3D localized in vivo 1H spectroscopy of human brain by using a hybrid of 1D-Hadamard with 2D-chemical shift imaging

Fast, regional three-dimensional hybrid (1D-Hadamard 2D-rosette) proton MR spectroscopic imaging in the human temporal lobes

1 H-MRSI is commonly performed with gradient phase encoding, due to its simplicity and minimal radio frequency (RF) heating (specific absorption rate). Its two well-known main problems-(i) "voxel bleed" due to the intrinsic point-spread function, and (ii) chemical shift displacement error (CSDE) when slice-selective RF pulses are used, which worsens with increasing volume of interest (VOI) size-have long become accepted as unavoidable. Both problems can be mitigated with Hadamard multislice RF encoding. This is demonstrated and quantified with numerical simulations, in a multislice phantom and in five healthy young adult volunteers at 3 T, targeting a 2-cm thick temporal lobe VOI through the bilateral hippocampus. This frequently targeted region (e.g. in epilepsy and Alzheimer's disease) is subject to strong, 1-2 ppm.cm-1 regional B0, susceptibility gradients that can dramatically reduce the signal-to-noise ratio (SNR) and water suppression effectiveness. The chemical shift imaging (CSI) sequence used a 3-ms Shinnar-Le Roux (SLR) 90° RF pulse, acquiring eight steps in the slice direction. The Hadamard sequence acquired two overlapping slices using the same SLR 90° pulses, under twofold stronger gradients that proportionally halved the CSDE. Both sequences used 2D 20 × 20 rosette spectroscopic imaging (RSI) for in-plane spatial localization and both used RF and gradient performance characteristics that are easily met by all modern MRI instruments. The results show that Hadamard spectroscopic imaging (HSI) suffered dramatically less signal bleed within the VOI compared with CSI (<1% vs. approximately 26% in simulations; and 5%-8% vs. >50%) in a phantom specifically designed to test these effects. The voxels' SNR per unit volume per unit time was also 40% higher for HSI. In a group of five healthy volunteers, we show that HSI with in-plane 2D-RSI facilitates fast, 3D multivoxel encoding at submilliliter spatial resolution, over the bilateral human hippocampus, in under 10 min, with negligible CSDE, spectral and spatial contamination and more than 6% improved SNR per unit time per unit volume.

Fast data acquisition techniques in magnetic resonance spectroscopic imaging

Magnetic resonance spectroscopic imaging (MRSI) is an important technique for assessing the spatial variation of metabolites in vivo. The long scan times in MRSI limit clinical applicability due to patient discomfort, increased costs, motion artifacts, and limited protocol flexibility. Faster acquisition strategies can address these limitations and could potentially facilitate increased adoption of MRSI into routine clinical protocols with minimal addition to the current anatomical and functional acquisition protocols in terms of imaging time. Not surprisingly, a lot of effort has been devoted to the development of faster MRSI techniques that aim to capture the same underlying metabolic information (relative metabolite peak areas and spatial distribution) as obtained by conventional MRSI, in greatly reduced time. The gain in imaging time results, in some cases, in a loss of signal-to-noise ratio and/or in spatial and spectral blurring. This review examines the current techniques and advances in fast MRSI in two and three spatial dimensions and their applications. This review categorizes the acceleration techniques according to their strategy for acquisition of the k-space. Techniques such as fast/turbo-spin echo MRSI, echo-planar spectroscopic imaging, and non-Cartesian MRSI effectively cover the full k-space in a more efficient manner per T_R . On the other hand, techniques such as parallel imaging and compressed sensing acquire fewer k-space points and employ advanced reconstruction algorithms to recreate the spatial-spectral information, which maintains statistical fidelity in test conditions (ie no statistically significant differences on voxel-wise comparisions) with the fully sampled data. The advantages and limitations of each state-of-the-art technique are reviewed in detail, concluding with a note on future directions and challenges in the field of fast spectroscopic imaging.

Three-dimensional Hadamard-encoded proton spectroscopic imaging in the human brain using time-cascaded pulses at 3 Tesla

PURPOSE

To reduce the specific-absorption-rate (SAR) and chemical shift displacement (CSD) of three-dimensional (3D) Hadamard spectroscopic imaging (HSI) and maintain its point spread function (PSF) benefits.

METHODS

A 3D hybrid of 2D longitudinal, 1D transverse HSI (L-HSI, T-HSI) sequence is introduced and demonstrated in a phantom and the human brain at 3 Tesla (T). Instead of superimposing each of the selective Hadamard radiofrequency (RF) pulses with its N single-slice components, they are cascaded in time, allowing N-fold stronger gradients, reducing the CSD. A spatially refocusing 180° RF pulse following the T-HSI encoding block provides variable, arbitrary echo time (TE) to eliminate undesirable short T2 species' signals, e.g., lipids.

RESULTS

The sequence yields 10-15% better signal-to-noise ratio (SNR) and 8-16% less signal bleed than 3D chemical shift imaging of equal repetition time, spatial resolution and grid size. The 13 ± 6, 22 ± 7, 24 ± 8, and 31 ± 14 in vivo SNRs for myo-inositol, choline, creatine, and N-acetylaspartate were obtained in 21 min from 1 cm(3) voxels at TE ≈ 20 ms. Maximum CSD was 0.3 mm/ppm in each direction.

CONCLUSION

The new hybrid HSI sequence offers a better localized PSF at reduced CSD and SAR at 3T. The short and variable TE permits acquisition of short T2 and J-coupled metabolites with higher SNR.

In vivo free induction decay based 3D multivoxel longitudinal hadamard spectroscopic imaging in the human brain at 3 T

We propose and demonstrate a full 3D longitudinal Hadamard spectroscopic imaging scheme for obtaining chemical shift maps, using adiabatic inversion pulses to encode the spins' positions. The approach offers several advantages over conventional Fourier-based encoding methods, including a localized point spread function; no aliasing, allowing for volumes of interest smaller than the object being imaged; an option for acquiring noncontiguous voxels; and inherent outer volume rejection. The latter allows for doing away with conventional outer volume suppression schemes, such as point resolved spectroscopy (PRESS) and stimulated echo acquisition mode (STEAM), and acquiring non-spin-echo spectra with short acquisition delay times, limited only by the excitation pulse's duration. This, in turn, minimizes T2 decay, maximizes the signal-to-noise ratio, and reduces J-coupling induced signal decay. Results are presented for both a phantom and an in vivo healthy volunteer at 3 T.

Non-spin-echo 3D transverse hadamard encoded proton spectroscopic imaging in the human brain

A non-spin-echo multivoxel proton MR localization method based on three-dimensional transverse Hadamard spectroscopic imaging is introduced and demonstrated in a phantom and the human brain. Spatial encoding is achieved with three selective 90° radiofrequency pulses along perpendicular axes: The first two create a longitudinal ±M(Z) Hadamard order in the volume of interest. The third pulse spatially Hadamard-encodes the ±M(Z)s in the volume of interest in the third direction while bringing them to the transverse plane to be acquired immediately. The approaching-ideal point spread function of Hadamard encoding and very short acquisition delay yield signal-to-noise-ratios of 20 ± 8, 23 ± 9, and 31 ± 10 for choline, creatine, and N-acetylaspartate in the human brain at 1.5 T from 1 cm(3) voxels in 21 min. The advantages of transverse Hadamard spectroscopic imaging are that unlike gradient (Fourier) phase-encoding: (i) the volume of interest does not need to be smaller than the field of view to prevent aliasing; (ii) the number of partitions in each direction can be small, 8, 4, or even 2 at no cost in point spread function; (iii) the volume of interest does not have to be contiguous; and (iv) the voxel profile depends on the available B1 and pulse synthesis paradigm and can, therefore, at least theoretically, approach "ideal" "1" inside and "0" elsewhere.

Brain metabolites B1-corrected proton T1 mapping in the rhesus macaque at 3 T

The accuracy of metabolic quantification in MR spectroscopy is limited by the unknown radiofrequency field and T(1). To address both issues in proton ((1)H) MR spectroscopy, we obtained radiofrequency field-corrected T(1) maps of N-acetylaspartate, choline, and creatine in five healthy rhesus macaques at 3 T. For efficient use of the 4 hour experiment, we used a new three-point protocol that optimizes the precision of T(1) in three-dimensional (1)H-MR spectroscopy localization for extensive, approximately 30%, brain coverage at 0.6 x 0.6 x 0.5 cm(3) = 180-microL spatial resolution. The resulting mean T(1)s in 700 voxels were N-acetylaspartate = 1232 +/- 44, creatine = 1238 +/- 23 and choline = 1107 +/- 56 ms (mean +/- standard error of the mean). Their histograms from all 140 voxels in each animal were similar in position and shape, characterized by standard errors of the mean of the full width at half maximum divided by their means of better than 8%. Regional gray matter N-acetylaspartate, choline, and creatine T(1)s (1333 +/- 43, 1265 +/- 52, and 1131 +/- 28 ms) were 5-10% longer than white matter: 1188 +/- 34, 1201 +/- 24, and 1082 +/- 50 ms (statistically significant for the N-acetylaspartate only), all within 10% of the corresponding published values in the human brain.

Detection of lactate with a hadamard slice selected, selective multiple quantum coherence, chemical shift imaging sequence (HDMD-SelMQC-CSI) on a clinical MRI scanner: Application to tumors and muscle ischemia

Lactate is an important metabolite in normal and malignant tissues detectable by NMR spectroscopy; however, it has been difficult to clinically detect the lactate methyl resonance because it is obscured by lipid resonances. The selective homonuclear multiple quantum coherence transfer technique offers a method for distinguishing lipid and lactate resonances. We implemented a three-dimensional selective homonuclear multiple quantum coherence transfer version with Hadamard slice selection and two-dimensional phase encoding (Hadamard encoded-selective homonuclear multiple quantum coherence transfer-chemical shift imaging) on a conventional clinical MR scanner. Hadamard slice selection is explained and demonstrated in vivo. This is followed by 1-cm(3) resolution lactate imaging with detection to 5-mM concentration in 20 min on a 3-T clinical scanner. An analysis of QSel gradient duration and amplitude effects on lactate and lipid signal is presented. To demonstrate clinical feasibility, a 5-min lactate scan of a patient with a non-Hodgkin's lymphoma in the superficial thigh is reported. The elevated lactate signal coincides with the T(2)-weighted image of this tumor. As a test of selective homonuclear multiple quantum coherence transfer sensitivity, a thigh tourniquet was applied to a normal volunteer and an increase in lactate was detected immediately after tourniquet flow constriction. In conclusion, the Hadamard encoded-selective homonuclear multiple quantum coherence transfer-chemical shift imaging sequence is demonstrated on a phantom and in two lipid-rich, clinically relevant, in vivo conditions.

Retrospective correction for T1-weighting bias in T2 values obtained with various spectroscopic spin-echo acquisition schemes

Localized tissue transverse relaxation time (T(2)) is obtained by fitting a decaying exponential to the signals from several spin-echo experiments at different echo times (TE). Unfortunately, time constraints in magnetic resonance spectroscopy (MRS) often mandate in vivo acquisition schemes at short repetition times (TR), that is, comparable with the longitudinal relaxation constant (T(1)). This leads to different T(1)-weighting of the signals at each TE. Unaccounted for, this varying weighting causes systematic underestimation of the T(2)'s, sometimes by as mush as 30%. In this article, we (i) analyze the phenomenon for common MRS spin-echo T(2) acquisition schemes; (ii) propose a general post hoc T(1)-bias correction for any (TR, TE) combination; (iii) show that approximate knowledge of T(1) is sufficient, since a 20% uncertainty in T(1) leads to under 3% bias in T(2); and consequently, (iv) efficient, precision-optimized short TR spin-echo T(2) measurement protocols can be designed and used without risk of accuracy loss. Tables of correction for single-refocusing (conventional) spin-echo and double refocusing, such as, PRESS acquisitions, are provided.

Simultaneous two-voxel localized (1)H-observed (13)C-edited spectroscopy for in vivo MRS on rat brain at 9.4T: Application to the investigation of excitotoxic lesions

(13)C spectroscopy combined with the injection of (13)C-labeled substrates is a powerful method for the study of brain metabolism in vivo. Since highly localized measurements are required in a heterogeneous organ such as the brain, it is of interest to augment the sensitivity of (13)C spectroscopy by proton acquisition. Furthermore, as focal cerebral lesions are often encountered in animal models of disorders in which the two brain hemispheres are compared, we wished to develop a bi-voxel localized sequence for the simultaneous bilateral investigation of rat brain metabolism, with no need for external additional references. Two sequences were developed at 9.4T: a bi-voxel (1)H-((13)C) STEAM-POCE (Proton Observed Carbon Edited) sequence and a bi-voxel (1)H-((13)C) PRESS-POCE adiabatically decoupled sequence with Hadamard encoding. Hadamard encoding allows both voxels to be recorded simultaneously, with the same acquisition time as that required for a single voxel. The method was validated in a biological investigation into the neuronal damage and the effect on the Tri Carboxylic Acid cycle in localized excitotoxic lesions. Following an excitotoxic quinolinate-induced localized lesion in the rat cortex and the infusion of U-(13)C glucose, two (1)H-((13)C) spectra of distinct (4x4x4mm(3)) voxels, one centred on the injured hemisphere and the other on the contralateral hemisphere, were recorded simultaneously. Two (1)H bi-voxel spectra were also recorded and showed a significant decrease in N-acetyl aspartate, and an accumulation of lactate in the ipsilateral hemisphere. The (1)H-((13)C) spectra could be recorded dynamically as a function of time, and showed a fall in the glutamate/glutamine ratio and the presence of a stable glutamine pool, with a permanent increase of lactate in the ipsilateral hemisphere. This bi-voxel (1)H-((13)C) method can be used to investigate simultaneously both brain hemispheres, and to perform dynamic studies. We report here the neuronal damage and the effect on the Tri Carboxylic Acid cycle in localized excitotoxic lesions.

Voxel-shift and interpolation for hadamard-encoded MR images

Although Fourier gradient phase-encoding and Hadamard radio-frequency encoding are two established spatial MR localization techniques, the absence of voxel-shift and interpolation postprocessing algorithms for the latter has always placed it at a discouraging disadvantage. This article presents a method for voxel-shift and interpolation of Hadamard-encoded data and demonstrates both theoretically and experimentally the similarities of the respective operations between the two localization methods.

Lactate imaging with Hadamard-encoded slice-selective multiple quantum coherence chemical-shift imaging

The ability to generate in vivo maps of lactate may have significant diagnostic utility in staging and treatment planning of a wide variety of cancers. The double selective multiple quantum filter technique (SelMQC) has been shown to be effective for nonlocalized detection of lactate with little or no interference from other signals. Here the SelMQC technique has been combined with longitudinal Hadamard slice selection and chemical shift imaging (CSI) to yield slice-selective images of lactate. The technique is shown to be effective in phantoms and in WSU-DLCL2 xenografts implanted in flanks of SCID mice. Tumors exhibited an annulus of elevated lactate concentration surrounding a necrotic tumor core.

Anteroposterior hippocampal metabolic heterogeneity: three-dimensional multivoxel proton 1H MR spectroscopic imaging--initial findings

PURPOSE

To quantify proton magnetic resonance (MR) spectroscopy-detectable metabolite concentrations along anteroposterior axis of hippocampus in healthy young and elderly subjects.

MATERIALS AND METHODS

Young (three women, three men; age range, 25-35 years) and elderly (four women, two men; age range, 68-72 years) groups underwent MR imaging and proton MR spectroscopic imaging at 3 T in this HIPAA-compliant prospective study and gave institutional review board-approved written consent. Volume of interest was centered on and tilted parallel to hippocampal anteroposterior plane. Absolute N-acetylaspartate (NAA), choline, and creatine levels were obtained in each voxel, with phantom replacement.

RESULTS

Mean NAA, creatine, and choline concentrations in the young group were higher in posterior hippocampus (12.9 mmol/L +/- 2.0 [standard deviation], 7.8 mmol/L +/- 1.2, 2.3 mmol/L +/- 0.4, respectively) than anterior hippocampus (8.0 mmol/L +/- 1.1, 6.0 mmol/L +/- 1.4, 1.5 mmol/L +/- 0.2; P = .005, .02, and .0002, respectively). In the elderly group, mean concentrations were higher in posterior hippocampus (8.6 mmol/L +/- 0.9, 5.6 mmol/L +/- 0.6, 1.5 mmol/L +/- 0.2, respectively) than anterior hippocampus (7.2 mmol/L +/- 1.0, 2.4 mmol/L +/- 0.3, 1.0 mmol/L +/- 0.2; P = .006, .0001, .04, respectively). Mean concentrations were significantly higher in the young group (13.2 mmol/L +/- 1.0, 7.4 mmol/L +/- 0.8, 2.1 mmol/L +/- 0.3, respectively) than in the elderly group (9.0 mmol/L +/- 1.0, 5.8 mmol/L +/- 0.8, 1.8 mmol/L +/- 0.3; P = .0001, .01, .05, respectively). Posteroanterior metabolic gradients differed: NAA decreased faster in the young group (-1.0 mmol/L x cm(-1)) than the elderly group (-0.7 mmol/L x cm(-1)); creatine and choline concentrations decreased faster in the elderly group (-0.8 and -0.058 mmol/L x cm(-1), respectively) than the young group (-0.16 and -0.008 mmol/L x cm(-1), respectively). No left-right metabolic differences were found.

CONCLUSION

Significant metabolic heterogeneity was observed between groups and along anteroposterior axis of healthy hippocampus in both groups. Age matching and consistent voxel placement are important for correct comparisons of both absolute metabolic levels and metabolite ratios in longitudinal intra- and intersubject cross-sectional studies.

Proton MR spectroscopic imaging of rhesus macaque brain in vivo at 7T

Due to the overall similarity of their brains' structure and physiology to its human counterpart, nonhuman primates provide excellent model systems for the pathogenesis of neurological diseases and their response to treatments. Its much smaller size, 80 versus 1250 cm(3), however, requires proportionally higher spatial resolution to study, nondestructively, as many analogous regions as efficiently as possible in anesthetized animals. The confluence of these requirements underscores the need for the highest sensitivity, spatial coverage, resolution, and exam speed. Accordingly, we demonstrate the feasibility of 3D multi-voxel, proton ((1)H) MRSI at (0.375 cm)(3)=0.05 cm(3) isotropic spatial resolution over 21 cm(3) (approximately 25%) of the anesthetized rhesus macaques brain at 7T in 25 min. These voxels are x10(2)-10(1) times smaller than the 8-1 cm(3) common to (1)H-MRS in humans, retaining similar proportions between the macaque and human brain. The spectra showed a signal-to-noise-ratio (SNR) approximately 9-10 for the major metabolites and the interanimal SNR spatial distribution reproducibility was in the +/-10% range for the standard error of their means (SEMs). Their metabolites' linewidths, 9+/-2 Hz, yield excellent spectral resolution as well. These results indicate that 3D (1)H-MRSI can be integrated into comprehensive MR studies in primates at such high fields.

Chemical-shift artifact reduction in Hadamard-encoded MR spectroscopic imaging at high (3T and 7T) magnetic fields

Proton MR spectroscopic imaging (MRSI) at higher magnetic fields (B(0)) suffers metabolite localization errors from different chemical-shift displacements (CSDs) if spatially-selective excitation is used. This phenomenon is exacerbated by the decreasing radiofrequency (RF) field strength, B(1), at higher B(0)s, precluding its suppression with stronger gradients. To address this, two new methods are proposed: 1) segmenting the volume-of-interest (VOI) into several slabs, allowing proportionally stronger slice-select gradients; and 2) sequentially cascading rather than superposing the components of the Hadamard selective pulses used for reasons of better point-spread function (PSF) to localize the few slices within each slab. This can reduce the peak B(1) to that of a single slice. Combining these approaches permits us to increase the selective gradient four- to eightfold per given B(1), to 12 or 18mT/m for 4- or 2-cm VOIs. This "brute force" approach reduces the CSD to under 0.05 cm/ppm at 7T, or less than half that at 3T.

Human brain-structure resolvedT2 relaxation times of proton metabolites at 3 tesla

The transverse relaxation times, T(2), of N-acetylaspartate (NAA), total choline (Cho), and creatine (Cr) obtained at 3T in several human brain regions of eight healthy volunteers are reported. They were obtained simultaneously in 320 voxels with three-dimensional (3D) proton MR spectroscopy ((1)H-MRS) at 1 cm(3) spatial resolution. A two-point protocol, optimized for the least error per given time by adjusting both the echo delay (TE(i)) and number of averages, N(i), at each point, was used. Eight healthy subjects (four males and four females, age = 26 +/- 2 years) underwent the hour-long procedure of four 15-min, 3D acquisitions (TE(1) = 35 ms, N(1) = 1; and TE(2) = 285 ms, N(2) = 3). The results reveal that across all subjects the NAA and Cr T(2)s in gray matter (GM) structures (226 +/- 17 and 137 +/- 12 ms, respectively) were 13-17% shorter than the corresponding T(2)s in white matter (WM; 264 +/- 10 and 155 +/- 7 ms, respectively). The T(2)s of Cho did not differ between GM and WM (207 +/- 17 and 202 +/- 8, respectively). For the purpose of metabolic quantification, these values justify to within +/-10% the previous use of one T(2) per metabolite for 1) the entire brain and 2) all subjects. These T(2) values (which to our knowledge were obtained for the first time at this field, spatial resolution, coverage, and precision) are essential for reliable absolute metabolic quantification.

Robust fully automated shimming of the human brain for high-field 1H spectroscopic imaging

Although a variety of methods have been proposed to provide automated adjustment of shim homogeneity, these methods typically fail or require large numbers of iterations in vivo when applied to regions with poor homogeneity, such as the temporal lobe. These limitations are largely due to 1) the limited accuracy of single evolution time measurements when full B0 mapping studies are used, and 2) inaccuracies arising from projection-based methods when the projections pass through regions where the inhomogeneity exceeds the order of the fitted parameters. To overcome these limitations we developed a novel B0 mapping method using multiple evolution times with a novel unwrapping scheme in combination with a user-defined ROI selection tool. We used these methods at 4T on 10 control subjects to obtain high-resolution spectroscopic images of glutamate from the bilateral hippocampi.

Optimizing the efficiency of high-field multivoxel spectroscopic imaging by multiplexing in space and time

A new strategy to yield information from the maximum number of voxels, each at the optimum signal-to-noise ratio (SNR) per unit time, in MR spectroscopic imaging (MRSI) is introduced. In the past, maximum acquisition duty-cycle was obtained by multiplexing in time several single slices each repetition time (TR), while optimal SNR was achieved by encoding the entire volume of interest (VOI) each TR. We show that optimal SNR and acquisition efficiency can both be achieved simultaneously by multiplexing in space and time several slabs of several slices, each. Since coverage of common VOIs in 3D proton MRSI in the human brain typically requires eight or more slices, at 3 T or higher magnetic fields, two or more slabs can fit into the optimum TR (approximately 1.6 s). Since typically four or less slices would then fit into each slab, Hadamard encoding is favored in that direction for slice profile reasons. It is demonstrated that per fixed examination length, the new method gives, at 3 T, twice as many voxels, each of the same SNR and size, compared with current 3D chemical shift imaging techniques. It is shown that this gain will increase for more extensive spatial coverage or higher fields.

Reducing voxel bleed in Hadamard-encoded MRI and MRS

The point spread function (PSF) of Hadamard encoding deviates from its ideal profile due to practical (as opposed to intrinsic) reasons. Finite radiofrequency (RF) pulse length and gradient strength cause slice profile imperfections that lead to cross-talk ("voxel bleed") as large as 17% for a 1-KHz bandwidth, 5.12-ms RF pulse under 3 mT/m. This could adversely affect localization and quantification, and consequently clinical usefulness. A simple modification of the Hadamard RF pulse synthesis that exploits its unique ability to encode noncontiguous slices is proposed and shown to markedly improve the PSF. Computer simulation, in vitro and in vivo experiments confirm the theoretical derivation of voxel bleed reduction from approximately 17% to below 5% per Hadamard-encoded direction.

Spatial localization in nuclear magnetic resonance spectroscopy

The ability to select a discrete region within the body for signal acquisition is a fundamental requirement of in vivo NMR spectroscopy. Ideally, it should be possible to tailor the selected volume to coincide exactly with the lesion or tissue of interest, without loss of signal from within this volume or contamination with extraneous signals. Many techniques have been developed over the past 25 years employing a combination of RF coil properties, static magnetic field gradients and pulse sequence design in an attempt to meet these goals. This review presents a comprehensive survey of these techniques, their various advantages and disadvantages, and implications for clinical applications. Particular emphasis is placed on the reliability of the techniques in terms of signal loss, contamination and the effect of nuclear relaxation and J-coupling. The survey includes techniques based on RF coil and pulse design alone, those using static magnetic field gradients, and magnetic resonance spectroscopic imaging. Although there is an emphasis on techniques currently in widespread use (PRESS, STEAM, ISIS and MRSI), the review also includes earlier techniques, in order to provide historical context, and techniques that are promising for future use in clinical and biomedical applications.

Robust analysis of short echo time (1)H MRSI of human brain

Short echo time proton MR Spectroscopic Imaging (MRSI) suffers from low signal-to-noise ratio (SNR), limiting accuracy to estimate metabolite intensities. A method to coherently sum spectra in a region of interest of the human brain by appropriate peak alignment was developed to yield a mean spectrum with increased SNR. Furthermore, principal component (PC) spectra were calculated to estimate the variance of the mean spectrum. The mean or alternatively the first PC (PC(1)) spectrum from the same region can be used for quantitation of peak areas of metabolites in the human brain at increased SNR. Monte Carlo simulations showed that both mean and PC(1) spectra were more accurate in estimating regional metabolite concentrations than solutions that regress individual spectra against the tissue compositions of MRSI voxels. Back-to-back MRSI studies on 10 healthy volunteers showed that mean spectra markedly improved reliability of brain metabolite measurements, most notably for myo-inositol, as compared to regression methods.

Acquisition time reduction in magnetic resonance spectroscopic imaging using discrete wavelet encoding

This paper describes a new magnetic resonance spectroscopic imaging (MRSI) technique based upon the discrete wavelet transform to reduce acquisition time and cross voxel contamination. Prototype functions called wavelets are used in wavelet encoding to localize defined regions in localized space by dilations and translations. Wavelet encoding in MRSI is achieved by matching the slice selective RF pulse profiles to a set of dilated and translated wavelets. Single and dual band slice selective excitation and refocusing pulses, with profiles resembling Haar wavelets, are used in a spin-echo sequence to acquire 2D-MRSI wavelet encoding data. The 2D space region is spanned up to the desired resolution by a proportional number of dilations (increases in the localization gradients) and translations (frequency shift) of the Haar wavelets (RF pulses). Acquisition time is reduced by acquiring successive MR signals from regions of space with variable size and different locations with no requirement for a TR waiting time between acquisitions. An inverse wavelet transform is performed on the data to produce the correct spatial MR signal distribution.

MR spectroscopy of brain tumors

MR Spectroscopy provides a means to characterize the metabolite profiles of tumoral and non-tumoral lesions in the brain. This article aims to provide tools to increase our sensitivity and specificity of neurodiagnosis, particularly in combination with other advanced MRI techniques such as perfusion MR imaging.

Improved spectral quality for 3D MR spectroscopic imaging using a high spatial resolution acquisition strategy

Spectral quality in (1)H MR spectroscopic imaging (MRSI) of the brain is often significantly degraded in regions subject to local magnetic susceptibility variations, which results in broadened and distorted spectral lineshapes. In this report, a modified acquisition strategy for volumetric echo-planar spectroscopic imaging (3D EPSI) is presented that extends the region of the brain that can be observed. The data are sampled at higher spatial resolution, then corrected for local B(0) shifts and reconstructed such that the final spatial resolution matches that of 3D EPSI data acquired with the conventional lower spatial resolution. Comparison of in vivo data obtained at 1.5 T with these two acquisition schemes shows that the high spatial resolution acquisition provides considerable reduction of spectral linewidths in many problematic brain regions, though with a reduction in signal-to-noise ratio by a factor of approximately 1.4 to 1.6 for the matrix sizes used in this study. However, the effect of the increased noise was largely offset by the improved spectral quality, leading to an overall improvement of the metabolite image quality obtained using automated spectral analysis.

High-resolution 3D proton spectroscopic imaging of the human brain at 3 T: SNR issues and application for anatomy-matched voxel sizes

In a systematic study on the interdependence of linewidth, signal-to-noise ratio (SNR), and spatial resolution in 3D proton spectroscopic imaging ((1)H-SI) at 3 T, we demonstrate reduced linewidths with increased spatial resolution due to reduced magnetic inhomogeneity within the brain. High-precision quantitative data (0.75-0.094 cm(3)) were obtained for all resolutions, enabling the creation of metabolic maps that display details such as the ventricles, sulci, and gyri. High-resolution (1)H-SI allows differences in metabolic ratios to be estimated for anatomically defined regions in gray (GM) and white matter (WM). Seven distinct regions in a healthy brain were anatomically segmented and their metabolic ratios were compared quantitatively. Data from a tumor patient are also presented to demonstrate potential clinical applications. Because of the high resolution, the metabolite ratios could be determined for distinct pathologic regions within the tumor and its surroundings. The method was additionally applied to a patient with patchy Pelizaeus Merzbacher disease (PMD), and compared to single-voxel spectroscopy performed in the same session. High-resolution SI data were demonstrated in our study to allow the direct matching of anatomic and metabolic images. This may enhance the clinical value of (1)H-SI.

SNR versus resolution in 3D 1H MRS of the human brain at high magnetic fields

It is commonly accepted that the signal-to-noise ratio (SNR = peak-signal/RMS-noise) per-unit-time of proton MR spectroscopy (1H-MRS) is linearly proportional to the voxel volume. Consequently, with a headcoil and 30-min acquisition, 1 cm3 is considered the SNR-limited spatial resolution barrier in the human brain. However, since local linewidths, Delta(upsilon*) = (piT2*)(-1), at high magnetic fields (B0), are dominated by regional inhomogeneities (DeltaB0), i.e., T2* << T2, reducing the voxel dimensions may increase T2*. This could compensate, in part, for signal loss with volume decrease. It is shown that for two cubic voxels of sides l1 and l2, l1 > l2, as the volume decreases by (l1/l2)3, their SNR ratio is reduced by only (l1/l2)2 due to a commensurate T2* increase of l1/l2. This is demonstrated in a phantom and the brains of volunteers, with 3D 1H-MRS in a headcoil at 4 T. It is shown that while the cubic voxels' dimensions were all halved, reducing their volume eightfold, their metabolites' SNR decreased only fourfold, due to their Delta(upsilon*s') twofold decrease. In other words, both spatial and spectral resolutions were doubled at a significantly, x2, smaller-than-expected SNR loss. This advantage was exploited to produce quality high spatial resolution, 0.75 x 0.75 x 0.75 cm3, metabolic maps in a 27-min acquisition.

Assessment of 3D proton MR echo-planar spectroscopic imaging using automated spectral analysis

For many clinical applications of proton MR spectroscopic imaging (MRSI) of the brain, diagnostic assessment is limited by insufficient coverage provided by single- or multislice acquisition methods as well as by the use of volume preselection methods. Additionally, traditional spectral analysis methods may limit the operator to detailed analysis of only a few selected brain regions. It is therefore highly desirable to use a fully 3D approach, combined with spectral analysis procedures that enable automated assessment of 3D metabolite distributions over the whole brain. In this study, a 3D echo-planar MRSI technique has been implemented without volume preselection to provide sufficient spatial resolution with maximum coverage of the brain. Using MRSI acquisitions in normal subjects at 1.5T and a fully automated spectral analysis procedure, an assessment of the resultant spectral quality and the extent of viable data in human brain was carried out. The analysis found that 69% of brain voxels were obtained with acceptable spectral quality at TE = 135 ms, and 52% at TE = 25 ms. Most of the rejected voxels were located near the sinuses or temporal bones and demonstrated poor B0 homogeneity and additional regions were affected by stronger lipid contamination at TE = 25 ms.

Chemical-shift imaging utilizing the positional shifts along the readout gradient direction

In this work, we describe a method that uses the linear phase acquired during the readout period due to chemical shift to generate individual magnetic resonance (MR) images of chemically shifted species. The method utilizes sets of Fourier (or k-space) data acquired with different directions of the readout gradient and a postprocessing algorithm to generate chemical shift images. The methodology is developed for both Cartesian data acquisition and for radial data acquisition. The method is presented here for two chemically shifted species but it can be extended to more species. In this work, we present the theory, show the results in phantoms and in human images, and discuss the artifacts and signal-to-noise ratio of the images obtained with the technique.

Dynamic imaging with multiple resolutions along phase-encode and slice-select dimensions

An implementation is reported of an imaging method to obtain MUltiple Resolutions along Phase-encode and Slice-select dimensions (MURPS), which enables dynamic imaging of focal changes using a graded, multiresolution approach. MURPS allows one to trade spatial resolution in part of the volume for improved temporal resolution in dynamic imaging applications. A unique method of Hadamard slice encoding is used, enabling the varying of the phase encode and slice resolution while maintaining a constant effective TR throughout the entire 3-D volume. MURPS was implemented using a gradient-recalled echo sequence, and its utility was demonstrated for MR temperature monitoring. In this preliminary work, it has been shown that changes throughout a large volume can be effectively monitored in times that would normally only permit dynamic imaging in one or a very few slices.

Two-voxel localization sequence for in vivo two-dimensional homonuclear correlation spectroscopy

The combination of localized 2D 1H MR correlation spectroscopy and Hadamard encoding allows the simultaneous acquisition of multiple volumes of interest without an increase in the experimental duration, compared to single-voxel acquisition. In the present study, 2D correlation spectra were acquired simultaneously within 20 to 40 min in two voxels located in each hemisphere of the rat brain. An intervoxel distance of 20% of the voxel size was sufficient to limit spatial contamination. The following cerebral metabolites gave detectable crosspeaks: N-acetylaspartate, the glutamate/glutamine pool, aspartate, phosphoethanolamine, glucose, glutathione, taurine, myo-inositols, lactate, threonine, gamma-aminobutyric acid, and alanine. Most of the metabolites were measured without contamination of other resonances.

Line scan echo planar spectroscopic imaging

A line scan echo planar spectroscopic imaging (LSEPSI) sequence is presented which can rapidly produce 2D chemical shift imaging (CSI) data with minimal relaxation weighting and motion-related artifacts. The technique is based on successive "snapshot" 1D CSI acquisitions of individual tissue columns, and avoids T(1) saturation problems associated with the short TR periods needed for very rapid scanning with either conventional or echo planar-based 2D CSI methods. Potential applications include rapid fat/water spectral quantitation in the abdomen and internally referenced temperature monitoring for interventional procedures.

Fast 3D T(2)-weighted MRI with Hadamard encoding in the slice select direction

A fast method to obtain 3-dimensional (3D) magnetic resonance imaging with long repetition times is presented. It can be used to obtain fast 3D MRI with for example T(2) or diffusion weighted imaging. The method uses a 3D multiple thin slab sequence with radio frequency encoding, preferably Hadamard encoding, in the slice select direction. The point-spread function of the Hadamard-encoded slices is close to ideal even at low encoding numbers. This allows the acquisition of 3D data volumes with tolerable image quality up to four times faster than is possible using Fourier phase encoding. The scope of the method includes both longitudinal and transverse encoding. Longitudinal encoding provides a better point spread function than transverse encoding, at the expense of having to discard one slice per slab. The method is demonstrated experimentally for 4th order longitudinal Hadamard encoding to obtain 3D T(2)-weighted images.

Elimination of residual lipid contamination in single volume proton MR spectra of human brain

Short echo time 1H NMR spectra of the human brain reveal signals from various metabolites. In addition, resonances from macromolecules are present that may provide further useful information in several brain diseases. The detection of all these signals is possible if excellent volume selection is obtained; even small lipid contamination from surrounding fat tissue leads to strong spectral contamination. It affects the macromolecule resonances in the 0.5 to 2.0 ppm region and some adjacent metabolite signals and jeopardizes their quantitative analysis. This paper demonstrates how spatial contamination from insufficiently dephased signals can be recognized, analysed, and removed in localized STEAM spectroscopy of the brain.

Multiple-voxel double-quantum lactate-edited spectroscopy using two-dimensional longitudinal Hadamard encoding

A conventional gradient-selected double-quantum lactate editing sequence was combined with fourth order two-dimensional longitudinal Hadamard encoding and slice-selective refocusing to acquire lactate-edited spectra in a 3 x 3 matrix of voxels. The performance of the sequence was verified in phantoms at 9.4 T and in focally ischemic rat brain at 7.0 T. Efficient suppression of water, lipid, and the singlet resonances of creatine, choline, and N-acetylaspartate was achieved, giving multi-voxel localized lactate-edited spectra with good signal-to-noise ratio.

Hadamard encoding with surface coils for high SNR MR spectroscopy

The advantages of Hadamard over phase encoding in magnetic resonance spectroscopy (MRS) applied with surface coils in the direction perpendicular to the coil are demonstrated experimentally. With the recently introduced, time-shifted adiabatic pulses, the application of Hadamard encoding with surface coils results with almost ideal point spread function for pixels up to a distance of a radius from the coil. Comparison to phase encoding with equal region of interest size shows the significant advantage of Hadamard encoding in slice sharpness, overlapping, and spatial contamination. In addition, since there is no aliasing in Hadamard space, the total experimental time for the same region of interest is much shorter. We conclude that the hybrid of Hadamard encoding in the direction perpendicular to the coil and phase encoding in other directions is the method of choice to obtain reliable high signal to noise ratio MRS in vivo.

Total brain N-acetylaspartate concentration in normal, age-grouped females: quantitation with non-echo proton NMR spectroscopy

The intra-individual and inter-individual variations of the global N-acetylaspartate (NAA) concentration were measured in a cohort of five 42+/-5 year-old normal females. The total NAA signal from the whole head was obtained with non-localized non-echo proton spectroscopy (1H-MRS) and converted into absolute mole amounts using phantom replacement. Since NAA is assumed to be present only in neurons, its concentration was obtained by dividing these mole amounts with the brains' volume, calculated from high resolution MRI. The key feature of the procedure is its near-complete suppression of the intense subcutaneous and bone marrow lipids' signals, whose chemical shifts neighbor and underlay the NAA. This was achieved by exploiting the lipids' much shorter T1s, compared to that of NAA, for destructive interference of their signals in co-addition following alternating, nonselective 180 degrees inversions. The average global, inter-individual NAA concentration in that group was found to be 10.63 mM with a 95% confidence interval of 10.43-10.82 mM.

3D multivoxel proton spectroscopy of human brain using a hybrid of 8th-order Hadamard encoding with 2D chemical shift imaging

Multivoxel 3D localized proton spectroscopy using a hybrid of 1D 8th-order transverse Hadamard spectroscopic imaging (HSI) with 2D chemical shift imaging (CSI) is demonstrated in human brain. The spatially selective HSI pulse incorporates naturally into the PRESS sequence (TE = 135 ms), which then both excites an 8 x 8 x 6 cm parallelepiped volume of interest (VOI) and subdivides it into eight slices. The planes of these slices are further partitioned into 16 x 16 voxel arrays using 2D CSI to yield 8 x 8 x 8 voxels within the VOI. Simultaneous 3D coverage yields good voxel signal-to-noise (8, 12, and 22 for choline, creatine, and N-acetylaspartate, respectively) from these 0.75-ml voxels, in approximately 45 min. The high spatial isolation allows localization to within less than 1 cm from the skull without fat contamination.