Block Regional Interpolation Scheme for k-Space (BRISK): A Rapid Cardiac Imaging Technique

We introduce an acquisition method, “block regional interpolation scheme for k-space” (BRISK), to reduce the acquisition time for cardiac imaging. The method exploits the high degree of correlation that exists between time-resolved cardiac images. For representative k-space data sets, Fourier analysis was applied along the cardiac phase dimension to reveal that different regions of k-space can be effectively sampled at different rates. A reduced sampling strategy was implemented, and unsampled points were generated by Fourier interpolation. Time savings of up to 75% are quite feasible and 25% BRISK scans compare well with 100% scans. Simulations and acquisitions using a normal volunteer and patients are presented.     

Common k-space acquisition: A method to improve myocardial grid-tag contrast

A current limitation of many myocardial tag acquisitions employing SPAMM encoding is the relatively rapid loss of tag contrast over the cardiac cycle. Acquisition schemes that apply line tags produce prolonged tag contrast compared with directly excited grid tags. However, a grid-tag series can be generated by combining two orthogonal line-tag series. To be time efficient, each line-tag series can be acquired with the read gradient oriented perpendicular to the line-tag direction. There are several disadvantages associated with this approach, including the requirement to avoid signal fold-over and that fat shift artifacts appear in different directions in each line-tag series. Presented here is a method of applying separate line tags that does not require interchanging the read and phase encoding gradients and does not extend the scan time compared with a conventional grid-tag acquisition. Additionally, the means of generating grid tags results in a doubling of the tag contrast to noise ratio compared a line-tag set. Computer simulations are presented along with phantom and volunteer scans.     

Fast three dimensional sodium imaging

An efficient scheme for fast three dimensional acquisition of sodium MR images is described. This scheme relies on the use of three dimensional k-space trajectories with constant sample density to achieve significant (60–70%) reductions in total data acquisition time over conventional projection imaging schemes. The performance of this data acquisition scheme is demonstrated with acquisition of sodium data sets on phantoms and normal human volunteers at 1.5 and 3.0 Tesla. The experimental results demonstrate that high quality three dimensional sodium images (0.2 cc voxel size, 10:1 signal-to-noise ratio) can be acquired at clinical field strengths (1.5 Tesla) in under 10 min.     

Spectrally weighted twisted projection imaging: Reducing T2 signal attenuation effects in fast three-dimensional sodium imaging

A scheme for the reduction of T₂ signal attenuation effects in three-dimensional twisted projection imaging is presented. By purposely reducing the sample density at the high spatial frequencies, a considerable reduction in readout time is achieved. The reduction in readout time leads to decreased T₂ signal attenuation which translates into improved signal-to-noise ratio (SNR). The SNR improvement is achieved without decreasing the image's resolution since the point spread function depends on the sample weighting as well as the T₂ attenuation. The results indicate that SNR improvements of up to 40% can be achieved using the proposed scheme.     

Projection-reconstruction methods: Fast imaging sequences and data processing

New, rapid two- and three-dimensional imaging sequences based on steady-state gradient echoes and projection-reconstruction (PR) techniques are proposed. Quantitative studies show that fast PR sequences and classical, fast gradient-echo Fourier transform sequences lead to identical contrasts. In order to minimize inhomogeneity effects, a particular focus has been placed on echo-time reduction. The use of a weighting window permits one to acquire severely truncated echoes; partial k-space scanning may be considered.     

Frequency resolved single-shot MR imaging using stochastic k-space trajectories

A new single-shot stochastic imaging technique with a random k-space path that provides very selective filtering with respect to chemical shift or off-resonance signals of the investigated tissue is proposed. It is demonstrated that in stochastic imaging only on-resonance compartments are visible whereas frequency shifted compartments cancel to noise that is distributed over the whole image. This method can be used as a single-shot chemical shift selective imaging technique and allows to calculate frequency resolved spectra for each spatial position of the image based on a single signal aquisition. The single-shot stochastic imaging sequence makes high demands on the gradient system and the theoretical k-space trajectory is distorted by imperfect gradient performance. Therefore an additional k-space guided imaging technique that uses the true, measured k-space trajectory to correct artifacts generated by eddy currents and delay times of the rapid switched gradients is presented. In vitro and in vivo measurements demonstrate the successful implementation of single-shot stochastic imaging on a conventional MR scanner with unshielded gradient systems.     

A fast spin echo technique with circular sampling

This paper presents a fast spin echo (FSE) imaging method that employs circular sampling of Jr-space. The technique has been implemented on a 2 Tesla imaging system and validated on both phantoms and living animals. Experimental studies have shown that circular sampling can produce artifact-free FSE images without the need of phase correction. Although not fully explored, preliminary results also show that circular sampling may have advantages over the conventional rectilinear FSE in signal-to-noise ratio and imaging efficiency. A major disadvantage is the increased sensitivity to off-resonance effects. The authors expect that the FSE technique with circular sampling will find its applications in magnetic resonance microscopy, neuro-functional imaging, and real-time dynamic studies.     

Black blood dual phase turbo FLASH MR imaging of the heart

An MR imaging scheme for dark blood cardiac images acquired simultaneously in e～d diastolic and end systolic phases, in breath-hold times, is presented. The sequence consists of a magnetization preparation period followed by two segmented k-space acquisitions. Image quality was investigated with respect to different sequence parameters (optimal values are indicated in brackets): (a) echo time (TE)/repetition time (TR)/flip angle (FA) (3/6.2 msec/20°); (b) number of lines/segments in the acquisition window (11 lines/segment); (c) location of acquisition windows and inversion time; and (d) thickness of slice ?reinverted”? during preparation (1.53 times the acquisition slice thickness). The image quality of the basal slices at end systole was critically dependent on the last parameter. High quality short axis views of the heart, with good blood signal suppression, were acquired with the optimized sequence on four volunteers from apex to base at two phases in 10 breath-holds.     

Three-dimensional projection imaging with half the number of projections

This study demonstrates a sampling scheme for three-dimensional projection imaging with half the number of projections. The angular distribution of projections is designed so that the oversampling of the low spatial frequencies, in tandem with a partial Fourier algorithm, can be used to recreate purposely missing projections. The performance of the sampling scheme and its associated reconstruction algorithm are illustrated with computer-simulated as well as experimental data sets. We find that this technique produces images of comparable quality to the conventional projection imaging scheme. Although a 30% loss of signal-to-noise ratio (SNR) results from its use, the algorithm should prove useful for applications where robustness against motion artifacts and reduced T₂* signal loss are desired.     

Reduced phase encoding in spectroscopic imaging

The effect of different spatial-encoding (k-space) sampling distributions are evaluated for magnetic resonance spectroscopic imaging (MRSI) using Fourier reconstruction. Previously, most MRSI studies have used square or cubic k-space functions, symmetrically distributed. These studies examine the conventional k-space distribution with spherical distribution, and 1/2 k-space acquisition, using computer simulation studies of the MRSI acquisition for three spatial dimensions and experimental results. Results compare the spatial response function, Gibbs ringing effects, and signal contamination for different spatial-encoding distribution functions. Results indicate that spherical encoding, in comparison with cubic encoding, results in a modest improvement of the re sponse function with approximately equivalent spatial resolution for the same acquisition time. For spin-echo acquired data, reduced acquisition times can readily be obtained using 1/2 k-space methods, with a concomitant reduction in signal to noise ratio.     

Fast Three Dimensional Magnetic Resonance Imaging

To reduce the scan time in three-dimensional (3D) imaging, the authors consider alternative trajectories for traversing k-space. They differ from traditional 3D trajectories, such as 3DFT, in that they employ time-varying gradients allowing longer readouts and in turn a reduced scan time. Some of these trajectories reduce by an order of magnitude the number of excitations compared with 3DFT and provide flexibility for trading off signal-to-noise ratio for scan time. Other concerns are the minimum echo time and flow/motion properties. As examples, the authors show two applications: A 3D data set of the head (field of view of 30 x 30 x 7.5 cm and resolution of 1.5 x 1.5 x 1.5 mm) acquired in 56 s using a stack of spirals in 3D k-space; and a 3D movie of the heart (20 x 20 x 20 cm field of view, 2 x 2 x 2 mm resolution, and 16 time frames per cardiac cycle) acquired in 11 min using a cones trajectory.     

Rapid cardiac-output measurement with ungated spiral phase contrast.

An ungated spiral phase-contrast (USPC) method was used to measure cardiac output (CO) rapidly and conveniently. The USPC method, which was originally designed for small peripheral vessels, was modified to assess CO by measuring flow in the ascending aorta (AA). The modified USPC used a 12-interleaf spiral trajectory to acquire full-image data every 283 ms with 2-mm spatial resolution. The total scan time was 5 s. For comparison, a triggered real-time (TRT) method was used to indirectly calculate CO by measuring left-ventricular (LV) volume. The USPC and TRT measurements from all normal volunteers agreed. In a patient with patent ductus arteriosus (PDA), high CO was measured with USPC, which agreed well with the invasive cardiac-catheterized measurement. In normal volunteers, CO dropped about 20-30% with Valsalva maneuvering, and increased about 100% after exercise. Continuous 28-s cycling between Valsalva maneuvering and free-breathing showed that USPC can temporally resolve physiological CO changes.     

Comparison of k-space sampling schemes for multidimensional MR spectroscopic imaging

For clinical ³¹P MR spectroscopic imaging (MRSI) studies, where signal averaging is necessary, some improvement of sensitivity and spatial response function may be achieved by acquiring data over a spherical k-space volume and varying the number of averages acquired in proportion to the desired spatial filter. Eight different k-space sampling schemes are compared through simulations that provide graphs of the spatial response functions (SRF), and tabulations of voxel volumes, relative signal-to-noise ratios (SNR), and relative data collection efficiencies (SNR per unit volume over the same time). All schemes were based on practical experiments, each of which could be implemented in the same length of time. The results show that in comparison with cubic k-space sampling with the same number of signal averages at each point, spherical and acquisition-weighted k-space sampling can be used to achieve reduced Gibbs ringing along the principal axes directions, and thus reduced contamination from adjacent tissue in these directions, without degradation of voxel volume or SNR.     

Doppler US gating of cardiac MR imaging

RATIONALE AND OBJECTIVES: Electrocardiographic (ECG) gating of cardiac magnetic resonance (MR) imaging has been problematic for many reasons. The purpose of this study was to demonstrate the feasibility of using Doppler ultrasound (US) gating, either directly off the moving cardiac wall or the systolic upstroke of the arterial signal from the great vessels in neck, in alternative gating modes. MATERIALS AND METHODS: A 2.5-MHz, range-gated Doppler US device was used with A-mode guidance for gating directly off left ventricular wall motion. A 4- or 8.1-MHz, continuous-wave (CW) Doppler US device was used for gating off the systolic upstroke from the great vessels in the neck. The subject undergoing imaging held the transducer against his chest for range-gated Doppler US and against his neck for 8.1-MHz CW Doppler US. The 4-MHz transducer was strapped to the subject's neck. Modified Doppler signals were fed back into the gating circuitry of the MR imager to achieve cardiac synchrony. RESULTS: Cardiac gating was achieved by using both the range-gated technique directly off the cardiac wall and the CW method off blood flow from the great vessels. Problems occurred with radiofrequency shielding during the range-gated method; however, these problems were almost completely removed by use of the CW Doppler probes. CONCLUSION: Doppler US gating of MR images is possible and potentially could overcome many shortcomings of ECG gating. Subsequent embodiments of the technique will require improved radiofrequency shielding in the range-gated technique.     

Magnetic resonance imaging with true fast imaging with steady-state precession and half-Fourier acquisition single-shot turbo spin-echo sequences in cases of suspected placenta accreta

Purpose: To present the magnetic resonance imaging (MRI) findings of placenta accreta in suspected cases of placenta accreta with true fast imaging with steady-state precession (True FISP) and half-Fourier acquisition single-shot turbo spin-echo (HASTE) sequences.

Material and Methods: Five patients underwent MRI with HASTE (n=5) and/or True FISP (n=4) sequences for suspected placenta accreta. Retrospective review of MRI was performed to define the location and extent of the implantation abnormality.

Results: The uteroplacental interface was visualized as three layers; inner low signal intensity layer, middle high signal intensity layer of myometrium, and outer low signal intensity layer of uterine serosa. Three cases were diagnosed with placenta accreta on MRI and focal non-visualization of the inner layer was demonstrated.

Conclusion: The finding of focal non-visualization of the inner layer between the placenta and myometrium by MRI with True FISP and HASTE sequences was the diagnostic finding for placenta accreta.     

CINE turbo spin echo imaging

High‐resolution turbo spin echo (TSE) images have demonstrated important details of carotid artery morphology; however, it is evident that pulsatile blood and wall motion related to the cardiac cycle are still significant sources of image degradation. Although ECG gating can reduce artifacts due to cardiac‐induced pulsations, gating is rarely used because it lengthens the acquisition time and can cause image degradation due to nonconstant repetition time. This work introduces a relatively simple method of converting a conventional TSE acquisition into a retrospectively ECG‐correlated cineTSE sequence. The cineTSE sequence generates a full sequence of ECG‐correlated images at each slice location throughout the cardiac cycle in the same scan time that is conventionally used by standard TSE sequences to produce a single image at each slice location. The cineTSE images exhibit reduced pulsatile artifacts associated with a gated sequence but without the increased scan time or associated nonconstant repetition time effects. Magn Reson Med, 2011. © 2011 Wiley Periodicals, Inc.     

Reduction of motion artifacts in cine MRI using variable-density spiral trajectories

Dynamic cardiac imaging in MRI is a very challenging task. To obtain high spatial resolution, temporal resolution, and signalto-noise ratio (SNR), single-shot imaging is not sufficient Use of multishot techniques resolves this problem but can cause motion artifacts because of data inconsistencies between views. Motion artifacts can be reduced by signal averaging at some cost in increased scan time. However, for the same increase in scan time, other techniques can be more effective than simple averaging in reducing the artifacts. If most of the energy of the inconsistencies is limited to a certain region of k-space, increased sampling density (oversampling) in this region can be especially effective in reducing motion artifacts. In this work, several variable-density spiral trajectories are designed and tested. Their efficiencies for artifact reduction are evaluated in computer simulations and in scans of normal volunteers. The SNR compromise of these trajectories is also investigated. The authors conclude that variable-density spiral trajectories can effectively reduce motion artifacts with a small loss in SNR as compared with a uniform density counterpart.     

Single-breathhold 3D-trueFISP cine cardiac imaging.

Cardiac MRI function measurements are typically based on multiple breathhold 2D sequences to acquire images of the entire heart. In the present study, the feasibility of a cine 3D TrueFISP technique in which several complete volumetric measurements may be obtained during a single breathhold is demonstrated. In contrast to 3D FLASH, the TrueFISP sequence offers an excellent contrast between the myocardium and the intraventricular cavity without the use of contrast agent. An ECG-gated 3D cine TrueFISP sequence was implemented with a repetition time of 2.4-2.8 ms, which allows imaging of the complete heart within a single breathhold throughout 20-46 heartbeats with a 3D frame rate of 8-13 volumes per cardiac cycle and a spatial resolution of about 1.5 x 3.5 x 3.5 mm(3). Breathhold volumetric cine imaging with the 3D TrueFISP technique holds promise for rapid and accurate evaluation of the cardiac regional wall motion and the calculation of cardiac volume and ejection fraction.