. Rutherford1
1Imperial
College London, London, UK
Thalamic
inputs are necessary for proper development of functional cortical columnar units.
Thalamic quantification in 80 T1-weighted images of preterm infants between 25
and 45 weeks postmenstrual age showed that the thalamic growth was 0.49
cm3/week. Further exploration in a group of longitudinally scanned infants
confirmed that both the total thalamic volume and rate of growth were
significantly reduced in the presence of significant cerebral lesions. This combination of absolute volume reduction
and failure of growth may explain the thalamic volume reductions seen in
preterm infants at term equivalent age as shown by a variety of automated and
manual methods.
14:00
3387.
Evidence of Thalamocortical Fibers Maturation in Early Human Brain
Development Assessed by Diffusion Tensor Imaging
Computer 40
Vincent
Denolin1, Xavier De Tiège2, Anne Pardou3,
Danielle Balériaux2, Patrick Van Bogaert2, Thierry Metens2,
Alec Aeby2
1Philips
Medical Systems Benelux, Brussels, Belgium; 2ULB-Hopital Erasme,
Brussels, Belgium; 3ULB-Hopital Erasme, Belgium
Early
studies on brain maturation with DTI were based on measurements in pre-defined
regions of interest (ROI), which introduce a bias due to a priori hypotheses
about the localization of maturational changes. Therefore we used a voxel-based
approach, as known as statistical parametric mapping (SPM), to detect
maturational changes in a population of preterm and term newborns, without
prior prescription of the brain area to be analyzed. We showed that, besides
regions previously identified using ROI analyses, i.e. motor and somatosensory
tracts, optic radiation and corpus callosum, significant maturational changes
also occur in the thalamus between 34 and 41 weeks gestational age.
14:30
3388.
Evolution of MR DTI Changes in Neonatal Rats After Mild Hypoxic-Ischemic
Insult
Computer 40
Jian
Yang1,2, Cheng Wang Jin2, Pek Lan Khong, Gang
Niu2, Ed X. Wu1
1Department
of Electrical and Electronic Engineering,The University of Hong Kong, Hong
Kong, People's Republic of China; 2The First Hospital of Xi'an
Jiaotong University, Xi'an, People's Republic of China
7-day-old
mild hypoxic-ischemic (H-I) rats (n=12) were studied at 7T with diffusion
tensor imaging (DTIs) in day 1, 3 and 7 post H-I insult. The apparent diffusion
coefficient (ADC), fractional anisotropy (FA) and directional diffusivities
(¦Ë// and ¦Ë¡Í) values were measured in white matter (WM) and gray matter (GM)
lesion and their contralateral side. In day 1, the ADC, FA and ¦Ë// values
showed significant decrease in GM lesion whilst WM lesion only showed a
significant ADC increase. In day 3, the significant FA decrease and ¦Ë¡Í
increase were found in GM lesion. Significant ADC and ¦Ë¡Í increase, and FA
decrease were observed in the ipsilateral WM. By day 7, all DTI values were not
statistically different between two hemispheres. Thus DTI is a sensitive method
to study the early transient GM and WM changes after mild H-I insult by probing
the microstructural changes.
15:00
3389.
In-Vivo Magnetization Transfer Brain MRI of Mice with Neonatal
Hypoxia-Ischemia
Computer 40
Ali
Fatemi1,2, Mary Ann Wilson1, Seth A. Smith1,
Susumu Mori1, Jiangyang Zhang1, Michael V. Johnston1,
Michael T. McMahon1
1Johns
Hopkins University, Baltimore, Maryland, USA; 2Massachusetts General
Hospital, Boston, Massachusetts, USA
The
objective of this study was to develop a quantitative in-vivo imaging technique
for assessment of outcome in mice with perinatal brain injury. Magnetization
Transfer (MT) Imaging was performed 3 weeks after a hypoxic-ischemic insult in
CD-1 pups and showed decreased MTR values on the affected hemisphere as well as
in the corpus callosum and internal capsule. This work is the first
demonstration that perinatal brain injuries can be highlighted in mice using MT
imaging.
Computer 41
Bindu
Setty1,2, Eva-Maria Ratai,23, Pallavi Sagar1,2,
Kalpathy S. Krishnamoorthy,2, Ellen P. Grant1,2
1Massachusetts
General Hospital, Boston, Massachusetts, USA; 2Harvard Medical
School, Boston, Massachusetts, USA; 3Massachusetts General Hospital
– A.A. Martinos Center for Biomedical Imaging , Charlestown, Massachusetts,
Eight
neonates with hypoxic ischemic injury were studied within the first 3
days of life using MRI (DWI and MRS). The objective of this study was to determine,
if the absence of lactate predicts good outcomes in patients with
decreased diffusion in the ventrolateral thalamus. No significant correlation
was obtained between lactate ratios in the deep gray nuclei
and
ADC values. In addition, there was no correlation between presence or absence
of lactate and the extent of DWI abnormality or clinical outcome. In
fact, 2 patients that died showed no lactate on MRS but had extensive areas
of decreased diffusion.
14:30
3391.
Temporal Changes in Brain Water Diffusivity in Neonatal Meningitis
Computer 41
Rakesh
K. Gupta1, Richa Trivedi1, Gyanendra K. Malik2,
Abhishek Yadav1, Kashi N. Prasad1, Ram KS Rathore3
1Sanjay
Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India; 2Chhatrapati
Shahuji Maharaj Medical University, Lucknow, India; 3Indian
Institute of Technology, Kanpur, India
Periventricular
white matter of neonatal brain is known to be vulnerable to oxidative and
hypoxic/ischemic injury secondary to neuro-infections. Serial diffusion weighted
imaging (DWI) was performed on 45 neonates with meningitis at two time points:
1) at the time of diagnosis; and 2) after 3 weeks of antibiotic treatment.
T1/T2 images showed abnormalities along with pseudo-normalization of apparent
diffusion coefficient (ADC) values on 2nd study in those patients (29 %), who
showed low ADC values with normal T2/T1 imaging on first MRI. The temporal
variation of ADC in this study is suggestive of hypoxia related injury and may
be responsible for long term neurological sequel in these neonates.
Hall D Monday 14:00-16:00
14:00
3392.
Axial Diffusivity in Acute and Isolated Optic Neuritis
Computer 42
Robert
T. Naismith1, Junqian Xu1, Abraham Snyder1,
Tammie Benzinger1, Joshua Shimony1, Kathryn Trinkaus1,
Anne H. Cross1, Sheng-Kwei Song1
1Washington
University, Saint Louis, Missouri, USA
Diffusion
Tensor Imaging was performed in 4 cases of acute and isolated optic neuritis in
order to determine whether axial and radial diffusivity have a prognostic role
for long-term clinical function.
14:30
3393.
Quantification of Glaucomatous Optic Atrophy Utilizing High Resolution
MRI of the Optic Nerve
Computer 42
Thorsten
Alexander Bley1,2, Mathias Weigel1, Miriam
Gaggl1, Robin Munk1, Mathias Langer1, Juergen
Hennig1, Wolf Lagreze1
1University
of Freiburg, Freiburg, Germany; 2University of Wisconsin - Madison,
Madison, USA
With
age and under certain pathologic conditions, such as glaucoma the diameter of
the optic nerve may change. This study evaluates MRI of the retro-bulbar optic
nerve as a surrogate marker for axonal atrophy in glaucoma. 3 Tesla MRI of the optic nerve in 47 patients
with glaucoma depicted the optic nerve and its sheath within the full
intra-orbital track with high contrast in 1.5 sec. acquisition time per slice.
Imaging findings correlated well with severity of glaucoma.
Computer 42
Yuan
Feng Man1, Chang Zhen Wang1, Hong Yong Jiao1
1Tongren
hospital, Beijing, People's Republic of China
In
this article we use MRI to explore the abnormalities of oculomotor and
extraocular muscles of congenital complex strabismus, the objective is to study
the relationship of the nerve and the corresponding extraocular muscles.
Hall D Monday 14:00-16:00
14:00
3395.
Rapid and High-Resolved Diffusion Tensor Imaging of Mouse Lower Brain
and Cervical Spinal Cord
Computer 43
Virginie
Callot1, Guillaume Duhamel1, Patrick J. Cozzone1
1Centre
de Résonance Magnétique Biologique et Médicale (CRMBM) - UMR CNRS 6612,
Marseille, France
Mouse
diffusion tensor imaging (DTI) of both the brain and the spinal cord (SC) may
reveal useful information on tissue damage consequent to inflammatory or
degenerative diseases. To rapidly assess the geometrical and functional extent
of the pathology, a large coverage and sufficient spatial resolution are
required. Standard pulse sequences preclude obtaining such requirements as they
do not allow achieving sufficient signal to noise ratio within an acceptable
scan-time. In this work, we propose to use a spin-echo (SE) EPI DTI-based
sequence to obtain spatially resolved (140x140x750 µm3) DTI metrics of both the
cerebellum and cervical SC within 30 minutes.
14:30
3396.
Accurate Measurement of Mean Axon Diameter with Q-Space Diffusion MRI
Computer 43
Henry
H. Ong1, Felix W. Wehrli1
1University
of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA
Accurate
knowledge of white matter (WM) mean axon diameter (MAD) would provide important
insight into injury and pathology. The signal decay at low q-values in q-space
diffusion experiments, which can be modeled as a Gaussian, contains MAD
information, but is complicated by signal from both extra- and intra-cellular
spaces. Here, we use one- and two-compartment models to fit decay curves from
WM tracts in healthy mouse spinal cords to measure MAD. The measured MAD values
showed excellent agreement with histology, particularly the two-compartment
fit, and demonstrate the feasibility of this method to extract accurate WM
architecture information non-destructively.
15:00
3397.
Mouse Lumbar Spinal Cord Blood Flow (SCBF) Measurements by Arterial Spin
Labeling
Computer 43
Guillaume
Duhamel1, Virginie Callot1, Frank Kober1,
Patrick J. Cozzone1
1CRMBM,
CNRS 6612, Faculté de Médecine, Université de la Méditerranée, Marseille,
France
A
recent work has demonstrated the feasibility of mouse SC blood flow (SCBF)
measurement with arterial spin labeling (ASL) using a flow-sensitive
alternating inversion recovery (FAIR) technique at the cervical level. Accurate
measurements of SCBF within structures of the cord were obtained. The
assessment of perfusion measurements at lower levels of the cord (thoracic, lumbar)
is of interest for numerous disease models (contusion, ischemia…) but
challenged by amplified bulk motion. The present study demonstrates the
feasibility of lumbar SCBF measurement by ASL. Good image quality was achieved
and cervical and lumbar SCBF values were compared.
15:30
3398.
A Diffusion and Perfusion EPI-Based MR Protocol for the Characterization
of Rodent Spinal Cord Diseases
Computer 43
Virginie
Callot1, Guillaume Duhamel1, Frank Kober1,
Patrick J. Cozzone1
1Centre
de Résonance Magnétique Biologique et Médicale (CRMBM) - UMR CNRS 6612, Faculté
de Médecine, Marseille, France
The
purpose of this work was to give an experimental MR procedure that may be used
to better characterize spinal cord (SC) diseases in which structural tissue
damage and deficient blood supply are involved. Such a characterization has not
been possible so far.
In
this study, we propose to perform spin-echo EPI Diffusion Tensor Imaging and
quantitative perfusion imaging of the mouse spinal cord, with an in-plane
resolution of 133x133 µm² and a total scan-time of 90 minutes. Imaging was
performed at the cervical level, where spinal cord injury, infarction or tumor
may easily occur.
Computer 44
Virginie
Callot1, Guillaume Duhamel1, Yann Le Fur1,
Patrick J. Cozzone1
1Centre
de Résonance Magnétique Biologique et Médicale (CRMBM) - UMR CNRS 6612, Faculté
de Médecine, Marseille, France
Spin-Echo
(SE) EPI technique applied to mouse cervical spinal cord (SC) imaging has
recently been proposed and a 3 to 4 acquisition-time gain-factor was
demonstrated as compared to conventional SE. However, the application of SE-EPI
technique at lower SC levels is challenged by higher field heterogeneities and
motion amplitude.
The
present study demonstrates that high-quality mouse SC DTI-EPI acquisitions at
the thoracic and lumbar levels can be performed. A complete dataset (2
b-values, 6 encoding directions, 9 slices, 172x172x750 µm3 resolution) can be
collected within 15 minutes. Moreover, by using outer volume suppression, 86x86
µm² in-plane resolutions can be achieved.
Computer 44
Jiangyang
Zhang1, Melina V. Jones1, Cynthia A. Deboy1,
Daniel S. Reich1, Paul N.
Hoffman1, Kazim A. Sheikh1, Susumu Mori1,
Peter A.Calabresi1
1Johns
Hopkins University School of Medicine, Baltimore, Maryland, USA
Diffusion
tensor images of postmortem rat spinal cords were acquired after dorsal root
axotomy. We found significant decreases in FA and axial diffusivity and a
significant increase in radial diffusivity during the first three days after
axotomy, followed by slow increase in radial diffusivity. MR findings were
correlated with pathology in the dorsal column from 18 hours to 30 days after
surgery. Our results suggest that axial diffusivity can detect very early
axonal degeneration at different locations in the spinal cord, and the changes
in radial diffusivity were complex and remain to be investigated.
15:00
3401.
High Field MRI of the Acute Stage of Cervical SCI in the Rat
Computer 44
Georgeta
Mihai1,2, Yvette S. Nout2, Petra Schmalbrock3,
Jacqueline C. Bresnahan2, Michael S. Beattie2
1The
Ohio State University , Columbus, USA; 2University of California San
Francisco, San Francisco, USA; 3The Ohio State University, Columbus,
USA
In
this study the characterization of acute stage of cervical SCI was archived by
imaging at 4.7T using a 3cm home built coil. Evolution of edema, cord swelling
and hemorrhage was compared between two injured groups of animals, one that
received 5% NaCl IV solution infusion and another one that received 0.9%NaCl
starting at 30 minutes post injury. The eight hours continuous MRI scans
revealed differences in term of injury evolution between the two groups. This
work suggests that MRI can be used to monitor therapies and treatments aimed at
reducing secondary injury in SCI.
15:30
3402.
Biplanar Spinal Cord MRI in MS - Depiction of Cord Pathology and
Improvement of Clinical Correlations
Computer 44
Katrin
Weier1, Alain Thoeni1, Yvonne Naegelin1,
Michael Amann1, Jochen G. Hirsch1, Hüseyin Duyar1,
Ludwig Kappos1, Wolfgang Steinbrich1, Ernst-Wilhelm Radue1,
Achim Gass1
1University
Hospital Basel, Basel, Switzerland
Biplanar
spinal cord (SC) MRI of MS patients was performed in 202 MS patients. PD and
T2-w in axial and sagittal planes were employed to visualize SC pathology
thoroughly. Typical pathological findings (focal lesions, extent of diffuse
hyperintensity, focal or generalised atrophy) were used to categorize patients
according to a new SC abnormality score. The resulting scores showed a good
correlation to clinical functional scores (EDSS and FS). Improved visualization
of the pathological changes and their integration in a new abnormality score
may overcome difficulties in detection and aid interpretation of SC MRI
results.
14:00
3403.
BOLD Signal Responses to Controlled Hypercapnia in Human Spinal Cord
Computer 45
Julien
Cohen-Adad1,2, Claudine Gauthier1,3,
Habib Benali2, Serge Rossignol1, Richard D. Hoge3
1GRSNC,
Department of Physiology, Faculty of Medicine, Université de Montréal,
Montreal, Canada; 2INSERM U678, Université Pierre et Marie Curie (Paris
VI), CHU Pitié-Salpêtrière, Paris, France; 3Unité de Neuroimagerie
Fonc
We
used hypercapnia to characterize the sensitivity of the BOLD method in human
spinal cord. Results suggest that BOLD signal is highly dependent on the gross
vascular anatomy of the spinal cord, limiting the sensitivity in grey matter
since the precise vascular organization may vary among subjects. Moreover,
negative BOLD was measured in various regions of the cord, possibly caused by
blood-stealing, or by an increased in signal dephasing due to higher flow
velocities in large vessels. The framework developed in this study will help
future investigations aiming at characterizing the BOLD signal in the spinal
cord.
14:30
3404.
Magnetization Transfer (MT) Asymmetry in Human Cervical Spinal Cord
Computer 45
Man
Cheuk Ng1, Jun Hua2, Peter C.M. van Zijl2,
Edward S. Yang3, Hong Hu1, Keith D. Luk1,
Edmund Y. Lam1
1The
University of Hong Kong, Pokfulam, Hong Kong, People's Republic of China; 2Johns
Hopkins University, Baltimore, USA; 3ASTRI, Science Park, Shatin,
Hong Kong, People's Republic of China
Magnetization
transfer (MT) effect includes the interaction between bulk water and semisolid
macromolecules (conventional MT) and the chemical exchange dependent saturation
transfer (CEST) effect. MT asymmetry was investigated in human cervical spinal
cord at 3T using low power off-resonance RF irradiation at different frequency
offsets. Our results showed that the z-spectrum in gray/white matter was
asymmetrical about the water resonance frequency (P < 0.001) with more
saturation effect at lower frequencies (negative frequency offset) far from
water and at higher frequencies (positive offset) close to water, which were
attributed mainly to the conventional MT and CEST effects respectively.
15:00
3405.
Computational MR Image Analysis for Spinal Cord Injury Studies
Computer 45
Kevin
Ming1,2, Rafeef Abugharbieh1, Claire F. Jones1,
Andrew Yung1, Piotr Kozlowski1, Wolfram Tetzlaff1,
Peter A. Cripton1
1University
of British Columbia, Vancouver, Canada
We
present a novel computational image analysis approach for studying and
quantifying spinal cord deformations, in vivo, using magnetic resonance imaging
(MRI) of rats. High field (7 Tesla) MR data is acquired and used to compute
deformations of the spinal cord in its original physiological environment, for
the first time, as opposed to traditional approaches in which animal spinal
cords are exposed and directly subjected to a mechanical force. Image data was
captured before and after a non-injurious deformation representative of that
associated with mild myelopathy (pressure on the cord in vivo) were acquired
using gradient echo FLASH T1-weighted sequence The proposed approach provides a
new framework through which the causes, mechanisms, and tolerance parameters of
sustained compression of the spinal cord and myelopathy, as well as the
measures used in the study of neuroprotection and regeneration of spinal cord
tissue, can be prospectively derived in a manner that ensures the bio-fidelity
of the cord.
15:30
3406.
Hyper-Acute Evaluation of Spared White Matter in Mouse Model Contusion
SCI Using in Vivo DTI
Computer 45
Joong
Kim1, Hsiao-Fang Liang1, Sheng-Kwei Song1
1Washington
University School of Medicine, St. Louis, Missouri, USA
Graded
contusion spinal cord injury was generated at T9 spinal cord level of C57 BL/6
female mice. In vivo DTI was performed immediately after injury at 4.7 T magnet
followed by perfusion and fixation. Spared white matter of contusion injured
spinal cord was determined non-invasively using axial diffusivity
thresholding. The neurofilament and
myelin basic protein positive immunohistochemistry staining was performed along
with the standard silver staining to assess spared white matter content. In
vivo DTI determined spared white matter showed good agreement with that
determined using histology.
Spectroscopy & Computer Aided Analysis of Brain Tumor
Hall D Monday 14:00-16:00
14:00
3407.
Amide Proton Transfer (APT) Imaging of Human Brain Tumors with B0
Inhomogeneity Correction
Computer 46
Jinyuan
Zhou1,2, Jaishri Blakeley1, Jun Hua1,2,
Mina Kim1,2, Peter C.M. van Zijl1,2
1Johns
Hopkins University, Baltimore, Maryland, USA; 2KKI, Baltimore,
Maryland, USA
B0
field inhomogeneity is known to interfere with the quality of CEST-type experiments,
especially for in vivo studies, such as APT imaging, because of the need to
assign the water frequency in each voxel. We show here a practical six-offset
scheme to acquire human brain high-SNR APT images. Eight brain tumor patients
were scanned at 3T. Combined with the water frequency offset map fitted from an
additional CEST spectrum, the method is able to correct for the artifacts
produced by B0 field inhomogeneity on standard two-offset APT images.
14:30
3408.
Spatial Correlation of Metabolic Abnormalities in Non-Enhancing Low
Grade Gliomas
Computer 46
Qian
Zhao1,2, Ying Lu1, Jiqian Fang3,
Wei Bian1, Sarah Nelson1, Tracy McKnight1
1University
of California, San Francisco, San Francisco, USA; 2School of Public
Health,Sun Yat-Sen University, Guangzhou, People's Republic of China; 3School
of Public Health,,Sun Yat-Sen University, Guangzhou, People's Re
Grade
2 astrocytoma (AS) and oligodendroglioma (OD) are often similarly non-enhancing
(NE) and appear although they convey different prognoses for patients. Studies have shown a link between the spatial
correlation of MRS features and growth activity in NE glioma. We compared the magnitude and correlation of
MRS metabolites in NE AS and OD. Compared with OD, the relative level of Cho
and Cre throughout AS lesions was more uniform. AS were further distinguished
by a negative correlation between NAA and Lac. The differences in the metabolic
characteristics within the two tumor types may reflect differences in the
underlying biology.
15:00
3409.
1H-MR Spectroscopy as a Cancer Biomarker for Anti-Angiogenic Treatment
in Glioblastoma
Computer 46
Hei-Soog
Kim1,2, Ciprian Catana2, Eva-Maria Ratai2,
Wei-Ting Zhang2, Priscilla Yeo2, Meiyun Wang2,
Tracy T. Batchelor2, Rakesh K. Jain2, A. Gregory Sorensen2
1Massachusetts
Institute of Technology, Cambridge, Massachusetts, USA; 2Massachusetts
General Hospital, Boston, Massachusetts, USA
A
previous study demonstrated the efficiency of anti-angiogenic clinical trial in
glioblastoma using MRI markers. However, there is a concern that the changes
observed using standard MR imaging do not directly relate to tumor cell
changes. This study suggests that 1H-MRS is a very promising method for
assessing the response of cancerous tissue to anti-angiogenic treatment.
NAA/Cho
measured in twenty patients showed no significant changes until day 28 and
decreased afterwards. The tumor is likely to regress after 28-56 days, later
than the enhancement changes on the MRI. One probable interpretation of our
data is that signs of tumor regression appear later, while those of tumor
recurrence earlier on 1H-MRS than on MRI.
15:30
3410.
Incorporation of MR Spectroscopic Imaging Into Gamma Knife Radiosurgery
Treatment Planning
Computer 46
Lawrence
Ryner1,2, Muoi Tran3, Herve Momo-Jeufack2,
Jordan Hovdebo1,2, Michael West2
1National
Research Council Institute for Biodiagnostics, Winnipeg, Canada; 2University
of Manitoba, Winnipeg, Canada; 3CancerCare Manitoba, Winnipeg,
Canada
Infiltrative
gliomas are known to extend beyond the area of contrast enhancement, however,
conventional Gamma Knife treatment planning typically involves identifying the
radiosurgical target volume on contrast-enhanced MR images. MRSI can show early
biochemical changes useful for targeting abnormal tissue. We present further
work in this area confirming the utility of MRSI in Gamma Knife treatment
planning, as observed in patient MRSI datasets showing spectra from MRSI voxels
(outside of and adjacent to the planned radiosurgical target volume) with
abnormal choline to NAA ratios as compared to age and brain region-matched
controls.
14:00
3411.
Interobserver Agreement for Cerebral Glioma Volumetrics on Conventional
MR Imaging
Computer 47
Gerard
Thompson1, John Robert Cain1, Alan Jackson1,
Samantha Jane Mills1
1University
of Manchester, Manchester, UK
A
number of groups have demonstrated that measurements of glioma tumour volume on
magnetic resonance imaging are more valuable prognostically than simple uni- or
bi-dimensional measurements such as the RECIST criteria. Such volumetric
parameters will only be of value in clinical use and in further investigation
if they can be reliably reproduced by different observers. This work tests for,
and demonstrates, good interobserver agreement between non-expert radiologists
employing a simple volumetric method on conventional MR imaging of cerebral
gliomas.
14:30
3412.
Towards Predicting Neoplastic Recurrence with Multi-Parametric MR
Computer 47
Evangelia
I. Zacharaki1, Ragini Verma2, Sanjeev Chawla2,
Elias R. Melhem2, Ronald Wolf2, Christos Davatzikos2
1University
of Pennsylvania , Philadelphia, Pennsylvania, USA; 2University of
Pennsylvania, Philadelphia, Pennsylvania, USA
Treatment
of brain neoplasms can greatly benefit from knowledge of the extent and degree
of neoplastic infiltration. The current work is a preliminary study aiming to
illustrate how the integration of multiple MRI parameters via sophisticated
nonlinear pattern classification methods could be applied to predict possible
neoplastic progression. The method incorporates high dimensional intensity
features created from multiple MRI acquisition protocols (structural MRI as
well as DTI) into a pattern classification framework, to obtain a voxel-wise
probabilistic spatial map that reflects the likelihood of a region presenting neoplastic
recurrence after resection.
Computer 47
John
Robert Cain1, Gerard Thompson1, Alan Jackson1,
Samantha Jane Mills1
1University
of Manchester, Manchester, UK
Tumour
border sharpness coefficient (TBSC), and subjective qualitative measures such
as descriptors of tumour margins have both shown potential in differentiating
between histological and genetic subtypes of glioma. These techniques will only
be of use if they are reproducible in the clinical setting. This study
demonstrates excellent interobserver reproducibility for TBSC in tumours of all
grades. Only T2W border descriptors showed poor reproducibility in grade IV
tumours and may reflect the location of these tumour and their oedema to
cortical and ventricular margins. All other qualitative descriptors
demonstrated acceptable reproducibility in both low and high grade tumours.
15:30
3414.
Computer-Aided Detection of Metastatic Brain Tumors Using Automated 3-D
Template Matching
Computer 47
Robert
Ambrosini1, Peng Wang2, Balasubramanya Kolar1,
Walter O'Dell1
1University
of Rochester, Rochester, USA; 2University of Michigan, Ann Arbor,
USA
Although
screening for brain metastases using MR imaging is a critical component of the
medical care for cancer patients, reading these images is both a time-consuming
and subjective process when performed without computer assistance. We have developed an automated detection
algorithm for brain metastases that is based upon 3-D template matching. The data collected on 22 patient datasets
(1320 coronal MR slices) containing 161 total brain metastases show that we can
achieve currently a sensitivity of 87.6% with a false positive rate of 0.58 per
image slice. These results demonstrate
our algorithm’s value as a clinical assist tool for radiologists.
14:00
3415.
Nosologic Imaging of the Brain: Combined MRI and MRSI Segmentation and
Classification
Computer 48
Jan
Luts1, Albert J. Idema2, Arend Heerschap2,
Dirk Vandermeulen1, Johan Suykens3, Sabine Van Huffel3
1Katholieke
Universiteit Leuven, Leuven , Belgium; 2University of Nijmegen,
University Medical Center, Nijmegen, Netherlands; 3Katholieke
Universiteit Leuven, Leuven, Belgium
In
this study advanced methods from image processing and pattern recognition are
applied to segment and classify brain tumours, thereby including spatial
information. Both MRSI and MRI data are combined to produce higher resolution
nosologic images. Furthermore, class probabilities are calculated for the
segmented tumour region. A leave-one-patient-out evaluation procedure is
performed. The segmentation and classification scheme are applied to patients
with grade II glioma tumours, grade III glioma tumours, glioblastomas or
meningiomas. The proposed method offers a new way to produce nosologic images,
representing tumour heterogeneity and class probabilities, which may help
clinicians in decision making.