Fast computation of full density matrix of multispin systems for spatially localized in vivo magnetic resonance spectroscopy

Frequency and phase correction of GABA-edited magnetic resonance spectroscopy using complex-valued convolutional neural networks

PURPOSE

To determine the significance of complex-valued inputs and complex-valued convolutions compared to real-valued inputs and real-valued convolutions in convolutional neural networks (CNNs) for frequency and phase correction (FPC) of GABA-edited magnetic resonance spectroscopy (MRS) data.

METHODS

An ablation study using simulated data was performed to determine the most effective input (real or complex) and convolution type (real or complex) to predict frequency and phase shifts in GABA-edited MEGA-PRESS data using CNNs. The best CNN model was subsequently compared using both simulated and in vivo data to two recently proposed deep learning (DL) methods for FPC of GABA-edited MRS. All methods were trained using the same experimental setup and evaluated using the signal-to-noise ratio (SNR) and linewidth of the GABA peak, choline artifact, and by visually assessing the reconstructed final difference spectrum. Statistical significance was assessed using the Wilcoxon signed rank test.

RESULTS

The ablation study showed that using complex values for the input represented by real and imaginary channels in our model input tensor, with complex convolutions was most effective for FPC. Overall, in the comparative study using simulated data, our CC-CNN model (that received complex-valued inputs with complex convolutions) outperformed the other models as evaluated by the mean absolute error.

CONCLUSION

Our results indicate that the optimal CNN configuration for GABA-edited MRS FPC uses a complex-valued input and complex convolutions. Overall, this model outperformed existing DL models.

Integrated Short-TE and Hadamard-edited Multi-Sequence (ISTHMUS) for advanced MRS

BACKGROUND

To examine data quality and reproducibility using ISTHMUS, which has been implemented as the standardized MR spectroscopy sequence for the multi-site Healthy Brain and Child Development (HBCD) study.

METHODS

ISTHMUS is the consecutive acquisition of short-TE PRESS (32 transients) and long-TE HERCULES (224 transients) data with dual-TE water reference scans. Voxels were positioned in the centrum semiovale, dorsal anterior cingulate cortex, posterior cingulate cortex and bilateral thalamus regions. After acquisition, ISTHMUS data were separated into the PRESS and HERCULES portions for analysis and modeled separately using Osprey. In vivo experiments were performed in 10 healthy volunteers (6 female; 29.5±6.6 years). Each volunteer underwent two scans on the same day. Differences in metabolite measurements were examined. T2 correction based on the dual-TE water integrals were compared with: 1) T2 correction based on the default white matter and gray matter T2 reference values in Osprey and 2) shorter WM and GM T2 values from recent literature.

RESULTS

No significant difference in linewidth was observed between PRESS and HERCULES. Bilateral thalamus spectra had produced significantly higher (p<0.001) linewidth compared to the other three regions. Linewidth measurements were similar between scans, with scan-to-scan differences under 1 Hz for most subjects. Paired t-tests indicated a significant difference only in PRESS NAAG between the two thalamus scans (p=0.002). T2 correction based on shorter T2 values showed better agreement to the dual-TE water integral ratio.

CONCLUSIONS

ISTHMUS facilitated data acquisition and post-processing and reduced operator workload to eliminate potential human error.

Integrated Short-TE and Hadamard-edited Multi-Sequence (ISTHMUS) for Advanced MRS

Background

To examine data quality and reproducibility using ISTHMUS, which has been implemented as the standardized MR spectroscopy sequence for the multi-site Healthy Brain and Child Development (HBCD) study.

Methods

ISTHMUS is the consecutive acquisition of short-TE PRESS (32 transients) and long-TE HERCULES (224 transients) data with dual-TE water reference scans. Voxels were positioned in the centrum semiovale, dorsal anterior cingulate cortex, posterior cingulate cortex and bilateral thalamus regions. After acquisition, ISTHMUS data were separated into the PRESS and HERCULES portions for analysis and modeled separately using Osprey. In vivo experiments were performed in 10 healthy volunteers (6 female; 29.5±6.6 years). Each volunteer underwent two scans on the same day. Differences in metabolite measurements were examined. T2 correction based on the dual-TE water integrals were compared with: 1) T2 correction based the default white matter and gray matter T2 reference values in Osprey; 2) shorter WM and GM T2 values from recent literature; and 3) reduced CSF fractions.

Results

No significant difference in linewidth was observed between PRESS and HERCULES. Bilateral thalamus spectra had produced significantly higher (p<0.001) linewidth compared to the other three regions. Linewidth measurements were similar between scans, with scan-to-scan differences under 1 Hz for most subjects. Paired t-tests indicated a significant difference only in PRESS NAAG between the two thalamus scans (p=0.002). T2 correction based on shorter T2 values showed better agreement to the dual-TE water integral ratio.

Conclusions

ISTHMUS facilitated and standardized acquisition and post-processing and reduced operator workload to eliminate potential human error.

Results of the 2023 ISBI challenge to reduce GABA-edited MRS acquisition time

PURPOSE

Use a conference challenge format to compare machine learning-based gamma-aminobutyric acid (GABA)-edited magnetic resonance spectroscopy (MRS) reconstruction models using one-quarter of the transients typically acquired during a complete scan.

METHODS

There were three tracks: Track 1: simulated data, Track 2: identical acquisition parameters with in vivo data, and Track 3: different acquisition parameters with in vivo data. The mean squared error, signal-to-noise ratio, linewidth, and a proposed shape score metric were used to quantify model performance. Challenge organizers provided open access to a baseline model, simulated noise-free data, guides for adding synthetic noise, and in vivo data.

RESULTS

Three submissions were compared. A covariance matrix convolutional neural network model was most successful for Track 1. A vision transformer model operating on a spectrogram data representation was most successful for Tracks 2 and 3. Deep learning (DL) reconstructions with 80 transients achieved equivalent or better SNR, linewidth and fit error compared to conventional 320 transient reconstructions. However, some DL models optimized linewidth and SNR without actually improving overall spectral quality, indicating a need for more robust metrics.

CONCLUSION

DL-based reconstruction pipelines have the promise to reduce the number of transients required for GABA-edited MRS.

The effects of residual dipolar coupling on carnosine in proton muscle spectra

Carnosine, an MR-visible dipeptide in human muscle, is well characterized by two peaks at ~8 and ~7 ppm from C2 and C4 imidazole protons. Like creatine and other metabolites, carnosine is subject to residual dipolar coupling in the anisotropic environment of muscle fibers, but the effects have not been studied extensively. Single-voxel TE 30-32 PRESS spectra from three different 3T studies were acquired from gastrocnemius medialis and soleus muscles in the human lower leg. In these studies, carnosine T2 values were measured, and spectra were obtained at three different foot angles. LCModel was used to fit the carnosine peaks with a basis set that was generated using shaped RF pulses and included a range of dipolar couplings affecting the C4 peak. A seven-parameter analytic expression was used to fit the CH2 doublets of creatine. It incorporated an optimized "effective TE" value to model the effect of shaped RF pulses. The fits confirm that the triplet C4 peak of carnosine is dipolar coupled to a pair of CH2 protons, with no need to include a contribution from a separate pool of freely rotating uncoupled carnosine. Moreover, the couplings experienced by carnosine C4 protons and creatine CH2 protons are strongly correlated (R2  =  0.88, P<0.001), exhibiting a similar 3cos2 θ - 1 dependence on the angle θ between fiber orientation and B0. T2 values for the singlet C2 peak of gastrocnemius carnosine are inversely proportional to the C4 dipolar coupling strength (R2  = 0.97, P < 0.001), which in turn is a function of foot orientation. This dependence indicates that careful positioning of the foot while acquiring lower leg muscle spectra is important to obtain reproducible carnosine concentrations. As proton magnetic resonance spectroscopy of carnosine is currently used to non-invasively estimate the muscle fiber typology, these results have important implications in sport science.

Glutamate measurements using edited MRS

PURPOSE

To demonstrate J-difference coediting of glutamate using Hadamard encoding and reconstruction of Mescher-Garwood-edited spectroscopy (HERMES).

METHODS

Density-matrix simulations of HERMES (TE 80 ms) and 1D J-resolved (TE 31-229 ms) of glutamate (Glu), glutamine (Gln), γ-aminobutyric acid (GABA), and glutathione (GSH) were performed. HERMES comprised four sub-experiments with editing pulses applied as follows: (A) 1.9/4.56 ppm simultaneously (ONGABA /ONGSH ); (B) 1.9 ppm only (ONGABA /OFFGSH ); (C) 4.56 ppm only (OFFGABA /ONGSH ); and (D) 7.5 ppm (OFFGABA /OFFGSH ). Phantom HERMES and 1D J-resolved experiments of Glu were performed. Finally, in vivo HERMES (20-ms editing pulses) and 1D J-resolved (TE 31-229 ms) experiments were performed on 137 participants using 3 T MRI scanners. LCModel was used for quantification.

RESULTS

HERMES simulation and phantom experiments show a Glu-edited signal at 2.34 ppm in the Hadamard sum combination A+B+C+D with no overlapping Gln signal. The J-resolved simulations and phantom experiments show substantial TE modulation of the Glu and Gln signals across the TEs, whose average yields a well-resolved Glu signal closely matching the Glu-edited signal from the HERMES sum spectrum. In vivo quantification of Glu show that the two methods are highly correlated (p < 0.001) with a bias of ∼10%, along with similar between-subject coefficients of variation (HERMES/TE-averaged: ∼7.3%/∼6.9%). Other Hadamard combinations produce the expected GABA-edited (A+B-C-D) or GSH-edited (A-B+C-D) signal.

CONCLUSION

HERMES simulation and phantom experiments show the separation of Glu from Gln. In vivo HERMES experiments yield Glu (without Gln), GABA, and GSH in a single MRS scan.

Linewidth-related bias in modelled concentration estimates from GABA-edited 1H-MRS

J-difference-edited MRS is widely used to study GABA in the human brain. Editing for low-concentration target molecules (such as GABA) typically exhibits lower signal-to-noise ratio (SNR) than conventional non-edited MRS, varying with acquisition region, volume and duration. Moreover, spectral lineshape may be influenced by age-, pathology-, or brain-region-specific effects of metabolite T2, or by task-related blood-oxygen level dependent (BOLD) changes in functional MRS contexts. Differences in both SNR and lineshape may have systematic effects on concentration estimates derived from spectral modelling. The present study characterises the impact of lineshape and SNR on GABA+ estimates from different modelling algorithms: FSL-MRS, Gannet, LCModel, Osprey, spant and Tarquin. Publicly available multi-site GABA-edited data (222 healthy subjects from 20 sites; conventional MEGA-PRESS editing; TE = 68 ms) were pre-processed with a standardised pipeline, then filtered to apply controlled levels of Lorentzian and Gaussian linebroadening and SNR reduction. Increased Lorentzian linewidth was associated with a 2-5% decrease in GABA+ estimates per Hz, observed consistently (albeit to varying degrees) across datasets and most algorithms. Weaker, often opposing effects were observed for Gaussian linebroadening. Variations are likely caused by differing baseline parametrization and lineshape constraints between models. Effects of linewidth on other metabolites (e.g., Glx and tCr) varied, suggesting that a linewidth confound may persist after scaling to an internal reference. These findings indicate a potentially significant confound for studies where linewidth may differ systematically between groups or experimental conditions, e.g. due to T2 differences between brain regions, age, or pathology, or varying T2* due to BOLD-related changes. We conclude that linewidth effects need to be rigorously considered during experimental design and data processing, for example by incorporating linewidth into statistical analysis of modelling outcomes or development of appropriate lineshape matching algorithms.

Impaired striatal glutamate/GABA regulation in violent offenders with antisocial personality disorder and psychopathy

Men with antisocial personality disorder (ASPD) with or without psychopathy (+/-P) are responsible for most violent crime in society. Development of effective treatments is hindered by poor understanding of the neurochemical underpinnings of the condition. Men with ASPD with and without psychopathy demonstrate impulsive decision-making, associated with striatal abnormalities in functional neuroimaging studies. However, to date, no study has directly examined the potential neurochemical underpinnings of such abnormalities. We therefore investigated striatal glutamate: GABA ratio using Magnetic Resonance Spectroscopy in 30 violent offenders (16 ASPD-P, 14 ASPD + P) and 21 healthy non-offenders. Men with ASPD +/- P had a significant reduction in striatal glutamate : GABA ratio compared to non-offenders. We report, for the first time, striatal Glutamate/GABA dysregulation in ASPD +/- P, and discuss how this may be related to core behavioral abnormalities in the disorders.

sLASER and PRESS perform similarly at revealing metabolite-age correlations at 3 T

PURPOSE

To compare the respective ability of PRESS and sLASER to reveal biological relationships, using age as a validation covariate at 3 T.

METHODS

MRS data were acquired from 102 healthy volunteers using PRESS and sLASER in centrum semiovale and posterior cingulate cortex (PCC). Acquisition parameters included TR/TE = 2000/30 ms, 96 transients, and 2048 datapoints sampled at 2 kHz. Spectra were analyzed using Osprey. SNR, FWHM linewidth of total creatine, and metabolite concentrations were extracted. A linear model was used to compare SNR and linewidth. Paired t-tests were used to assess differences in metabolite measurements between PRESS and sLASER. Correlations were used to evaluate the relationship between PRESS and sLASER metabolite estimates, as well as the strength of each metabolite-age relationship. Coefficients of variation were calculated to assess inter-subject variability in each metabolite measurement.

RESULTS

SNR and linewidth were significantly higher (p < 0.01) for sLASER than PRESS in PCC. Paired t-tests showed significant differences between PRESS and sLASER in most metabolite measurements. PRESS-sLASER measurements were significantly correlated (p < 0.05) for most metabolites. Metabolite-age relationships were consistently identified using both methods. Similar coefficients of variation were observed for most metabolites.

CONCLUSION

The study results suggest strong agreement between PRESS and sLASER in identifying relationships between brain metabolites and age in centrum semiovale and PCC data acquired at 3 T. sLASER is technically desirable due to the reduced chemical shift displacement artifact; however, PRESS performed similarly in homogeneous brain regions at clinical field strength.

Quantification of spatially localized MRS by a novel deep learning approach without spectral fitting

PURPOSE

To propose a novel end-to-end deep learning model to quantify absolute metabolite concentrations from in vivo J-point resolved spectroscopy (JPRESS) without using spectral fitting.

METHODS

A novel encoder-decoder-style neural network was created, which was trained to predict metabolite concentrations and individual component signals concurrently from 3T JPRESS data in the time domain. The training data set contained 100 000 samples created by spin-density simulations using experimentally used RF pulses. Concentrations, phase, frequencies, linewidths, and T2 relaxation times in the training data set were varied over a large range with uniform distributions. Random synthesized noise and extraneous signals were added to the data set. Two thousand validation samples were created similarly to the training data set but with mean concentrations close to in vivo values. An in vivo test was conducted with 20 samples acquired from the human brain.

RESULTS

Both validation and in vivo test results showed that the proposed model successfully predicted metabolite concentrations as well as individual metabolite signals without involving spectral fitting, while extraneous peaks or unregistered signals were filtered out. Compared with the short-TE spectral fitting by LCModel, the proposed method had the advantage that the undesired correlations between the estimated concentrations and noise levels and between metabolites were eliminated or substantially reduced.

CONCLUSION

The proposed method provides a working deep learning model that directly maps in vivo JPRESS data to metabolite concentrations. Because spectral fitting is not used, the trained model does not depend on the assumptions associated with parameter tuning when applied to in vivo data.

In vivo magnetic resonance spectroscopy by transverse relaxation encoding with narrowband decoupling

Cell pathology in neuropsychiatric disorders has mainly been accessible by analyzing postmortem tissue samples. Although molecular transverse relaxation informs local cellular microenvironment via molecule-environment interactions, precise determination of the transverse relaxation times of molecules with scalar couplings (J), such as glutamate and glutamine, has been difficult using in vivo magnetic resonance spectroscopy (MRS) technologies, whose approach to measuring transverse relaxation has not changed for decades. We introduce an in vivo MRS technique that utilizes frequency-selective editing pulses to achieve homonuclear decoupled chemical shift encoding in each column of the acquired two-dimensional dataset, freeing up the entire row dimension for transverse relaxation encoding with J-refocusing. This results in increased spectral resolution, minimized background signals, and markedly broadened dynamic range for transverse relaxation encoding. The in vivo within-subject coefficients of variation for the transverse relaxation times of glutamate and glutamine, measured using the proposed method in the human brain at 7 T, were found to be approximately 4%. Since glutamate predominantly resides in glutamatergic neurons and glutamine in glia in the brain, this noninvasive technique provides a way to probe cellular pathophysiology in neuropsychiatric disorders for characterizing disease progression and monitoring treatment response in a cell type-specific manner in vivo.

Vespa: Integrated applications for RF pulse design, spectral simulation and MRS data analysis

PURPOSE

The Vespa package (Versatile Simulation, Pulses, and Analysis) is described and demonstrated. It provides workflows for developing and optimizing linear combination modeling (LCM) fitting for ¹ H MRS data using intuitive graphical user interface interfaces for RF pulse design, spectral simulation, and MRS data analysis. Command line interfaces for embedding workflows in MR manufacturer platforms and utilities for synthetic dataset creation are included. Complete provenance is maintained for all steps in workflows.

THEORY AND METHODS

Vespa is written in Python for compatibility across operating systems. It embeds the PyGAMMA spectral simulation library for spectral simulation. Multiprocessing methods accelerate processing and visualization. Applications use the Vespa database for results storage and cross-application access. Three projects demonstrate pulse, sequence, simulation, and data analysis workflows: (1) short TE semi-LASER single-voxel spectroscopy (SVS) LCM fitting, (2) optimizing MEGA-PRESS (MEscher-GArwood Point RESolved Spectroscopy) flip angle and LCM fitting, and (3) creating a synthetic short TE dataset.

RESULTS

The LCM workflows for in vivo basis set creation and spectral analysis showed reasonable results for both the short TE semi-LASER and MEGA-PRESS. Examples of pulses, simulations, and data fitting are shown in Vespa application interfaces for various steps to demonstrate the interactive workflow.

CONCLUSION

Vespa provides an efficient and extensible platform for characterizing RF pulses, pulse design, spectral simulation optimization, and automated LCM fitting via an interactive platform. Modular design and command line interface make it easy to embed in other platforms. As open source, it is free to the MRS community for use and extension. Vespa source code and documentation are available through GitHub.

Harmonization of multi-scanner in vivo magnetic resonance spectroscopy: ENIGMA consortium task group considerations

Magnetic resonance spectroscopy is a powerful, non-invasive, quantitative imaging technique that allows for the measurement of brain metabolites that has demonstrated utility in diagnosing and characterizing a broad range of neurological diseases. Its impact, however, has been limited due to small sample sizes and methodological variability in addition to intrinsic limitations of the method itself such as its sensitivity to motion. The lack of standardization from a data acquisition and data processing perspective makes it difficult to pool multiple studies and/or conduct multisite studies that are necessary for supporting clinically relevant findings. Based on the experience of the ENIGMA MRS work group and a review of the literature, this manuscript provides an overview of the current state of MRS data harmonization. Key factors that need to be taken into consideration when conducting both retrospective and prospective studies are described. These include (1) MRS acquisition issues such as pulse sequence, RF and B0 calibrations, echo time, and SNR; (2) data processing issues such as pre-processing steps, modeling, and quantitation; and (3) biological factors such as voxel location, age, sex, and pathology. Various approaches to MRS data harmonization are then described including meta-analysis, mega-analysis, linear modeling, ComBat and artificial intelligence approaches. The goal is to provide both novice and experienced readers with the necessary knowledge for conducting MRS data harmonization studies.

MRSCloud: A cloud-based MRS tool for basis set simulation

PURPOSE

The purpose of this study is to present a cloud-based spectral simulation tool "MRSCloud," which allows MRS users to simulate a vendor-specific and sequence-specific basis set online in a convenient and time-efficient manner. This tool can simulate basis sets for GE, Philips, and Siemens MR scanners, including conventional acquisitions and spectral editing schemes with PRESS and semi-LASER localization at 3 T.

METHODS

The MRSCloud tool was built on the spectral simulation functionality in the FID-A software package. We added three extensions to accelerate computation (ie, one-dimensional projection method, coherence pathways filters, and precalculation of propagators). The RF waveforms were generated based on vendors' generic pulse shapes and timings. Simulations were compared within MRSCloud using different numbers of spatial resolution (21 × 21, 41 × 41, and 101 × 101). Simulated metabolite basis functions from MRSCloud were compared with those generated by the generic FID-A and MARSS, and a phantom-acquired basis set from LCModel. Intraclass correlation coefficients were calculated to measure the agreement between individual metabolite basis functions. Statistical analysis was performed using R in RStudio.

RESULTS

Simulation time for a full PRESS basis set is approximately 11 min on the server. The interclass correlation coefficients ICCs were at least 0.98 between MRSCloud and FID-A and were at least 0.96 between MRSCloud and MARSS. The interclass correlation coefficients between simulated MRSCloud basis spectra and acquired LCModel basis spectra were lowest for glutamine at 0.68 and highest for N-acetylaspartate at 0.96.

CONCLUSIONS

Substantial reductions in runtime have been achieved. High ICC values indicated that the accelerating features are running correctly and produce comparable and accurate basis sets.

Multinuclear MRI in Drug Discovery

The continuous development of magnetic resonance imaging broadens the range of applications to newer areas. Using MRI, we can not only visualize, but also track pharmaceutical substances and labeled cells in both in vivo and in vitro tests. ¹H is widely used in the MRI method, which is determined by its high content in the human body. The potential of the MRI method makes it an excellent tool for imaging the morphology of the examined objects, and also enables registration of changes at the level of metabolism. There are several reports in the scientific publications on the use of clinical MRI for in vitro tracking. The use of multinuclear MRI has great potential for scientific research and clinical studies. Tuning MRI scanners to the Larmor frequency of a given nucleus, allows imaging without tissue background. Heavy nuclei are components of both drugs and contrast agents and molecular complexes. The implementation of hyperpolarization techniques allows for better MRI sensitivity. The aim of this review is to present the use of multinuclear MRI for investigations in drug delivery.

Monte Carlo study of metabolite correlations originating from spectral overlap

Monte Carlo simulations and a mathematical model of spectral fitting were used to study the correlations between metabolites with overlapping resonances. The dependence of the polarity and the magnitude of cross-correlation coefficients between overlapping metabolites on the spectral patterns of MRS signals was investigated. The results demonstrate the importance of quantifying metabolite correlations originating from spectral overlap as they may confound determination of correlations of biological origin. The findings also indicate that it is possible to minimize unwanted metabolite correlations by altering spectral patterns in the presence of significant spectral overlap.

Comparison of seven modelling algorithms for γ-aminobutyric acid-edited proton magnetic resonance spectroscopy

Edited MRS sequences are widely used for studying γ-aminobutyric acid (GABA) in the human brain. Several algorithms are available for modelling these data, deriving metabolite concentration estimates through peak fitting or a linear combination of basis spectra. The present study compares seven such algorithms, using data obtained in a large multisite study. GABA-edited (GABA+, TE = 68 ms MEGA-PRESS) data from 222 subjects at 20 sites were processed via a standardised pipeline, before modelling with FSL-MRS, Gannet, AMARES, QUEST, LCModel, Osprey and Tarquin, using standardised vendor-specific basis sets (for GE, Philips and Siemens) where appropriate. After referencing metabolite estimates (to water or creatine), systematic differences in scale were observed between datasets acquired on different vendors' hardware, presenting across algorithms. Scale differences across algorithms were also observed. Using the correlation between metabolite estimates and voxel tissue fraction as a benchmark, most algorithms were found to be similarly effective in detecting differences in GABA+. An interclass correlation across all algorithms showed single-rater consistency for GABA+ estimates of around 0.38, indicating moderate agreement. Upon inclusion of a basis set component explicitly modelling the macromolecule signal underlying the observed 3.0 ppm GABA peaks, single-rater consistency improved to 0.44. Correlation between discrete pairs of algorithms varied, and was concerningly weak in some cases. Our findings highlight the need for consensus on appropriate modelling parameters across different algorithms, and for detailed reporting of the parameters adopted in individual studies to ensure reproducibility and meaningful comparison of outcomes between different studies.

GABAB receptor modulation of visual sensory processing in adults with and without autism spectrum disorder

[Figure: see text].

In vivo spectral editing of phosphorylethanolamine

PURPOSE

To demonstrate J-difference editing of phosphorylethanolamine (PE) with chemical shifts at 3.22 (PE3.22 ) and 3.98 (PE3.98 ) ppm, and compare the merits of two editing strategies.

METHODS

Density-matrix simulations of MEGA-PRESS (Mescher-Garwood PRESS) for PE were performed at TEs ranging from 80 to 200 ms in steps of 2 ms, applying 20-ms editing pulses (ON/OFF) at (1) 3.98/7.5 ppm to detect PE3.22 and (2) 3.22/7.5 ppm to detect PE3.98 . Phantom experiments were performed using a PE phantom to validate simulation results. Ten subjects were scanned using a Philips 3T MRI scanner at TEs of 90 ms and 110 ms to edit PE3.22 and PE3.98 . Osprey was used for data processing, modeling, and quantification.

RESULTS

Simulations show substantial TE modulation of the intensity and shape of the edited signals due to coupling evolution. Simulated and phantom integrals suggest that TEs of 110 ms and 90 ms were optimal for the edited detection of PE3.22 and PE3.98 , respectively. Phantom results indicated strong agreement with the simulated spectra and integrals. In vivo quantification of the PE3.22 /total creatine and PE3.98 /total creatine concentration ratio yielded values of 0.26 ± 0.04 (between-subject coefficient of variation [CV]: 15.4%) and 0.18 ± 0.04 (CV: 22.8%), respectively, at TE = 90 ms, and 0.24 ± 0.02 (CV: 8.2%) and 0.23 ± 0.04 (CV: 18.0%), respectively, at TE = 110 ms.

CONCLUSION

Simulations and in vivo MEGA-PRESS of PE demonstrate that both PE3.22 and PE3.98 are potential candidates for editing, but PE3.22 at TE = 110 ms yields lower variation across TEs.

Simulated basis sets for semi-LASER: the impact of including shaped RF pulses and magnetic field gradients

Objective

To study the need for inclusion of shaped RF pulses and magnetic field gradients in simulations of basis sets for the analysis of proton MR spectra of single voxels of the brain acquired with a semi-LASER pulse sequence.

Materials and methods

MRS basis sets where simulated at different echo times with hard RF pulses as well as with shaped RF pulses without or with magnetic field gradients included. The influence on metabolite concentration quantification was assessed using both phantom and in vivo measurements. For comparison, simulations and measurements were performed with the PRESS pulse sequence.

Results

The effect of including gradients in the simulations was smaller for semi-LASER than for PRESS, however, still noticeable. The difference was larger for strongly coupled metabolites and at longer echo times. Metabolite quantification using semi-LASER was thereby less dependent on the inclusion of gradients than PRESS, which was seen in both phantom and in vivo measurements.

Discussion

The inclusion of the shaped RF pulses and magnetic field gradients in the simulation of basis sets for semi-LASER is only important for strongly coupled metabolites. If computational time is a limiting factor, simple simulations with hard RF pulses can provide almost as accurate metabolite quantification as those that include the chemical-shift related displacement.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10334-020-00900-1.

FSL-MRS: An end-to-end spectroscopy analysis package

PURPOSE

We introduce FSL-MRS, an end-to-end, modular, open-source MRS analysis toolbox. It provides spectroscopic data conversion, preprocessing, spectral simulation, fitting, quantitation, and visualization.

METHODS

The FSL-MRS package is modular. Its programs operate on data in a standard format (Neuroimaging Informatics Technology Initiative [NIfTI]) capable of storing single-voxel and multivoxel spectroscopy, including spatial orientation information. The FSL-MRS toolbox includes tools for preprocessing of raw spectroscopy data, including coil combination, frequency and phase alignment, and filtering. A density matrix simulation program is supplied for generation of basis spectra from simple text-based descriptions of pulse sequences. Fitting is based on linear combination of basis spectra and implements Markov chain Monte Carlo optimization for the estimation of the full posterior distribution of metabolite concentrations. Validation of the fitting is carried out on independently created simulated data, phantom data, and three in vivo human data sets (257 single-voxel spectroscopy and 8 MRSI data sets) at 3 T and 7 T. Interactive HTML reports are automatically generated by processing and fitting stages of the toolbox. The FSL-MRS package can be used on the command line or interactively in the Python language.

RESULTS

Validation of the fitting shows low error in simulation (median error of 11.9%) and in phantom (3.4%). Average correlation between a third-party toolbox (LCModel) and FSL-MRS was high (0.53-0.81) in all three in vivo data sets.

CONCLUSION

The FSL-MRS toolbox is designed to be flexible and extensible to new forms of spectroscopic acquisitions. Custom fitting models can be specified within the framework for dynamic or multivoxel spectroscopy. It is available as part of the FMRIB Software Library.

What is the optimal schedule for multiparametric MRS? A magnetic resonance fingerprinting perspective

Clinical magnetic resonance spectroscopy (MRS) mainly concerns itself with the quantification of metabolite concentrations. Metabolite relaxation values, which reflect the microscopic state of specific cellular and sub-cellular environments, could potentially hold additional valuable information, but are rarely acquired within clinical scan times. By varying the flip angle, repetition time and echo time in a preset way (termed a schedule), and matching the resulting signals to a pre-generated dictionary - an approach dubbed magnetic resonance fingerprinting - it is possible to encode the spins' relaxation times into the acquired signal, simultaneously quantifying multiple tissue parameters for each metabolite. Herein, we optimized the schedule to minimize the averaged root mean square error (RMSE) across all estimated parameters: concentrations, longitudinal and transverse relaxation time, and transmitter inhomogeneity. The optimal schedules were validated in phantoms and, subsequently, in a cohort of healthy volunteers, in a 4.5 mL parietal white matter single voxel and an acquisition time under 5 minutes. The average intra-subject, inter-scan coefficients of variation (CVs) for metabolite concentrations, T₁ and T₂ relaxation times were found to be 3.4%, 4.6% and 4.7% in-vivo, respectively, averaged over all major singlets. Coupled metabolites were quantified using the short echo time schedule entries and spectral fitting, and reliable estimates of glutamate+glutamine, glutathione and myo-inositol were obtained.

Magnetic resonance Spectrum simulator (MARSS), a novel software package for fast and computationally efficient basis set simulation

The aim of this study was to develop a novel software platform for the simulation of magnetic resonance spin systems, capable of simulating a large number of spatial points (128³ ) for large in vivo spin systems (up to seven coupled spins) in a time frame of the order of a few minutes. The quantum mechanical density-matrix formalism is applied, a coherence pathway filter is utilized for handling unwanted coherence pathways, and the 1D projection method, which provides a substantial reduction in computation time for a large number of spatial points, is extended to include sequences of an arbitrary number of RF pulses. The novel software package, written in MATLAB, computes a basis set of 23 different metabolites (including the two anomers of glucose, seven coupled spins) with 128³ spatial points in 26 min for a three-pulse experiment on a personal desktop computer. The simulated spectra are experimentally verified with data from both phantom and in vivo MEGA-sLASER experiments. Recommendations are provided regarding the various assumptions made when computing a basis set for in vivo MRS with respect to the number of spatial points simulated and the consideration of relaxation.

Osprey: Open-source processing, reconstruction & estimation of magnetic resonance spectroscopy data

BACKGROUND

Processing and quantitative analysis of magnetic resonance spectroscopy (MRS) data are far from standardized and require interfacing with third-party software. Here, we present Osprey, a fully integrated open-source data analysis pipeline for MRS data, with seamless integration of pre-processing, linear-combination modelling, quantification, and data visualization.

NEW METHOD

Osprey loads multiple common MRS data formats, performs phased-array coil combination, frequency-and phase-correction of individual transients, signal averaging and Fourier transformation. Linear combination modelling of the processed spectrum is carried out using simulated basis sets and a spline baseline. The MRS voxel is coregistered to an anatomical image, which is segmented for tissue correction and quantification is performed based upon modelling parameters and tissue segmentation. The results of each analysis step are visualized in the Osprey GUI. The analysis pipeline is demonstrated in 12 PRESS, 11 MEGA-PRESS, and 8 HERMES datasets acquired in healthy subjects.

RESULTS

Osprey successfully loads, processes, models, and quantifies MRS data acquired with a variety of conventional and spectral editing techniques.

COMPARISON WITH EXISTING METHOD(S)

Osprey is the first MRS software to combine uniform pre-processing, linear-combination modelling, tissue correction and quantification into a coherent ecosystem. Compared to existing compiled, often closed-source modelling software, Osprey's open-source code philosophy allows researchers to integrate state-of-the-art data processing and modelling routines, and potentially converge towards standardization of analysis.

CONCLUSIONS

Osprey combines robust, peer-reviewed data processing methods into a modular workflow that is easily augmented by community developers, allowing the rapid implementation of new methods.

Determination of Brain Metabolite T 1 Without Interference From Macromolecule Relaxation

BACKGROUND

J-coupled metabolites are often measured at a predetermined echo time in the presence of macromolecule signals, which complicates the measurement of metabolite T₁ .

PURPOSE

To evaluate the feasibility and benefits of measuring metabolite T₁ relaxation times without changing the overlapping macromolecule baseline signals.

STUDY TYPE

Prospective.

SUBJECTS

Five healthy volunteers (three females and two males; age = 27 ± 7 years).

FIELD STRENGTH/SEQUENCE

7T scanner using a point resolved spectroscopy (PRESS)-based spectral editing MR spectroscopy (MRS) sequence with inversion recovery (IR).

ASSESSMENT

F-tests were performed to evaluate if the new approach, which fitted all the spectra together and used the same baselines for the three different IR settings, significantly reduced the variances of the metabolite T₁ values compared to a conventional fitting approach.

STATISTICAL TESTS

Cramer-Rao lower bound (CRLB), within-subject coefficient of variation, and F-test.

RESULTS

The T₁ relaxation times of N-acetylaspartate (NAA), total creatine (tCr), total choline (tCho), myo-inositol (mI), and glutamate (Glu) were determined with CRLB values below 6%. Glutamine (Gln) T₁ was determined with a 17% CRLB, and the T₁ of γ-aminobutyric acid (GABA) was determined with a 34% CRLB. The new approach significantly reduced the variances (F-test P < 0.05) of NAA, Glu, Gln, tCr, tCho, and mI T₁ s compared to the conventional approach.

DATA CONCLUSION

Keeping macromolecule signals intact by using only long IR times allowed the use of a single macromolecule spectral model for different IR settings and significantly reduced the variances of NAA, Glu, Gln, tCr, tCho, and mI T₁ s.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY STAGE: 1.

Signal enhancement of glutamine and glutathione by single-step spectral editing

A single-step spectral editing approach using an always-on editing pulse was proposed to enhance the signals of strongly coupled spins. Specifically, a single-step spectral editing sequence with an always-on editing pulse applied at 2.12 ppm was used to enhance glutamine (Gln) and glutathione (GSH) signals at TE = 56 ms on a 7 T scanner. Density matrix simulations demonstrated that the current method (TE = 56 ms) led to large signal enhancement of at least 61% for Gln and 51% for GSH compared to a previous single-step method (TE = 106 ms). Monte Carlo simulations showed that the current method reduced noise-originated variations by 31% for Gln and 26% for GSH compared to a previous three-step spectral editing method from which the present method was derived.

Effects of noise and linewidth on in vivo analysis of glutamate at 3 T

Magnetic resonance spectroscopy (MRS) can noninvasively detect metabolites in vivo, including glutamate (Glu). However, quantification is known to be affected by the overlaps among metabolite resonance lines and background macromolecule signals. We found that adding a moderate amount of noise or line broadening (2 Hz) caused large variations in concentration of Glu and other metabolites, when determined by LCModel analysis of in vivo short-echo time (TE) spectra. Theses variations were largely attributed to strong spectral baselines in short TE spectra, especially near 2.35 ppm, as well as overlapping metabolite resonance lines. To address this issue, we acquired in vivo data at 3 T using both short-TE and the multiple echo time J-resolved point-resolved spectroscopy (JPRESS) MRS techniques. We found that one-dimensional (1D) JPRESS, by simultaneously fitting the two cross-sections of JPRESS at J = 0 and J = 7.5 Hz, was highly resistant to variations in noise levels and spectral linewidths. Our results demonstrate that LCModel analysis of short-TE data is highly sensitive to variations in noise levels and spectral linewidths and this sensitivity is greatly reduced by 1D JPRESS given its substantially reduced baselines and enhanced spectral resolution.

Simultaneous edited MRS of GABA, glutathione, and ethanol

The aim of this work was to develop simultaneous edited MRS of γ-aminobutyric acid (GABA), glutathione (GSH), and ethanol (EtOH) using Hadamard encoding and reconstruction of MEGA-edited spectroscopy (HERMES) at 3T. Density-matrix simulations of HERMES were carried out and compared with phantom experiments. In vivo experiments were performed in six healthy volunteers about 30 min after alcohol consumption. Simulations of HERMES showed GABA-, GSH-, and EtOH-edited spectra with low levels of crosstalk and excellent agreement with phantom spectra. In vivo experiments showed well edited GABA signals at 3.0 ppm, GSH at 2.95 ppm, and EtOH at 1.18 ppm in the respective Hadamard combination spectra. Measured integral ratios were 0.082 ± 0.012 for GABA/Cr, 0.037 ± 0.006 for GSH/Cr, and 0.305 ± 0.129 for EtOH/Cr. Simulated, phantom, and in vivo measurements of HERMES show excellent separation of GABA-, GSH-, and EtOH-edited signals with negligible levels of crosstalk. HERMES allows a threefold acceleration of editing while maintaining spectral quality compared with sequentially acquired MEGA-PRESS measurements.

Quantification of in vivo transverse relaxation of glutamate in the frontal cortex of human brain by radio frequency pulse-driven longitudinal steady state

PURPOSE

The principal excitatory neurotransmitter glutamate plays an important role in many central nervous system disorders. Because glutamate resides predominantly in glutamatergic neurons, its relaxation properties reflect the intracellular environment of glutamatergic neurons. This study developed an improved echo time-independent technique for measuring transverse relaxation time and demonstrated that this radio frequency (RF)-driven longitudinal steady state technique can reliably measure glutamate transverse relaxation in the frontal cortex, where structural and functional abnormalities have been associated with psychiatric symptoms.

METHOD

Bloch and Monte Carlo simulations were performed to improve and optimize the RF-driven, longitudinal, steady-state (MARzss) technique to significantly shorten scan time and increase measurement precision. Optimized four-flip angle measurements at 0°,12°, 24°, and 36° with matched repetition time were used in nine human subjects (6F, 3M; 27-49 years old) at 7 Tesla. Longitudinal and transverse relaxation rates for glutamate were measured from a 2 x 2 x 2 cm3 voxel placed in three different brain regions: gray matter-dominated medial prefrontal lobe, white matter-dominated left frontal lobe, and gray matter-dominated occipital lobe.

RESULTS

Compared to the original MARzss technique, the scan time per voxel for measuring glutamate transverse relaxation was shortened by more than 50%. In the medial frontal, left frontal, and occipital voxels, the glutamate T2 was found to be 117.5±12.9 ms (mean ± standard deviation, n = 9), 107.3±12.1 (n = 9), and 124.4±16.6 ms (n = 8), respectively.

CONCLUSIONS

The improvements described in this study make the MARZSS technique a viable tool for reliably measuring glutamate relaxation from human subjects in a typical clinical setting. It is expected that this improved technique can be applied to characterize the intracellular environment of glutamatergic neurons in a variety of brain disorders.

Advanced Hadamard-encoded editing of seven low-concentration brain metabolites: Principles of HERCULES

PURPOSE

To demonstrate the framework of a novel Hadamard-encoded spectral editing approach for simultaneously detecting multiple low-concentration brain metabolites in vivo at 3T.

METHODS

HERCULES (Hadamard Editing Resolves Chemicals Using Linear-combination Estimation of Spectra) is a four-step Hadamard-encoded editing scheme. 20-ms editing pulses are applied at: (A) 4.58 and 1.9 ppm; (B) 4.18 and 1.9 ppm; (C) 4.58 ppm; and (D) 4.18 ppm. Edited signals from γ-aminobutyric acid (GABA), glutathione (GSH), ascorbate (Asc), N-acetylaspartate (NAA), N-acetylaspartylglutamate (NAAG), aspartate (Asp), lactate (Lac), and likely 2-hydroxyglutarate (2-HG) are separated with reduced signal overlap into distinct Hadamard combinations: (A+B+C+D); (A+B-C-D); and (A-B+C-D). HERCULES uses a novel multiplexed linear-combination modeling approach, fitting all three Hadamard combinations at the same time, maximizing the amount of information used for model parameter estimation, in order to quantify the levels of these compounds. Fitting also allows estimation of the levels of total choline (tCho), myo-inositol (Ins), glutamate (Glu), and glutamine (Gln). Quantitative HERCULES results were compared between two grey- and white-matter-rich brain regions (11 min acquisition time each) in 10 healthy volunteers. Coefficients of variation (CV) of quantified measurements from the HERCULES fitting approach were compared against those from a single-spectrum fitting approach, and against estimates from short-TE PRESS data.

RESULTS

HERCULES successfully segregates overlapping resonances into separate Hadamard combinations, allowing for the estimation of levels of seven coupled metabolites that would usually require a single 11-min editing experiment each. Metabolite levels and CVs agree well with published values. CVs of quantified measurements from the multiplexed HERCULES fitting approach outperform single-spectrum fitting and short-TE PRESS for most of the edited metabolites, performing only slightly to moderately worse than the fitting method that gives the lowest CVs for tCho, NAA, NAAG, and Asp.

CONCLUSION

HERCULES is a new experimental approach with the potential for simultaneous editing and multiplexed fitting of up to seven coupled low-concentration and six high-concentration metabolites within a single 11-min acquisition at 3T.

Effects of carrier frequency mismatch on frequency-selective spectral editing

OBJECTIVES

This study sought to investigate the effects of carrier frequency mismatch on spectral editing and its correction by frequency matching of basis functions.

MATERIALS AND METHODS

Full density matrix computations and Monte-Carlo simulations based on magnetic resonance spectroscopy (MRS) data collected from five healthy volunteers at 7 T were used to analyze the effects of carrier frequency mismatch on spectral editing. Relative errors in metabolite quantification were calculated with and without frequency matching of basis functions. The algorithm for numerical computation of basis functions was also improved for higher computational efficiency.

RESULTS

We found significant errors without frequency matching of basis functions when carrier frequency mismatch was generally considered negligible. By matching basis functions with the history of frequency deviation, the mean errors in glutamate, glutamine, γ-aminobutyric acid, and glutathione concentrations were reduced from 3.90%, 1.85%, 11.53%, and 3.43% to 0.18%, 0.34%, 0.40%, and 0.51%, respectively.

CONCLUSION

Matching basis functions to frequency deviation history was necessary even when frequency deviations during frequency-selective spectral editing were fairly small. Basis set frequency matching significantly improved accuracy in the quantification of glutamate, glutamine, γ-aminobutyric acid, and glutathione concentrations.

The application of magnetic resonance fingerprinting to single voxel proton spectroscopy

Magnetic resonance fingerprinting has been proposed as a method for undersampling k-space while simultaneously yielding multiparametric tissue maps. In the context of single voxel spectroscopy, fingerprinting can provide a unified framework for parameter estimation. We demonstrate the utility of such a magnetic resonance spectroscopic fingerprinting (MRSF) framework for simultaneously quantifying metabolite concentrations, T₁ and T₂ relaxation times and transmit inhomogeneity for major singlets of N-acetylaspartate, creatine and choline. This is achieved by varying T_R , T_E and the flip angle of the first pulse in a PRESS sequence between successive excitations (i.e. successive T_R values). The need for multiparametric schemes such as MRSF for accurate medical diagnostics is demonstrated with the aid of realistic in vivo simulations; these show that certain schemes lead to substantial increases to the area under receiver operating characteristics of metabolite concentrations, when viewed as classifiers of pathologies. Numerical simulations and phantom and in vivo experiments using several different schedules of variable length demonstrate superior precision and accuracy for metabolite concentrations and longitudinal relaxation, and similar performance for the quantification of transverse relaxation.

Simultaneous measurement of glutamate, glutamine, GABA, and glutathione by spectral editing without subtraction

PURPOSE

To simultaneously measure glutamate, glutamine, γ-aminobutyric acid (GABA), and glutathione using spectral editing without subtraction at 7T.

METHODS

A novel spectral editing approach was proposed to simultaneously measure glutamate, glutamine, GABA, and glutathione using a TE of 56 ms at 7T. By numerical optimization of sequence timing in the presence of an editing pulse, the 4 metabolites all form relatively intense pseudo singlets with maximized peak amplitudes and minimized peak linewidths in 1 of the 3 interleaved spectra. For measuring glutamate, glutamine, and glutathione, the editing pulse targets the H3 protons of these metabolites near 2.12 parts per million. Both GABA H2 and H4 resonances are fully utilized in spectral fitting.

RESULTS

Concentration levels (/[total creatine]) of glutamate, glutamine, GABA, and glutathione from an 8 mL voxel in the pregenual anterior cingulate cortex of 5 healthy volunteers were found to be 1.26 ± 0.13, 0.33 ± 0.06, 0.13 ± 0.03, and 0.27 ± 0.03, respectively, with within-subject coefficient of variation at 3.2%, 8.2%, 7.1%, and 10.2%, respectively. The total scan time was less than 4.5 min.

CONCLUSIONS

The proposed new technique does not require data subtraction. The 3 major metabolites of the glutamatergic and GABAergic systems and the oxidative stress marker glutathione were all measured in 1 short scan with high precision.