Kelly C McPhee¹ and Alan H Wilman^2
1Physics, University of Alberta, Edmonton, Alberta, Canada, ²Biomedical Engineering, University of Alberta, Edmonton, Alberta, Canada

Proton density and T₂-weighted fast spin echo images are frequently acquired in clinical MRI exams. T₂ quantification is commonly performed in the clinic via applying an exponential fit to these two images, despite recent evidence that an exponential fit is insufficient to correctly quantify T₂. We present T₂ quantification by implementing stimulated echo compensated T₂ fitting using only a proton density-weighted and a T₂-weighted fast spin echo image. This method is demonstrated in the human brain in comparison to a standard quantitative multi-echo spin echo approach.

Richard Stephen Nicholas^1,2, Stephen J Gandy^1,2, and Richard A Lerski^1,2
1Medical Physics, NHS Tayside, Dundee, Scotland, United Kingdom, ²University of Dundee, Dundee, Scotland, United Kingdom

This study compares a single and dual acquisition method for simultaneously measuring T1 and T2 in phantoms using an IR SSFP sequence. Measurements were made over a combination of RR intervals and flip angles, and compared with calibrated reference values. For the single acquisition method, significant deviations in both T1 and T2 were observed, particularly at shorter RR intervals with longer T1 phantoms. The dual acquisition method was demonstrated to be superior for measuring T1 but over-estimated T2. The increased accuracy may offset the requirement for a second acquisition, but this would need further evaluation in vivo.

Xialing Ulrich¹ and Dmitriy A Yablonskiy^1
1Mallinckrodt Institute of Radiology, Washington University in St. Louis, St. Louis, Missouri, United States

Gibbs ringing artifacts especially pronounced in the presence of magnetic field inhomogeneities adversely affect different aspects of MRI. In quantitative MRI they bias and often corrupt measurements. The VSF (voxel spread function) method provides a platform for correcting images and can be applied to a variety of quantitative gradient-echo-based MRI techniques. Elimination of Gibbs artifacts in VSF approach requires solution of high-rank matrix equation (VSF equation) and relies on certain approximations. Herein we propose a method for solution of VSF equation allowing for substantially improving visibility of sharp edges and demonstrate improved contrast of blood vessels in the human brain.

Anna-Katinka Bracher¹, Erich Hell², Johannes Ulrici², and Volker Rasche^1
1Internal Medicine II, University Hospital of Ulm, Ulm, BW, Germany, ²Sirona Dental Systems, HE, Germany

For tissues with very short T2* values ultrashort echo time T2* (UTE -T2*) mapping is used instead of conventional multi echo approach because in the latter case echo time spacing is too large for the fast signal decay in rigid bodys. Since a full sampled image is measured for every TE value UTE-T2* mapping is a very time consuming approach. Using a contrast manipulating mixed k-space (COMMIX) approach provides multiple TE images in a single k-space data set.

Chuan Huang¹, Maria I Altbach², Quanzheng Li¹, Xiaomeng Zhang¹, and Georges El Fakhri^1
1Center for Advanced Medical Imaging Sciences, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States, ²Medical Imaging, The University of Arizona, Tucson, AZ, United States

multiple spin-echo is generally used due for T2 mapping to the long acquisition time required by single spin-echo strategies. However, because of pulse imperfection, the signal acquired by MSE sequence is generally contaminated by indirect echoes. Several groups have investigated T2 fitting techniques incorporating the indirect echoes into the signal model. One common aspect of these model-based indirect echo compensated T2 estimation techniques is that the flip angles of the refocusing pulses are also being fitted to the acquired TE images for each pixel along with T2 values. However, due to the non-convex nature of the cost function in the fitting, there is an ambiguity issue for the refocusing flip angles. Breikreutz et al proposed to use separately acquired rFA map to improve the accuracy. However, the additional acquisition time for rFA mapping is not desirable, and the approach also suffers from subject motion between these two acquisitions. In this work, we propose an iterative refocusing flip angle map constrained T2 decay model fitting technique with indirect echo compensated to solve the ambiguity problem without additional acquisition.

Rui Pedro A. G. Teixeira^1,2, Shaihan J. Malik^1,3, and Joseph V. Hajnal^1,3
1Division of Imaging Sciences and Biomedical Engineering, King's College London, London, London, United Kingdom, ²Institute of Biophysics and Biomedical Engineering, Faculty of Sciences, Lisbon, Lisbon, Portugal, ³Centre for the Developing Brain, King's College London, London, London, United Kingdom

The Driven Equilibrium Single Pulse Observation of T1 & T2 (DESPOT) relaxometry technique aims to provide high quality 3D maps of T1 and T2 within a clinical time frame. We propose an optimization method built on the Cramér-Rao Lower Bound to guarantee low estimation standard deviation over a defined range of relaxation times and use it to design optimal single compartment DESPOT acquisitions for adult and neonatal brain.

Ken-Pin Hwang^1,2, Jong Bum Son², Marcel Warntjes³, Ersin Bayram¹, and Jingfei Ma^2
1Global MR Applications and Workflow, GE Healthcare, Houston, TX, United States, ²Department of Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, United States, ³Center for Medical Imaging Science and Visualisation, University of Linkoping, Linkoping, Sweden

The Fast Triple Echo Dixon fat-water separation technique is combined with QMAP, to obtain in a single sequence the relative signal fraction of each species, in addition to the T1, T2, and PD information provided by QMAP. The FSE sequence is modified by a replacing the standard readout pulse with a three-echo bipolar readout. The fat/water separation is first performed, and relaxation parameter fitting is then applied separately to the fat and water images. Not only is more information provided in this technique, but quantitation could potentially be improved in mixed voxel species.

Casey P. Johnson¹, Daniel R. Thedens¹, and Vincent A. Magnotta^1
1Radiology, University of Iowa, Iowa City, IA, United States

Quantitative T1ρ mapping has tremendous potential to probe a variety of brain diseases. However, T1ρ mapping has largely been limited to 2D single-slice acquisitions. 3D mapping would be a major advance to provide high spatial resolution maps of the entire brain, but acquisition time is a significant challenge. Here we simulate tradeoffs of imaging parameters for recently-described rapid segmented 3D T1ρ mapping sequences to determine which sets of parameters provide the shortest acquisition times to achieve target T1ρ precision thresholds. This is a critical step toward fulfilling the need to efficiently acquire 3D T1ρ maps of the brain.

Guan Wang^1,2, Abdel-Monem M. El-Sharkawy², and Paul A. Bottomley^1,2
1Electrical & Computer Engineering, Johns Hopkins University, Baltimore, MD, United States, ²Russell H. Morgan Dept. of Radiology & Radiological Sciences, Johns Hopkins University, Baltimore, MD, United States

T₁, T₂ and proton density contain almost all of the information that MRI routinely uses in clinical diagnosis and research, but are seldom imaged directly. Moreover, measurements depend critically on B₁-field homogeneity making field mapping essential at higher field strengths. Here, a novel method that measures T₁, T₂, PD and B₁ with only four acquisitions–the minimum possible–is presented. This ‘Tri-FA’ method encodes T₁ with three varied flip-angles, and T₂ via long 0° BIR-4 pre-pulses instead of spin-echoes. 2D and 3D ‘Tri-FA’ imaging is demonstrated in vitro and in human brain studies at 3 Tesla.

Ningzhi Li¹, Mark Bolding², and Donald B Twieg^3
1Center for Neuroscience and Regenerative Medicine, Bethesda, Maryland, United States, ²Department of Radiology, University of Alabama at Birmingham, AL, United States, ³Department of Biomedical Engineering, University of Alabama at Birmingham, AL, United States

Rapid R2 mapping is of significant interests to MRI scientists and can be applied in many research areas. The recent introduced SE-SS-PARSE (single-shot parameter assessment by retrieval from signal encoding) is an ultrafast and direct R2 mapping technique. By modeling signal changes during the sampling, SE-SS-PARSE promises a more accurate and robust data measurement. This study evaluates the accuracy of R2 mapping in SE-SS-PARSE by comparing it with fast SE, a commonly used rapid R2 mapping method. Conventional SE provided references. We found that SE-SS-PARSE provided accurate R2 mapping with a much shorter acquisition time compared to fast SE.

Cyril Lorenzato¹, Chris Oerlemans¹, Baudouin Denis de Senneville^1,2, Chrit Moonen¹, and Clemens Bos^1
1Imaging Division, University Medical Center Utrecht, Utrecht, Netherlands, ²UMR 5251 CNRS / Université Bordeaux 1 / INRIA, Institut de Mathématiques de Bordeaux, Talence, France

Thermosensitive liposomes (TSL) can be prepared to contain drugs as well as paramagnetic MR-contrast agents. This would allow heat induced local drug release, while detecting the release by imaging relaxivity changes. A method is presented to measure T₁, T₂* and temperature simultaneously and is applied to monitor release from gadolinium containing TSL during heating. The results indicated the need to correct for temperature related apparent proton density changes in the T1 measurements. The method will be an aid to monitor the release of MR-contrast agents from TSL in local drug delivery studies, as a correlate of drug release.

Jason Su^1,2 and Brian Rutt^2
1Electrical Engineering, Stanford University, Stanford, CA, United States, ²Radiology, Stanford University, Stanford, CA, United States

An open source computational framework that calculates the CRLB for arbitrary signal equations to machine precision with minimal effort is developed and applied to the analysis of steady-state relaxometry with the SPGR and bSSFP pulse sequences. The unbiased protocol that achieves the highest precision efficiency estimate of T1 and T2 is found allowing the free variables: flip angle, phase cycle, and acquisition time fraction. The optimum PCVFA protocol consists only of 0 and π-phase-cycled bSSFP at 4 flip angles. This provides over a factor of 2.1x gain in efficiency for simulated WM and GM compared to previous related DESPOT2 protocols.

Mathieu Boudreau¹, Christine Tardif², Nikola Stikov¹, and G. Bruce Pike^1,3
1Montreal Neurological Institute, McGill University, Montreal, Quebec, Canada, ²Department of Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Saxony, Germany, ³Hotchkiss Brain Institute, Faculty of Medicine, University of Calgary, Calgary, Alberta, Canada

B₁ maps are an essential part of most quantitative MRI protocols, including Variable Flip Angle (VFA) T₁ mapping. To achieve whole brain quantitative imaging in reasonable scan times, several novel rapid B₁ methods have been introduced. This work compared VFA T₁ maps in white matter produced with four B1 methods: Double angle, Bloch-Siegert, AFI, and double angle EPI. Six healthy adult subjects were scanned with a 3T Siemens Tim Trio MRI using a 32-channel receive-only head coil. All B₁ methods resulted in comparable WM T₁ maps, and all rapid methods strongly correlated with the reference DA map.

Jinxia Zhu¹, Markus Klarhöfer¹, Francesco Santini¹, Klaus Scheffler^2,3, and Oliver Bieri^1
1Radiological physics, University Hospital Basel, basel, basel, Switzerland, ²High-Field Magnetic Resonance Center, Max-Planck Institute for Biological Cybernetics, Tübingen, Germany, ³Department of Biomedical Magnetic Resonance, University Hospital Tübingen, Tübingen, Germany

Basic spin-echo (SE) based protocols were used to estimate T1 and T2 relaxation times in human brain in the presence of multiple field strengths (0.35T, 1.5T, 3T, 7T, and 9.4 T). Experiments were performed on the same two subjects at all B0 using the same acquisition parameters. The T1 dependence on B0 is further analyzed.

Govind Nair¹, Qi Duan¹, Daniel S Reich¹, and Souheil Inati^2
1NINDS, National Institutes of Health, Bethesda, MD, United States, ²NIMH, National Institutes of Health, Bethesda, MD, United States

B1-mapping techniques are seldom applied to T1-mapping of the human brain at high field strengths due to the high SAR resulting in long scan times. Background-field-corrected Bloch-Siegert B1-mapping scheme is used here for whole-brain T1-mapping at 1 mm isotropic resolution on a 7T scanner. The total scan time was less than 30 minutes. The corrected T1 map was uniform and had a distinct bimodal distribution with peaks at 1350 and 1950 ms, and repeat calculation at non-optimal shimming yielded an average change of 0.1 ± 0.1%. This technique is being applied in quantitative lesion-evolution study in multiple sclerosis.

Ovidiu Cristian Andronesi¹, Dylan M. Tisdall¹, and Andre J. van der Kouwe^1
1Martinos Center for Biomedical Imaging, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States

T1 relaxation in the rotating frame (T1rho) is sensitive to molecular dynamics of water molecules interacting with macromolecules and has been shown to be a valuable imaging biomarker in stroke, neurodegenerative diseases (Alzheimer’s Disease, Parkinson Disease), cancer, liver cirrhosis and cartilage damage. Mapping of T1rho relaxation constant requires the acquisition of a time series of images with increasing rotating frame relaxation weighting. Perfect alignment of the images in the time series is critical for the accurate fitting of the T1rho constant. Because of large changes in the contrast among images acquired at different preparation (weighting) times, postprocessing motion correction algorithms have difficulties to accurately coregister serial volumes and may introduce false displacements. Here, we show that real-time motion correction improves the quality of T1rho mapping when subjects move.

Matthias Alexander Dieringer^1,2, Jeanette Schulz-Menger², and Thoralf Niendorf^1,2
1Berlin Ultrahigh Field Facility (B.U.F.F.), Max-Delbrueck Center for Molecular Medicine, Berlin, Germany, ²Experimental and Clinical Research Center (ECRC), Charité Medical Faculty and Max-Delbrueck Center for Molecular Medicine, Berlin, Germany

2D variable flip angle (VFA) T1 mapping has been proposed for rapid T1 quantification. RF pulse imperfections evoke non-uniform excitation flip angle distributions that deem T1 quantification with 2D VFA inaccurate. Sharp and uniform slice excitations could render considerations on complex slice profile corrections obsolete and could afford a direct T1 calculation of using simple gradient echo signal equations. In this work, we use the minimum-time variable-rate-selective-excitation (VERSE) algorithm to produce high quality slice profiles in combination with rapid two-point 2D VFA T1 mapping measurements in phantom and in the human brain.

Mary McLean¹, Andrew Patterson², Reem Bedair², Martin Graves², Scott Reid³, John Griffiths¹, and Fiona Gilbert^2
1CRUK Cambridge Institute, University of Cambridge, Cambridge, Cambridgeshire, United Kingdom, ²Radiology, Cambridge University Hospitals NHS Foundation Trust, Cambridgeshire, United Kingdom, ³GE Healthcare, Hertfordshire, United Kingdom

We implemented B1-corrected T1 measurements in the breast, using the Bloch-Siegert method of B1 mapping and DESPOT1. Orthogonal phase-encoding directions were acquired to compensate for cardiac motion. Flip angles higher than prescribed were consistently found to be delivered on the left side, leading to artifactual elevation in estimated T1. Compensation for B1 led to a reduction in the magnitude of asymmetry (|L-R/R|) from approximately 70% to 5% in both fat and parenchyma. Following correction, average T1 was 1368 ms in breast parenchyma and 458 ms in fat.

Ives R Levesque^1,2, Manojkumar Saranathan¹, Thomas Tourdias^1,3, Jason Su^1,4, James A Rioux¹, and Brian K Rutt^1
1Radiology, Stanford University, Stanford, CA, United States, ²Medical Physics, Oncology and RI-MUHC, McGill University, Montreal, QC, Canada,³Department of Neuroradiology, Bordeaux University Hospital, Bordeaux, France, ⁴Electrical Engineering, Stanford University, Stanford, CA, United States

Fast T₁ mapping is of interest notably for segmentation of brain structures and myelin imaging. Whole-brain, high-resolution T₁ maps can be obtained from 3 MP-RAGE images with carefully selected inversion times in a clinically relevant time and free from B₁ heterogeneity effects. In this work, we validated the method for in vivo human applications, and compared the accuracy and blur of the 3-TI-MP method for different k-space ordering and parallel imaging factors. Our results show high accuracy and improved map quality from a 2D-centric k-space ordering scheme.

Mathies Breithaupt¹, Sebastian Flassbeck¹, Moritz C. Berger¹, and Armin M. Nagel^1
1Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany

In this work, a modification to the DESPOT1-HIFI approach for applications at ultra-high magnetic field strengths of 7T is presented. By simulation of the mprage sequence via numerically solving the Bloch equations, the magnetization prior to the inversion can be approximated. This way, one parameter can be called from a lookup table inside the non-linear least squares minimization routine simplifying the target function. Three-dimensional T1 maps of the whole head with an isotropic resolution of 1 mm can be acquired within a 25 minute timeframe, revealing T1 values of 124643 ms (gray matter) and 190448 ms (white matter).

John W Grinstead¹, Dingxin Wang², Himanshu Bhat³, Vibhas Deshpande⁴, Stephen F Cauley⁵, Kawin Setsompop⁵, Thomas Benner⁶, Valerie C Anderson⁷, and William D Rooney^7
1Siemens Healthcare, Portland, OR, United States, ²Siemens Healthcare, MN, United States, ³Siemens Healthcare, MA, United States, ⁴Siemens Healthcare, TX, United States, ⁵A.A. Martinos Center for Biomedical Imaging, MA, United States, ⁶Siemens Healthcare, Erlangen, Germany, ⁷Oregon Health & Science University, OR, United States

Parametric T₁ mapping using inversion recovery has competing requirements for speed, signal-to-noise ratio, spatial resolution, anatomical coverage, and adequate sampling of the longitudinal magnetization recovery. Slice-accelerated multi-slice techniques utilize RF pulses which excite multiple 2D slices simultaneously. We describe the implementation and application of simultaneous multi-slice techniques to accelerate quantitative T₁ mapping. Compared to previous approaches, this provides extra flexibility in the number of slices and inversion recovery samples possible within a given measurement time, and has the potential of making quantitative T₁ mapping much more feasible and clinically practical.

James A. Rioux¹, Manojkumar Saranathan¹, and Brian K. Rutt^1
1Radiology, Stanford University, Stanford, CA, United States

The MP3RAGE sequence is a straightforward modification of MP2RAGE that acquires three images with different T₁ contrast following an inversion pulse. These images can be used for direct T₁ lookup, enabling accurate high-resolution mapping in clinically relevant scan times (10 minutes or less). We demonstrate that, with an appropriate choice of parameters, MP3RAGE T₁ maps can be made almost totally insensitive to B₁ effects, a key requirement for high-field imaging. These T₁ maps have excellent accuracy compared to a reference map. The MP3RAGE data can also be used to generate a whole-brain flip angle map at the same resolution.

Christoph Birkl¹, Christian Langkammer¹, Johannes Haybaeck², Christina Ernst², Johannes Petzold², Franz Fazekas¹, and Stefan Ropele^1
1Department of Neurology, Medical University of Graz, Graz, Austria, ²Department of Neuropathology, Institute of Pathology, Medical University of Graz, Graz, Austria

Tissue fixation with formaldehyde can significantly affects the relaxation properties of the underlying tissue which hinders direct comparison with in vivo conditions. We here present a simple correction scheme that is based on the temperature dependency of T1, T2 and T2* in fixed and unfixed tissue. The proposed correction scheme was validated with corresponding fixed and unfixed data from literature.

Kasper Hansen^1,2, Esben Szocska Hansen¹, Martin Colliander Kristensen³, and Michael Pedersen^1,4
1MR Research Centre, Aarhus University Hospital, Aarhus N, DK, Denmark, ²Dept. of Clinical Medicine, Aarhus University Hospital, Aarhus N, DK, Denmark, ³Dept. of Procurement & Clinical Engineering, Region Midt, Aarhus N, Denmark, ⁴Dept. of Clinical Medicine, Aarhus University Hospital, Aarhus N, Denmark

We have build an MR imaging compatible pressure chamber system for use in basic barometrical research. Relaxation times (T1 and T2) have been measured during hyperbaric oxygen, helium, and nitrogen (relevant in diving applications) exposure (1-10 ATA) of liquid-phantoms consisting of manganese or gadolinium enriched water or olive oil. Molecular oxygen is paramagnetic and can be used as a MRI contrast through dipol-dipol interactions with protons, a property currently exploited for non-invasive oxymetry. Here, we extrapolate this property through hyperbaric oxygen exposure.

Antonella Meloni¹, Gennaro Restaino², Gianluca Valeri³, Massimiliano Missere², Vincenzo Positano¹, Stefano Pulini⁴, Michele Centra⁵, Massimo Lombardi¹, and Alessia Pepe^1
1CMR Unit, Fondazione G.Monasterio CNR-Regione Toscana and Institute of Clinical Physiology, Pisa, Italy, ²Dipartimento di Radiologia, Un. Cattolica del Sacro Cuore - Centro di Ricerca e Formazione ad Alta Tecnologia "G. Paolo II", Campobasso, Italy, ³Dipartimento di Radiologia, Azienda Ospedaliero-Universitaria Ospedali Riuniti "Umberto I-Lancisi-Salesi", Ancona, Italy, ⁴U.O. Ematologia Clinica, Osped. Civile “Spirito Santo”, Pescara, Italy, ⁵Servizio Trasfusionale, OO.RR. Foggia, Foggia, Italy

The bone marrow contains reticuloendothelial cells and, of consequence, it is among the first organs to be affected by iron overload, which can be non-invasively assessed by MRI T2* technique. In thalassemia major patients bone marrow T2* values increased with age. Males showed significantly higher T2* values. Bone marrow T2* values were associated with heart, liver and spleen T2* values. Splenectomised patients showed higher bone marrow T2* values.

Antonella Meloni¹, Gennaro Restaino², Massimiliano Missere², Emilio Lozupone³, Vittorio Semeraro³, Vincenzo Positano¹, Massimo Lombardi¹, Giuseppina Sallustio², and Alessia Pepe^1
1CMR Unit, Fondazione G.Monasterio CNR-Regione Toscana and Institute of Clinical Physiology, Pisa, Italy, ²Dipartimento di Radiologia, Un. Cattolica del Sacro Cuore - Centro di Ricerca e Formazione ad Alta Tecnologia "G. Paolo II", Campobasso, Italy, ³Policlinico A. Gemelli, Roma, Italy

Bone marrow T2* measurements are feasible, reproducible, non-time-consuming and can be limited to a circular region of interest. Bone marrow T2* values in healthy subjects are independent of age but are significantly higher in females. Gender-specific lower limits of normal have been defined.

Jeam Haroldo Oliveira Barbosa¹, Antônio Carlos Santos², José Eymard Homem Pittella², Luciano Neder Serafini², Oswaldo Baffa Filho¹, and Carlos Ernesto Garrido Salmon^1
1FFCLRP- D.Physics, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil, ²FMRP, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil

Quantification paramagnetic ions in postmortem brain using Electron Spin Resonance (ESR) and its correlation with R2 and R2* maps. Fe+3 and Cu+2 paramagnetic ions are found in regions with considerable increased R2 and R2* values. The determination of the influence from paramagnetic ions can help understanding the origins of changed R2 and R2* values in the basal ganglia.

Matthew Tarasek¹, Oguz Akin², Jeannette Roberts³, Thomas Foo¹, and Desmond Yeo^1
1Diagnostics & Biomedical Technologies, GE Global Research, Niskayuna, NY, United States, ²Radiology, Memorial Sloan Kettering Cancer Center, NY, United States, ³Biomedical Imaging & Physiology, GE Global Research, Niskayuna, NY, United States

Longitudinal relaxation time (T1), and transverse relaxation time (T2) are examples of imaging parameters that change with temperature. It is challenging to use these descriptors to measure temperature in vivo because the way they change is very tissue dependant. Here we use T1, and T2 to determine salient characteristics of ex vivo rat breast adenocarcinoma and rat prostate carcinoma tumors and their surrounding muscle tissues at steady-state temperatures from ~0°C to 22°C. Findings indicate a link to improved MR imaging visualization or characterization of tumors with heat-induced T1/T2 relaxation contrast types.

Florence Colliez¹, Julie Magat¹, Marta M Safronova², Bénédicte F Jordan¹, Thierry Duprez³, and Bernard Gallez^1
1Louvain Drug Research Institute, Biomedical Magnetic Resonance Research Group, University of Louvain, Brussels, Belgium, ²Service de Radiologie, Cliniques universitaires Saint-Luc, Brussels, Belgium, ³Department of Radiology and Medical Imaging, Cliniques universitaires Saint-Luc, Brussels, Belgium

There is a critical need for dynamic and noninvasive methods able to monitor brain oxygenation in clinical practice. Variations in T1 and T2* are potentially valuable MRI tools to detect changes in tumor oxygenation. We hereby (i) test the ability of ‘MOBILE’, a method of oxygenation mapping based on the changes in the relaxation properties of the tissue lipids, to map brain oxygenation in brain and cerebellum (ii) evaluate its capability to follow an increase in oxygenation on healthy volunteers submitted to a 100% O2 breathing (iii) and compare sensitivities of the ‘MOBILE’ and ‘Oxygen enhanced MRI’ techniques.

John Thomas Spear^1,2, Zhongliang Zu^2,3, and John C. Gore^2,4
1Physics & Astronomy, Vanderbilt University, Nashville, TN, United States, ²Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States, ³Radiology, Vanderbilt University, Nashville, TN, United States, ⁴Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

The dispersion of the spin-lattice relaxation rate in the rotating frame, R₁, has recently been shown to be sensitive to diffusion through magnetically inhomogeneous media with internal susceptibility gradients. In most systems in vivo, chemical exchange will significantly affect this dispersion as well, especially at high field. A method to analyze both effects simultaneously has been proposed and tested using blood with varying oxygen saturation levels as a model system. The contribution of diffusion decreases with oxygen saturation as the internal gradients diminish, leaving only chemical exchange to influence R₁ dispersion.

John Thomas Spear^1,2, Zhongliang Zu^2,3, and John C. Gore^2,4
1Physics & Astronomy, Vanderbilt University, Nashville, TN, United States, ²Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States, ³Radiology, Vanderbilt University, Nashville, TN, United States, ⁴Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

MRI detection of 2-Deoxy-D-Glucose in tissues has the potential to allow imaging of regional metabolism. Adding 2DG to tissue should add to the background R₁ dispersion by an amount dependent on the concentration. R₁ dispersion was measured in rat brain tissue homogenates with various concentrations of 2DG to demonstrate this effect and show how the mean exchange rate shifts. Exchange Rate Contrast (ERC) images can be made so the image intensity scales with 2DG concentration, which is established by calculating mean pixel intensities as a function of concentration.

L. Sorina Truica¹, J. Keiko McCreary¹, Adam R. Neumann¹, Maurice A. Needham¹, Albert R. Cross¹, and Ian Q. Whishaw^1
1Neuroscience, University of Lethbridge, Lethbridge, Alberta, Canada

The use of manganese (Mn) as a contrast agent for neuro-anatomical and functional mapping has been studied in animal models for the last decade. Mn contrast enhancement is directly proportional to the changes in relaxation rates in brain tissue. However, the binding of Mn to proteins may alter its relaxation. Thus protein levels and metabolic processes that produce them in the brain and cerebrospinal fluid (CSF) will alter Mn contrast. The aim of this study was to investigate the relaxation rate of Mn in artificial CSF (aCSF), when different levels of protein are present. We found that the concentration of protein used affected both the T1 and T2 relaxation times.

Maria Engström^1,2, Jan B M Warntjes^2,3, Anders Tisell^2,4, Anne-Marie Landtblom^2,5, and Peter Lundberg^2,4
1Radiology, Linköping University, Linköping, Sweden, ²Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden, ³Clinical Physiology, Linköping University, Linköping, Sweden, ⁴Radiation Physics, Linköping University, Linköping, Sweden, ⁵Neurology, Linköping University, Linköping, Sweden

The purpose was to improve objective measures for brain tissue characterization and visualization in groups using quantitative MRI in combination with brain normalization. Nineteen multiple sclerosis (MS) patients and 20 healthy controls were investigated with quantitative MRI, measuring the relaxation rates, R1 and R2, and the proton density. Voxel-based statistical analysis showed shared tissue aberrations in the MS patients and aberrations that were related to MS disability. Quantitative MRI together with voxel-based statistical analysis and multi-parametric visualization is an up-and-coming method to study neuropathology and its neuroanatomical correlation to clinical disability measures that are common for an entire group.

Alexandre Vignaud¹, Xavier Violas², Alain Rahmouni³, Philippe Robert², and Alexis Amadon^1
1UNIRS/NeuroSpin/I2BM/DSV, CEA, Gif Sur Yvette, France, ²Experimental Imaging, Guerbet Group, Aulnay Sous Bois, France, ³Imagerie Médicale Service, Henri Mondor Albert Chenevier group, AP-HP, Creteil, France

Several papers reported contrast agent (CA) relaxivity measurements at several clinical magnetic fields. None have investigated them on a large physiologically relevant range including first pass equivalent concentrations. Thus B1 and B0 heterogeneities make this data difficult to extract at Ultra High Field (UHF). In this work, using state of the art relaxometry methods, we measured r1 and r2 as a function of B0 (1.5, 3 and 7T) for a large CA concentration window

Sidyarth Garimall¹, Anup Singh¹, Kevin D'Aquilla¹, Mark Elliott¹, Hari Hariharan¹, and Ravinder Reddy^1
1Radiology, University of Pennsylvania, Philadelphia, PA, United States

The objective of this study was to evaluate the reproducibility of high resolution T1� maps of human knee cartilage at ultra-field MRI scanners. Ten subjects were scanned, at two time points, on a 7T whole-body MRI scanner for T1� data of Knee cartilages. Intraclass correlation coefficient for T1� values in knee cartilage was statistically significant. Similarly, coefficient of variation for T1� values between two repetitions was under 4%, which confirm a high degree of reproducibility of T1� maps of knee cartilage at 7T.

Peter Andrew Hardy^1,2, Vrinda Sardana³, Bruce Spottiswoode⁴, Vince Sorrell³, and Steve Leung^3
1Radiology, University of Kentucky, Lexington, Kentucky, United States, ²MRISC, University of Kentucky, Lexington, Kentucky, United States,³Cardiology, University of Kentucky, Lexington, Kentucky, United States, ⁴Siemens Medical Solution, Chicago, Illinois, United States

We investigated the dependence of the myocardial T1 measured using MOLLI upon the infiltration of the myocardium by fat. The off-resonance frequency of the lipid protons distorts the estimate of the relaxation time.

James Goldfarb^1,2 and Wenguo Zhao^1
1St Francis Hospital, Roslyn, NY, United States, ²Program in Biomedical Engineering, SUNY Stony Brook, Stony Brook, NY, United States

Myocardial fibrosis alters Gd-contrast agent dynamics. Myocardial partition coefficients calculated from a multipoint slope calculation will vary in healed myocardial infarction based on the selection of samples. This includes the use of pre-contrast measurements and samples early after contrast agent administration. Partition coefficient calculations are insensitive to data sampling effects in viable myocardium.

Neville D Gai¹, Fabio Raman¹, Christopher Sibley¹, Songtao Liu¹, and David Bluemke^1,2
1Radiology & Imaging Sciences, NIH, Bethesda, Maryland, United States, ²NIBIB, Bethesda, Maryland, United States

T1 mapping and derived values such as partition coefficient and extracellular volume (ECV) have been used to differentiate normal myocardium from myocardium with diffuse fibrosis. There is scant theoretical and experimental evidence to-date favoring one measure over the other. In this study, we systematically compare the sensitivity and precision of T1 mapping measures. Using error analysis and measurements from phantoms, we show that post-contrast myocardium T1 (T1_cm) has a higher precision than derived values like ECV. Comparing T1_cm of normal volunteers with heart failure patients, we show that T1_cm has higher sensitivity to disease discrimination compared with ECV.

Meredith Ireene Taylor¹, Haonon Wang¹, James Badal¹, Kevin Perkins¹, Daniel J. Park¹, Joshua Kaggie², Jonathan Wisco³, and Neal K. Bangerter^1,2
1Electrical and Computer Engineering, Brigham Young University, Provo, UT, United States, ²Utah Center for Advanced Imaging Research, University of Utah, Salt Lake City, UT, United States, ³Department of Physiology and Developmental Biology, Brigham Young University, Provo, UT, United States

Models of vocal fold biomechanics and vibration can be used to study voice physics and to guide clinicians as they attempt to restore or preserve voice quality in persons with vocal or speech disorders. To assist in the design of these accurate models we created a custom 2-channel phased-array receive-only coil for high-resolution imaging of the larynx. T1 and T2 relaxometry was performed on an excised pig larynx to understand the MR signal characteristics of various tissues. These values were then used to choose optimal sequence parameters for maximizing contrast between two important tissues for modeling the larynx.

Chaitra Badve¹, Dan Ma², Yun Jiang², Alice Yu³, Jeffrey Sunshine¹, Vikas Gulani^1,2, and Mark Griswold^1,2
1Radiology, Case Western Reserve University and University Hospitals Cleveland, Cleveland, Ohio, United States, ²Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio, United States, ³School of Medicine, Case Western Reserve University, Cleveland, Ohio, United States

This study applies magnetic resonance fingerprinting for quantitative analysis of gliomas. Five glioma patients were scanned. T1, T2 quantification of solid tumor parenchyma and white matter at various locations was performed. T1, T2 values of tumor were significantly different than contralateral white matter. T1, T2 values of tumor were significantly different than perilesional white matter. These results demonstrate that quantitative analysis with MRF can distinguish glial tumors from peritumoral white matter changes and uninvolved white matter. MRF may also help to identify regions of infiltrative peritumoral edema in higher-grade tumors.

Claudia Fellner¹, Niklas Verloh¹, Michael Haimerl¹, Miriam Rabea Kubach², Marcel D. Nickel², Mona Schlabeck¹, Christian Stroszczynski¹, and Philipp Wiggermann^1
1Institute of Radiology, University Hospital Regensburg, Regensburg, Germany, ²MR Applications Development, Siemens AG, Healthcare Sector, Erlangen, Germany

Based on a 3D VIBE sequence with 3 flip angles and a preceding B1 map, T1 maps of the whole liver were acquired in 84 patients at 3 T before and after the application of Gd-EOB-DTPA. B1 map as well as the T1 mapping sequence were measured during breath hold (14s, 17s). A single slice technique (2D TurboFLASH with 2 inversion times) was used additionally. 3D VIBE yielded reliable results for T1 – even in critical anatomical regions like the left liver lobe. The reduction rate of T1 relaxation times before and after Gd-EOB-DTPA correlated well between both T1 mapping techniques.

Byeong-Yeul Lee¹, Xiao-Hong Zhu¹, Xiufeng Li¹, and Wei Chen^1
1Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, Minnesota, United States

The corpus callosum (CC) facilitates the intense neural pathways connecting two cerebral hemispheres in mammals, and it contains numerous intra- and inter-hemispheric myelinated axonal projections. Imaging of callosal microstructures is of importance to understand its functional and anatomical connectivity to cortical areas. We hypothesized that the parametric T1 measure could be sensitive to the CC microstructure and its change associated with the fiber size; and it should be logically correlated to the CC myelin content mapped by T1/T2 ratio image. To test this hypothesis, we incorporated T1 mapping, myelin content mapping and collosal parcellation mapping for differentiating and correlating the fiber size and myelin content in the CC of healthy human at 7T. In comparison with histology report, our results indicate a positive correlation between the T1 value and CC fiber size, in contrast, it approximately correlates inversely to the CC myelin content. The overall results show the utility of complementary parametric T1 and myelin imaging approaches to quantitatively assess the fiber microstructure of human corpus callosum. This hybrid imaging approach could provide a robust and useful imaging tool for detection of fiber abnormality in the human white matter.

Hiromi Sano^1,2, Takayuki Obata³, Hiroshi Kawaguchi¹, Kuniaki Nabatame², Satoshi Obara², Jeff Kershaw¹, Keiichi Akahane², Yoshiya Shimada², and Hiroshi Ito^1
1Molecular Imaging Center, National Institute of Radiological Sciences, Chiba, Chiba, Japan, ²Medical Exposure Research Project, National Institute of Radiological Sciences, Chiba, Chiba, Japan, ³Research Centre for Charged Particle Therapy, National Institute of Radiological Sciences, Chiba, Chiba, Japan

Irradiation dose can be three-dimensionally visualized with polymer gel dosimetry and multi-slice R2 mapping. However, the high number of RF pulses may increase the temperature of the polymer gel, which causes errors in dose estimation because the R2 depends on temperature as well as the radiation dose. In this study, we investigated the relationship between changes to R2 and temperature caused by RF pulses in a polymer gel phantom. A temperature rise of 1 °C caused a 0.099 s^-1 reduction in R2, which suggests that temperature control and/or correction is needed to reduce errors in polymer gel dosimetry with MRI

Weronika Piedzia¹, Krzysztof Jasinski¹, Katarzyna Kalita¹, and Wladyslaw P. Weglarz^1
1Department of MRI, Institute of Nuclear Physics, Polish Academy of Sciences, Kraków, Malopolska, Poland

Sang-Young Kim¹, Hyeonman Baek², Jung-Hoon Lee¹, Do-Wan Lee¹, Jin-Young Jung¹, and Bo-Young Choe^1
1Department of Biomedical Engineering, The Catholic University of Korea, Seoul, Seoul, Korea, ²Korea Basic Science Institute, Seoul, Korea

We present the feasibility of use of the parallel RF transmission with multiple RF source (MultiTransmit) imaging in MRI-based polymer gel dosimetry. The image quality and B1 field inhomogeneity between conventional and MultiTransmit MR imaging were compared. Finally, the estimated R2 uncertainty, ¥ò(R2) and dosimetric performance (dose resolution) between two methods were compared. Improved image quality and B1 field homogeneity results in lower ¥ò(R2) on the MultiTransmit images than on the conventional images. We demonstrated the feasibility of MultiTransmit MR imaging for introducing gel dosimetry into clinical routine, suggesting that MultiTransmit MR imaging has potential benefits for 3D gel dosimetry.

Weitian Chen¹, Edwin Oei^2,3, and Garry E Gold^2,4
1Global MR Applications & Workflow, GE Healthcare, Menlo Park, CA, United States, ²Radiology, Stanford University, CA, United States, ³Radiology, Erasmus MC Rotterdam, Netherlands, ⁴Bioengineering and Orthopedic Surgery, Stanford University, CA, United States

Intermediate T2-weighted and PD-weighted FSE images are commonly used in clinical MRI for detection of joint abnormalities. There are increasing interests in using T2 and T1rho as biomarkers for improved diagnosis of osteoarthritis and related diseases. T2 or T1rho relaxometry is typically performed in 2D or 3D plane with inferior resolution compared to anatomical imaging due to scan time limitations. In this work, we investigated the use of a single scan to acquire T2-weighted, PD-weighted anatomy images and relaxometry map with 3D isotropic resolution of 0.6mm in a clinically feasible scan time. Both relaxometry map and anatomical images can be subsequently reformatted to arbitrary plane for post-analysis and diagnosis.

Anna Barnes¹, Catherine Morgan², Alan Bainbridge³, Lorna Smith⁴, Scott Rice⁵, David Atkinson⁶, and Shonit Punwani^5
1Nuclear Medicine, University College Hospital, London, England, United Kingdom, ²Centre of Medical Imaging and Computing, University College London, london, England, United Kingdom, ³Medical Physics, University College Hospital, London, England, United Kingdom, ⁴Radiology, University College Hospital, London, England, United Kingdom, ⁵Radiology, University College Hospital, london, England, United Kingdom, ⁶Centre of Medical Imaging, University College London, london, England, United Kingdom

This abstract describes the work done to establish whole body B1 corrected T1 mapping optimised for use with only 2 flip angles. The authors present results that show a flip angle pair of 2.5 and 20 provides a robust and reproducible T1 map across a number of subjects and time points when compared to using all 8 flip angles to calculate T1. The mDixon sequence is expected to have SNR benefits over a single gradient echo since it combines multiple echo times.

Md Nasir Uddin¹, R Marc Lebel^1,2, and Alan H Wilman^1
1Biomedical Engineering, University of Alberta, Edmonton, Alberta, Canada, ²Applied Science Laboratory, GE Healthcare, Calgary, Alberta, Canada

Quantitative transverse relaxation measurements and the low and field strengths difference measures can be used to assess iron levels in the human brain. We compared the R2, R2*, R2’ and multi-slice FDRI mapping approaches for iron sensitivity measurements in deep gray matter using 1.5 T and 4.7 T. Results show R2* is most sensitive at iron at high field while R2’ and FDRI provide specific brain iron measures.