Ioannis Angelos Giapitzakis^1,2, Roland Kreis ³, and Anke Henning ^1,4
1Max Planck Institute for Biological Cybernetics, Tübingen, Germany, ²IMPRS for Cognitive and Systems Neuroscience, University of Tuebingen, Tuebingen, Germany, ³Depts. Radiology and Clinical Research, University of Bern, Bern, Switzerland, ⁴Institute of Biomedical Engineering, University and ETH, Zürich, Switzerland

Macromolecular resonances (MM) overlap with metabolites resulting in inaccurate quantification of the metabolites due to baseline distortion. This effect becomes even more severe in case of short echo times (TE). The purpose of this study was the development of an adiabatic pulse for double inversion recovery and investigation of impact to include MM into quantification of 9.4T MRS data of human brain. This is the first study where MC-STEAM is combined with a double inversion technique. The results showed the advantages of UHF and MC as well as the necessity of the inclusion of MM baseline in the basis set.

Nicolas Geades¹, Carrie Wismans², Mariska Damen², Penny Gowland¹, Hans Hoogduin², Vincent Boer², Dennis Klomp², and Jannie Wijnen^2
1Sir Peter Mansfield Imaging Centre, University of Nottingham, Nottingham, United Kingdom, ²Department of Radiology, University Medical Centre Utrecht, Utrecht, Netherlands

The regional, spectral and relaxation differences of macromolecules (MM) in the human brain were investigated using T1 mapping, metabolite nulling and high resolution MRSI with a crusher coil at 7T. Differences between macromolecular signal of GM and WM were observed by all three methods. The T1 mapping showed different T1 relaxation time of MM in GM and WM. Metabolic maps created by fitting an averaged WM spectrum showed differences in M1 and M2. The macromolecules in the metabolite nulled data showed a different M4 in GM and WM. Some of these differences can be explained by differences in T1 relaxation.

Sandeep K Ganji¹, Zhongxu An¹, Vivek Tiwari¹, Marco Pinho², Edward Pan³, Bruce Mickey⁴, Elizabeth Maher⁵, and Changho Choi^1
1Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, TX, United States, ²Radiology, UT Southwestern Medical Center, Dallas, TX, United States, ³Neurology and Neurotherapeutic, UT Southwestern Medical Center, Dallas, TX, United States, ⁴Neurological Surgery, UT Southwestern Medical Center, Dallas, TX, United States, ⁵Internal Medicine, UT Southwestern Medical Center, Dallas, TX, United States

2-hydroxyglutarate (2HG) has become an important biomarker in the diagnosis and management of glioma patients as well as in the workup of an undiagnosed mass. The 1H MRS signals of 2HG are extensively overlapped with other metabolite signals. Specifically, uncertainty in 2HG evaluation arising from the spectral overlap of the 2HG 2.25-ppm signal with the GABA 2.29-ppm resonance may be a major obstacle when the 2HG level is relatively low. Here we report a novel triple-refocusing difference editing that provides complete differentiation between 2HG and GABA signals at 7T. 

Chencai Wang¹, Chaohsiung Hsu¹, Stephanie Wolohan¹, and Yung-Ya Lin^1
1Department of Chemistry and Biochemistry, UCLA, Los Angeles, CA, United States

A general indirect-detection and spin-amplification scheme has been developed to enhance the sensitivity of heteronuclear MRS and MRI based on dynamic instability of the solvent proton magnetization under collective feedback fields of radiation damping and the distant dipolar field. The heteronuclear solute spins are first detected by the solvent proton spins through various magnetization transfer mechanisms and serve as small “input” signals to perturb the solvent proton magnetization, which is prepared in an unstable state. The weakly detected signal is then amplified through subsequent nonlinear evolution of the solvent proton magnetization to achieve 10x SNR improvement for 13C MRS and MRI.

Brice Tiret^1,2, Maria-Angeles Carrillo-de Sauvage^1,2, Huu Phuc Nguyen^3,4, Nicole El Massioui^5,6, Valérie Doyère^5,6, Vincent Lebon^1,2, Emmanuel Brouillet^1,2, and Julien Valette^1,2
1CEA/DSV/I2BM/MIRCen, Fontenay-aux-Roses, France, ²CNRS Université Paris-Saclay UMR 9199, Fontenay-aux-Roses, France, ³Institute of Medical Genetics and Applied Genomics, University of Tuebingen, Tuebingen, Germany, ⁴Centre for Rare Diseases, University of Tuebingen, Tuebingen, Germany, ⁵Paris-Saclay Institute of Neuroscience, Université Paris-Sud, UMR 9197, Orsay, France, ⁶Centre National de la Recherche Scientifique, Orsay, France

Localized ³¹P MRS with progressive magnetization transfer (MT) is performed in the BACHD transgenic rat model of Huntington’s disease to assess energy metabolism. Localized measurements of the ATP formation rate through creatine kinase and oxidative phosphorylation (ATPsynthase) are performed in the rat brain for the first time. Results show that ATPsynthase rate is reduced by a factor 2, which is partly compensated by higher cerebral concentrations of phosphocreatine to generate ATP via creatine kinase.

Sreenath P Kyathanahally¹, Victor Mocioiu², Nuno Miguel Pedrosa de Barros³, Johannes Slotboom³, Alan J Wright⁴, Margarida Julià-Sapé ², Carles Arús², and Roland Kreis^1
1Depts. Radiology and Clinical Research, University of Bern, Bern, Switzerland, ²Centro de Investigación Biomédica en Red en Bioingeniería, Biomateriales y Nanomedicina (CIBER-BBN), Universitat Autònoma de Barcelona, Barcelona, Spain, ³DRNN, Institute of Diagnostic and Interventional Neuroradiology/SCAN, University Hospital Bern, Bern, Switzerland, ⁴CRUK Cambridge Institute, University of Cambridge, Cambridge, United Kingdom

Despite many potential applications of MR spectroscopy in the clinic, its usage is limited – and the need for human experts to identify bad quality spectra may contribute to this. Previous studies have shown that machine learning methods can be developed to accept or reject a spectrum automatically. In this study, we extend this to different machine learning methods on 1916 spectra from the eTUMOUR and INTERPRET databases. The RUSBoost classifier, which handles unbalanced data, improved specificity and accuracy compared to other classifiers, in particular in combination with an extended feature set and multi-class labels.

Nuno Miguel Pedrosa de Barros^1,2, Urspeter Knecht¹, Richard McKinley¹, Jonathan Giezendanner¹, Roland Wiest¹, and Johannes Slotboom^1
1Institute for Diagnostic and Interventional Neuroradiology, Inselspital, Bern, Switzerland, ²University of Bern, Bern, Switzerland

MRSI-data frequently contains bad-quality spectra which strongly limits its clinical-use. Current clinical practice in our institute is that these bad-quality spectra are filtered out by an MRS-expert, at the expense of long processing times. In this work we present a new method for automatic quality assessment of both long and short-TE MRSI brain tumour data. This method is based upon a novel set of spectral features, and it is as accurate as an expert but considerably faster (3/4 minutes vs 3seconds).

Junyu Guo¹, Zoltan Patay¹, and Wilbrun E. Reddick^1
1St Jude Children's Research Hospital, Memphis, TN, United States

We present a novel, simple and fast MR spectroscopic imaging technique and show its conceptual validation with simulations and demonstrate proof-of-principle with phantom and human studies. First, compared to the conventional spectroscopic imaging in the time-domain, our method acquires data in the frequency domain, allowing flexible non-uniform sampling to speed up the acquisition. Second, using ultra-small RF pulses offers intrinsic water and fat suppression, greatly simplifying the scanning procedures. Third, this new technique has hundreds of times lower energy deposition than conventional MRI scans. We believe our method could allow spectroscopic imaging to play a larger role in clinical applications.

Michal Považan^1,2, Gilbert Hangel¹, Bernhard Strasser¹, Eva Heckova¹, Lukas Hingerl¹, Stephan Gruber¹, Siegfried Trattnig^1,2, and Wolfgang Bogner^1
1High Field MR Center, Department of Biomedical Imaging and Image-guided Therapy, Medical University Vienna, Vienna, Austria, ²Christian Doppler Laboratory for Clinical Molecular MR Imaging, Vienna, Austria

Ultra-short echo/acquisition delay MRS spectra have a strong characteristic background consisting of macromolecule (MM) resonances superimposed on the signal of metabolites. Typically a single metabolite-nulled MM spectrum is included into quantification routine to account for this. To detect  more prominent regional and pathologic changes, we replaced this single MM spectrum by individual MM peaks. We found that the MM peaks in a 2.3-0.5 ppm region are higher in gray matter compared to white matter, whereas the MM peaks from 2.9 to 3.2 ppm were significantly higher in white matter of healthy volunteers and one MS patient.

Assaf Tal^1
1Chemical Physics, Weizmann Institute of Science, Rehovot, Israel

A stochastic excitation and corresponding dictionary matching  scheme is presented for  quantifying metabolite concentrations, longitudinal relaxation times and transmit inhomogeneity in single voxel proton magnetic resonance spectroscopy in the human brain.