Reproducibility of human 3D fMRI brain maps acquired during a motor task

This study is an investigation into the reproducibility of brain activation in the sensorimotor cortex obtained with 3D "PRESTO" fMRI on eleven normal subjects. During one session, two series of functional scans were acquired while the subjects performed a finger opposition task (2 Hz). Nine subjects were tested once more on a different day. Each individual motor trial was analyzed separately, with a conservative zt-based method. Using these results, the agreement between repeated series was examined in a number of ways, comparing the two series within one session, and the two series across sessions. In 28 of the 31 series (90%) significant signal change was found in the contralateral primary sensorimotor cortex (PSM). Overall, 0.20% of all voxels (total about 11,000) in the scanned volume reached significance, and approximately 60% of the significant positive signal changes were located in the PSM (P<5x10(-7) for a chance occurrence). Comparisons within and across sessions yielded similar results: there was a 20-30% overlap of the clusters of activated voxels in the PSM (chance overlap within the PSM: P<0.01). The mean distance between zt-weighted centers of mass was 4.0-4.4 mm (chance distance within the PSM: P=0.033 and 0.058, respectively). No significant difference was found between series in the magnitude of significant signal change. Whereas the number of activated voxels in the PSM was not consistently correlated between series, the ratio of this number over the total number of activated voxels in the scanned volume was significantly correlated (rho=0.75-0.79, P<0.05). These results indicate that activation in sensorimotor cortex associated with oppositional finger movement is reliably mapped with 3D PRESTO fMRI.

Carbon-13 nuclear magnetic resonance study on the dynamics of the conformation of reduced flavin

Several flavin model compounds in the reduced state have been investigated by 13C NMR techniques. The NMR spectra were recorded in dependence of temperature, in the range of 30 to -100 degrees C. The results show that the activation barrier for the ring inversion ("butterfly" motion) is too low to be observed directly. In order to be able to detect the barrier of the ring inversion, it was coupled with a side-chain rotation. In this way, the intrinsic barrier for the ring inversion is increased by the barrier of the side-chain rotation, which allowed detection of the former barrier. It is shown that the intrinsic barrier for the ring inversion is less than 20 kJ/mol. Moreover, it is shown that previous results of Tauscher et al. [Tauscher, L., Ghisla, S., & Hemmerich, P. (1973) Helv. Chim. Acta 56, 630-649] are incorrect and nitrogen inversion is not observed. Symmetry arguments in the dynamic processes are discussed. From the low activation barrier for the ring inversion, it can be concluded that the conformation of the reduced flavin can be easily influenced upon binding to apoflavoproteins. This aspect might be of importance in the regulation of the function of the flavin prosthetic group in biological systems.

Thermal therapies in interventional MR imaging. Focused ultrasound

MR image-guided focused ultrasound (FUS) provides an entirely noninvasive approach for local thermal therapies. MR imaging allows target definition and continuous temperature mapping. Therefore, the heating procedure can be controlled spatially and temporally based on automatic feedback to the FUS apparatus. Phased-array ultrasound technology will further help the development. MR imaging/FUS may be applied not only for tissue ablation, but also for local drug delivery, gene therapy, and drug activation.

Diffusion tensor MRI of the human kidney

This study characterizes the diffusion anisotropy of the human kidney using a diffusion-weighted, single-shot echo planar imaging (EPI) sequence in order to access the full apparent diffusion tensor (ADT) within one breathhold. The fractional anisotropy (FA) of the cortex and the medulla were found to be 0.22 +/- 0.12 and 0.39 +/- 0.11, respectively (N = 10), which emphasizes the need for rotationally invariant diffusion measurements for clinical applications. Additional limitations for clinical diffusion imaging on the kidney are the strong susceptibility variations within the abdomen that restrict the use of imaging techniques employing long echo trains, and the severe motion sensitivity that limits the available imaging time to one breath-hold. To overcome these problems an isotropic, diffusion-weighted, segmented EPI protocol that facilitates the acquisition of high-resolution diffusion-weighted images within a single breath-hold was implemented. Using this method, the apparent diffusion coefficient (ADC) of the cortex and medulla were found to be 2.89 +/- 0.28. 10(-9) m2/s and 2.18 +/- 0.36. 10(-9) m2/s (N = 10).

On-line correction and visualization of motion during MRI-controlled hyperthermia

Displacement of tissue during MRI-controlled hyperthermia therapy can cause significant problems. Errors in calculated temperature may result from motion-related image artifacts and inter-image object displacement, leading to incorrect spatial temperature reference. Here, cyclic navigator echoes were incorporated in rapid gradient-echo MRI sequences, used for temperature mapping based on the proton resonance frequency. On-line evaluation of navigator information was used in three ways. First, motion artifacts were minimized in echo-shifted (TE > TR) gradient-echo images using the phase information of the navigator echo. Second, navigator profiles were matched for a quantitative evaluation of displacement. Together with a novel processing method, this information was employed to correct the reference temperature maps, thereby avoiding persistence of motion-related temperature errors throughout the hyperthermic period. Third, on-line visualization of displacement, together with temperature maps and thermal dose images, was developed, allowing physician intervention at all times. Examples are given of on-line corrections during hyperthermia procedures with focused ultrasound and radiofrequency heat sources. Magn Reson Med 45:128-137, 2001.

On-line correction and visualization of motion during MRI-controlled hyperthermia

Displacement of tissue during MRI-controlled hyperthermia therapy can cause significant problems. Errors in calculated temperature may result from motion-related image artifacts and inter-image object displacement, leading to incorrect spatial temperature reference. Here, cyclic navigator echoes were incorporated in rapid gradient-echo MRI sequences, used for temperature mapping based on the proton resonance frequency. On-line evaluation of navigator information was used in three ways. First, motion artifacts were minimized in echo-shifted (TE > TR) gradient-echo images using the phase information of the navigator echo. Second, navigator profiles were matched for a quantitative evaluation of displacement. Together with a novel processing method, this information was employed to correct the reference temperature maps, thereby avoiding persistence of motion-related temperature errors throughout the hyperthermic period. Third, on-line visualization of displacement, together with temperature maps and thermal dose images, was developed, allowing physician intervention at all times. Examples are given of on-line corrections during hyperthermia procedures with focused ultrasound and radiofrequency heat sources. Magn Reson Med 45:128-137, 2001.

Diffusion tensor MRI of the spinal cord

Apparent diffusion tensor (ADT) measurements on the spinal cord using a pulsed-field-gradient (PFG) multi-shot echo-planar imaging (EPI) sequence are presented. In a study of 10 healthy volunteers, the obtained rotationally invariant anisotropy information is compared to the results obtained by the rotationally dependent methods. The water diffusivity in the direction parallel to the fibers was found to be almost 2.5 times higher than the average diffusivity in directions perpendicular to the fibers and showed cylindrically symmetric anisotropy characteristics. The influence of partial volume effects and the point spread function on the measured results was evaluated, and it was the concluded that a resolution of 1 mm in the read and phase directions is required to obtain unbiased values. Possible clinical implications were demonstrated by investigating the diffusion characteristics of 10 patients suffering from narrowing of the cervical canal. The changes in the diffusion characteristics were found to be large enough to allow a robust detection of diffusion changes in the spine, even in cases in which conventional T(2) and T(1) weighted images were unable to detect any lesion.

Local hyperthermia with MR-guided focused ultrasound: spiral trajectory of the focal point optimized for temperature uniformity in the target region

The objective of hyperthermia treatment is to deliver a similar therapeutic thermal dose throughout the target volume within a minimum amount of time. We describe a noninvasive approach to this goal based on magnetic resonance imaging (MRI)-guided focused ultrasound (FUS) with a spherical transducer that can be moved along two directions inside the bed of a clinical MR imager and that has an adjustable focal length in the third dimension. Absorption of FUS gives rise to a highly localized thermal buildup, which then spreads by heat diffusion and blood perfusion. A uniform temperature within a large target volume can be obtained using a double spiral trajectory of the transducer focal point together with constant and maximum FUS power. Differences between the real and target temperatures during the first spiral are evaluated in real time with temperature MRI and corrected for during the second spiral trajectory employing FUS focal point velocity modulation. Once a uniform temperature distribution is reached within the entire volume, FUS heating is applied only at the region's boundaries to maintain the raised temperature levels. Heat conduction, together with the design and timing of the trajectories, therefore ensures a similar thermal dose for the entire target region. Good agreement is obtained between theory and experimental results in vitro on gel phantoms, ex vivo on meat samples, and in vivo on rabbit thigh muscle. Edema in muscle was visible 1 hour after hyperthermia as a spatially uniform rise of the signal intensity in T(2)-weighted images.

Magnetic resonance temperature imaging for guidance of thermotherapy

Continuous thermometry during a hyperthermic procedure may help to correct for local differences in heat conduction and energy absorption, and thus allow optimization of the thermal therapy. Noninvasive, three-dimensional mapping of temperature changes is feasible with magnetic resonance (MR) and may be based on the relaxation time T(1), the diffusion coefficient (D), or proton resonance frequency (PRF) of tissue water. The use of temperature-sensitive contrast agents and proton spectroscopic imaging can provide absolute temperature measurements. The principles and performance of these methods are reviewed in this paper. The excellent linearity and near-independence with respect to tissue type, together with good temperature sensitivity, make PRF-based temperature MRI the preferred choice for many applications at mid to high field strength (>/= 1 T). The PRF methods employ radiofrequency spoiled gradient-echo imaging methods. A standard deviation of less than 1 degrees C, for a temporal resolution below 1 second and a spatial resolution of about 2 mm, is feasible for a single slice for immobile tissues. Corrections should be made for temperature-induced susceptibility effects in the PRF method. If spin-echo methods are preferred, for example when field homogeneity is poor due to small ferromagnetic parts in the needle, the D- and T(1)-based methods may give better results. The sensitivity of the D method is higher that that of the T(1) methods provided that motion artifacts are avoided and the trace of D is evaluated. Fat suppression is necessary for most tissues when T(1), D, or PRF methods are employed. The latter three methods require excellent registration to correct for displacements between scans.

Rapid three-dimensional MR imaging method for tracking a bolus of contrast agent through the brain

A gradient-echo three-dimensional magnetic resonance imaging technique (principles of echo shifting with a train of observations, or PRESTO) is presented for use in tracking a bolus of paramagnetic contrast agent through the brain. The approach combines a segmented echo-planar type of acquisition with echo shifting, which leads to echo times that are longer than the repetition time. Unlike echo-planar imaging, the method maintains image resolution despite drastic T2* changes and frequency shifts.

Hyperthermia by MR-guided focused ultrasound: accurate temperature control based on fast MRI and a physical model of local energy deposition and heat conduction

Temperature regulation in MR-guided focused ultrasound requires rapid MR temperature mapping and automatic feedback control of the ultrasound output. Here, a regulation method is proposed based on a physical model of local energy deposition and heat conduction. The real-time evaluation of local temperature gradients from temperature maps is an essential element of the control system. Each time a new image is available, ultrasound power is adjusted on-the-fly in order to obtain the desired evolution of the focal point temperature. In vitro and in vivo performance indicated fast and accurate control of temperature and a large tolerance of errors in initial estimates of ultrasound absorption and heat conduction. When using correct estimates for the physical parameters of the model, focal point temperature was controlled within the measurement noise limit. Initial errors in absorption and diffusion parameters are compensated for exponentially with a user-defined response time, which is suggested to be on the order of 10 sec.

Fast lipid-suppressed MR temperature mapping with echo-shifted gradient-echo imaging and spectral-spatial excitation

The water proton resonance frequency (PRF) is temperature dependent and can thus be used for magnetic resonance (MR) thermometry. Since lipid proton resonance frequencies do not depend on temperature, fat suppression is essential for PRF-based temperature mapping. The efficacy of echo-shifted (TE > TR) gradient-echo imaging with spectral-spatial excitation is demonstrated, resulting in accurate and rapid, lipid-suppressed, MR thermometry. The method was validated on phantoms, fatty duck liver, and rat thigh, demonstrating improvements in both the speed and precision of temperature mapping. Heating of a rat thigh with focused ultrasound was monitored in vivo with an accuracy of 0.37 degree C and a time resolution of 438 msec.

An echo-shifted gradient-echo MRI method for efficient diffusion weighting

A segmented magnetic resonance imaging (MRI) method is introduced with time-efficient diffusion weighting resulting in total imaging times similar to those of single-shot methods. The approach is based on the principles of echo shifting with a train of observations (PRESTO) MRI sequence. The time efficiency of the sequence is based on the use of diffusion gradient pulses that also serve to shift the echo train to the next TR period, resulting in TE > TR. Each diffusion gradient is therefore used twice, for dephasing one set of spins as well as rephasing a second set of spins. Diffusion weighting and acquisition are thus achieved simultaneously. The sequence is validated in vitro and in vivo on rat kidney.

Real-time control of focused ultrasound heating based on rapid MR thermometry

RATIONALE AND OBJECTIVES

Real-time control of the heating procedure is essential for hyperthermia applications of focused ultrasound (FUS). The objective of this study is to demonstrate the feasibility of MRI-controlled FUS.

METHODS

An automatic control system was developed using a dedicated interface between the MR system control computer and the FUS wave generator. Two algorithms were used to regulate FUS power to maintain the focal point temperature at a desired level.

RESULTS

Automatic control of FUS power level was demonstrated ex vivo at three target temperature levels (increase of 5 degrees C, 10 degrees C, and 30 degrees C above room temperature) during 30-minute hyperthermic periods. Preliminary in vivo results on rat leg muscle confirm that necrosis estimate, calculated on-line during FUS sonication, allows prediction of tissue damage. CONCLUSIONS. The feasibility of fully automatic FUS control based on MRI thermometry has been demonstrated.

Phase navigator correction in 3D fMRI improves detection of brain activation: quantitative assessment with a graded motor activation procedure

Motion poses severe problems for BOLD fMRI, particularly in clinical studies, as patients exhibit more involuntary movements than controls. This study focuses on the merits of a motion correction technique incorporated in multishot fMRI scans, so-called phase navigator correction. The technique entails real-time assessment and off-line elimination of signal fluctuations caused by subject motion. The purpose of this study was to quantify and characterize the effect of this type of improvement on 3D fMRI brain activity maps. For imaging, the 3D PRESTO method was used, with a relatively simple finger opposition task. The followed strategy was guided by the notion that application of any fMRI imaging tool in clinical studies requires several qualities, such as high and spatially homogeneous sensitivity to brain activity, and low sensitivity to motion. A graded motor activation protocol in 10 healthy subjects revealed that image stability was improved by approximately 20%, by the use of phase navigator correction. As a result, sensitivity for task-related BOLD signal change was enhanced considerably in the brain activity maps. Implications for use of this fMRI technique in patient studies are discussed.

Functional magnetic resonance imaging brain mapping in psychiatry: methodological issues illustrated in a study of working memory in schizophrenia

Functional magnetic resonance imaging (fMRI) is a potential paradigm shift in psychiatric neuroimaging. The technique provides individual, rather than group-averaged, functional neuroimaging data, but subtle methodological confounds represent unique challenges for psychiatric research. As an exemplar of the unique potential and problems of fMRI, we present a study of 10 inpatients with schizophrenia and 10 controls performing a novel "n back" working memory (WM) task. We emphasize two key design steps: (1) the use of an internal activation standard (i.e., a physiological control region) to address activation validity, and (2) the assessment of signal stability to control for "activation" artifacts arising from unequal signal variance across groups. In the initial analysis, all but one of the patients failed to activate dorsolateral prefrontal cortex (DLPFC) during the working memory task. However, some patients (and one control) also tended to show sparse control region activation in spite of normal motor performance, a result that raises doubts about the validity of the initial analysis and concerns about unequal subject motion. Subjects were then matched for signal variance (voxel stability), producing a subset of six patients and six controls. In this comparison, the internal activation standard (i.e., motor activation) was similar in both groups, and five of six patients, including two whom were neuroleptic-naive, failed to activate DLPFC. In addition, a tendency for overactivation of parietal cortex was seen. These results illustrate some of the promise and pitfalls of fMRI. Although fMRI generates individual brain maps, a specialized survey of the data is necessary to avoid spurious or unreliable findings, related to artifacts such as motion, which are likely to be frequent in psychiatric patients.

Magnetization transfer imaging of rat brain under non-steady-state conditions. Contrast prediction using a binary spin-bath model and a super-lorentzian lineshape

Magnetization transfer contrast imaging using turbo spin echo and continuous wave off-resonance irradiation was carried out on rat brain in vivo at 4.7 T. By systematically varying the off-resonance irradiation power and the offset-frequency, the signal intensities obtained under steady-state for both transverse and longitudinal magnetization were successfully analyzed with a simple binary spin-bath model taking into account a free water compartment and a pool of protons with restricted motions bearing a super-Lorentzian lineshape. Due to important RF power deposition, such experimental conditions are not practical for routine imaging on humans. An extension of the model was derived to describe the system for shorter off-resonance pulse duration, i.e., when the longitudinal magnetization of the free protons has not reached a steady-state. Data sets obtained for three regions of interest, namely the corpus callosum, the basal ganglia, and the temporal lobe, were correctly interpreted for off-resonance pulse durations varying from 0.3 to 3 s. The parameter sets obtained from the calculations made it possible to predict the contrast between the different regions as a function of the pulse power, the offset frequency, and pulse duration. Such an approach could be extended to contrast prediction for human brain at 1.5 T.

On the feasibility of MRI-guided focused ultrasound for local induction of gene expression

Gene therapy is a promising approach to the treatment of many forms of disease, including cancer. Of critical concern in its implementation is the ability to control the location, duration, and level of expression of the therapeutic gene. Here, we propose the use of local heat in combination with a heat-sensitive promoter to help accomplish this. Certain members of the family of heat shock protein (hsp) promoters display a regulation that depends strongly on temperature. We present a study of natural hsp70 induction in rat leg by MRI-guided focused ultrasound to investigate the hsp70 promoter as a possible candidate for use in control of gene expression with local heat. A temperature increase of 5-8 degrees C in the focal region for 45 minutes led to a differential expression of the hsp70 mRNA between the focal region and the surrounding tissue ranging from a factor of 3 to 67.

Neurochemistry of brain lesions determined by spectroscopic imaging in systemic lupus erythematosus

OBJECTIVE

The significance and etiology of focal brain lesions in systemic lupus erythematosus (SLE) are unknown. Our purpose was to determine whether the neurochemistry of focal lesions and normal appearing brain tissues in SLE were consistent with neuronal loss, demyelination, or ischemia.

METHODS

Patients with SLE (n = 14) and controls (n = 13) were studied using magnetic resonance imaging (MRI) and spectroscopic imaging (SI) at 1.5 Tesla.

RESULTS

MRI detected fixed focal brain lesions (n = 16) and SI measured brain metabolites, including N-acetylaspartate (NAA), creatine (Cre), choline (Cho), and lactate (Lac). NAA/Cre of normal appearing brain was decreased in patients with SLE compared to controls: grey matter (1.74 +/- 0.16 vs 1.92 +/- 0.18; p = 0.01), occipital white matter (1.98 +/- 0.22 vs 2.23 +/- 0.16; p = 0.004), and periventricular white matter (2.00 +/- 0.23 vs 2.33 +/- 0.23; p = 0.001). Lesions were characterized by markedly decreased NAA/Cre relative to normal appearing tissues in the same patient (1.67 +/- 0.22 vs 1.88 +/- 0.14; p = 0.0002). Elevated Cho/Cre was observed in 25% of focal lesions and 21% of normal appearing tissues. No elevation of lactate was observed in lesions or normal appearing tissues.

CONCLUSION

SI detects focal and generalized brain abnormalities in SLE characterized by decreased NAA, elevated choline, and normal lactate. These findings are consistent with widespread neuronal injury and demyelination, but are not consistent with anaerobic metabolism.

Determination of cerebral glucose transport and metabolic kinetics by dynamic MR spectroscopy

A new in vivo nuclear magnetic resonance (NMR) spectroscopy method is introduced that dynamically measures cerebral utilization of magnetically labeled [1-13C]glucose from the change in total brain glucose signals on infusion. Kinetic equations are derived using a four-compartment model incorporating glucose transport and phosphorylation. Brain extract data show that the glucose 6-phosphate concentration is negligible relative to glucose, simplifying the kinetics to three compartments and allowing direct determination of the glucose-utilization half-life time [t1/2 = ln2/(k2 + k3)] from the time dependence of the NMR signal. Results on isofluorane (n = 5)- and halothane (n = 7)-anesthetized cats give a hyperglycemic t1/2 = 5.10 +/- 0.11 min-1 (SE). Using Michaelis-Menten kinetics and an assumed half-saturation constant Kt = 5 +/- 1 mM, we determined a maximal transport rate Tmax = 0.83 +/- 0.19 mumol.g-1.min-1, a cerebral metabolic rate of glucose CMRGlc = 0.22 +/- 0.03 mumol.g-1.min-1, and a normoglycemic cerebral influx rate CIRGlc = 0.37 +/- 0.05 mumol.g-1.min-1. Possible extension of this approach to positron emission tomography and proton NMR is discussed.

Unraveling diffusion constants in biological tissue by combining Carr-Purcell-Meiboom-Gill imaging and pulsed field gradient NMR

A diffusion-weighted multi-spin-echo pulse sequence is presented, which allows for simultaneous measurement of T2, the fractional amplitude, and the diffusion constant of different fractions. Monte Carlo simulations demonstrate an improvement of this sequence with respect to the accuracy of diffusion constant and fractional amplitude for slow exchange. Examples are shown for a simple phantom containing two fractions. In addition, experiments on cat brain in healthy condition and following occlusion of the middle cerebral artery show that the fractional amplitude and the diffusion constant of cerebral spinal fluid and normal brain tissue can be analyzed within each pixel with acceptable accuracy.

Regionally specific pattern of neurochemical pathology in schizophrenia as assessed by multislice proton magnetic resonance spectroscopic imaging

OBJECTIVE

Several single-voxel proton magnetic resonance spectroscopy (1H-MRS) studies of patients with schizophrenia have found evidence of reductions of N-acetyl-aspartate (NAA) concentrations in the temporal lobes. Multislice proton magnetic resonance spectroscopy imaging (1H-MRSI) permits simultaneous acquisition and mapping of NAA, choline-containing compounds (CHO), and creatine/phosphocreatine (CRE) signal intensities from multiple whole brain slices consisting of 1.4-ml single-volume elements. We have used 1H-MRSI to assess the regional specificity of previously reported changes of metabolite signal intensities in schizophrenia. Hippocampal volume was also measured to test the relationship between 1H-MRSI findings and tissue volume in this region.

METHOD

Ratios of areas under the metabolite peaks of the proton spectra were determined (i.e., NAA/CRE, NAA/CHO, CHO/CRE) for multiple cortical and subcortical regions in 10 inpatients with schizophrenia.

RESULTS

Patients showed significant reductions of NAA/CRE and NAA/CHO bilaterally in the hippocampal region and in the dorsolateral prefrontal cortex. There were no significant changes in CHO/CRE or in NAA ratios in any other area sampled. No significant correlation was found between metabolite ratios in the hippocampal region and its volume.

CONCLUSIONS

NAA-relative signal intensity reductions in schizophrenia appear to be remarkably localized, involving primarily the hippocampal region and the dorsolateral prefrontal cortex, two regions implicated prominently in the pathophysiology of this disorder.

Fast 3D functional magnetic resonance imaging at 1.5 T with spiral acquisition

A new method to perform rapid 3D fMRI in human brain is introduced and evaluated in normal subjects, on a standard clinical scanner at 1.5 Tesla. The method combines a highly stable gradient echo technique with a spiral scan method, to detect brain activation related changes in blood oxygenation with high sensitivity. A motor activation paradigm with a duration of less than 5 min, performed on 10 subjects, consistently showed significant changes in signal intensity in the area of the motor cortex. In all subjects, these changes survived high statistical thresholds.

Functional mapping of human sensorimotor cortex with 3D BOLD fMRI correlates highly with H2(15)O PET rCBF

Positron emission tomography (PET) functional imaging is based on changes in regional cerebral blood flow (rCBF). Functional magnetic resonance imaging (fMRI) is based on a variety of physiological parameters as well as rCBF. This study is aimed at the cross validation of three-dimensional (3D) fMRI, which is sensitive to changes in blood oxygenation, with oxygen-15-labeled water (H2(15)O) PET. Nine normal subjects repeatedly performed a simple finger opposition task during fMRI scans and during PET scans. Within-subject statistical analysis revealed significant ("activated") signal changes (p < 0.05, Bonferroni corrected for number of voxels) in contralateral primary sensorimotor cortex (PSM) in all subjects with fMRI and with PET. With both methods, 78% of all activated voxels were located in the PSM. Overlap of activated regions occurred in all subjects (mean 43%, SD 26%). The size of the activated regions in PSM with both methods was highly correlated (rho = 0.87, p < 0.01). The mean distance between centers of mass of the activated regions in the PSM for fMRI versus PET was 6.7 mm (SD 3.0 mm). Average magnitude of signal change in activated voxels in this region, expressed as z-values adapted to timeseries, zt, was similar (fMRI 5.5, PET 5.3). Results indicate that positive blood oxygen level-dependent (BOLD) signal changes obtained with 3D principles of echo shifting with a train of observations (PRESTO) fMRI are correlated with rCBF, and that sensitivity of fMRI can equal that of H2(15)O PET.

Single-shot diffusion MRI of human brain on a conventional clinical instrument

A single-shot diffusion MRI technique on a standard clinical 1.5T scanner is presented. The method incorporates the following elements: (a) an inversion RF pulse followed by a delay of 1.3 s to null cerebral spinal fluid (CSF) signal, (b) a stimulated echo sequence (TE = 56 ms, TM = 100 ms) to obtain strong diffusion weighting, (c) a single-shot gradient- and spin-echo (GRASE) sequence for imaging with a modified k-space trajectory and Carr-Purcell Meiboom-Gill (CPMG)-phase cycle. The trace of the diffusion coefficient obtained with this approach is in good agreement with values reported for animal brain, and for recent human studies. It is demonstrated that single-shot diffusion imaging of human brain is feasible on an unmodified standard instrument without high-gradient slew rate or extreme field homogeneity.

Reproducibility of proton MR spectroscopic imaging findings

PURPOSE

To study the intraindividual, interindividual, and intraregional reproducibility of multisection proton MR spectroscopic imaging in healthy adults.

METHODS

Six subjects were studied three times with proton MR spectroscopic imaging. Multisection long-echo-time proton MR spectroscopic imaging permits simultaneous acquisition of N-acetylaspartate (NAA), choline-containing compounds (Cho), and creatine plus phosphocreatine (Cr) signal intensities from four 15-mm-thick sections divided into 0.84-mL single-volume elements. Regions of interest were the frontal cortex, the occipital cortex, the parietal cortex, the insular cortex, the cingulate gyrus, the centrum semiovale, the thalamus, and the caudate. Statistical evaluation was performed by analyses of variance and components of variance method.

RESULTS

The ratio NAA/Cr showed the lowest overall coefficient of variation (CV, %) in most of the regions of interest (range, 8.9 to 26.1). Interregional differences in the overall CV were present. Interindividual CVs ranged from 4.2 to 8.7 for NAA/Cr, from 6.8 to 17.4 for NAA/Cho, and from 5.0 to 13.6 for Cho/Cr. Intraindividual CVs ranged from 8.2 to 22.2 for NAA/Cr, from 12.8 to 25.8 for NAA/Cho, and from 4.5 to 21.0 for Cho/Cr. Intraregional CVs ranged from 12.3 to 21.2 for NAA/Cr, from 13.0 to 20.4 for NAA/Cho, and from 12.2 to 18.9 for Cho/Cr.

CONCLUSIONS

Proton MR spectroscopic imaging showed good overall reproducibility. The finding of interregional variations of CV indicates that care is needed when using this imaging technique for follow-up studies.

Three-dimensional functional magnetic resonance imaging of human brain on a clinical 1.5-T scanner

Functional magnetic resonance imaging (fMRI) is a tool for mapping brain function that utilizes neuronal activity-induced changes in blood oxygenation. An efficient three-dimensional fMRI method is presented for imaging brain activity on conventional, widely available, 1.5-T scanners, without additional hardware. This approach uses large magnetic susceptibility weighting based on the echo-shifting principle combined with multiple gradient echoes per excitation. Motor stimulation, induced by self-paced finger tapping, reliably produced significant signal increase in the hand region of the contralateral primary motor cortex in every subject tested.

Brain regional distribution pattern of metabolite signal intensities in young adults by proton magnetic resonance spectroscopic imaging

Proton magnetic resonance spectroscopy (1H-MRS) is evolving from single-volume localized acquisitions to multiple-volume acquisitions using magnetic resonance spectroscopic imaging (1H-MRSI). The normal regional patterns of 1H-MRSI-detectable metabolite signal intensities have yet to be established. We studied 13 healthy young adults with a multiple-section 1H-MRSI technique. The metabolite signals measured were N-acetylaspartate (NA), choline-containing compounds (CHO), creatine-phosphocreatine (CRE), and lactate. Ten neuroanatomic regions (nine bilateral) were identified in gray matter, white matter, and basal nuclei. Analysis of the data led to the following conclusions: (1) NA and CHO signals from centrum semiovale (CSO) can be used as a normalizing factor to reduce intersubject variability due to external causes; (2) in normal human brain, there is no left versus right asymmetry in the regions studied; (3) statistically significant patterns of signal distribution of NA, CHO, and CRE can be identified in normal human brain; and (4) CSO-normalized metabolite signal intensities and metabolite ratios complement each other for the detection of significant regional differences.

Analytical solution for phase modulation in BURST imaging with optimum sensitivity

In order to improve the efficiency of BURST imaging, a new phase-modulation scheme of the DANTE-type RF excitation is proposed. It is shown that analytical optimization of the phase-modulation scheme with optimal SNR can be found for an arbitrary number N of subpulses in the RF excitation. Both theory and experiment indicate a maximum attainable gain in efficiency of the square root of N, compared to nonmodulated (constant-phase) excitation schemes.

Susceptibility insensitive single shot MRI combining BURST and multiple spin echoes

A single shot MR imaging technique insensitive to magnetic susceptibility effects is introduced. The method allows multislice imaging in areas with poor magnetic field homogeneity, and can be implemented on standard clinical scanners. The design is based on the combination of a BURST excitation with multiple RF refocusing pulses. Images were obtained at 1.5 T on phantoms and human brain with a matrix size of 64 x 54 and a resolution of 4 x 4 mm in 230 ms.

Brain mapping with functional MR imaging: comparison of gradient-echo--based exogenous and endogenous contrast techniques

PURPOSE

To compare directly the two most widely used methods of functional magnetic resonance (MR) imaging--dynamic contrast material-enhanced MR imaging and blood oxygenation level-dependent (BOLD) MR imaging.

MATERIALS AND METHODS

Five healthy volunteers underwent dynamic contrast-enhanced and BOLD MR imaging with a conventional 1.5-T MR unit during visual stimulation and a dark control state. BOLD studies were performed with a gradient-echo sequence, and dynamic MR imaging was performed with an echo-shifted gradient-echo sequence after intravenous administration of a bolus of gadopentetate dimeglumine.

RESULTS

A significantly greater percentage signal change was found with dynamic MR imaging than with the BOLD technique. The extent of area activated was also significantly greater.

CONCLUSION

With standard clinical imagers and these gradient-echo-based techniques, greater percentage activation and area of activation can be achieved with dynamic MR imaging than with BOLD MR imaging.

Incorporation of lactate measurement in multi-spin-echo proton spectroscopic imaging

An improved multi-slice, multi-spin-echo proton spectroscopic imaging method for human brain is presented. The technique allows the reconstruction of lactate images, along with choline plus creatine, N-acetylaspartate, and lipid images from one single data set processed in three separate ways. The discrimination between resonances of lipid protons and lactate methyl protons is based on homonuclear spin-spin coupling. The reliability of the separation of the lipid and lactate contribution depends on the T2 of the lipid resonances. Measurements were performed on a standard 1.5 Tesla clinical scanner on healthy volunteers and a patient with high grade CNS lymphoma, demonstrating the ability to obtain high quality metabolite maps within 11 min.

Multislice proton magnetic resonance spectroscopic imaging in X-linked adrenoleukodystrophy

Multislice proton magnetic resonance spectroscopic imaging permits metabolic analysis of brain tissue in vivo by data acquisition in four oblique axial slices, each 15-mm thick and divided into 0.8-ml single-volume elements. We applied this technique to the systematic study of 25 patients with adrenoleukodystrophy: 3 with the severe childhood or adult cerebral form of the disease, 5 with adrenomyeloneuropathy, 12 with no demonstrable neurological involvement, and 5 women heterozygous for adrenoleukodystrophy who had some degree of neurological disability. Abnormalities on magnetic resonance spectroscopic imaging included a reduction in N-acetyl aspartate, an increase in choline-containing compounds, and at times, an increase in lactate. Five patients showed abnormalities in the presence of normal-appearing magnetic resonance images, and in 8 other patients the alterations on spectroscopic images were more severe than those demonstrable by magnetic resonance imaging. Correlation with clinical course suggests that an increase in the choline-containing compounds is associated with an active demyelinative process, whereas such compounds are not elevated in lesions that are stable. We conclude that magnetic resonance spectroscopic imaging is a more sensitive indicator of early neurological involvement than is magnetic resonance imaging, and that the character of abnormalities detected by the former technique may serve as a gauge of the degree of activity of the demyelinating process and as a guide to the selection and evaluation of therapeutic approaches.

3D bolus tracking with frequency-shifted BURST MRI

OBJECTIVE

Our goal was to develop and test a 3D bolus-tracking MR technique for perfusion imaging of normal and pathological (infarcted) human brain.

MATERIALS AND METHODS

All experiments were performed on standard 1.5 T GE/Signa clinical scanners. Five normal volunteers and one patient with a subacute brain infarct were studied. Modified [frequency-shifted (FS)] BURST MRI was performed during injection of a bolus of Gd-DTPA (0.13 mmol/kg) in the antecubital vein. The 3D datasets were acquired with a time resolution of 2.2 s and an effective spatial resolution of 4.3 x 4.3 x 6.4 mm. Three-dimensional maps of blood volume and bolus arrival time were determined by fitting a synthetic curve to the intensity time course on a voxel-by-voxel basis.

RESULTS

Both relative cerebral blood volume and arrival time maps demonstrated sensitivity to regional differences in blood supply in both normal brain and in the subacute brain infarction. The transit time maps showed arrival time delays of 5-7 s within and around the infarct and confirmed the diagnosis of left middle cerebral artery occlusion.

CONCLUSION

The results of the measurements on both normal and diseased human brain demonstrated the ability to acquire valuable 3D information about brain perfusion using FS BURST MRI.

Fast volume scanning with frequency-shifted BURST MRI

We introduce a modified BURST imaging technique with reduced saturation effects and improved signal-to-noise ratio. The method applies a frequency shift to the RF excitation pulse on successive repetitions. It allows collection of 3D datasets of human brain within a few seconds, on a standard clinical scanner at 1.5 Tesla.

19F magnetic resonance imaging of cerebral blood flow with 0.4-cc resolution

19F magnetic resonance imaging techniques were used to determine "wash-in" and "wash-out" curves of the inert, diffusible gas CHF3 from 0.4-cc voxels in the cat brain, and mass spectrometer gas detection was used to determine the CHF3 concentration in expired air. These two sets of data were used to calculate cerebral blood flow values in the 0.4-cc voxels, and the blood flow images were registered with high-resolution 1H magnetic resonance images. Data were collected both during the wash-in and wash-out phases of the experiment, but the two sets of data were analyzed separately to obtain independent estimates of the blood flow during the two phases, i.e., Qin and Qout. Repeated determinations of cerebral blood flow images were performed in individual animals, and the entire protocol was repeated on five different animals. The average values of Qin and Qout for a typical 0.4-cc voxel in the parietal cortex were 83 ml 100 g-1 min-1 and 72 ml 100 g-1 min-1, respectively. Monte Carlo calculations utilizing the noise in the 19F NMR signal from this voxel predict an average standard deviation for Qin and Qout of +/- 10%. The average standard deviation for repeated measurements (in the same animal) of Qin and Qout in this voxel was +/- 14%. We conclude that 19F magnetic resonance imaging approaches have the potential to image cerebral blood flow in humans.

3-dimensional functional imaging of human brain using echo-shifted FLASH MRI

A 3-dimensional MRI method has been developed for functional mapping of the human brain, based on blood oxygenation level dependent (BOLD) contrast mechanisms. The method uses recently introduced principles of echo-shifted FLASH to acquire a single 3D data set in 20 s. The technique was tested on a conventional 1.5 Tesla clinical scanner with a standard head coil using visual stimulation with a 8 Hz flashing white light, or a varying checkerboard pattern. Areas of increased signal intensity were identified in the visual cortex, consistent with the known functional organization.

Inflow versus deoxyhemoglobin effects in BOLD functional MRI using gradient echoes at 1.5 T

Modified gradient-echo MR techniques were applied to study the effects of inflow on functional brain imaging studies using visual and motor cortex stimulation. The results demonstrate that the large signal changes, seen in previously reported gradient-echo studies at 1.5-2.0 T, are dominated by direct inflow effects, in particular when using a large flip angle and a thin slice. The findings suggest that inflow-based functional imaging, along with Blood Oxygenation Level Dependent (BOLD) functional MRI, may play an important role in future research towards the functional organization of the human brain.

Evaluation of restricted diffusion in cylinders. Phosphocreatine in rabbit leg muscle

Interpretation of NMR diffusion data is complicated in the presence of diffusion barriers. In this study, restricted diffusion in cylinders is evaluated. A method is presented to determine the diameter of the cylinder and the unrestricted diffusion coefficient from the dependence of the trace of the diffusion tensor with diffusion time. This method is valid even if the orientation of the cylinders is unknown and varies within the sample. An example is given for the diffusion of phosphocreatine in the cylindrically shaped fibers of rabbit skeletal muscle.

Water diffusion and acute stroke

The occlusion of the middle cerebral artery was used as an experimental acute stroke model in 30 cats. The diffusion of water was followed by diffusion-sensitized MRI between 1 and 15 h after induction of stroke. It is demonstrated that images representing the trace of the diffusion tensor provide a much more accurate delineation of affected area than images representing the diffusion in one direction only. The reason is that the strong contrast caused by the anisotropy and orientation of myelin fibers is completely removed in the trace of the diffusion tensor. The trace images show a small contrast between white and gray matter. The diffusion coefficient of white matter is decreased in acute stroke to approximately the same extent as gray matter. It is further shown that the average lifetime of water in extra and intracellular space is shorter than 20 ms both for healthy and ischemic tissue indicating that myelin fibers are permeable to water. The anisotropy contrast did not change before or after induction of stroke, nor after sacrifice. Together, these observations are consistent with the view that the changes in water diffusion during acute stroke are directly related to cytotoxic oedema, i.e., to the change in relative volume of intra- and extracellular spaces. Changes in membrane permeability do not appear to contribute significantly to the changes in diffusion.

Functional brain MR imaging based on bolus tracking with a fast T2*-sensitized gradient-echo method

Dynamic physiological scanning, based on temporary changes in local field homogeneity during the passage of a contrast agent bolus, has been performed hitherto with echo-planar imaging (EPI) or conventional gradient-recalled techniques (FLASH). Here, it is shown that the T2* sensitivity of conventional FLASH techniques can be improved drastically on a conventional whole body instrument by delaying the gradient-echo until the subsequent TR-period without increasing total imaging time. Examples are given for a full k-space matrix (128 x 256) obtained within 2 s with a TE of 25 ms, resulting in images free of artifacts. The method is applied to bolus tracking through the brain of healthy volunteers during visual stimulation and in the dark. An average increase of regional cerebral blood volume (rCBV) in the visual cortex of 10.9% (n = 9, p = .001) was found.

A functional MRI technique combining principles of echo-shifting with a train of observations (PRESTO)

We present a fast MRI technique sensitized to microscopic susceptibility effects. The method combines elements of echo-shifted gradient-recalled MR imaging (TE > TR) with the acquisition of multiple k-space lines within a single TR-period. The sequence results in a much reduced imaging time as compared with conventional gradient-echo MRI methods. The feasibility of the method is demonstrated for susceptibility bolus tracking in the cat brain using an imaging time of 153 ms. The relative cerebral blood volume maps created with this method are comparable with those obtained with conventional methods.

In vivo proton spectroscopy and spectroscopic imaging of [1-13C]-glucose and its metabolic products

Metabolism of [1-13C]-glucose was studied in situ in cat brain using gradient-enhanced proton-detected heteronuclear spectroscopy. Proton detection of [1-13C]-glucose, [3-13C]-lactate, 4-[13C]-glutamine, 4-[13C]-glutamate and the combined signals 2-[13C]-glutamate/glutamine and 3-[13C]-glutamate/glutamine was achieved, despite the fact that some of the associated proton resonances are close to the water signal. Two-dimensional [1H-13C]-spectra demonstrate the possibility of in situ spectral assignment with 1H sensitivity and 13C resolution. Spectroscopic images of glucose and its metabolic products were also acquired, showing the possibility to study spatial dependence of metabolism.

Fast proton spectroscopic imaging of human brain using multiple spin-echoes

We introduce a multi-echo multi-slice MR proton spectroscopic imaging method, which allows for a dramatic reduction of the measurement time by acquiring multiple spin-echoes within a single repetition time. In the multi-echo multi-slice experiment discussed in this paper, a threefold reduction in measurement time is obtained by sacrificing some spectral resolution. Signal-to-noise ratio and spatial resolution are preserved. Metabolite images of N-acetyl aspartate, and total choline+total creatine from multiple slices through the human brain are presented and compared with images obtained with a conventional single-echo multi-slice method.