1H Spectroscopy without Solvent Suppression: Characterization of Signal Modulations at Short Echo Times (original) (raw)
Related papers
Journal of Magnetic Resonance, 2002
Most Magnetic Resonance Spectroscopy (MRS) localization methods can generate gradient vibrations at acoustic frequencies and/or magnetic field oscillation, which can cause a time-varying magnetic field superimposed onto the static one. This effect can produce frequency modulations of the spectral resonances. When localized MRS data are acquired without water suppression, the associated frequency modulations are manifested as a manifold of spurious peaks, called sidebands, which occur symmetrically around the water resonance. These sidebands can be larger than the small metabolite resonances and can present a problem for the quantitation of the spectra, especially at short echo times. Furthermore, the resonance lineshapes may be distorted if any low frequency modulations are present. A simple solution is presented which consists of selecting the modulus of the acquired Free Induction Decay (FID) signal. Since the frequency modulations affect only the phase of the FID signal, the obtained real spectrum of the modulus is free from the spurious peaks where quantitative results may be directly obtained. Using this method, the distortions caused by the sidebands are removed. This is demonstrated by processing proton MRS spectra acquired without water suppression collected from a phantom containing metabolites at concentrations comparable to those in human brain and from a human subject using two different localization methods (PRESS and Chemical Shift Imaging PRESS-(CSI)). The results obtained illustrate the ability of this approach to remove the spurious peaks. The corrected spectra can then be fit accurately. This is confirmed by the results obtained from both the relative and the absolute metabolites concentrations in phantoms and in-vivo.
Toward quantitative short-echo-time in vivo proton MR spectroscopy without water suppression
Magnetic Resonance in Medicine, 2006
A methodological development for quantitative short-echotime (TE) in vivo proton MR spectroscopy (MRS) without water suppression (WS) is described that integrates experimental and software approaches. Experimental approaches were used to eliminate frequency modulation sidebands and first-order phase errors. The dominant water signal was modeled and extracted by the matrix pencil method (MPM) and was used as an internal reference for absolute metabolite quantification. Spectral fitting was performed by combining the baseline characterization by a wavelet transform (WT)-based technique and time-domain (TD) parametric spectral analysis using full prior knowledge of the metabolite model spectra. The model spectra were obtained by spectral simulation instead of in vitro measurements. The performance of the methodology was evaluated by Monte Carlo (MC) studies, phantom measurements, and in vivo measurements on rat brains. More than 10 metabolites were quantified from spectra measured at TE ؍ 20 ms on a 4.7 T system.
Observation of coupled1H metabolite resonances at long TE
Magnetic Resonance in Medicine, 2005
A PRESS localization 1 H MRS acquisition sequence with a Carr-Purcell train of refocusing pulses (CP-PRESS) has been implemented using global refocusing "sandwich" pulses. The CP pulse train minimized the effects of J-coupled dephasing in metabolites with strongly coupled, multiplet resonance groups as demonstrated in both phantom data and in vivo single-voxel spectroscopy in normal volunteers. Metabolites with multiplet resonance patterns were maintained with greater signal to noise and a simpler resonance pattern at long echo times. T 2 decay times for metabolites with singlet and multiplet resonances were similar to published values, except for the NAA multiplet at 2.5 ppm, which had a significantly shorter T 2 value (147 ms) than that typically reported for the singlet at 2.01 ppm. Metabolite-nulled spectra were acquired in normal volunteers to evaluate the effects of CP-PRESS on baseline signal contributions from residual water, lipids, and macromolecules. The T 2 decay times in four baseline regions in data acquired with the CP-PRESS sequence showed longer decays than corresponding regions in metabolite-nulled spectra from a standard PRESS sequence, but were significantly diminished long before the metabolites of interest were gone. The spectral analysis for spectra with longer TE times also showed less variability due the higher metabolite SNR, simpler spectral patterns, and the decreased baseline contributions.
In vivo proton spectroscopy without solvent suppression
Concepts in Magnetic Resonance, 2001
In 1 H MR spectroscopy of the human brain, it is common practice to suppress the solvent signal prior to acquisition. This reduces the large dynamic range which is otherwise required of the MR receiver and digitizer in order to detect the dilute metabolite resonances in the presence of the much larger water signal. However, complete solvent suppression is not always obtainable, particularly over large volumes and in superficial regions containing large susceptibility gradients. In this work, it demonstrated that modern commercial MR scanners possess the dynamic range necessary to adequately resolve the 1 H metabolites in unsuppressed spectra. Moreover, a postacquisition method is presented which can completely remove the intact water signal and accurately quantitate the metabolite peaks. Preserving the water signal in in vivo spectroscopy has several useful benefits, such as providing a high signal-to-noise ratio internal concentration, frequency, and line shape reference. Comparison is made between suppressed and unsuppressed spectra from both a phantom and the human brain acquired at 4 T.
Journal of Magnetic Resonance (1969), 1988
Noninvasive measurement of absolute metabolite concentrations is essential for understanding the physiology of living systems. Surface-coil NMR spectroscopy can achieve this by double tuning the coil to a reference nucleus of known concentration (usually water 'H), as suggested by Thulbom and Ackerman in 1983. The spatial sensitivities of the two nuclei are matched by using identical nutation angles at the coil center for each nucleus; the ratio of signals from each nucleus is then proportional to the ratio of concentrations and independent of the particular sample geometry. Water 'H concentration is relatively constant from one tissue to another. Coil loading affects Q, sensitivity, and nutation angle. Circuit analysis of these effects is experimentally confirmed. The value of the variable matching capacitance gives Q, pulse length, and sensitivity for any particular sample. A practical scheme for measuring "P concentrations, suitable for use by biologists, is presented. In vivo measurements of "P metabolite concentrations in rat tissue are in good agreement with in vitro values measured by chemical analysis. Measurement of absolute concentrations without a reference nucleus is possible if the volume of interest can be made to fall entirely within the tissue, for example by using gradient localization.
Magnetic Resonance in Medicine, 2010
Quantification of short-echo time proton magnetic resonance spectroscopy results in >18 metabolite concentrations (neurochemical profile). Their quantification accuracy depends on the assessment of the contribution of macromolecule (MM) resonances, previously experimentally achieved by exploiting the several fold difference in T 1. To minimize effects of heterogeneities in metabolites T 1 , the aim of the study was to assess MM signal contributions by combining inversion recovery (IR) and diffusion-weighted proton spectroscopy at high-magnetic field (14.1 T) and short echo time (58 msec) in the rat brain. IR combined with diffusion weighting experiments (with d/D 5 1.5/200 msec and b-value 5 11.8 msec/ mm 2) showed that the metabolite nulled spectrum (inversion time 5 740 msec) was affected by residuals attributed to creatine, inositol, taurine, choline, N-acetylaspartate as well as glutamine and glutamate. While the metabolite residuals were significantly attenuated by 50%, the MM signals were almost not affected (<8%). The combination of metabolite-nulled IR spectra with diffusion weighting allows a specific characterization of MM resonances with minimal metabolite signal contributions and is expected to lead to a more precise quantification of the neurochemical profile. Magn Reson Med 64:939-946, 2010. V
Quantitative Proton Magnetic Resonance Spectroscopy in Presence of Sidebands
IEEE International Symposium on Biomedical Imaging, 2007
We perform proton magnetic resonance spectroscopy (MRS) without water suppression in contrast to traditional water-suppressed MRS. The preserved water signal can be used as a reference for the absolute quantification of metabolites, reducing the total measurement time. However, the non-water-suppressed spectra can be contaminated by gradient-induced frequency modulations resulting in sidebands in the spectrum. Sidebands may obscure the metabolite resonances
Metabolite‐specific NMR spectroscopy in vivo
NMR in Biomedicine, 1997
An outline is presented of metabolite-specific in vivo NMR spectroscopy (particularly in brain). It reviews from a physical spectroscopist's perspective, the need for and the methods of observation of, individual metabolite resonances. © Abbreviations used: AX, Two single spins weakly coupled; AB, Two single spins strongly coupled; AX 3 , A single spin weakly coupled to a group of three spins; A 2 M 2 X 2 , Three weakly coupled spin pairs; AMNPQ, A single spin weakly coupled to two internally strongly coupled pairs of spins, the coupling between which is also strong; B 0 , Static magnetic field . A limited region (2 ppm to 3 ppm) of the 300 MHz proton spectrum from an acid extract of cat brain, reproduced by kind permission of Dr C. C. Hanstock.
Magnetic Resonance in Medicine, 1991
NMR spectroscopy using stimulated echoes (STEAM) has been used to study metabolites in different proximal skeletal muscles of normal volunteers at rest. Single scan water-suppressed proton NMR spectra obtained at 1.5 and 2.0 T (Siemens Magnetom) from a 64-ml volume-of-interest (VOI) yield resonances due to triglycerides, phosphocreatine plus a minor contribution from creatine, and betaines comprising camitine and choline-containing compounds. The observation of the pH-dependent resonances of carnosine required multiple acquisitions and echo times as short as 20 ms. T , and T2 relaxation times of muscle metabolites were obtained by varying the repetition time and echo time of the STEAM sequence, respectively. Although rather long T2 values such as 180 ms for (phospho-) creatine correspond to natural resonance linewidths of only 2 Hz, the observed linewidths of typically 10-12 Hz are entirely determined by the short T2 relaxation times (25-30 ms) of the water protons used for shimming. The spectroscopic results from 24 muscle studies on 17 young male volunteers show remarkable intra-and interindividual differences in the absolute signal intensities of mobile lipids. Further metabolic variations were observed for the relative concentrations of betaines (by a factor of 2) and carnosine (by a factor of 3 ) when total creatine is assumed to be constant. o
Magnetic Resonance Spectroscopy Data Analysis for Clinical Applications
Proton magnetic resonance spectroscopy (MRS) is provided as an option by most manufacturers and is becoming more common in clinical practice, particularly for prostate and neurological applications. Although MRS can be performed on nuclei such as 31P and 13C, proton (1H) MRS is the easiest and least expensive spectroscopy for all MRI system because requires only a test phantom and a common software package which automate acquisition sequences and post- processing for metabolites quantification. Non-proton spectroscopy indeed requires radio frequency (RF) coils tuned to the Larmor frequency of other nuclei plus matching preamplifiers, hybrids, and a broadband power amplifier. Sometimes however, the efficient implementation of MRS acquisition protocols is beyond the expectations for most MR technologists, therefore MR physicists are often called in to perform MRS procedures to evaluate whether problems with proton MRS are due to equipment malfunctions, software problems, or operator errors. In this thesis we focus on some of these problems.