Chloé Najac¹, Vincent Boer², Hermien E. Kan¹, Andrew G. Webb¹, and Itamar Ronen^1
1C.J. Gorter Center for High Field MRI, Department of Radiology, Leiden University Medical Center, Leiden, Netherlands, ²Danish Research Centre for Magnetic Resonance, Copenhagen University Hospital Hvidovre, Copenhagen, Denmark

Spin-echo sequences suffer from loss of signal due to J-modulation for coupled spins. Previously, it was shown that adding a π/2 pulse between two π pulses in a double spin-echo sequence partially refocuses some of the J-evolution. However, the potential of such sequence at ultrahigh field is limited due to the large chemical shift displacement error. Here, we propose a J-refocused variant of the sLASER sequence (J-sLASER) to improve quantification of J-coupled metabolites at ultrahigh field. Significant improvement in quantitation of J-coupled metabolites is illustrated using simulation, phantom and in vivo measurements in the human brain.

William T Clarke¹, Jamie Near², Uzay E Emir^3,4, and Saad Jbabdi^1
1Wellcome Centre for Integrative Neuroimaging, NDCN, University of Oxford, Oxford, United Kingdom, ²Douglas Mental Health University Institute, and Department of Psychiatry, McGill University, Montreal, QC, Canada, ³School of Health Sciences, Purdue University, West Lafayette, IN, United States, ⁴Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, United States

FSL-MRS is a new Python-based MRS fitting tool. The model is based on a linear combination of basis spectra, and fitting is achieved with a sampling approach which gives full posterior distributions of fitted parameters. Results are stored both as text files and an interactive HTML report. On publication the tool will be open source and free as part of the FSL package. The tool will be developed to handle MRSI and model fMRS as time-series. Here we demonstrate an initial implementation in phantom data and three brain regions of 37 subjects and compare against another fitting software.

Kyung Min Nam¹, Arjan Hendriks¹, Vincent O. Boer², Dennis D.J. Klomp¹, Jannie P. Wijnen¹, and Alex Bhogal^1
1Imaging & Oncology, UMC Utrecht, Utrecht, Netherlands, ²Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Hvidovre, Denmark

Metabolic mapping at ultra-high field benefits from increased SNR and large chemical shift dispersion, enabling high spatial resolution and more resolved metabolite resonances. By exploiting increased SNR, MRSI acquisitions can be accelerated using echo-planar readout gradients (EPSI) or parallel imaging techniques. However, signal contamination from extra-cranial lipids can degrade the spectral quality and/or introduce fold-over artefacts. Therefore, suppression of extra-cranial lipids during acquisition is imperative for fast spectroscopic imaging. This work aimed to use a “crusher coil” to suppress extracranial lipid signals while acquiring metabolic information using a pulse-acquire acquisition with EPSI readout.

Yibo Zhao^1,2, Yudu Li^1,2, Jiahui Xiong^1,2, Rong Guo^1,2, Yao Li^3,4, and Zhi-Pei Liang^1,2
1Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ⁴Med-X Research Institute, Shanghai Jiao Tong University, Shanghai, China

J-resolved MRSI using hybrid FID/SE signals has recently been proposed as an effective approach to achieving rapid, high-resolution mapping of brain metabolites and neurotransmitters, an elusive goal of the MRSI community. The new data acquisition scheme poses new problems for data processing, from removal of nuisance signals to reconstruction of spatiospectral distributions. This paper presents an effective method to address these problems, utilizing a union-of-subspaces framework to absorb complementary and prior  information. The proposed method has been evaluated using experimental data, producing high-quality spatiospectral distributions of metabolites and neurotransmitters from hybrid FID/SE J-resolved MRSI data with limited coverage of (k,t,t_J)-space.

Karl Landheer¹, Martin Gajdosik¹, and Christoph Juchem^1,2
1Biomedical Engineering, Columbia University, New York, NY, United States, ²Radiology, Columbia University, New York, NY, United States

The aim of this work was to develop an optimized sLASER sequence which is capable of acquiring artefact-free data with an echo time as short as 20.1 ms on a whole-body clinical 3 T MR system. This was achieved through the use of specialized pulses and optimizing the crusher scheme and phase cycling schemes. High quality spectra were obtained and quantified in 6 healthy volunteers, both in the prefrontal and the occipital cortex.

Karl Landheer¹, Ralph Noeske², Michael Garwood³, and Christoph Juchem^1,4
1Biomedical Engineering, Columbia University, New York, NY, United States, ²GE Healthcare, Berlin, Germany, ³Center for Magnetic Resonance Research and Radiology, University of Minnesota, Twin Cities, MN, United States, ⁴Radiology, Columbia University, New York, NY, United States

A previously developed magnetic resonance spectroscopy method is improved upon which is now able to obtain spectra with an echo time as short as 4 ms, while recovering the entirety of the available magnetization. This method obtains full 3D spatial localization through a 2D adiabatic inversion pulse which is cycled “on” and “off” every other repetition, in combination with a slice-selective excitation pulse. Both 1D and 2D spectra with an ultrashort echo time of 4 ms are demonstrated at 3T. High quality spectra were obtained for all experiments.

Martyna Dziadosz¹, Wolfgang Bogner², and Roland Kreis^1
1Departments of Radiology and Biomedical Research, University of Bern, Bern, Switzerland, ²Department of Biomedical Imaging and Image-guided Therapy, Medical University Vienna, High-field MR Center, Vienna, Austria

Most clinical MRS studies concentrate on the upfield part of the spectrum, neglecting the downfield region, since it is mainly composed of labile protons with intensities depending on the water suppression scheme. For reliable and sensitive quantification, a non-water-suppressed or even better a non-water-excitation (NWE) technique is required.

We introduce a non-echo technique based on ISIS localization for optimal detection of (moderately) fast exchanging protons at 3T with minimal signal loss due to T₂ or exchange and with optimized water magnetization path for longitudinal relaxation enhancement.

Andres Saucedo¹, Manoj Sarma¹, and M. Albert Thomas^1
1Radiological Sciences, University of California, Los Angeles, Los Angeles, CA, United States

In this pilot study, we implement a radial echo-planar spectroscopic imaging (rEPSI) sequence and compare its performance to the Cartesian (EPSI) sequence under multiple rates of retrospective undersampling, for the case of 2D acquisitions in which a single phase-encoded direction is undersampled for EPSI, and a reduced number of radial profiles using golden angle view ordering are reconstructed for rEPSI. Fully sampled data from a healthy volunteer brain and from a brain phantom are acquired. The retrospectively undersampled data is reconstructed using compressed sensing with Total Variation regularization, and the performance of both methods are compared.

Sophie M. Peereboom¹ and Sebastian Kozerke^1
1Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland

Cardiac triglyceride levels can be assessed using proton MR spectroscopy. While metabolite cycling has already been applied to cardiac proton MRS, the combination of metabolite cycling and spectroscopic imaging for cardiac applications remains to be demonstrated. In this work metabolite-cycled echo-planar spectroscopic imaging is proposed to asses triglyceride-to-water (TG/W) ratios in different regions of the human heart. Results were compared to conventional single-voxel measurements in the interventricular septum. Although SNR is limited for metabolite-cycled EPSI, the method allows for detection of regional TG/W and therefore holds promise to provide insights into regional TG variations.

Vasiliki Mallikourti¹, Sai Man Cheung¹, Tanja Gagliardi ^2,3, Nicholas Senn¹, Yazan Masannat⁴, Trevor McGordlrick⁵, Ravi Sharma⁵, Steven Heys^1,4, and Jiabao He^1
1Institute of Medical Sciences, School of Medicine, University of Aberdeen, Aberdeen, United Kingdom, ²Department of Clinical Radiology, Aberdeen Royal Infirmary, Aberdeen, United Kingdom, ³Department of Radiology, Royal Marsden Hospital, London, United Kingdom, ⁴Breast Unit, Aberdeen Royal Infirmary, Aberdeen, United Kingdom, ⁵Department of Oncology, Aberdeen Royal Infirmary, Aberdeen, United Kingdom

Signal combination algorithms, designed for 1D MRS, were adapted for lipid composition spectra acquired using 2D MRS of double quantum filtered correlation spectroscopy (DQF-COSY). The algorithms of adaptively optimised combination (AOC), noise decorrelated combination (nd-comb), and whitened singular value decomposition (WSVD) were evaluated on lipid composition spectra from healthy volunteers and patients with breast cancer (tumour and peritumoural adipose tissue). WSVD provided maximal SNR uniformly across all spectral peaks. WSVD, through the elimination of additional acquired reference spectrum, shortens acquisition in patients to less than 11 min instead of 17 min while maintaining sensitivity.

Patrick J Bolan¹, Gregory J Metzger¹, Dinesh Deelchand¹, and Malgorzata Marjanska^1
1Radiology / Center for Magnetic Resonance Research, University of Minnesota-Twin Cities, Minneapolis, MN, United States

Measurements of metabolite T₂ relaxation constants can be valuable biomarkers of aging and disease. The conventional method for measuring multiple metabolite T₂s is to independently fit spectra from a multi-TE and then separately fit the amplitudes to an exponential decay to estimate T₂. In this work we implement a simultaneous fitting approach to fit all of the multi-TE spectra at the same time by incorporating the transverse relaxation in the model. This approach greatly reduces the degrees of freedom and enable T₂ estimation in noisy data, which may be used to shorten acquisition times and/or measure smaller regions.

Sofie Tapper^1,2, Mark Mikkelsen^1,2, Blake E. Dewey^2,3, and Richard A. E. Edden^1,2
1Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ²F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, ³Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, United States

Frequency-and-phase correction is an important step in the processing of single-voxel magnetic resonance spectroscopy data, and is required for J-difference editing, which relies on subtraction to reveal a low-SNR signal. We investigated an approach for frequency-and-phase correction using deep learning. Our networks were trained using simulated spectra manipulated with different frequency-and-phase offsets. During validation, the network returned spectra that were corrected to within 1.76 ± 1.19 degrees of phase and 0.09 ± 0.05 Hz of frequency, giving a difference spectrum very similar to the true unmanipulated spectrum. Frequency-and-phase correction is a promising application for deep learning in in vivo MRS.

Chi-Hyeon Yoo^1,2, Hyeon-Man Baek², and Bo-Young Choe^1
1The Catholic University of Korea, Seoul, Republic of Korea, ²Gacheon University, Incheon, Republic of Korea

The aim of this study was to apply the atlas-based brain segmentation to high-resolution 3D MRI of the mice brain, to assess a partial volume effect in the voxel. The altas-based segmentation successfully decomposed high-resolution 3D MRI into various anatomical compartments, and by volumetrically analyzing binary masks of the tissue compartment and voxels, a true contribution of the intended tissue compartment in the localized voxel can be assessed. By analyzing the metabolite-specific volume masks, an agreement of the localization between metabolites was evaluated. By evaluating the true contribution of the intended compartment and metabolite-specific agreement, localization reliability can be improved.

Georg Oeltzschner^1,2, Helge Jörn Zöllner^1,2, and Richard Anthony Edward Edden^1,2
1Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, United States, ²F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

Magnetic resonance spectroscopy (MRS) offers a wide range of methods to study in vivo metabolism. As a result of this diversity, data acquisition and analysis are far from standardized, and researchers frequently develop customized and therefore highly heterogeneous pipelines, often interfacing with commercial or closed-source external fitting software. Here, we present “Osprey”, an all-in-one software package that incorporates all steps of state-of-the-art pre-processing, linear-combination modeling, tissue correction, quantification, and visualization of MRS data from the human brain. Osprey provides a modular, fully open-source framework, allowing the community to rapidly incorporate future methodological developments, accelerate their adaptation, and foster standardization.

Clara J. Fallone¹, Anthony G. Tessier ^1,2, Catherine J. Field³, and Atiyah Yahya^1,2
1Department of Oncology, University of Alberta, Edmonton, AB, Canada, ²Department of Medical Physics, Cross Cancer Institute, Edmonton, AB, Canada, ³Department of Agricultural, Food, and Nutritional Science, University of Alberta, Edmonton, AB, Canada

Omega-3 (ω-3) fat content in adipose tissue is relevant to the study of disease. It is challenging to quantify with Magnetic Resonance Spectroscopy (MRS) due to its low concentration in vivo. In addition, its resonance is difficult to resolve from that of the non-ω-3 protons, even at 9.4T.  A PRESS (Point RESolved Spectroscopy) echo time of 109ms was determined to yield resolved ω-3 (≈ 0.98ppm) and non-ω-3 (≈ 0.9ppm) methyl peaks at 9.4T. The efficacy of the sequence was verified on oil phantoms and on rats fed a high ω-3 fat diet. 

Lorenz Pfleger^1,2, Albrecht Ingo Schmid^2,3, Stefan Wampl^2,3, Martin Gajdošík^1,2,4, Siegfried Trattnig^2,5, Michael Krebs¹, and Martin Krššák^1,2,5
1Division of Endocrinology and Metabolism, Department of Medicine III, Medical University of Vienna, Vienna, Austria, ²High-field MR Centre, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria, ³Center for Medical Physics and Biomedical Engineering, Medical University of Vienna, Vienna, Austria, ⁴Department of Biomedical Engineering, Columbia University, New York, NY, United States, ⁵Christian Doppler Laboratory for Clinical Molecular MR Imaging, Vienna, Austria

This study focuses on the in vivo determination of the longitudinal relaxation time (T₁) of phosphatidylcholine (PtdC) in the gallbladder at 7T and its challenges. In most cases, it was possible to assess T₁ with a 1D-ISIS sequence. In cases where the gallbladder is further away from the coil a STEAM sequence, which does not depend on 180° RF pulses, can be used. The T₁ values of PtdC ranged between 290 and 480ms.

Hidehiro Watanabe¹, Nobuhiro Takaya¹, and Fumiyuki Mitsumori^1
1Center for Environmental Measurement and Analysis, National Institute for Environmental Studies, Tsukuba, Ibaraki, Japan

The method for absolute quantitation of ¹H MRS in a human brain at high B₀ field was proposed. A water phantom as a concentration reference and a human brain area measured separately. The ratio of the reception sensitivities between uniform areas in the phantom and in the human brain can be computed from measured B₁⁺s. The ratio between the VOI and the uniform area in the human brain can be computed by our previously reported ratio map method. Then, concentration in the VOI can be calculated. Our method was demonstrated in the phantom experiments.

Shinichi Shoda¹, Fuminori Hyodo¹, Norikazu Koyasu¹, Yoko Tachibana², Hinako Eto², and Masayuki Matsuo^1
1Gifu University, Gifu, Japan, ²Kyushu University, Fukuoka, Japan

  Oxidative stress is implicated in various diseases such as inflammation, neurodegenerative disorders (Alzheimer’s disease, Parkinson’s disease), atherosclerosis, diabetic and cancer. Excess reactive oxygen species (ROS) are produced during altered cellular metabolism in various diseases. Among ROSs, hydroxyl radicals (•OH) is one of the most reactive molecules in biological systems. Therefore, it is considered that monitoring of •OH is could be useful technologies for evaluation of redox mechanism and oxidative diseases. In this study, we developed the hydroxyl radical imaging technique using combining DNP-MRI and DMPO.

Navid Pourramzangandji¹, Christopher Sica², Byeong-Yeul Lee³, Hannes Wiesner³, Maryam Sarkarat⁴, Michael Lanagan⁴, Xiao-Hong Zhu³, Wei Chen³, and Qing Yang^1
1Department of Neurosurgery, PennState University College of Medicine, Hershey, PA, United States, ²Department of Radiology, PennState University College of Medicine, Hershey, PA, United States, ³CMRR, Department of Radiology, University of Minnesota, Minneapolis, MN, United States, ⁴Department of Engineering Science and Mechanics, Pennsylvania State University, State College, PA, United States

We investigated the effect of ultrahigh dielectric constant (uHDC) materials on reducing the noise and enhancing B1 efficiency at 78-MHz for low-γ nuclear applications (²³Na and ¹³C at 7T or 61-MHz for ¹⁷O at 10.5 T). By employing uHDC materials, the conservative E-field can be reduced in the sample, which makes the non-conservative field dominant and increases the transmit efficiency and receive sensitivity. We calculated the normalized noise level and extracted the B1⁺ efficiency map for baseline (no uHDC) and when uHDC is used, and show experimentally that noise level is reduced considerably in the presence of the UHDC material.

Jullie W Pan¹, Victor Yushmanov¹, Chan H Moon¹, Frank Lieberman², and Hoby P Hetherington^1
1Radiology, University of Pittsburgh, Pittsburgh, PA, United States, ²Neurology, UPMC, Pittsburgh, PA, United States

The increased SNR at 7T and high efficiency of spatial encoding from fast readout trajectories means that substantially accelerated spectroscopic imaging acquisitions are practical. In this report, we implement targeted multi-slice Hadamard encoded cascaded and simultaneous acquisitions. With the cascaded rosette, minimal CSDE is incurred, generating 4slices in 8min12s. The J-refocused sequence achieves excellent CRLB in <2.3min for a single slice. With a B1 reduction strategy that allows multiple simultaneous Hadamard encoding, the J-refocused sequence is also demonstrated with 4slice acquisitions. As performed in brain tumor patients, the targeted Hadamard rosette is well tolerated and sensitive for detection of pathology.

Chathura Kumaragamage¹, Henk M De Feyter¹, Peter B Brown¹, Scott McIntyre¹, Terence W Nixon¹, and Robin A de Graaf^1
1Radiology and Biomedical Imaging, Yale University, New Haven, CT, United States

Βeta-hydroxybutyrate (BHB) and lactate (Lac) are MRS visible metabolites important for brain energy metabolism. In this work we investigate the feasibility to detect elevated levels of BHB with proton MRSI following oral consumption of an energy drink containing 25g of a keto-ester. A gradient-modulated ECLIPSE-OVS method with minimal chemical shift displacement was implemented for lactate and BHB edited MRSI of the human brain. Elevated levels of BHB following oral intake of a keto-ester energy drink were observed in an 8 mL nominal volume MRSI acquired within  9-min  at 4 T.

Chathura Kumaragamage¹, Henk M De Feyter¹, Peter B Brown¹, Scott McIntyre¹, Terence W Nixon¹, and Robin A de Graaf^1
1Radiology and Biomedical Imaging, Yale University, New Haven, CT, United States

At high magnetic field strengths (≥3T), the utility of MRSI is impeded by challenges such as increased B₁ field heterogeneity, increased RF power requirements for reduced chemical shift displacement errors (CSDE), and water/lipid contamination. In this work, the utilization of gradient modulated (GOIA) RF pulses, is investigated, with elliptical localization with pulsed second order fields (ECLIPSE) for MRSI. A 15 kHz GOIA pulse based ECLIPSE double-spin-echo MRSI sequence was developed. In vivo data demonstrate that the GOIA based MRSI method provides full-intensity metabolite spectra in edge of the brain voxels, and undetectable extracranial lipid signals within or outside the brain.

Maria Yanez Lopez¹ and Peter J Lally^2
1Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom, ²Department of Brain Sciences, Faculty of Medicine, Imperial College London, London, United Kingdom

The aim of this work was to investigate the effect of receiver bandwidth and number of points in the fitting of MRS spectra acquired in healthy controls, using a LASER MRS sequence at 3T. Our results highlight the need for harmonisation between different studies in terms of BW/number of points wherever possible, since variations across clinical populations might not be detected due to systematic errors induced from variations in these parameters alone. Optimisation of BW/number of points is also encouraged when setting up a new study.

Ana Gogishvili^1,2, Michael Schwerter^1,3, Ezequiel Farrher¹, Ketevan Kotetishvili², and N. Jon Shah^1,4,5,6
1Institute of Neuroscience and Medicine 4, Foschungszentrum Jülich, Jülich, Germany, ²Engineering Physics Department, Georgian Technical University, Tbilisi, Georgia, ³Faculty of Mathematics, Computer Science and Natural Sciences, RWTH Aachen University, Aachen, Germany, ⁴Department of Neurology, Faculty of Medicine, RWTH Aachen University, Aachen, Germany, ⁵JARA - BRAIN - Translational Medicine, RWTH Aachen University, Aachen, Germany, ⁶Institute of Neuroscience and Medicine 11, Foschungszentrum Jülich, Jülich, Germany

Proton MRS only benefits from ultra-high field strengths if good B₀ homogeneity is achieved. For this, projection-based B₀ shim techniques, e.g. FASTMAP, are commonly used in single-voxel acquisitions. However, FASTMAP and its variants typically only use up to 2^nd order shim coils and only provide shim values for a specific voxel position. In contrast, fieldmap-based approaches for B₀ shimming have no shim order restrictions and can be used to calculate multiple shim settings over different ROIs from a single acquisition. Here we compare projection and fieldmap-based shim approaches in brain areas known to be problematic in terms of B₀ homogeneity.

Mark Mikkelsen^1,2 and Richard A. E. Edden^1,2
1Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, United States, ²F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

Measurement of lower-concentration metabolites at ≤3T using edited MRS requires relatively long scan times (~10 min for a 27-mL voxel). We investigated whether scan times can be halved to ~5 min without a significant increase in group-level variance. GABA-edited MEGA-PRESS data from the Big GABA repository were analyzed by quantifying GABA+/Cr levels for complete datasets (320 averages; ~10 min) versus the first half of each (160 averages; ~5 min). There were no significant differences in group-level variance or GABA+/Cr levels. It appears that for a large, high-SNR voxel, GABA editing scan times of ~5 min are feasible.

Ludger Starke¹, Joao dos Santos Periquito¹, Christian Prinz¹, Andreas Pohlmann¹, Thorald Niendorf^1,2, and Sonia Waiczies^1
1Berlin Ultrahigh Field Facility (B.U.F.F.), Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association, Berlin, Berlin, Germany, ²Charité Campus Buch, Experimental and Clinical Research Center (ECRC), Berlin, Germany

Fluorine-19 (¹⁹F) MRI is an established tool for tracking inflammatory cells in vivo due to its excellent detection specificity. Nonetheless, low signal-to-noise ratios remain a major challenge, especially when studying inflammatory cell distribution in the brain. It was shown that compressed sensing (CS) increases ¹⁹F MRI sensitivity, yet prospective undersampling using dedicated CS sequences has only been reported in proof of concept experiments. Since false positives were observed in CS reconstructions, this work provides a thorough assessment of detection performance of in vivo CS as a tool for enhancing ¹⁹F MRI sensitivity.

Li An¹, Maria Ferraris Araneta¹, Milalynn Victorino¹, and Jun Shen^1
1National Institute of Mental Health, National Institutes of Health, Bethesda, MD, United States

A single-step spectral editing MRS sequence is described for simultaneously measuring glutamate, glutamine, and glutathione at 7 T. Same as a previously described three-step editing technique, the editing pulse induces the collapse of targeted multiplets into high intensity pseudo singlets at a medium echo time. No data subtraction is necessary. In the current example, the editing pulse acts on the H3 protons of glutamate, glutamine, and glutathione near 2.12 ppm to generate sharp and high intensity pseudo singlets of glutamate, glutamine, and glutathione H4 protons. Signal enhancement of the three metabolites is analyzed in detail.

Helge J. Zöllner^1,2, Georg Oeltzschner^1,2, Michal Povazan^1,2, and Richard A. E. Edden^1,2
1Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, United States, ²F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

Contemporary linear-combination modeling (LCM) methods for MRS data are usually implemented in compiled, ‘black-box’ fashion. The ability to modify underlying algorithms and/or introduce novel quantification approaches may improve the transparency, robustness and accuracy of metabolite estimation, but there is no established framework for rapid prototyping of modeling algorithms and benchmarking their performance. Here, we use a large, publicly available 3T PRESS dataset from multiple sites and vendors to assess the performance of a new open-source LCM algorithm, featured in a new MRS data analysis toolbox (‘Osprey’, github.com/schorschinho/osprey). Quantification of four major metabolites is compared to the widely used LCModel algorithm.

Manoj Kumar Sarma¹, Andres Saucedo¹, Uzay E. Emir², and M. Albert Thomas^1
1Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA, United States, ²School of Health Sciences, Purdue University – West Lafayette, West Lafayette, IL, United States

We implemented a semi-LASER localized total correlated spectroscopic imaging (EP-TOCSI) sequence on a 3T MRI scanner.  Compared to L-COSY the coherence transfer 900 RF pulse was replaced by a spin-lock module consisting of adiabatic train of RF pulses using MLEV4 phase cycling; also, multi-voxel encoding included EPI read-out combined with phase-encoding. After evaluating the performance of the 4D EP-TOCSI sequence using a corn oil phantom, we investigated calf muscle of 2 healthy and 1 type 2 diabetic subjects.  There were relayed cross peaks in addition to COSY cross peaks conventionally recorded by L-COSY or EP-COSI. 

Kimberly Chan¹, Theresia Ziegs², and Anke Henning^1,2
1Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, United States, ²Max Planck Institute for Biological Cybernetics, Tuebingen, Germany

It has been shown that neural networks combined with variable k-space undersampling (MultiNet GRAPPA) is superior to a conventional GRAPPA reconstruction at 9.4T.  Here, the feasibility of performing MultiNet GRAPPA for 1H FID-MRSI at 7T is investigated with and without novel modifications to the original acquisition/reconstruction scheme.  In this study, it is shown that MultiNet GRAPPA is shown to be feasible for 1H MRSI acceleration at 7T with a new k-space undersampling scheme for higher signal-to-noise and increased map reliability and use of a novel technique to increase SNR retention using semi-synthetic calibration data without an increase in acquisition time.

Rama Jayasundar¹, Aruna Singh¹, and Dushyant Kumar^1
1Department of NMR, All India Institute of Medical Sciences, New Delhi, India

Given the alarming increase in obesity and associated NCDs, use of nutritional and medicinal plants in obesity management have come into focus. There is growing interest in pungent  thermogenic plants with anti-obesity activity. With this in view, the study has explored the potential of NMR metabolomics to identify plants with anti-obesity activity based on their chemosensory and thermogenic properties, the latter evaluated by Differential Scanning Calorimeter. Multivariate analysis of NMR data discriminated thermogenic (pungent) and non-thermogenic (sweet) plants and also correlated well with the calorimetry analyses. NMR can identify quickly plants with anti-obesity potential using their chemosensory associated thermogenic properties.

Andrey Pravdivtsev¹, Frank D. Sönnichsen², and Jan-Bernd Hövener^1
1Medical faculty, MOIN CC, Kiel University, Kiel, Germany, ²Otto Diels Institute for Organic Chemistry, Kiel University, Kiel, Germany

MRS provides a non-invasive window into an in vivo biochemistry – basically a virtual biopsy. Here, we present two novel MRS sequences (SISTEM-1 and 2) and method to selectively prepare and detect the signal of endogenous metabolites. The long-lived signal of NAA-CH2 protons was isolated by the strong suppression of H2O and lipid signals using broad-band saturation or double-quantum filter. The frequency of the modulated NAA-CH2 signal was measured and associated with a pH value. Non-invasive measurement of pH value, imaging of drug distribution are only some of the possible applications.

Dunja Simicic^1,2, Veronika Rackayova², and Cristina Cudalbu^2
1LIFMET, EPFL, Lausanne, Switzerland, ²CIBM, EPFL, Lausanne, Switzerland

At short echo times 1H-MR spectra contain the contribution of mobile macromolecules (MM), i.e. broader resonances characterized by shorter relaxation times (T₁,T₂) which underlie the narrower peaks of metabolites. There are very few studies assessing MM T₂ relaxation times, with only one study reporting T₂-s of individual MM peaks in the full ppm range at 9.4T in the human brain. In this work we present a new approach: single inversion recovery with an optimized inversion time combined with AMARES post-processing. Using this technique we quantified 10-MM components and estimated T₂ relaxation times (for 7-MM components) in rat brain at 9.4T.

Reyhaneh Nosrati^1,2, Mukund Balasubramanian^1,2, and Robert Mulkern^1,2
1Radiology, Boston Children's Hospital, Boston, MA, United States, ²Harvard Medical School, Boston, MA, United States

Several studies have reported strong correlations between transverse relaxation rate (R₂) of the brain metabolites and neurological/psychiatric diseases. Typically, R₂ quantification requires several PRESS acquisitions at different echo-times leading to long scan times. Here, we evaluated the feasibility of a novel technique based on a single-echo PRESS acquisition with full echo sampling for R₂ estimation of major brain metabolites (NAA, Cr and Cho) and compared those to the values obtained using five PRESS acquisitions. The proposed method allowed for R₂ estimation with 87.7±6.8% accuracy compared to the multiple PRESS acquisitions representing a five-fold decrease in acquisition time in this case.

Li An¹, Maria Ferraris Araneta¹, Milalynn Victorino¹, and Jun Shen^1
1National Institute of Mental Health, National Institutes of Health, Bethesda, MD, United States

Optimal detection of many metabolites requires a specific TE, at which significant macromolecule signals are often present. The presence of macromolecule signals complicates the measurement of metabolite T₁ because different spectral models may be necessary for macromolecules at different inversion recovery stage, therefore, potentially introducing additional errors. To minimize interference from macromolecules, we chose inversion times in such a way that the metabolite signals were changed by as much as 60% while the macromolecule signals were essentially unchanged. This avoided the T₁ relaxation effect of macromolecules, and thus simplified data acquisition and post-processing.

Peter Truong¹, Napapon Sailasuta^1,2, and Sofia Chavez^1,2
1Research Imaging Centre - MRI, Centre for Addiction and Mental Health, Toronto, ON, Canada, ²Psychiatry, University of Toronto, Toronto, ON, Canada

Higher-numbered phased-array coils have been associated with significant increases in SNR. For this gain, one would expect more accurate and precise MRS measurements for a regular PRESS sequence processed with a fully automated program such as LCModel. However, the effects of increased SNR on a J-edited technique such as MEGA-PRESS for GABA measurements are not as straight-forward. In this study, we wish to compare the reproducibility of GABA+ and Glx measurements from an 8-channel and a 32-channel head coil in two cortical regions of interest: Anterior Cingulate Cortex (ACC) and Left-Dorsolateral Prefrontal Cortex (Lt-DLPFC).

Alireza Abaei¹, Dinesh K Deelchand², Francesco Roselli³, and Volker Rasche^1
1Core Facility Small Animal Imaging, Ulm University, Medical Center, ulm, Germany, ²Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States, ³German Center for Neurodegenerative Diseases (DZNE), ulm, Germany

Using sub-microlitre voxel size in magnetic resonance spectroscopy (MRS) is advantageous to investigate specific small brain region without any partial volume effect. The aim of this study was to determine the intra-individual coefficient-of-variation of neurochemical concentrations when using sub-microlitre MRS voxel measured during 3 consecutive days. This study was performed in adult mice at 11.7T. The intra-individual CV for the major and high concentration brain metabolites was ≤ 9% for MRS data collected in pre-clinically feasible scan times. This in turn shows that sub-microlitre can be used in pre-clinical application to detect subtle changes in neurochemical profile in diseased brain.

Dinesh K Deelchand¹ and Pierre-Gilles Henry^1
1Center for Magnetic Resonance Research, Radiology, University of Minnesota, Minneapolis, MN, United States

This study investigates the effect of water suppression and metabolite cycling on metabolite quantification and macromolecules in ¹H MRS. Single-voxel semi-LASER spectra and metabolite-nulled spectra (macromolecules) were acquired in the posterior cingulate cortex in the human brain at 3T using three different schemes: VAPOR, metabolite-cycling (MC), and metabolite-cycling combined with a single water suppression pulse. Results show that macromolecule signal and several metabolites (ascorbate, total creatine, glutamate, and glutathione) have higher in concentration when using MC compared to VAPOR. Combining MC with a single water suppression pulse results in concentrations in between those obtained with VAPOR and MC.

Srijyotsna Volety¹, Elizabeth Seaquist², Gulin Oz³, and Uzay Emir^1,4
1Health Sciences, Purdue University, West Lafayette, IN, United States, ²Department of Medicine, Medical School, University of Minnesota, Minneapolis, MN, United States, ³Center for Magnetic Resonance Research, Department of Radiology, University of Minnesota, Minneapolis, MN, United States, ⁴Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, United States

MRS quantification tools such as LCModel require the use of prior knowledge and can be time-consuming. Therefore, we propose an untargeted metabolomics approach to in vivo 1H-MRS using pattern recognition and machine learning to analyze spectral features. The ability of our approach to measure changes in brain glucose while blood glucose levels were increased was studied using high quality spectra and reliable data acquisition methods. Results showed similar time-course glucose signals and sensitivity to changes in glucose concentrations for both LCModel and our pattern recognition analysis. Thus, demonstrating that untargeted metabolomics techniques can be used for in vivo MRS quantification.

Steve CN Hui^1,2, Muhammad G Saleh^1,2, Georg Oeltzschner^1,2, Mark Mikkelsen^1,2, Sofie Tapper^1,2, and Richard AE Edden^1,2
1Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University, Baltimore, MD, United States, ²F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

An improved editing scheme was proposed for the Hadamard Editing Resolves Chemicals Using Linear-combination Estimation of Spectra (HERCULES). The original editing pulse at 4.18 ppm was replaced by a new wider-band pulse at 4.04 ppm to facilitate signal acquisition for aspartate, ascorbate and lactate. Furthermore, the new scheme (HERCULES-2) was implemented within the semi-LASER and PRESS sequences. HERCULES-2 with semi-LASER yielded increases in overall signal intensities in most targeted metabolites, robust to B₁ inhomogeneity, facilitating the precise quantification of edited spectra.

Hikari Yoshihara¹, Nicolas Kunz², Jean-Claude Martinou³, and Hongxia Lei^2
1Laboratory for Metabolic Imaging, Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ²Animal Imaging and Technology Core (AIT), Center for Biomedical Imaging (CIBM), Ecole Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, ³Faculty of Science, University of Geneva, Geneva, Switzerland

Dynamic shimming updates with first- and second-order shims are essential for dual-voxel spectroscopy of mouse cortex and hippocampus at high magnetic field. We show that DSU is feasible for studying pharmacological effects in transgenic mice at 14.1 T.

Karl Landheer¹, Hetty Prinsen², Ognen A Petroff³, Douglas L Rothman², and Christoph Juchem^1,4
1Biomedical Engineering, Columbia University, New York, NY, United States, ²Radiology and Biomedical Imaging, Yale University, New Haven, CT, United States, ³Neurology, Yale University, New Haven, CT, United States, ⁴Radiology, Columbia University, New York, NY, United States

The aim of this work was to develop a novel method to assess accurately both the relative concentrations of homocarnosine as well as GABA free from overlapping creatine, homocarnosine and macromolecule signal. This was achieved via the combination of short echo time STEAM and MEGA-sLASER experiments at 7T. The metabolites GABA and homocarnosine were measured in 6 healthy control subjects, and in a single subject medicated with isoniazid. It was found that (16.6 ± 10.2)% of the supposed GABA signal originated from homocarnosine, and that isoniazid caused a significantly elevated concentration of GABA and homocarnosine in a single subject.

Song-I Lim¹ and Lijing Xin^1
1Centre d'Imagerie BioMédicale (CIBM), Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland

The aim of the study is to improve the efficiency of GABA editing and macromolecule suppression at 7T. An asymmetric adiabatic pulse with broad inversion bandwidth was applied at 1.9 ppm to invert 1.5 – 1.9 ppm range and a narrow Gaussian pulse with 95 Hz of bandwidth (0.95 < M_z) was applied at 1.7 ppm to suppress macromolecule. In the GABA and lysine phantom tests, this scheme shows ± 9 Hz of signal loss threshold in comparison with highly selective inversion pulse without macromolecule suppression (± 6 Hz of threshold) and symmetric pulses (± 3 Hz of threshold).

Zahra Shams¹, Vincent Oltman Boer², Dennis W.J. Klomp¹, Jannie P. Wijnen¹, and Evita Wiegers^1
1Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ²Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Hvidovre, Hvidovre, Denmark

This study compares different crushing schemes for short-TE semi-LASER at 7T; minimal crushers, DOTCOPS and newly designed Adaptive crusher scheme. Maintaining a short echo time is traded for optimal spoiling of unwanted coherences.

Vasiliki Mallikourti¹, Sai Man Cheung¹, Tanja Gagliardi ^2,3, Yazan Masannat⁴, Steven Heys^1,4, and Jiabao He^1
1Institute of Medical Sciences, School of Medicine, University of Aberdeen, Aberdeen, United Kingdom, ²Department of Clinical Radiology, Aberdeen Royal Infirmary, Aberdeen, United Kingdom, ³Department of Radiology, Royal Marsden Hospital, London, United Kingdom, ⁴Breast Unit, Aberdeen Royal Infirmary, Aberdeen, United Kingdom

Phased-array signal combination algorithms of adaptively optimised combination (AOC), noise decorrelated combination (nd-comb), whitened singular value decomposition (WSVD), S/N², S/N, and Signal Weighting, were evaluated for enhancing the SNR of polyunsaturated fatty acids spectra acquired using MQC MRS from breast cancer. Data were acquired from seventeen breast tumours excised from patients, fifteen healthy volunteers and five patients with breast cancer. AOC yielded the maximum SNR improvement and was independent from voxel volume, PUFA content, or water/fat ratio. AOC improved SNR by 24% (relative to non-noise decorrelated algorithms) and shortened acquisition time in patients by 1.5 times while maintaining SNR. 

Yuqing Huang¹, Haolin Zhan¹, and Zhong Chen^1
1Department of Electronic Science, Xiamen University, Xiamen, China

Measurements on heterogeneous samples, such as semisolid food and biological tissues, constitute a significant research topic in various fields. However, measurements on heterogeneous samples by conventional 1D 1H NMR generally suffer from two challenges, namely magnetic field inhomogeneity caused by anisotropic interactions and spectral congestion induced by crowded NMR resonances. In this report, we introduce a spatially-selective pure shift NMR approach for high-resolution measurements on heterogeneous samples based on the suppression on effects of magnetic field inhomogeneity and J couplings, thus useful for investigating heterogeneous sample systems with extensive chemical and biological applications.

Rudy Rizzo¹ and Roland Kreis^1
1Department of Radiology and Biomedical Research, University of Bern, Bern, Switzerland

The benefits of multi-echo single-shot (MESS) spectroscopy are explored aiming at simultaneous determination of metabolite content and T₂ times through simultaneous linear-combination model fitting of partially sampled echoes. Cramer-Rao lower bounds (CRLB) and Monte-Carlo simulations are used to judge this benefit. The novel scheme was compared with traditional multi-echo multi-shot and single-echo methods, exploring different TE settings for spectra of the major brain metabolites. Results indicate that MESS outperforms older methods for simultaneous determinations of T₂s and concentrations, with improvements ranging at 20-30% for T₂s and 30-50% for areas.  However, for concentrations alone traditional single-echo sequences are more sensitive.

Mohammed Goryawala¹, Sulaiman Sheriff¹, Ronald Ouwerkerk², Hari Hariharan³, Peter Barker⁴, Hyunsuk Shim⁵, and Andrew Maudsley^1
1Radiology, University of Miami, Miami, FL, United States, ²Biomedical and Metabolic Imaging Branch, The National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), Bethesda, MD, United States, ³Center for Magnetic Resonance & Optical Imaging, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States, ⁴Department of Radiology and Oncology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ⁵Department of Radiation Oncology, Emory University School of Medicine, Atlanta, GA, United States

Whole-brain Magnetic Resonance Spectroscopic Imaging (MRSI) is an effective technique for non-invasive quantification of brain metabolite levels ^1-3 that can be used to create maps for the study of both regional or diffuse metabolic alterations in various pathologies. Improved spatial resolution can result in significantly better mapping but is often limited by the signal-to-noise ratio (SNR). This abstract presents a whole-brain 3D MRSI acquisition scheme that uses hypergeometric dual-band (HGDB) pulses for lipid suppression over the brain volume, real-time frequency drift correction, and novel post-processing methods to generate whole-brain metabolite maps in humans at 3T.

Jullie W Pan¹, Victor Yushmanov¹, Chan H Moon¹, and Hoby P Hetherington^1
1Radiology, University of Pittsburgh, Pittsburgh, PA, United States

We have developed and implemented a fast spectroscopic imaging acquisition for selective homonuclear polarization transfer sequence for the detection of GABA at 7T. With the high SNR available at 7T, whole slice studies of 1.65cc nominal resolution are achievable with acquisition durations (5 to 10min) conventionally used in single voxel studies. As this editing is performed with longitudinal magnetization, this method is single shot and is flexible for evaluation of multiple coupled spins.

Alexandra Lipka^1,2, Lukas Hingerl¹, Gilbert Hangel¹, Eva Heckova¹, Stanislav Motyka¹, Verena Endmayr³, Romana Höftberger³, Simon Hametner⁴, Siegfried Trattnig^1,2, and Wolfgang Bogner^1,2
1Highfield MR Centre, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria, ²Christian Doppler Laboratory for Clinical Molecular MR Imaging, Vienna, Austria, ³Institute for Neurology, Medical University of Vienna, Vienna, Austria, ⁴Center for Brain Research, Institute of Neuroimmunology, Medical University of Vienna, Vienna, Austria

MRS has rarely been applied for postmortem measurements except for single-voxel studies, which can be explained by the fact that rapid degradation of metabolite levels makes scanning of fresh specimen highly critical.SVS contrary to MRSI cannot cover whole affected regions, while pathological changes may not be visible on anatomical images, yet. To investigate changes ex-vivo, a true whole-brain MRSI approach is critical. Our study shows the potential of ultra-highfield whole-brain FID-MRSI for metabolic fingerprinting of fresh postmortem brains for correlation with histology, as the link between neurochemical changes and underlying cell-types in many demyelinating/neurodegenerative disorders is still not fully understood.

Edwin Versteeg¹, Kyungmin Nam¹, Dennis Klomp¹, Jeroen Hendrikse¹, Alex Bhogal¹, Jeroen Siero^1,2, and Jannie Wijnen^1
1Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ²Spinoza Centre for Neuroimaging Amsterdam, Amsterdam, Netherlands

Echo planar spectroscopic imaging (EPSI) can be used for fast spectroscopic imaging, however, the gradient hardware greatly limits the spectral bandwidth. Moreover, the associated fast switching gradient produces uncomfortable levels of acoustic noise. In this work, we show the feasibility of spectroscopic imaging at high spectral bandwidth using a silent gradient axis driven at 20 kHz and a segmented EPSI sequence. We report results on the silent segmented EPSI compared to the conventional EPSI approach on a phantom containing NAA, creatine and glutamate and show the first silent and high bandwidth in-vivo measurements.

Seung-Cheol Lee¹, Hari Hariharan¹, Fernando Arias-Mendoza¹, Gabor Mizsei¹, Kavindra Nath¹, Sanjeev Chawla¹, Stephen J. Schuster¹, Ravinder Reddy¹, and Jerry Glickson^1
1University of Pennsylvania, Philadelphia, PA, United States

Lactate is an important marker of tumor metabolism. We have previously reported the Hadamard slice selected, selective multiple quantum coherence chemical shift imaging sequence for a 3T clinical scanner to image lactate in the human body. Here we report modification and optimization of the sequence for reducing motion effect and increasing lipid suppression while retaining lactate. A coherence selection gradient ratio of a:-1:2 with nonzero a provided improved suppression of lipid resonances at 1.3 ppm and 0.9 ppm compared to the previous 0:-1:2 ratio. Optimization of RF pulses blocked occurrence of MQC between lipid 2.0 ppm and 5.3 ppm resonances.

Dinesh K Deelchand¹ and Pierre-Gilles Henry^1
1Center for Magnetic Resonance Research, Radiology, University of Minnesota, Minneapolis, MN, United States

The goal of this study was to compare the spectral quality and metabolite concentrations obtained with short echo-time 2D spiral MRSI and single-voxel spectroscopy (SVS) in the human brain at 3T. Semi-LASER was used for localization (SVS) or pre-localization (MRSI). MRSI was acquired in a transverse slice through the posterior cingulate cortex (PCC), and SVS was from PCC. Results show comparable spectral quality between acquisitions. Differences in concentrations between the two techniques is likely due to the MRSI point-spread function. This study shows feasibility of acquiring high-quality spiral MRSI data similar in spectral quality to SVS within a few minutes.

Zepeng Wang^1,2 and Fan Lam^1,2
1Bioengineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States

This work presents a new method for B₀ inhomogeneitycorrected reconstruction of J-resolved (multi-TE) MRSI datawith limited k-space coverage. A constrained reconstruction formulation was proposed to simultaneously reconstruct the data at all TEs, using a joint low-rank constraint that exploits the spatial-spectral-TE correlation in the data and a spatial constraint that exploits prior edge information from high-resolution anatomical images. Simulation and in vivo experiments using a customized 3D J-resolved EPSI acquisition have been performed to evaluate the proposed method. Experimental results demonstrated improved lineshape and SNR achieved by the proposed method over existing B₀ correction methods.

Puneet Bagga¹, Hari Hariharan¹, Walter R Witschey¹, and Ravinder Reddy^1
1Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States

Downfield MRS provides a great tool to detect and quantify metabolites downfield to water including NAD and ATP. Acquisition of downfield MR spectrum in vivo without water suppression leads to the possibility of detection of many cross-relaxing/exchanging resonances. The T2 of many species occurring downfield of water are relatively short which require short echo time MRS methods. EBURP pulses have been shown to reliably excite narrow spectral region with minimum phase response. In this study, we generated new pulse sequences for single voxel and 3D spectral selective EBURP excitation to detect cerebral NAD from human brain at 7T.

Sunitha Thakur¹, Olivia Sutton¹, Eduardo Coello², Alexander Lin², Ralph Noeske³, Ricardo Otazo¹, and Robert Young^1
1Memorial Sloan Kettering Cancer Center, New York, NY, United States, ²Brigham and Women’s Hospital, Boston, MA, United States, ³MR Applied Science Lab, GE Healthcare, Berlin, Germany

 The aim of this study was to develop and test rapid MR spectroscopic imaging (MRSI) in patients with brain tumors. We implemented and optimized rapid MRSI at 3T using semi-localization by adiabatic selective refocusing pulses (semi-LASER), fast spatial encoding using symmetric echo planar spectroscopic imaging (EPSI), and robust water suppression with variable power and optimized relaxation delays. This rapid imaging approach was tested in a phantom and subsequently in a volunteer and a patient. In vivo studies demonstrated that high quality MRSI data with voxel volumes less than 0.5 mL can be collected within a clinically feasible time (2 minutes).

Carlijn Tenbergen¹, Loreen Ruhm², Arend Heerschap¹, Anke Henning^2,3, and Tom Scheenen^1
1Radiology and Nuclear Medicine, Radboud University Medical Center, Nijmegen, Netherlands, ²High-Field MR Center, Max Planck Institute for Biological Cybernetics, Tübingen, Germany, ³Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, United States

In in vivo MRSI spatial inhomogeneities of the main magnetic field cause spectral line broadening and location dependent frequency shifts in the spectra. An over-discretized reconstruction method is applied to existing prostate ¹H-MRSI data, consisting of spatial over-discretization of the MRSI grid, frequency shift correction according to a high resolution B0map and weighted spatial averaging to the target resolution. Intravoxel B0 variations are corrected for and resampling to a new target resolution results in improved localization of signals. This could lead to detection of signals with lower spectral intensity and to improved visualization of smaller cancer lesions in the prostate.

Tiffany Bell^1,2,3, Dana Goerzen⁴, Jamie Near⁴, and Ashley D Harris^1,2,3
1Department of Radiology, University of Calgary, Calgary, AB, Canada, ²Hotchkiss Brain Institute, Calgary, AB, Canada, ³Alberta Children's Hospital Research Institute, Calgary, AB, Canada, ⁴Douglas Mental Health Institute, Centre d’Imagerie Cerebrale, Montreal, QC, Canada

Separating glutamate (Glu) from its precursor, glutamine (Gln) at 3T is challenging due to signal overlap. Multiple sequence optimisations have been suggested, with simulated data used to assess accuracy. Here, using 7T data as a reference, we investigate the ability of four PRESS sequences with different echo times of separating Glu and Gln. We found that Glu measured at 3T using TE=20ms shows the highest agreement with 7T data. Gln measured using TE=20ms, 30ms and 40ms all agree with Gln measured at 7T. Therefore, we conclude PRESS with TE=20ms is the most effective at separating and quantifying Glu and Gln.

Eduardo Coello¹, Nicolas R. Bolo², Huijun Liao¹, Angelina Awad³, Margaret A. Niznikiewicz³, and Alexander Lin^1
1Radiology, Brigham and Women's Hospital, Boston, MA, United States, ²Psychiatry, Beth Israel Deaconess Medical Center, Boston, MA, United States, ³Psychiatry, Veterans Affairs Boston Healthcare System, Boston, MA, United States

This work describes a methodology to perform reproducible functional MR spectroscopy (FMRS) experiments to study auditory processing at 7T. A processing pipeline and quality control steps are proposed to remove distortions and temporal instabilities in the dynamic experiment. The designed experiment was successfully tested, producing highly correlated results in healthy volunteers and patients with schizophrenia scanned multiple times.

Yasmin Mzayek^1,2, Francesca Branzoli³, Thomas Troalen⁴, Yann Le Fur^1,2, Patrick Viout^1,2, Tangi Roussel^1,2, Stanislas Rapacchi^1,2, Maxime Guye^1,2, Itamar Ronen⁵, Wafaa ZAARAOUI^1,2, and Jean-Philippe RANJEVA^1,2
1Aix-Marseille Université, CNRS, CRMBM, Marseille, France, ²APHM, Hôpital de la Timone, Pôle d’Imagerie Médicale, CEMEREM, Marseille, France, ³Brain and Spine Institute - ICM, Centre for NeuroImaging Research - CENIR, Paris, France, ⁴Siemens Healthineers SAS, Saint-Denis, France, ⁵Department of Radiology Leiden University Medical Center, Leiden, Netherlands

We implemented a diffusion-weighted semi-LASER single voxel spectroscopy sequence with no water suppression to measure mean diffusivity and fractional anisotropy within two volumes-of-interest in the corpus callosum. Without water suppression, we were able to decrease acquisition time and use the water signal for phase, frequency, and eddy current corrections. We measured the signal decay of N-acetylaspartate (tNAA), creatine (tCr), choline (tCho), and water to derive diffusion outcomes and compare between the two volumes-of-interest. Our implementation was done in feasible time for in vivo measurements. Further development can lead to more specific assessments of brain microstructure and pathophysiology.

Ariane Fillmer¹, Layla Tabea Riemann¹, Frank Seifert¹, Semiha Aydin¹, Harald Pfeiffer¹, and Bernd Ittermann^1
1Physikalisch-Technische Bundesanstalt (PTB), Braunschweig und Berlin, Germany

To investigate underlying mechanisms of psychiatric diseases, such as schizophrenia, magnetic resonance spectroscopy is a powerful tool. Here, the medial prefrontal cortex is a region of high interest, however, high quality measurements are technically challenging, and proper absolute quantification is hindered by the lack of literature values for metabolite relaxation times. This work presents the acquisition of high-quality MR spectra from the medial prefrontal cortex, and the preliminary calculation of metabolite T₂ relaxation times, using an 8Tx/8Rx channel coil with a dedicated phase set optimized for B₁ efficiency in the frontal cortex.

Liangjie Lin^1,2 and Zhong Chen^2
1Philips Healthcare, Beijing, China, ²Department of Electronic Science, Xiamen University, Xiamen, China

Two-dimensional (2D) MR spectroscopy enables feasible detection of low-concentration scalar-coupled metabolites in human tissues. However, the traditional acquisition of 2D MRS can be time consuming, and cross peaks which contain crucial spectral information in 2D correlated spectra (COSY) usually get contaminated by nearby strong diagonal peaks. Here, a pulse sequence is developed for ultrafast recording of diagonal-free 2D MR correlated spectra, thus allowing improved recognition and measurement of scalar-coupled metabolites.

James M Joers¹, Dinesh K Deelchand¹, Bin Guo², Lynn E Eberly², Young Woo Park¹, Adam Berrington³, Michal Povazan⁴, Andre van der Kouwe⁵, Khalaf Bushara⁶, Chiadi U Onyike⁷, Jeremy Schmahmann⁸, Peter B Barker^4,9, Eva Ratai⁵, and Gülin Öz^1
1CMRR/Radiology, University of Minnesota, Minneapolis, MN, United States, ²Division of Biostatistics, University of Minnesota, Minneapolis, MN, United States, ³Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy, University of Nottingham, Nottingham, United Kingdom, ⁴Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ⁵A.A. Martinos Center for Biomedical Imaging / Radiology, Massachusetts General Hospital, Harvard Medical School, Charlestown, MA, United States, ⁶Neurology, University of Minnesota Medical School, Minneapolis, MN, United States, ⁷Psychiatry and Behavioral Sciences, Johns Hopkins University, Baltimore, MD, United States, ⁸Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States, ⁹Kennedy Krieger Institute, Johns Hopkins University School of Medicine, Baltimore, MD, United States

The reproducibility of an advanced single voxel MRS protocol was tested in a clinical cohort and control subjects at 3T in a multi-site setting. A standardized MRS protocol was implemented on two MR system vendors (Siemens, Philips) over three sites. This protocol utilized automated voxel placement and an sLASER pulse sequence with identical radiofrequency waveforms, pulse durations, interpulse delays and gradient spoilers. Spectral quality, metabolite concentrations and test-retest reproducibility results from three voxels are reported from both the healthy control group and subjects with genetically-confirmed spinocerebellar ataxia type 3 (SCA3), a hereditary movement disorder.