ISMRM 21st Annual Meeting & Exhibition 20-26 April 2013 Salt Lake City, Utah, USA

Perfusion & Permeability Measured with Contrast Agents
Thursday 25 April 2013
Room 151 AG  10:30 - 12:30 Moderators: Christopher Chad Quarles, Steven P. Sourbron

10:30 0582.   A Radial 3D GRE-EPI Pulse Sequence with kz Blip Encoding for Whole-Brain Isotropic 3D Perfusion Using DSC-MRI Bolus Tracking with Sliding Window Reconstruction (3D RAZIR)
Sumeeth Vijay Jonathan1, Parmede Vakil1, Yong Jeong1, Sameer A. Ansari2, Michael Hurley2, Bernard R. Bendok3, and Timothy J. Carroll1,2
1Biomedical Engineering, Northwestern University, Chicago, IL, United States, 2Radiology, Northwestern University, Chicago, IL, United States, 3Neurological Surgery, Northwestern University, Chicago, IL, United States

3D RAZIR obtains 76-slice whole-brain relative perfusion measurements with DSC-MRI bolus tracking at 1.7 mm3 isotropic voxel resolution. Consecutive 3D volumes are acquired in 10.3 s, but we recover dynamic bolus information at 160 ms per frame using sliding window reconstruction. 3D RAZIR uses in-plane radial sampling and through-plane Cartesian sampling with 3D GRE-EPI readouts to allow sliding window reconstruction, which permits increased volume coverage without sacrificing SNR for bolus tracking. Radial view-dependent N/2 ghosting artifacts are corrected using inline phase correction scans. 3D RAZIR obtains whole-brain perfusion maps of rCBF, rCBV, and MTT with good reference standard agreement.

10:42 0583.   High Resolution Dynamic Susceptibility Contrast Perfusion Imaging Using Multi-Echo Spirals and Temporal Compressed Sensing-Based Reconstruction
Julia V. Velikina1, Youngkyoo Jung2, Aaron A. Field3, and Alexey A. Samsonov3
1Medical Physics, University of Wisconsin - Madison, Madison, WI, United States, 2Radiology, Wake Forest University, Winston-Salem, NC, United States, 3Radiology, University of Wisconsin - Madison, Madison, WI, United States

We propose a novel approach to dynamic contrast susceptibility perfusion-weighted imaging using a combination of multi-echo spiral acquisition and compressed sensing-type regularized reconstruction based on 2nd derivative in temporal dimension to overcome limitations of standard EPI-based perfusion imaging technique. The proposed technique allows for a significant increase of in-plane spatial resolution and improved whole-head coverage (voxel size 1.375x1.375x5 mm) without compromising temporal resolution (1.35 ms). Availability of four different echo times allows for unbiased R2* mapping and more accurate/artifact-free estimation of relevant haemodynamic perfusion parameters (cerebral blood volume/flow and mean transit time).

10:54 0584.   Slice Accelerated EPI for Dynamic-Susceptibility Contrast Enhanced (DSC) MRI
Dingxin Wang1,2, Charles G. Cantrell3, Bruce S. Spottiswoode4, Vibhas Deshpande5, Timothy J. Carroll3, and Keith A. Heberlein6
1Siemens Medical Solution USA, Inc., Minneapolis, MN, United States, 2Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States, 3Department of Radiology, Northwestern University, Chicago, IL, United States, 4Siemens Medical Solutions USA, Inc., Chicago, IL, United States, 5Siemens Medical Solutions USA, Inc., Austin, TX, United States,6Siemens Healthcare USA, Charlestown, MA, United States

Our study demonstrates the feasibility of using slice accelerated EPI for DSC-MRI measurement and shows the evidence of association between sampling TR and perfusion parameters. The MTT and Tmax maps with faster TR sampling (509 ms) of perfusion data provide more image contrast than slower sampling rate (1527 ms). The difference in MTT also contributes to the variation of CBF spatial pattern. Faster data acquisition should reduce discretization errors in the perfusion measurement, especially for Tmax, as the measured Tmax is rounded-off to a value multiple of TR.

11:06 0585.   
Optimized Processing of Quantitative Susceptibility Mapping-Based Gadolinium Perfusion MRI: Correction of Bulk Susceptibility Effects and Comparison of Arterial Input Function Selection from Capital Greek DeltaR2* and QSM Data
David Bonekamp1, Xu Li1,2, Richard Leigh3, Peter C.M. van Zijl1,2, and Peter B. Barker1,2
1The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Baltimore, Maryland, United States, 2FM Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, Maryland, United States, 3Neurology, Johns Hopkins University, Baltimore, Maryland, United States

Recently, we developed the feasibility of dynamic Quantitative Susceptibility Mapping (QSM) for mapping of cerebral blood flow (CBF). The induced susceptibility effects are independent of intra- or extra-vascular contrast location, an advantage over existing methods. We present improvements of the method and compare the use of arterial input functions (AIF) derived from the QSM (Capital Greek Deltalower case Greek chi) and Capital Greek DeltaR2* data. We find good quantitative agreement between CBF perfusion images obtained using Capital Greek DeltaR2*, and Capital Greek Deltalower case Greek chi with both AIF approaches. Quantification of CBV and CBF is improved in the gray matter compared to our prior reports.

11:18 0586.   The Choice of Arterial Input Function in Perfusion-Weighted MR Imaging - Size Matters in Terms of Saturation-Like Effects
Peter Brunecker1, Ann-Christin Ostwaldt1, Ivana Galinovic1, Jochen B. Fiebach1, and Martin Ebinger1,2
1Center for Stroke Research Berlin (CSB), Charité - Universitätsmedizin Berlin, Berlin, -, Germany, 2Klinik und Poliklinik Hochschulambulanz für Neurologie, Charité - Universitätsmedizin Berlin, Berlin, -, Germany

DSC-MRI is the most widely used method for perfusion imaging (PI) in brain, especially for stroke patients. However, the relationship between tracer concentration and signal change is only strictly valid for a network of capillaries and not in case of estimating the arterial input function (AIF). To investigate that impact, an analysis of 211 PI measurements regarding peak signal drop and relative concentration was performed. We found a descending linearity between the micro- and macrovascular signal with an increasing size of the AIF-defining artery. We conclude that in case of standard imaging techniques more distal arteries are preferable.

11:30 0587.   
in vivo Assessment of the Tissue Cellularity Index (TCI) in Brain Tumors Using Multi-Echo DSC-MRI
Natenael B. Semmineh1, Jack T. Skinner1, and Christopher C. Quarles1
1Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States

Multi-echo DSC-MRI data in brain tumors contains a wealth of information since the T1 and T2* effects can be separated and quantified. Recently, we proposed that T1 and T2* leakage effects can be leveraged to derive a cellularity metric, which we termed the tissue cellularity index (TCI). The goal of this study was to evaluate and characterize the TCI using simulations, brain tumor animal models and in a preliminary clinical study. Simulations and the in vivo studies revealed that TCI increases in tumors with higher cellular density.

11:42 0588.   
Combined Spin and Gradient Echo Imaging Following Injection of USPIOs in Humans
Thomas Christen1, Deqiang Qiu1, Wendy W. Ni1, Heiko Schmiedeskamp1, Roland Bammer1, Michael E. Moseley1, and Greg Zaharchuk1
1Radiology, Stanford University, Stanford, California, United States

In the present study, we analysed the spin- and gradient echo contrast variations following injection of Ferumoxytol (an FDA-approved ultra-small paramagnetic iron oxide [USPIO] compound) in the human brain. The high magnetic susceptibility of ferumoxytol allow the acquisition of high quality spin-echo perfusion maps as well as parametric maps describing the vessel diameter and vessel density of the microvasculature. The long half-life of the contrast agent permits acquisitions at high spatial resolution.

11:54 0589.   Determination of the Vascular Input Function Using Magnitude or Phase-Based MRI: Influence on Dynamic Contrast-Enhanced MRI Model Parameters in Carotid Plaques
Raf van Hoof1,2, Martine Truijman1,3, Evelien Hermeling1,2, Robert J. van Oostenbrugge2,4, R.J. van der Geest5, A.H. Schreuder6, A.G.G.C. Korten7, N.P. van Orshoven8, Be Meens9, M.J.A.P. Daemen2,10, Joachim E. Wildberger1,2, Walter H. Backes1, and M.E. Kooi1,2
1Radiology, Maastricht University Medical Center, Maastricht, Netherlands, 2Cardiovascular Research Institute Maastricht (CARIM), Maastricht University, Maastricht, Netherlands, 3Clinical Neurophysiology, Maastricht University Medical Center, Maastricht, Netherlands, 4Neurology, Maastricht University Medical Center, Maastricht, Netherlands, 5Radiology, Leiden University Medical Center, Leiden, Netherlands, 6Neurology, Atrium Medical Centre, Heerlen, Netherlands, 7Neurology, Laurentius Medical Centre, Roermond, Netherlands, 8Neurology, Orbis Medical Centre, Sittard, Netherlands, 9Neurology, VieCuri Medical Centre, Venlo, Netherlands, 10Pathology, Academic Medical Centre, Amsterdam, Netherlands

A reliable vascular input function (VIF) is important for quantitative analysis of atherosclerotic carotid plaque microvasculature using dynamic contrast-enhanced (DCE) MRI. The purpose is 1) to compare magnitude-based VIF and phase-based VIF and 2) to investigate the influence of different VIFs on DCE MRI model parameters in carotid plaques. It is shown that magnitude-based VIF is strongly influenced by flow artefacts, leading to an underestimation of the peak Gadolinium concentration. Therefore, a phase-based VIF should be used for quantitative DCE MRI analysis.

12:06 0590.   Investigation of the Necessity of Pre-Contrast T1-Determination in DCE MRI; Simulations and Clinical Data
Christopher Larsson1, Magne Kleppestø1, and Atle Bjørnerud1,2
1The Intervention Centre, Oslo University Hospital, Rikshospitalet, Oslo, Oslo, Norway, 2Department of Physics, University of Oslo, Oslo, Oslo, Norway

Dynamic contrast-enhanced MRI is an established method to assess blood-brain barrier integrity and brain hemodynamics. Standard tracer kinetic models are established for estimation of kinetic parameters. Quantification requires an accurate estimation of the CA induced change in T1 relaxation rate in tissue and blood which in turn require knowledge of baseline T1 values (T1,0) . The need for T1,0-data results in additional scan-time, and raises challenges related to image co-registration and additional image processing steps. The added value of using T1,0 maps in DCE analysis has thus been questioned. The purpose of this study was to compare Ktrans values in primary brain tumors obtained using fixed T1,0-values compared to using calculated pixel-wise T1-values through simulations and clinical data. A secondary aim was to assess the variation in baseline T1-values observed in brain tumors.

12:18 0591.   Free-Breathing Vs. Breath-Hold Pulmonary Perfusion MRI: Quantification and Reproducibility -permission withheld
Michael Ingrisch1, Daniel Maxien1, Felix Schwab1, Maximilian F. Reiser1, Konstantin Nikolaou1, and Olaf Dietrich1
1Institute for Clinical Radiology, Ludwig-Maximilians-University Hospital, Munich, Germany

Pulmonary perfusion can be assessed with a DCE MRI measurement which is usually performed during breath hold. However, pulmonary perfusion depends strongly on the phase of the breathing cycle during which breath hold is performed. This leads to a poor reproducibility of quantitative estimates of pulmonary perfusion. Since it was recently demonstrated that pulmonary perfusion can also be assessed from a measurement during free breathing, this volunteer study therefore investigates whether an acquisition during free breathing, which also has a better patient compliance, leads to parameter estimates with a better reproducibility than an acquisition during inspiratory breath hold.