Fast scan and echo planar MR imaging technology.

The development of imaging technology for acquiring ever-faster images has been an ongoing activity since the early years of MR imaging development. This article gives the reader a sense of the range of technical challenges faced by the developers of faster imaging hardware and some of their solutions. Before reviewing the technical issues, however, the authors briefly discuss the marketplace realities that have led to technology decisions and, to a certain extent, the prolonged pace of development of the hardware.

Superconducting open-configuration MR imaging system for image-guided therapy.

PURPOSE: To develop a superconducting magnetic resonance (MR) imager that provides direct access to the patient and permits interactive MR-guided interventional procedures. MATERIALS AND METHODS: A 0.5-T superconducting magnet that allows a region of vertical access to the patient was designed and constructed. This magnet was integrated with newly designed shielded gradient coils, flexible surface coils, and nonmagnetic displays and with position-monitoring probes and device-tracking instrumentation. RESULTS: The magnet homogeneity was 12.3 ppm, and the gradient field was linear to within 1% over an imaging region 30 cm in diameter. The signal-to-noise ratio was 10% higher than in a comparable 0.5-T superconducting imager. Images were obtained in several anatomic regions with use of routine pulse sequences. Interactive image plane selection and near real-time imaging, with use of fast gradient-recalled echo sequences, were demonstrated at a rate of one image every 1.5 seconds. CONCLUSION: MR-guided interventional procedures can be performed with full patient access with use of an open-configuration, superconducting MR magnet with near real-time imaging and interactive image plane control.

Electronics for a high temperature superconducting receiver system for magnetic resonance microimaging.

We describe an MRI (magnetic resonance imaging) receiver system that incorporates a high temperature superconducting (HTS) resonator as a surface coil. Techniques for measuring the Q of the HTS resonator in a 7 Tesla field are discussed. A method for coupling a room temperature copper resonant circuit to the HTS element is covered, and it is shown that such a coupling scheme preserves the signal-to-noise (SNR) gain afforded by the HTS coil. A preamplifier with a noise figure (NF) of < 0.15 dB at 300 MHz is described.

Real-time acquisition, display, and interactive graphic control of NMR cardiac profiles and images.

A highly interactive MRI scanner interface has been developed that allows, for the first time, real-time graphic control of one-dimensional (1D) and two-dimensional (2D) cardiac MRI exams. The system comprises a Mercury array processor (AP) in a Sun SPARCserver with two connections to the MRI scanner, a data link that passes the NMR data directly to the AP as they are collected, and a control link that passes commands from the Sun to the scanner to redirect the imaging pulse sequence in real time. In the 1D techniques, a cylinder or "pencil" of magnetization is repeatedly excited using gradient-echo or spin-echo line-scan sequences, with the magnetization read out each time along the length of the cylinder, and a scrolling display generated on the Sun monitor. Rubber-band lines drawn on the scout image redirect the pencil or imaging slice to different locations, with the changes immediately visible in the display. M-mode imaging, 1D flow imaging, and 2D fast cardiac imaging have been demonstrated on normal volunteers using this system. This platform represents an operator-"friendly" way of directing real-time imaging of the heart.

A high-temperature superconducting receiver for nuclear magnetic resonance microscopy.

A high-temperature superconducting-receiver system for use in nuclear magnetic resonance (NMR) microscopy is described. The scaling behavior of sources of sample and receiver-coil noise is analyzed, and it is demonstrated that Johnson, or thermal, noise in the receiver coil is the factor that limits resolution. The behavior of superconductors in the environment of an NMR experiment is examined, and a prototypical system for imaging biological specimens is discussed. Preliminary spin-echo images are shown, and the ultimate limits of the signal-to-noise ratio of the probe are investigated.

An NMR phased array for human cardiac 31P spectroscopy.

A four-coil phased-array 31P NMR receiver was designed and tested for human cardiac applications, to determine whether the combination of relatively high signal-to-noise ratio (SNR) and large field of view produced in 1H imaging is also realized for in vivo 31P spectroscopy. Spectra were acquired in parallel from an array of four overlapping 6.5-cm surface coils using one- and two-dimensional phase-encoding pulse sequences and were optimally combined to yield composite spectroscopic images. The phased array was found to generate useful 31P spectra from a 2.5-fold wider lateral region around the anterior myocardium than a single receiver of the same size as the array elements, with no increase in imaging time. In addition, the sensitive depth was increased by up to 2 cm over that of a single coil. Spectra could be acquired in roughly 15 min from a region extending to the middle of the heart, with voxel sizes of 2 x 2 x 4 cm3. For the average heart voxel, the SNR of the combined spectrum was higher than that of the best spectrum from any one coil in the array by 30%, with some voxels showing an increase as high as 60%.

Rapid NMR cardiography with a half-echo M-mode method.

A real-time NMR cardiac profiling pulse sequence has been developed that incorporates two-dimensional (2D) selective excitation and a half-echo readout. The time resolution has been improved by a factor of two relative to the previous flow-compensated, full-echo version. The technique produces a 2D plot of "beam"-axis position versus time, analogous to M-mode echocardiography. In human subjects, details of valve leaflet motion, intracardiac flow, wall motion, and wall thickening may be observed along optimal lines of sight selected interactively. The pulse sequence uses a low-tip-angle 2D selective-excitation pulse derived from a spiral k-space trajectory to excite a narrow cylinder of magnetization, followed by a half-echo readout gradient oriented along the axis of the cylinder. One-dimensional Fourier transformation of the acquired signal results in a magnetization profile along the length of the cylinder, or beam. The pulse sequence is effectively flow compensated without any additional gradient lobes, because the rapid oscillation in the gradient wave forms of the 2D excitation pulse produces relatively small net gradient moments, and the shortened readout gradient has minimal first-order moment relative to center echo. The signal from moving blood can alternatively be velocity encoded by the addition of bipolar gradients along any of the three axes, producing Doppler-like traces of intracardiac blood flow.

Volume imaging with MR phased arrays.

A volume MR phased array was constructed with two coils placed anteriorly and two coils posteriorly. Data acquired simultaneously from the four coils on a phantom were combined into a single image having a signal-to-noise ratio 80% better than that from the body coil. Additional comparisons of the four-coil phased array with a two-coil phased array and a Helmholtz pair having the same overall dimensions show how variations of signal amplitude and phase in the individual coils affect the composite SNR. Images of the male and female pelvis demonstrate how the improved SNR can be used to reduce the number of excitations, decrease the field of view, increase the echo time, or reduce the slice thickness.

Noise correlations in data simultaneously acquired from multiple surface coil arrays.

Multiple images acquired simultaneously from an array of surface coils can be combined to give a composite image with an improved signal-to-noise ratio (SNR) and a large field of view. The composite images' SNR can be optimized by taking advantage of noise correlations between coils and phase shifts induced by surface coil reception. Methods are derived for making optimal composite images with uniform noise or with uniform sensitivity. A simplified model is used to provide an intuitive understanding of the interaction of noise correlation and phase shift phenomena.

The NMR phased array.

We describe methods for simultaneously acquiring and subsequently combining data from a multitude of closely positioned NMR receiving coils. The approach is conceptually similar to phased array radar and ultrasound and hence we call our techniques the "NMR phased array." The NMR phased array offers the signal-to-noise ratio (SNR) and resolution of a small surface coil over fields-of-view (FOV) normally associated with body imaging with no increase in imaging time. The NMR phased array can be applied to both imaging and spectroscopy for all pulse sequences. The problematic interactions among nearby surface coils is eliminated (a) by overlapping adjacent coils to give zero mutual inductance, hence zero interaction, and (b) by attaching low input impedance preamplifiers to all coils, thus eliminating interference among next nearest and more distant neighbors. We derive an algorithm for combining the data from the phased array elements to yield an image with optimum SNR. Other techniques which are easier to implement at the cost of lower SNR are explored. Phased array imaging is demonstrated with high resolution (512 x 512, 48-cm FOV, and 32-cm FOV) spin-echo images of the thoracic and lumbar spine. Data were acquired from four-element linear spine arrays, the first made of 12-cm square coils and the second made of 8-cm square coils. When compared with images from a single 15 x 30-cm rectangular coil and identical imaging parameters, the phased array yields a 2X and 3X higher SNR at the depth of the spine (approximately 7 cm).

Phosphate metabolite imaging and concentration measurements in human heart by nuclear magnetic resonance.

Cardiac-gated phosphorus (31P) nuclear magnetic resonance (NMR) spectroscopic imaging with surface coils resolves in three dimensions the spatial distribution of high energy phosphate metabolites in the human heart noninvasively. 31P spectra derive from 6- to 14-cm3 volumes of myocardium in the anterior left ventricle, septum, and apex, at depths of up to about 8 cm from the chest, as identified by proton (1H) NMR anatomical images acquired without moving the subject. Spectroscopic images are acquired in 9 to 21 min at 1.5 T. Metabolite concentrations are quantified with reference to a standard located outside the chest, yielding normal in vivo concentrations of phosphocreatine and adenosine triphosphate of about 11.0 +/- 2.7 (SD) and 6.9 +/- 1.6 mumol/g of wet heart tissue, respectively. High energy phosphate contents did not vary significantly with location in the normal myocardium, but 2,3-diphosphoglycerate signals from blood varied with subject and location.

Problems and expediencies in human 31P spectroscopy. The definition of localized volumes, dealing with saturation and the technique-dependence of quantification.

Several technological problems in in vivo localized spectroscopy of metabolism are discussed in the context of comparing data obtained by different means. Deficiencies in spectroscopy localization methods can produce spectra that are dominated by artefactual signals derived from outside of selected volumes. Such artefacts are not usually correctly accounted for by representations of the profiles of the transverse magnetization alone. Selected sensitive volumes should be defined in terms of the size of tissue contributing the major fraction of signal to an observed spectrum, which is the integrated response from the sample including any phase cancellation effects. Phase cancellation in one-dimensional localization techniques employing excitation by an RF field with uniform phase distribution and surface coil detection such as depth resolved surface coil spectroscopy, chemical shift imaging (CSI) and rotating frame zeugmatography (RFZ) can significantly alter the effective radius of the sensitive volumes depending on the sample distribution and the extent of the homogeneous region of the magnet. Also, discrete spatial sampling in RFZ and CSI can radiate signal artefacts of around 25% into adjacent elements depending on the location and distribution of signal sources. Acquisition delays between excitation and detection and partial saturation are other major sources of systematic error. Saturation factors for metabolites are not easily obtainable on localized volumes during clinical exams on an individual basis, but may be expediently obtained as larger-volume tissue-averages. Better documentation of saturation effects, acquisition delays and localized volume sizes is needed to compare and validate clinical results and performance.

Proton-decoupled, Overhauser-enhanced, spatially localized carbon-13 spectroscopy in humans.

Spatially localized, natural abundance, carbon (13C) NMR spectroscopy has been combined with proton (1H) decoupling and nuclear Overhauser enhancement to improve 13C sensitivity up to five-fold in the human leg, liver, and heart. Broadhand-decoupled 13C spectra were acquired in 1 s to 17 min with a conventional 1.5-T imaging/spectroscopy system, an auxiliary 1H decoupler, an air-cooled dual-coil coplanar surface probe, and both depth-resolved surface coil spectroscopy (DRESS) and one-dimensional phase-encoding gradient NMR pulse sequences. The surface coil probe comprised circular and figure-eight-shaped coils to eliminate problems with mutual coupling of coils at high decoupling power levels applied during 13C reception. Peak decoupler RF power deposition in tissue was computed numerically from electromagnetic theory assuming a semi-infinite plane of uniform biological conductor. Peak values at the surface were calculated at 4 to 6 W/kg in any gram of tissue for each watt of decoupler power input excluding all coil and cable losses, warning of potential local RF heating problems in these and related experiments. The average power deposition was about 9 mW/kg per watt input, which should present no systemic hazard. At 3 W input, human 13C spectra were decoupled to a depth of about 5 cm while some Overhauser enhancement was sustained up to about 3 cm depth, without ill effect. The observation of glycogen in localized natural abundance 13C spectra of heart and muscle suggests that metabolites in the citric acid cycle should be observable noninvasively using 13C-labeled substrates.

Rapid 31P spectroscopy on a 4-T whole-body system.

Initial 31P spectroscopy results from a 4-T whole-body system are reported. Localized spectra from the human head, liver, and calf were obtained using DRESS, slice-interleaved DRESS, and volume 3DFT spectroscopic imaging techniques. Substantial reductions in data acquisition times to 10 s-4 min were achieved relative to previous similar experiments at 1.5 T. Some gain in spectral resolution (as measured in ppm) was also realized in the head.

Human in vivo phosphate metabolite imaging with 31P NMR.

Phosphorus (31P) spectroscopic images showing the distribution of high-energy phosphate metabolites in the human brain have been obtained at 1.5 T in scan times of 8.5 to 34 min at 27 and 64 cm3 spatial resolution using pulsed phase-encoding gradient magnetic fields and three-dimensional Fourier transform (3DFT) techniques. Data were acquired as free induction decays with a quadrature volume NMR detection coil of a truncated geometry designed to optimize the signal-to-noise ratio on the coil axis on the assumption that the sample noise represents the dominant noise source, and self-shielded magnetic field gradient coils to minimize eddy-current effects. The images permit comparison of metabolic data acquired simultaneously from different locations in the brain, as well as metabolite quantification by inclusion of a vial containing a standard of known 31P concentration in the image array. Values for the NMR visible adenosine triphosphate in three individuals were about 3 mM of tissue. The ratio of NMR detectable phosphocreatine to ATP in brain was 1.15 +/- 0.17 SD in these experiments. Potential sources of random and systematic error in these and other 31P measurements are identified.