Helge Jörn Zöllner^1,2, Tao Gong^3,4, Steve C. N. Hui^1,2, Yulu Song^1,2, Weibo Chen⁵, Guangbin Wang^3,4, Georg Oeltzschner^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, ³Departments of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, China, ⁴Departments of Radiology, Shandong Provincial Hospital, Cheeloo College of Medicine, Jinan, Shandong, China, ⁵Philips Healthcare, Shanghai, China

Recent expert consensus emphasizes the use of measured MM spectra for reliable linear-combination modeling (LCM) of short-TE MRS data. Here we compare the metabolite estimates from short-TE PRESS data modeled with a parameterized and measured MM basis functions in a large dataset.

Meng Gu¹, Ralph Hurd¹, and Daniel Spielman^1
1Radiology, Stanford University, Stanford, CA, United States

Reliable in-vivo measurement of phosphocreatine using 1H MRS offers valuable information for understanding brain energy metabolism. At 3T, separate quantification of phosphocreatine and creatine has been challenging using LCModel. This problem exacerbates in studies of deep hypothermic circulatory arrest (DHCA), a technique used in many cardiac surgeries, due to temperature-dependent chemical shifts. Furthermore, total creatine quantification is hampered using a 37°C-only basis set as it fails to account for the increased polarization at lower temperatures. By using a semi-laser sequence with temperature dependent LCModel basis sets, reliable quantification of phosphocreatine and total creatine was achieved for the DHCA studies.

Christopher D. Kroenke¹, Natalie M. Zahr^2,3, Evan Kittle³, and Adolf Pfefferbaum^2,3
1Oregon Health & Science University, Portland, OR, United States, ²Stanford University, Palo Alto, CA, United States, ³SRI International, Palo Alto, CA, United States

This work implements a method for measuring cerebral kinetics of glucose metabolism using ¹H magnetic resonance spectroscopy (MRS) in rats using a small-animal 7T MRI system. The method has been demonstrated to be readily applied to nonhuman primates and human subjects using clinical MRI systems. The extension to small animals requires intravenous administration of small quantities of 13C-labeled glucose, and adaptation of data analysis procedures to data acquired at high magnetic field strength. We demonstrate robust quantification of ¹³C enrichment within the glutamate at the C4 position. This methodology will facilitate future translational research involving preclinical models of human disease.

Lily Qiu¹, Marco Palombo^2,3, Jason Lerch^1,4,5, and Clémence Ligneul^1
1Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom, ²Cardiff University Brain Research Imaging Centre (CUBRIC), School of Psychology, Cardiff University, Cardiff, United Kingdom, ³School of Computer Science and Informatics, Cardiff University, Cardiff, United Kingdom, ⁴Mouse Imaging Centre, The Hospital for Sick Children, Toronto, ON, Canada, ⁵Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada

Monitoring non-invasively the development of brain cells in the neonate and infant brain could be of interest for neurodevelopmental disorders.  In this work, we assess whether diffusion-weighted MR spectroscopy (DWMRS) has the potential to follow cerebellar and thalamic cells development in the healthy rat brain from P5 to P30.  Results show that  DWMRS is sensitive to microstructural changes in the developing brain. The modeling of the data shows a cell growth and complexification (from P15 to P30), and underlies the need of different modeling hypotheses at earlier time points.

Tiffany Bell^1,2,3, Kate J Godfrey^1,2,3, Ashley L Ware^2,3,4,5, Keith Owen Yeates^2,3,4, and Ashley D Harris^1,2,3
1Department of Radiology, University of Calgary, Calgary, AB, Canada, ²Hotchkiss Brain Institute, University of Calgary, Calgary, AB, Canada, ³Alberta Children's Hospital Research Institute, University of Calgary, Calgary, AB, Canada, ⁴Department of Psychology, University of Calgary, Calgary, AB, Canada, ⁵Department of Neurology, University of Utah, Salt Lake City, UT, United States

Multisite magnetic resonance spectroscopy (MRS) studies are becoming increasingly more common, however data collection across multiple sites introduces non-biological variability that can mask true biological effects. Using PRESS and MEGA-PRESS data obtained from the BIG GABA repository and linear modelling, we show that (1) scanner vendor and data collection site is significantly associated with metabolite levels and (2) data harmonisation using ComBat successfully removes non-biological variance to reveal biological effects of interest. We therefore recommend ComBat as an approach to harmonise multi-site MRS data.

Laura Beghini^1,2, Francesca Saviola², Stefano Tambalo², and Jorge Jovicich^2
1Department of Physics, Norwegian University of Science and Technology, Trondheim, Norway, Trondheim, Norway, ²CIMeC, Center for Mind/Brain Sciences, University of Trento, Rovereto (Trento), Italy, Trento, Italy

There is increased interest in measuring brain GABA dynamics with MR spectroscopy, yet there is no consensus about robust and fast ways to do this. In this study, we evaluated how averaging affects GABA estimates when using different modelling strategies (Gannet, Osprey). Our results show that 60 MEGA-PRESS averages give valid concentration estimates, consistent with literature reports. Moreover, Gannet gives a GABA+/Cr measurement which should be more sensitive to concentration variations across regions and give lower temporal variability at rest relative to the GABA+/tCr measure in Osprey.

Chloé Najac¹, André Döring², William Clarke³, Guglielmo Genovese⁴, Nathalie Just⁵, Roland Kreis^6,7, Henrik Lundell⁵, Jessie Mosso^8,9,10, Eloïse Mougel¹¹, Georg Oeltzschner^12,13, Marco Palombo^2,14, and Clémence Ligneul^3
1C.J. Gorter Center for High-Field MRI, Department of Radiology, Leiden University Medical Center, Leiden, Netherlands, ²Cardiff University Brain Research Imaging Centre (CUBRIC), School of Psychology, Cardiff University, Cardiff, United Kingdom, ³Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom, ⁴Center for Magnetic Resonance Research and Department of Radiology, University of Minnesota, Minneapolis, MN, United States, ⁵Danish Research Centre for Magnetic Resonance, Centre for Functional and Diagnostic Imaging and Research, Copenhagen University Hospital Amager and Hvidovre, Copenhagen, Denmark, ⁶Magnetic Resonance Methodology, Institute for Diagnostic and Interventional Neuroradiology, University Bern, Bern, Switzerland, ⁷Translational Imaging Center, sitem-insel, Bern, Switzerland, ⁸CIBM Center for Biomedical Imaging, Lausanne, Switzerland, ⁹Animal Imaging and Technology, EPFL, Lausanne, Switzerland, ¹⁰LIFMET, EPFL, Lausanne, Switzerland, ¹¹Commissariat à l’Energie Atomique et aux Energies Alternatives (CEA), Centre National de la Recherche Scientifique (CNRS), Molecular Imaging Research Center (MIRCen), Laboratoire des Maladies Neurodégénératives, Université Paris-Saclay, Fontenay-aux-Roses, France, ¹²Russell H. Morgan Department of Radiology and Radiological Science, The Johns Hopkins University School of Medicine, Baltimore, MD, United States, ¹³F. M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, ¹⁴School of Computer Science and Informatics, Cardiff University, Cardiff, United Kingdom

A processing and fitting challenge was initiated for the “Best practices & Tools for Diffusion MR Spectroscopy” workshop held at the Lorentz Center (Leiden, NL) in September 2021. Our goal was to assess variabilities in processing and fitting pipelines and possibly identify the most robust steps to analyze diffusion-weighted MRS data. A “brain-like” dataset of diffusion-weighted spectra was simulated and participants were asked to analyze the data with their routinely used pipelines. Results showed that variabilities between participants (n=8) particularly increased at higher b-values and for J-coupled metabolites (Ins, Glu) as a result of lower SNR and higher CRLB values.

Rudy Rizzo^1,2, Martyna Dziadosz^1,2, Sreenath Pruthviraj Kyathanahally³, and Roland Kreis^1,2
1Magnetic Resonance Methodology, Institute of Diagnostic and Interventional Neuroradiology, University of Bern, Bern, Switzerland, ²Translational Imaging Center, sitem-insel, Bern, Switzerland, ³Department Systems Analysis, Integrated Assessment and Modelling, Data Science for Environmental Research group, Dübendorf, Switzerland

Deep Learning has introduced the possibility to speed up quantitation in Magnetic Resonance Spectroscopy. However, questions arise about how to access and relate to prediction uncertainties. Distributions of predictions and Monte-Carlo dropout are here used to investigate data and model related uncertainties, exploiting ground truth knowledge (in-silico set up). It is confirmed that DL is a dataset-biased technique, showing higher uncertainties toward the edges of its training set. Surprisingly, metabolites present in high concentrations suffer from comparable high uncertainties as when present in low concentrations. Evaluating and respecting fitting uncertainties is equally crucial for DL and traditional approaches.

Duanghathai Pasanta^1,2, Georg Oeltzschner³, Talitha Ford⁴, David J. Lythgoe⁵, and Nicolaas A. Puts^1,6
1Department of Forensic and Neurodevelopmental Sciences, Sackler Institute for Translational Neurodevelopment, Institute of Psychiatry, Psychology, and Neuroscience, King's College London, London, United Kingdom, ²Department of Radiologic Technology, Faculty of Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand, ³Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ⁴Cognitive Neuroscience Unit, School of Psychology, Deakin University, Geelong, Australia, ⁵Department of Neuroimaging, King’s College London, London, United Kingdom, ⁶MRC Centre for Neurodevelopmental Disorders, King's College London, London, United Kingdom

Functional magnetic resonance spectroscopy (fMRS) can be used to investigate the neurometabolic responses to external stimuli in-vivo, but findings to date are inconsistent. We performed a systematic review and meta-analysis on 49 human fMRS studies on Glutamate, Glx (Glutamate + Glutamine) and GABA. Small to moderate effect sizes of 0.29-0.47 (p < 0.05) were observed for Glu/Glx regardless of stimulus domain but not for GABA. Results suggest Glu/Glx and GABA responses vary with time course and are unique to stimulus domain or task. This analysis will inform study design in future work.

Lauriane Jugé^1,2, Iain Ball³, and Caroline D Rae^1,2
1Neuroscience Research Australia, Sydney, Australia, ²School of Medical Sciences, University of New South Wales, Sydney, Australia, ³Philips Australia & New Zealand, North Ryde, Australia

Here, we combined two ¹H-MRS methods that have shown utility for detecting lactate in the brain, MEGA- editing and use of an asymmetric-PRESS and compared this approach (MEGA-APRESS) with MEGA-PRESS and A-PRESS in phantoms and eight individuals. MEGA-APRESS showed improved precision compared to MEGA-PRESS. Both editing approaches were greatly superior to A-PRESS alone. 

Sandra Zimmermann¹, Dario Pfyffer¹, Roland Kreis^2,3, Kadir Simsek^2,3, Patrick Freund^1,4, and Maryam Seif^1,4
1Spinal Cord Injury Center, Balgrist University Hospital, Zurich, Switzerland, ²Magnetic Resonance Methodology, Institute of Diagnostic and Interventional Neuroradiology, University of Bern, Bern, Switzerland, ³Translational Imaging Center, sitem-insel, Bern, Switzerland, ⁴Neurophysics Department, Max Plank Institute, Leipzig, Germany

This work uses magnetic resonance spectroscopy data obtained in the hippocampus of 28 patients with traumatic spinal cord injury (SCI) to investigate SCI-induced metabolic changes. Structural magnetic resonance imaging was performed for hippocampal volumetric assessment. Eighteen healthy controls underwent the same imaging protocol. Study participants underwent a functional assessment by testing the visuospatial and verbal memory performance to check for cognitive impairments. In SCI patients without cognitive impairments, hippocampal metabolites did not differ from healthy controls. This study does not support evidence of degeneration and inflammation in the hippocampus of animal models of SCI that show impaired spatial memory performance.

Fan Lam^1,2,3, Yahang Li^1,2, Yibo Zhao^2,4, and Justin Haldar^5
1Department of Bioengineering, 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, ³Cancer Center at Illinois, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁴Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁵Ming Hsieh Department of Electrical and Computer Engineering, University of Southern California, Los Angeles, CA, United States

One major challenge for ¹H-MRSI with localized excitation is minimizing interference from undesired regions, particularly the subcutaneous lipids. In this work, we exploit the complementariness of a new technique called region-optimized virtual coils (ROVir) that is capable of generating a set of virtual channels sensitized to specific regions and spatiospectral processing for lipid removal. Using brain ¹H-MRSI data, we demonstrated improved lipid removal by combining ROVir and a state-of-the-art union-of-subspaces method. We expect this integration to be useful for MRSI applications where VOIs are often lateralized (e.g., stroke and tumor) and/or cerebral lipids are of interest.