Quantitative susceptibility map reconstruction from MR phase data using bayesian regularization: validation and application to brain imaging

The diagnosis of many neurologic diseases benefits from the ability to quantitatively assess iron in the brain. Paramagnetic iron modifies the magnetic susceptibility causing magnetic field inhomogeneity in MRI. The local field can be mapped using the MR signal phase, which is discarded in a typical image reconstruction. The calculation of the susceptibility from the measured magnetic field is an ill-posed inverse problem. In this work, a bayesian regularization approach that adds spatial priors from the MR magnitude image is formulated for susceptibility imaging. Priors include background regions of known zero susceptibility and edge information from the magnitude image. Simulation and phantom validation experiments demonstrated accurate susceptibility maps free of artifacts. The ability to characterize iron content in brain hemorrhage was demonstrated on patients with cavernous hemangioma. Additionally, multiple structures within the brain can be clearly visualized and characterized. The technique introduces a new quantitative contrast in MRI that is directly linked to iron in the brain.

Nonlinear regularization for per voxel estimation of magnetic susceptibility distributions from MRI field maps

Magnetic susceptibility is an important physical property of tissues, and can be used as a contrast mechanism in magnetic resonance imaging (MRI). Recently, targeting contrast agents by conjugation with signaling molecules and labeling stem cells with contrast agents have become feasible. These contrast agents are strongly paramagnetic, and the ability to quantify magnetic susceptibility could allow accurate measurement of signaling and cell localization. Presented here is a technique to estimate arbitrary magnetic susceptibility distributions by solving an ill-posed inversion problem from field maps obtained in an MRI scanner. Two regularization strategies are considered: conventional Tikhonov regularization and a sparsity promoting nonlinear regularization using the l(1) norm. Proof of concept is demonstrated using numerical simulations, phantoms, and in a stroke model in a rat. Initial experience indicates that the nonlinear regularization better suppresses noise and streaking artifacts common in susceptibility estimation.

Free-breathing 3-dimensional steady-state free precession coronary magnetic resonance angiography: comparison of four navigator gating techniques

This work compared the performance of four navigator gating algorithms [accept/reject (A/R), diminishing variance algorithm (DVA), phase ordering with automatic window selection (PAWS) and retrospective gating (RETRO)] in suppressing respiratory motion artifacts in free-breathing 3D balanced steady-state free precession coronary MRA. In 10 volunteers, the right coronary artery (RCA) or the left anterior descending artery (LAD) was imaged (both if time permitted) at 1.5 T with the four gating techniques in random order. Vessel signal, vessel contrast and motion suppression were scored by the consensus of two blinded readers. In 15 imaged vessels (nine RCA and six LAD), PAWS provided significantly better image quality than A/R (P<.05), DVA (P=.02) and RETRO (P=.002). While the quality difference between A/R and DVA was not statistically significant, both algorithms yielded significantly better image quality than RETRO. PAWS and DVA were the most efficient algorithms, providing an approximately 20% and 40% relative increase in average navigator efficiency compared to A/R and RETRO, respectively.

In vivo quantification of femoral-popliteal compression during isometric thigh contraction: Assessment using MR angiography

PURPOSE

To quantify femoral-popliteal vessel deformation during thigh contraction.

MATERIALS AND METHODS

Eleven subjects underwent a magnetic resonance (MR) examination of the femoral-popliteal vasculature on a 1.5 T system. A custom 3D balanced steady-state free precession (SSFP) sequence was implemented to image a 15-20-cm segment of the vasculature during relaxation and voluntary isometric thigh contraction. The arterial and venous lumina were outlined using a semiautomated method. For the artery, this outline was fit to an ellipse whose aspect ratio was used to describe arterial deformation, while venous deformation was characterized by its cross-sectional area.

RESULTS

Focal compression of the femoral-popliteal artery during contraction was observed 94-143 mm superior to the condyle that corresponds to the distal adductor canal (AC) immediately superior to the adductor hiatus. This was illustrated by a significant reduction (P < or = 0.05) in aspect ratio from 0.88 +/- 0.06 during relaxation to 0.77 +/- 0.09 during contraction. A negligible change in arterial aspect ratio was observed inferior to the AC and in the proximal AC. Similarly, venous area was dramatically reduced in the distal AC region during contraction.

CONCLUSION

Rapid 3D SSFP MR angiography of the femoral-popliteal vasculature during thigh contraction demonstrated focal compression of the artery in the distal AC region. This may help explain the high stent failure rate and the high likelihood of atherosclerotic disease in the AC. J. Magn. Reson.

Calculation of susceptibility through multiple orientation sampling (COSMOS): a method for conditioning the inverse problem from measured magnetic field map to susceptibility source image in MRI

Magnetic susceptibility differs among tissues based on their contents of iron, calcium, contrast agent, and other molecular compositions. Susceptibility modifies the magnetic field detected in the MR signal phase. The determination of an arbitrary susceptibility distribution from the induced field shifts is a challenging, ill-posed inverse problem. A method called "calculation of susceptibility through multiple orientation sampling" (COSMOS) is proposed to stabilize this inverse problem. The field created by the susceptibility distribution is sampled at multiple orientations with respect to the polarization field, B(0), and the susceptibility map is reconstructed by weighted linear least squares to account for field noise and the signal void region. Numerical simulations and phantom and in vitro imaging validations demonstrated that COSMOS is a stable and precise approach to quantify a susceptibility distribution using MRI.

Effective motion-sensitizing magnetization preparation for black blood magnetic resonance imaging of the heart

PURPOSE

To investigate the effectiveness of flow signal suppression of a motion-sensitizing magnetization preparation (MSPREP) sequence and to optimize a 2D MSPREP steady-state free precession (SSFP) sequence for black blood imaging of the heart.

MATERIALS AND METHODS

Using a flow phantom, the effect of varying field of speed (FOS), b-value, voxel size, and flow pattern on the flow suppression was investigated. In seven healthy volunteers, black blood images of the heart were obtained at 1.5T with MSPREP-SSFP and double inversion recovery fast spin echo (DIR-FSE) techniques. Myocardium and blood signal-to-noise ratio (SNR) and myocardium-to-blood contrast-to-noise ratio (CNR) were measured. The optimal FOS that maximized the CNR for MSPREP-SSFP was determined.

RESULTS

Phantom data demonstrated that the flow suppression was induced primarily by the velocity encoding effect. In humans, FOS=10-20 cm/s was found to maximize the CNR for short-axis (SA) and four-chamber (4C) views. Compared to DIR-FSE, MSPREP-SSFP provided similar blood SNR efficiency in the SA basal and mid-views and significantly lower blood SNR efficiency in the SA apical (P=0.02) and 4C (P=0.01) views, indicating similar or better blood suppression.

CONCLUSION

Velocity encoding is the primary flow suppression mechanism of the MSPREP sequence and 2D MSPREP-SSFP black blood imaging of the heart is feasible in healthy subjects.

In vivo quantification of contrast agent concentration using the induced magnetic field for time-resolved arterial input function measurement with MRI

For pharmacokinetic modeling of tissue physiology, there is great interest in measuring the arterial input function (AIF) from dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) using paramagnetic contrast agents. Due to relaxation effects, the measured signal is a nonlinear function of the injected contrast agent concentration and depends on sequence parameters, system calibration, and time-of-flight effects, making it difficult to accurately measure the AIF during the first pass. Paramagnetic contrast agents also affect susceptibility and modify the magnetic field in proportion to their concentration. This information is contained in the MR signal phase which is discarded in a typical image reconstruction. However, quantifying AIF through contrast agent susceptibility induced phase changes is made difficult by the fact that the induced magnetic field is nonlocal and depends upon the contrast agent spatial distribution and thus on organ and vessel shapes. In this article, the contrast agent susceptibility was quantified through inversion of magnetic field shifts using a piece-wise constant model. Its feasibility is demonstrated by a determination of the AIF from the susceptibility-induced field changes of an intravenous bolus. After in vitro validation, a time-resolved two-dimensional (2D) gradient echo scan, triggered to diastole, was performed in vivo on the aortic arch during a bolus injection of 0.1 mmol/kg Gd-DTPA. An approximate geometrical model of the aortic arch constructed from the magnitude images was used to calculate the spatial variation of the field associated with the bolus. In 14 subjects, Gd concentration curves were measured dynamically (one measurement per heart beat) and indirectly validated by independent 2D cine phase contrast flow rate measurements. Flow rate measurements using indicator conservation with this novel quantitative susceptibility imaging technique were found to be in good agreement with those obtained from the cine phase contrast measurements in all subjects. Contrary to techniques that rely on intensity, the accuracy of this signal phase based method is insensitive to factors influencing signal intensity such as flip angle, coil sensitivity, relaxation changes, and time-of-flight effects extending the range of pulse sequences and contrast doses for which quantitative DCE-MRI can be applied.

Free-breathing 3D steady-state free precession coronary magnetic resonance angiography: comparison of diaphragm and cardiac fat navigators

PURPOSE

To compare the performance of the conventional diaphragm navigator (DNAV) and the recently developed cardiac fat navigator (FatNAV) in suppressing respiration-induced cardiac motion in free-breathing 3D balanced steady-state free precession coronary MRA (SSFP CMRA).

MATERIALS AND METHODS

In 16 healthy volunteers the right coronary artery (RCA) was imaged at 1.5T using a navigator-gated 3D SSFP CMRA sequence. DNAV and FatNAV gating were performed in random order. Image quality difference was scored by three experienced readers blinded to the gating technique. Blood signal-to-noise ratio (SNR), blood-to-myocardium contrast-to-noise ratio (CNR), and navigator efficiency were calculated.

RESULTS

Diagnostically interpretable CMRA was obtained successfully in all 16 subjects with FatNAV gating (0% failure rate) and only 14 subjects with DNAV gating (12% failure rate). Compared to DNAV gating, FatNAV gating provided similar SNR and CNR, better image quality (P < 0.01), and 28% improvement in navigator efficiency (P = 0.002).

CONCLUSION

FatNAV gating provides more effective motion suppression and better image quality than DNAV gating for free-breathing 3D SSFP CMRA of the RCA in healthy subjects.

Three-dimensional cine imaging using variable-density spiral trajectories and SSFP with application to coronary artery angiography

A single breath-hold 3D cardiac phase resolved steady-state free precession (SSFP) sequence was developed, allowing 3D visualization of the moving coronary arteries. A 3D stack of spirals was acquired continuously throughout the cardiac cycle, and a sliding window reconstruction was used to achieve high temporal resolution. A coil specific field of view reconstruction technique was combined with Parallel Imaging with Localized Sensitivities (PILS) to allow acquisition of a reduced field of view. A view ordering incorporating fat suppression was employed to allow use of sliding window reconstruction. The technique was evaluated on healthy volunteers (n=8), yielding images with 102 ms temporal resolution and 1.35 mm in-plane resolution, and reasonable visualization of the left and right coronary arteries was achieved.

Optimal coil array design: the two-coil case

The optimization problem for coil arrays is largely unsolved, even for the case of a two-coil system. This paper reports a systematic computer simulation to investigate the maximal achievable signal-to-noise ratio (SNR) with a two-coil receiver system where, using cancellation circuitry, mutual inductance is made zero. Both symmetrical and asymmetrical solutions with respect to two-coil geometry are considered. SNR is measured at a single point at a certain depth and also along a longitudinal or transverse line at the same depth. The conducting medium containing these regions of interest is assumed to be an infinite half space, an infinite cylinder or a finite sphere. The previous coil array design using a "magical" overlap only approximates the optimal solution for the infinite half space. For the infinite cylinder and the finite sphere, optimal solutions can be quite different from the "magical" overlap.

On the noise correlation matrix for multiple radio frequency coils

Noise correlation between multiple receiver coils is discussed using principles of statistical physics. Using the general fluctuation-dissipation theorem we derive the prototypic correlation formula originally determined by Redpath (Magn Res Med 1992;24:85-89), which states that correlation of current spectral noise depends on the real part of the inverse impedance matrix at a given frequency. A distinct correlation formula is also derived using the canonical partition function, which states that correlation of total current noise over the entire frequency spectrum depends on the inverse inductance matrix. The Kramers-Kronig relation is used to equate the inverse inductance matrix to the spectral integral of the inverse impedance matrix, implying that the total noise is equal to the summation of the spectral noise over the entire frequency spectrum. Previous conflicting arguments on noise correlation may be reconciled by differentiating between spectral and total noise correlation. These theoretical derivations are verified experimentally using two-coil arrays.

Contrast-Enhanced Magnetic Resonance Angiography with Biodegradable (Gd-DTPA)-Cystamine Copolymers: Comparison with MS-325 in a Swine Model

The purpose of this study is to evaluate the use of (Gd-DTPA)-cystamine copolymers (GDCC), a novel biodegradable intravascular polydisulfide-based macromolecular gadolinium(III) contrast agent, for first-pass and steady-state contrast-enhanced magnetic resonance angiography (MRA) in a swine model. A breath-hold background-suppressed 3D MRA of the thorax was performed for first-pass imaging and repeated every 10 min after GDCC injection to monitor the tissue enhancement time course. A navigator-gated 3D MRA of the coronary arteries was performed during steady state following the first-pass imaging. Imaging with intravascular agent MS-325 approximately 1 h after GDCC injection was also included for comparison. Experimental results indicated that GDCC provided significant blood signal-to-noise ratio (SNR) improvement, approximately 1633% for first-pass and 33% for steady-state contrast-enhanced MRA. Compared to MS-325, GDCC provided similar blood enhancement for first-pass and steady-state imaging but with a different tissue enhancement time course. The blood SNR enhancement half-time was 10 +/- 6 min for GDCC and 46 +/- 33 min for MS-325. GDCC provided less enhancement in the liver, bone growth plates, and muscle than MS-325.

Improved signal-to-noise ratio in parallel coronary artery magnetic resonance angiography using graph cuts based Bayesian reconstruction

High resolution 3D coronary artery MR angiography is time-consuming and can benefit from accelerated data acquisition provided by parallel imaging techniques without sacrificing spatial resolution. Currently, popular maximum likelihood based parallel imaging reconstruction techniques such as the SENSE algorithm offer this advantage at the cost of reduced signal-to-noise ratio (SNR). Maximum a posteriori (MAP) reconstruction techniques that incorporate globally smooth priors have been developed to recover this SNR loss, but they tend to blur sharp edges in the target image. The objective of this study is to demonstrate the feasibility of employing edge-preserving Markov random field priors in a MAP reconstruction framework, which can be solved efficiently using a graph cuts based optimization algorithm. The preliminary human study shows that our reconstruction provides significantly better SNR than the SENSE reconstruction performed by a commercially available scanner for navigator gated steady state free precession 3D coronary magnetic resonance angiography images (n = 4).

Reduction of reconstruction time for time-resolved spiral 3D contrast-enhanced magnetic resonance angiography using parallel computing

Time-resolved 3D MRI with high spatial and temporal resolution can be achieved using spiral sampling and sliding-window reconstruction. Image reconstruction is computationally intensive because of the need for data regridding, a large number of temporal phases, and multiple RF receiver coils. Inhomogeneity blurring correction for spiral sampling further increases the computational work load by an order of magnitude, hindering the clinical utility of spiral trajectories. In this work the reconstruction time is reduced by a factor of >40 compared to reconstruction using a single processor. This is achieved by using a cluster of 32 commercial off-the-shelf computers, commodity networking hardware, and readily available software. The reconstruction system is demonstrated for time-resolved spiral contrast-enhanced (CE) peripheral MR angiography (MRA), and a reduction of reconstruction time from 80 min to 1.8 min is achieved.

Cardiac fat navigator-gated steady-state free precession 3D magnetic resonance angiography of coronary arteries

Motion artifacts and the lack of accurate detection of cardiac motion present a major challenge for high-resolution cardiac MRI. Recently a multidimensional cardiac fat navigator was proposed to provide a fast and direct measurement of bulk cardiac motion. The objective of this study was to demonstrate the feasibility of employing the cardiac fat navigator in balanced steady-state free precession (SSFP) free-breathing 3D coronary MRA (CMRA). The cardiac fat navigator echo is optimized to provide both motion monitoring and epicardial fat suppression. Steady-state magnetization preparation, which is needed for SSFP CMRA, is optimized by comparing three preparation schemes: alpha/2, linear ramp with 20 RF pulses (20LR), and Kaiser ramp with six RF pulses (6KR). The present preliminary human study shows that the 6KR preparation provides better image quality than both the alpha/2 (P<0.0025) and the 20LR preparations (P<0.025) for free-breathing SSFP 3D CMRA (N=11).

Motion artifact suppression in breath hold 3D contrast enhanced magnetic resonance angiography using ECG ordering

Vascular pulsation and cardiac motion compromise image quality in contrast enhanced magnetic resonance angiography (CE-MRA) in the thorax, resulting in blurring and ghosting artifacts. The use of ECG gating has been proposed in the past to mitigate these artifacts but previous methods suffered from increased scanner time because only a fraction of the cardiac cycle was used for image acquisition and from loss of the study when gating failed. We propose a robust ECG ordering of k-space for breath hold CE-MRA that acquires the central part of k-space in a motion-free portion of diastole and fills in from the periphery of k-space at all other times. To make maximal use of the contrast enhancement, data is acquired continuously even when the ECG signal is lost. The proposed sequence is shown to allow thoracic and pulmonary MRA with a higher resolution when compared to the conventional gated sequence.

Multiprocessor scheduling implementation of the simultaneous multiple volume (SMV) navigator method

The simultaneous multiple volume (SMV) approach in navigator-gated MRI allows the use of the whole motion range or the entire scan time for the reconstruction of final images by simultaneously acquiring different image volumes at different motion states. The motion tolerance range for each volume is kept small, thus SMV substantially increases the scan efficiency of navigator methods while maintaining the effectiveness of motion suppression. This article reports a general implementation of the SMV approach using a multiprocessor scheduling algorithm. Each motion state is regarded as a processor and each volume is regarded as a job. An efficient scheduling that completes all jobs in minimal time is maintained even when the motion pattern changes. Initial experiments demonstrated that SMV significantly increased the scan efficiency of navigator-gated MRI.

View ordering for magnetization prepared steady state free precession acquisition: Application in contrast-enhanced MR angiography

Magnetization prepared segmented acquisition requires a view order that maximizes signal contrast during the acquisition of the central portion of k-space. Steady state free precession (SSFP) acquisition further requires a view order that minimizes changes in phase-encoding gradients from one repetition to the next in order to minimize eddy current artifacts. In this article, optimal view ordering schemes satisfying these two requirements are formulated and applied to inversion prepared 3D SSFP contrast-enhanced MR angiography (MRA). Experiments on phantoms and pigs demonstrated improved background suppression and reduced image artifacts.

Improved magnetization preparation for navigator steady-state free precession 3D coronary MR angiography

The purpose of this work was to investigate a new magnetization preparation scheme for navigator steady-state free precession (SSFP) 3D coronary MR angiography (MRA) that executes the navigator and fat saturation pulses in steady state after the dummy RFs in order to minimize the delay between the magnetization preparation and the image echoes. Compared to the previous preparation scheme that executes the navigator and fat saturation pulses before the dummy RFs, the new scheme was found to provide more effective motion suppression, significantly improved blood-to-myocardium contrast-to-noise ratio (46%, P < 0.001) at slightly but insignificantly decreased blood signal-to-noise ratio (SNR) (2%, P = 0.73), significantly reduced fat SNR (32%, P < 0.001), and better overall image quality (P = 0.05; Wilcoxon paired sample signed rank test).