Hoby Patrick Hetherington¹, Tiejun Zhao², Victor Yushmanov¹, and Jullie Pan^3
1Radiology, University of Pittsburgh, Pittsburgh, PA, United States, ²Siemens Medical Systems, New York, NY, United States, ³Neurology, University of Pittsburgh, Pittsburgh, PA, United States

To provide near whole brain coverage for both anatomical imaging and MRSI we used an 8x2 transceiver array with 8 independent RF channels and eight 1 to 2 splitters. This configuration provided a homogeneous RF distribution (<12% SD, 750Hz peak B1) while enabling 3D RF shimming based outer volume suppression to minimize extra-cerebral lipid signals. MRSI data was acquired at 7T from control subjects and patients with mTBI with a multi-band MRSI sequence (four simultaneous slices) using two RF distributions. Increases in choline/NAA were seen in both the anterior frontal lobe and the hippocampi.

Chao Ma¹, Fan Lam¹, Qiegen Liu¹, and Zhi-Pei Liang^1,2
1Beckman Institute, University of Illinois Urbana-Champaign, Urbana, IL, United States, ²Electrical and Computer Engineering, University of Illinois Urbana-Champaign, Urbana, IL, United States

Multidimensional spectroscopy increases spectral dispersion and enables accurate detection of more metabolites (e.g., Glu and GABA in 1H-MRSI of the brain) whose spectra largely overlap with other metabolites. However, the additional dimension of spectral information is obtained at the cost of increased data acquisition time, limiting the practical utility of in vivo multidimensional MRSI. This work presents a novel tensor-based approach to accelerated high-resolution multidimensional 1H-MRSI. The proposed method has been validated using  phantom and in vivo J-resolved 2D 1H-MRSI experimental studies on a 3T scanner, producing encouraging results. The method should enhance the practical utility of multidimensional MRSI.

Eren Kizildag¹, Jason P Stockmann², Borjan Gagoski^2,3,4, Bastien Guerin^2,4, P. Ellen Grant^2,3,4, Lawrence L. Wald^2,4, and Elfar Adalsteinsson^1,5,6
1Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States, ²A. A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, United States, ³Boston Children’s Hospital, Boston, MA, United States, ⁴Harvard Medical School, Boston, MA, United States, ⁵Harvard-MIT Health Sciences and Technology, Cambridge, MA, United States, ⁶Institute for Medical Engineering and Science, Cambridge, MA, United States

Severe B₀ inhomogeneity manifests itself in the in vivo brain Chemical Shift Imaging (CSI) by broadening the lineshapes and diminishing the quality of the observed spectra. We mitigate this problem by employing a 32-channel integrated RF-shim coil array which uses an optimal combination of local B₀ fields from each coil to cancel higher order local field inhomogeneities in the CSI volume. We observed 50% reduction in  ΔσB₀ over the slab as compared with 2nd order shimming, corresponding to pronounced improvements in the linewidths of 13 out of 24 CSI voxels while modestly worsening in only 3 voxels. 

Aline Seuwen¹, Markus Wick², Franek Hennel¹, Aileen Schroeter¹, and Markus Rudin^1,3
1Institute for biomedical engineering, ETH & University of Zürich, Zürich, Switzerland, ²Bruker BioSpin MRI GmbH, Ettlingen, Germany, ³Institute for pharmacology and toxicology, University of Zürich, Zürich, Switzerland

In order to increase the volume coverage of 2D FID based spectroscopic imaging in mice i.e. the simultaneous measurement of several brain slices, we implemented a dynamic shimming approach involving the separate optimization of first and second order shim terms for volumes of interest in individual slices. When acquiring two slices covering cortical and thalamic regions similar spectra quality has been observed in both slices using dynamic shimming as compared to measuring each slice individually. This allows simultaneous acquisition of metabolite signal changes in several brain regions associated with stimulus evoked neural activity upon sensory stimulation.

Irene Marco-Rius¹, Peng Cao¹, Cornelius von Morze¹, Matthew Merritt², Karlos X Moreno³, Gene-Yuan Chang⁴, Michael A Ohliger¹, David Pearce⁴, John Kurhanewicz¹, Peder EZ Larson¹, and Daniel B Vigneron^1
1Department of Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, ²Department of Biochemistry and Molecular Biology, University of Florida, Gainesville, FL, United States, ³Department of Chemistry, Engineering, Pre-Pharmacy, and Physics, South Texas College, Weslaco, TX, United States, ⁴Department of Medicine, Division of Nephrology, University of California San Francisco, San Francisco, CA, United States

¹³C-MR spectra of hyperpolarized [2-¹³C]dihydroxyacetone (DHAc), a new agent for imaging gluconeogenesis, was acquired using specialized acquisition methods in the rat liver and kidney in vivo. Because the resonances originating from the metabolism of [2-¹³C]DHAc have a large frequency distribution, we designed a novel spectral-spatial (SPSP), multi-band excitation pulse that corrects for chemical shift misregistration, resulting in accurate spatial-spectral selectivity. The metabolic products phosphoenolpyruvate (PEP) and glycerol 3-phosphate (G3P) were detected, evidencing metabolism of the hyperpolarized substrate towards the glycolytic pathway and activity of the enzyme glycerol 3-phosphate dehydrogenase.

Rohini Vidya Shankar¹, Shubhangi Agarwal¹, and Vikram D Kodibagkar^1
1Biomedical Engineering, Arizona State University, Tempe, AZ, United States

Lactate plays a key role in the development and progression of tumors and its spatial profile can be mapped using magnetic resonance spectroscopic imaging (MRSI). However, the long scan time involved in MRSI acquisitions is a deterrent to its inclusion in routine clinical protocols. A MRSI sequence containing lactate editing components combined with prospective compressed sensing acquisitions was developed for fast mapping of lactate metabolism, particularly in response to treatment. Results from in vivo experiments demonstrate a reduction in acquisition time by up to 80%, with the accelerated MRSI datasets maintaining high fidelity with the fully sampled reference dataset.

Ipshita Bhattacharya¹ and Mathews Jacob^1
1Department of Electrical and Computer Engineering, The University of Iowa, Iowa City, IA, United States

A novel compartmental low rank algorithm and data acquisition method for high resolution MR spectroscopic imaging without the use of any lipid suppression methods is introduced. The field inhomogeneity compensated data is modeled as the sum of a lipid dataset and a metabolite dataset using the spatial compartmental information obtained from the water reference data. These datasets are modelled to be low-rank subspaces and are assumed to be mutually orthogonal. The high resolution spiral acquisition method achieves in plane resolution of upto 1.8x1.8 mm² in 7.2 mins. Recovery from these measurements is posed as a low rank recovery problem. Experiments on in-vivo data demonstrates comparable results for both lipid suppressed and lipid unsuppressed data.

Fan Lam¹, Hanbing Lu², Yihong Yang², Bryan Clifford^1,3, Chao Ma¹, Gene E Robinson⁴, and Zhi-Pei Liang^1,3
1Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Neuroimaging Research Branch, National Institute on Drug Abuse, Baltimore, MD, United States, ³Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁴Carl R. Woese Institute for Genomic Biology, University of Illinois at Urbana-Champaign, Urbana, IL, United States

We present a multislice short-TE 1H-MRSI method to achieve fast, ultrahigh-resolution metabolic imaging of rats on a 9.4 Tesla animal scanner. The proposed method uses a subspace-based hybrid data acquisition strategy and a low-rank-model-based image reconstruction scheme. In vivo experiments have been performed to demonstrate the feasibility of the proposed method. We are able to produce high-SNR, spatially resolved metabolic profiles from the rat brain with 1x1x2mm³ nominal resolution in 16 minutes.

Eduardo Coello^1,2, Martin Janich², Timo Schirmer², Ralf Noeske³, Tamas Borbath², Axel Haase¹, and Rolf Schulte^2
1Technische Universität München, Munich, Germany, ²GE Global Research, Garching, Germany, ³GE Healthcare, Potsdam, Germany

An overdiscrete reconstruction for in-vivo 3D Echo-Planar Spectroscopic Imaging (EPSI) data is used for SNR improvement and voxel bleeding reduction. We propose the estimation of a B₀ field map, which is needed for the reconstruction, using the residual water signal in the dataset. A mean SNR enhancement of a factor of 2.8 was achieved for NAA and comparable reconstruction results were obtained with both the measured and the estimated B₀ field maps.

Gilbert Hangel¹, Bernhard Strasser², Michal Považan², Lukas Hingerl¹, Marek Chmelík², Stephan Gruber², Siegfried Trattnig^2,3, and Wolfgang Bogner^2
1MR Centre of Excellence, Medical University of Vienna, Vienna, Austria, ²MRCE, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria, ³Christian Doppler Laboratory for Clinical Molecular MR Imaging, Vienna, Austria

Increasing the resolution of MRSI is desirable to delineate small structures and pathologic deviations such as Multiple Sclerosis lesions and increase local B0-homogeneity per voxel. We show that using an FID-MRSI sequence with short TR and L2-regularisation for lipid contamination removal, the major brain metabolites can be mapped with a 128x128 matrix over a whole brain slice with unprecedented detail, with a nominal voxel volume of 1.7×1.7×8 mm³. The additional application of parallel imaging allows reducing measurement times enough for potential clinical applications.