Joint water-fat separation and deblurring for spiral imaging.

PURPOSE: Most previous approaches to spiral Dixon water-fat imaging perform the water-fat separation and deblurring sequentially based on the assumption that the phase accumulation and blurring as a result of off-resonance are separable. This condition can easily be violated in regions where the B0 inhomogeneity varies rapidly. The goal of this work is to present a novel joint water-fat separation and deblurring method for spiral imaging. METHODS: The proposed approach is based on a more accurate signal model that takes into account the phase accumulation and blurring simultaneously. A conjugate gradient method is used in the image domain to reconstruct the deblurred water and fat iteratively. Spatially varying convolutions with a local convergence criterion are used to reduce the computational demand. RESULTS: Both simulation and high-resolution brain imaging have demonstrated that the proposed joint method consistently improves the quality of reconstructed water and fat images compared with the sequential approach, especially in regions where the field inhomogeneity changes rapidly in space. The loss of signal-to-noise-ratio as a result of deblurring is minor at optimal echo times. CONCLUSIONS: High-quality water-fat spiral imaging can be achieved with the proposed joint approach, provided that an accurate field map of B0 inhomogeneity is available. Magn Reson Med 79:3218-3228, 2018. © 2017 International Society for Magnetic Resonance in Medicine.

Multi-frequency interpolation in spiral magnetic resonance fingerprinting for correction of off-resonance blurring.

Magnetic resonance fingerprinting (MRF) pulse sequences often employ spiral trajectories for data readout. Spiral k-space acquisitions are vulnerable to blurring in the spatial domain in the presence of static field off-resonance. This work describes a blurring correction algorithm for use in spiral MRF and demonstrates its effectiveness in phantom and in vivo experiments. Results show that image quality of T1 and T2 parametric maps is improved by application of this correction. This MRF correction has negligible effect on the concordance correlation coefficient and improves coefficient of variation in regions of off-resonance relative to uncorrected measurements.

ISMRM Raw data format: A proposed standard for MRI raw datasets.

PURPOSE: This work proposes the ISMRM Raw Data format as a common MR raw data format, which promotes algorithm and data sharing. METHODS: A file format consisting of a flexible header and tagged frames of k-space data was designed. Application Programming Interfaces were implemented in C/C++, MATLAB, and Python. Converters for Bruker, General Electric, Philips, and Siemens proprietary file formats were implemented in C++. Raw data were collected using magnetic resonance imaging scanners from four vendors, converted to ISMRM Raw Data format, and reconstructed using software implemented in three programming languages (C++, MATLAB, Python). RESULTS: Images were obtained by reconstructing the raw data from all vendors. The source code, raw data, and images comprising this work are shared online, serving as an example of an image reconstruction project following a paradigm of reproducible research. CONCLUSION: The proposed raw data format solves a practical problem for the magnetic resonance imaging community. It may serve as a foundation for reproducible research and collaborations. The ISMRM Raw Data format is a completely open and community-driven format, and the scientific community is invited (including commercial vendors) to participate either as users or developers. Magn Reson Med 77:411-421, 2017. © 2016 Wiley Periodicals, Inc.

Sliding-slab three-dimensional TSE imaging with a spiral-In/Out readout.

PURPOSE: T2 -weighted imaging is of great diagnostic value in neuroimaging. Three-dimensional (3D) Cartesian turbo spin echo (TSE) scans provide high signal-to-noise ratio (SNR) and contiguous slice coverage. The purpose of this preliminary work is to implement a novel 3D spiral TSE technique with image quality comparable to 2D/3D Cartesian TSE. METHODS: The proposed technique uses multislab 3D TSE imaging. To mitigate the slice boundary artifacts, a sliding-slab method is extended to spiral imaging. A spiral-in/out readout is adopted to minimize the artifacts that may be present with the conventional spiral-out readout. Phase errors induced by B0 eddy currents are measured and compensated to allow for the combination of the spiral-in and spiral-out images. A nonuniform slice encoding scheme is used to reduce the truncation artifacts while preserving the SNR performance. RESULTS: Preliminary results show that each of the individual measures contributes to the overall performance, and the image quality of the results obtained with the proposed technique is, in general, comparable to that of 2D or 3D Cartesian TSE. CONCLUSION: 3D sliding-slab TSE with a spiral-in/out readout provides good-quality T2 -weighted images, and, therefore, may become a promising alternative to Cartesian TSE.

Arterial spin labeled perfusion imaging using three-dimensional turbo spin echo with a distributed spiral-in/out trajectory.

PURPOSE: The three-dimensional (3D) spiral turbo spin echo (TSE) sequence is one of the preferred readout methods for arterial spin labeled (ASL) perfusion imaging. Conventional spiral TSE collects the data using a spiral-out readout on a stack of spirals trajectory. However, it may result in suboptimal image quality and is not flexible in protocol design. The goal of this study is to provide a more robust readout technique without such limitation. METHODS: The proposed technique incorporates a spiral-in/out readout into 3D TSE, and collects the data on a distributed spirals trajectory. The data set is split into the spiral-in and -out subsets that are reconstructed separately and combined after image deblurring. RESULTS: The volunteer results acquired with the proposed technique show no geometric distortion or signal pileup, as is present with GRASE, and no signal loss, as is seen with conventional spiral TSE. Examples also demonstrate the flexibility in changing the imaging parameters to satisfy various criteria. CONCLUSION: The 3D TSE with a distributed spiral-in/out trajectory provides a robust readout technique and allows for easy protocol design, thus is a promising alternative to GRASE or conventional spiral TSE for ASL perfusion imaging.

Dixon water-fat separation in PROPELLER MRI acquired with two interleaved echoes.

PURPOSE: To propose a novel combination of robust Dixon fat suppression and motion insensitive PROPELLER (periodically rotated overlapping parallel lines with enhanced reconstruction) MRI. METHODS: Two different echoes were acquired interleaved in each shot enabling water-fat separation on individual blades. Fat, which was blurred in standard PROPELLER because the water-fat shift (WFS) rotated with the blades, was shifted back in each blade. Additionally, field maps obtained from the water-fat separation were used to unwarp off-resonance-induced shifts in each blade. PROPELLER was then applied to the water, corrected fat, or recombined water-fat blades. This approach was compared quantitatively in volunteers with regard to motion estimation and signal-to-noise ratio (SNR) to a standard PROPELLER acquisition with minimal WFS and fat suppression. RESULTS: Shifting the fat back in each blade reduced errors in the translation correction. SNR in the proposed Dixon PROPELLER was 21% higher compared with standard PROPELLER with identical scan time. High image quality was achieved even when the volunteers were moving during data acquisition. Furthermore, sharp water-fat borders and image details were seen in areas where standard PROPELLER suffered from blurring when acquired with a low readout bandwidth. CONCLUSION: The proposed method enables motion-insensitive PROPELLER MRI with robust fat suppression and reduced blurring. Additionally, fat images are available if desired.

Analytical three-point Dixon method: With applications for spiral water-fat imaging.

PURPOSE: The goal of this work is to present a new three-point analytical approach with flexible even or uneven echo increments for water-fat separation and to evaluate its feasibility with spiral imaging. THEORY AND METHODS: Two sets of possible solutions of water and fat are first found analytically. Then, two field maps of the B0 inhomogeneity are obtained by linear regression. The initial identification of the true solution is facilitated by the root-mean-square error of the linear regression and the incorporation of a fat spectrum model. The resolved field map after a region-growing algorithm is refined iteratively for spiral imaging. The final water and fat images are recalculated using a joint water-fat separation and deblurring algorithm. RESULTS: Successful implementations were demonstrated with three-dimensional gradient-echo head imaging and single breathhold abdominal imaging. Spiral, high-resolution T1 -weighted brain images were shown with comparable sharpness to the reference Cartesian images. CONCLUSION: With appropriate choices of uneven echo increments, it is feasible to resolve the aliasing of the field map voxel-wise. High-quality water-fat spiral imaging can be achieved with the proposed approach.

Graphical programming interface: A development environment for MRI methods.

PURPOSE: To introduce a multiplatform, Python language-based, development environment called graphical programming interface for prototyping MRI techniques. METHODS: The interface allows developers to interact with their scientific algorithm prototypes visually in an event-driven environment making tasks such as parameterization, algorithm testing, data manipulation, and visualization an integrated part of the work-flow. Algorithm developers extend the built-in functionality through simple code interfaces designed to facilitate rapid implementation. RESULTS: This article shows several examples of algorithms developed in graphical programming interface including the non-Cartesian MR reconstruction algorithms for PROPELLER and spiral as well as spin simulation and trajectory visualization of a FLORET example. CONCLUSION: The graphical programming interface framework is shown to be a versatile prototyping environment for developing numeric algorithms used in the latest MR techniques.

The effects of SENSE on PROPELLER imaging.

PURPOSE: To study how sensitivity encoding (SENSE) impacts periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) image quality, including signal-to-noise ratio (SNR), robustness to motion, precision of motion estimation, and image quality. METHODS: Five volunteers were imaged by three sets of scans. A rapid method for generating the g-factor map was proposed and validated via Monte Carlo simulations. Sensitivity maps were extrapolated to increase the area over which SENSE can be performed and therefore enhance the robustness to head motion. The precision of motion estimation of PROPELLER blades that are unfolded with these sensitivity maps was investigated. An interleaved R-factor PROPELLER sequence was used to acquire data with similar amounts of motion with and without SENSE acceleration. Two neuroradiologists independently and blindly compared 214 image pairs. RESULTS: The proposed method of g-factor calculation was similar to that provided by the Monte Carlo methods. Extrapolation and rotation of the sensitivity maps allowed for continued robustness of SENSE unfolding in the presence of motion. SENSE-widened blades improved the precision of rotation and translation estimation. PROPELLER images with a SENSE factor of 3 outperformed the traditional PROPELLER images when reconstructing the same number of blades. CONCLUSION: SENSE not only accelerates PROPELLER but can also improve robustness and precision of head motion correction, which improves overall image quality even when SNR is lost due to acceleration. The reduction of SNR, as a penalty of acceleration, is characterized by the proposed g-factor method.

Revised motion estimation algorithm for PROPELLER MRI.

PURPOSE: To introduce a new algorithm for estimating data shifts (used for both rotation and translation estimates) for motion-corrected PROPELLER MRI. The method estimates shifts for all blades jointly, emphasizing blade-pair correlations that are both strong and more robust to noise. THEORY AND METHODS: The heads of three volunteers were scanned using a PROPELLER acquisition while they exhibited various amounts of motion. All data were reconstructed twice, using motion estimates from the original and new algorithm. Two radiologists independently and blindly compared 216 image pairs from these scans, ranking the left image as substantially better or worse than, slightly better or worse than, or equivalent to the right image. RESULTS: In the aggregate of 432 scores, the new method was judged substantially better than the old method 11 times, and was never judged substantially worse. CONCLUSION: The new algorithm compared favorably with the old in its ability to estimate bulk motion in a limited study of volunteer motion. A larger study of patients is planned for future work.

Spiral trajectory design: a flexible numerical algorithm and base analytical equations.

PURPOSE: Spiral-based trajectories for magnetic resonance imaging can be advantageous, but are often cumbersome to design or create. This work presents a flexible numerical algorithm for designing trajectories based on explicit definition of radial undersampling, and also gives several analytical expressions for charactering the base (critically sampled) class of these trajectories. THEORY AND METHODS: Expressions for the gradient waveform, based on slew and amplitude limits, are developed such that a desired pitch in the spiral k-space trajectory is followed. The source code for this algorithm, written in C, is publicly available. Analytical expressions approximating the spiral trajectory (ignoring the radial component) are given to characterize measurement time, gradient heating, maximum gradient amplitude, and off-resonance phase for slew-limited and gradient amplitude-limited cases. Several numerically calculated trajectories are illustrated, and base Archimedean spirals are compared with analytically obtained results. RESULTS: Several different waveforms illustrate that the desired slew and amplitude limits are reached, as are the desired undersampling patterns, using the numerical method. For base Archimedean spirals, the results of the numerical and analytical approaches are in good agreement. CONCLUSION: A versatile numerical algorithm was developed, and was written in publicly available code. Approximate analytical formulas are given that help characterize spiral trajectories.

Fast, variable system delay correction for spiral MRI.

PURPOSE: Time-varying system delays and eddy currents can substantially reduce the image quality of spiral images. A new method is proposed to estimate variable system delays for spiral-based trajectories. METHODS: This approach requires a minor modification of a conventional stack-of-spirals sequence and analyzes data collected on three overlapping orthogonal cylinders. RESULTS: Initial results are presented for acquired and synthesized phantom data and in vivo data. CONCLUSION: The proposed method includes gradient coupling effects, requires no phantom measurements or specialized hardware, is robust to off-resonance effects, and estimates independent continuous delays for each gradient channel over the data-sampling period.

Multidirectional high-moment encoding in phase contrast MRI.

The use of phase contrast MRI to measure vascular flow provides a unique method for acquiring quantitative estimates of flow as well as morphological imaging. The quantitative aspects of phase contrast magnetic resonance angiography (PC-MRA) provide unique relationships between measurement parameters and resulting signal to noise ratio of the velocity measurements. This article introduces a new method to exploit these relationships providing increased efficiency, and therefore, higher vessel conspicuity. Signal to noise ratio gains in high-moment PC-MRA are limited by the ability to unalias phase measurements that fall outside the -π to π interval. Unaliasing phase on a per pixel basis is limited by errors in the measurements due to noise and intravoxel flow distributions. Current dual-VENC methods have been shown to be robust to these errors and provide high velocity to noise ratio gains, however, the collection of a required high-VENC set can be inefficient. The presented method provides more time efficient gains in velocity to noise ratio compared to a dual-VENC approach by eliminating the high-VENC acquisitions and using shared information between nonorthogonal measurements. Simulations, phantom, and in vivo angiography are used to characterize the noise performance of each method. The velocity to noise ratio efficiency of the proposed method is shown to be ∼1.7 times greater than the dual-VENC method at the same gradient moment.

Efficient sample density estimation by combining gridding and an optimized kernel.

The reconstruction of non-Cartesian k-space trajectories often requires the estimation of nonuniform sampling density. Particularly for 3D, this calculation can be computationally expensive. The method proposed in this work combines an iterative algorithm previously proposed by Pipe and Menon (Magn Reson Med 1999;41:179-186) with the optimal kernel design previously proposed by Johnson and Pipe (Magn Reson Med 2009;61:439-447). The proposed method shows substantial time reductions in estimating the densities of center-out trajectories, when compared with that of Johnson. It is demonstrated that, depending on the trajectory, the proposed method can provide reductions in execution time by factors of 12 to 85. The method is also shown to be robust in areas of high trajectory overlap, when compared with two analytical density estimation methods, producing a 10-fold increase in accuracy in one case. Initial conditions allow the proposed method to converge in fewer iterations and are shown to be flexible in terms of the accuracy of information supplied. The proposed method is not only one of the fastest and most accurate algorithms, it is also completely generic, allowing any arbitrary trajectory to be density compensated extemporaneously. The proposed method is also simple and can be implemented on parallel computing platforms in a straightforward manner.

A new design and rationale for 3D orthogonally oversampled k-space trajectories.

A novel center-out 3D trajectory for sampling magnetic resonance data is presented. The trajectory set is based on a single Fermat spiral waveform, which is substantially undersampled in the center of k-space. Multiple trajectories are combined in a "stacked cone" configuration to give very uniform sampling throughout a "hub," which is very efficient in terms of gradient performance and uniform trajectory spacing. The fermat looped, orthogonally encoded trajectories (FLORET) design produces less gradient-efficient trajectories near the poles, so multiple orthogonal hub designs are shown. These multihub designs oversample k-space twice with orthogonal trajectories, which gives unique properties but also doubles the minimum scan time for critical sampling of k-space. The trajectory is shown to be much more efficient than the conventional stack of cones trajectory, and has nearly the same signal-to-noise ratio efficiency (but twice the minimum scan time) as a stack of spirals trajectory. As a center-out trajectory, it provides a shorter minimum echo time than stack of spirals, and its spherical k-space coverage can dramatically reduce Gibbs ringing.

Motion corrected intracranial MRA using PROPELLER with RF quadratic encoding.

A new motion corrected Time-of-Flight MRA technique named Variable Pitch PROPELLER is presented. This technique employs the PROPELLER acquisition and reconstruction scheme for in-plane bulk motion correction. A non- Fourier through-plane encoding mechanism called quadratic encoding boosts SNR, relative to conventional 2D MRA, in lieu of traditional 3D encoding. Partial Fourier encoding is applied in the slice direction for a further reduction in scan time. This work details the construction and optimization of this technique. VPPROP MRAs are compared with a clinical MOTSA protocol. Initial results show promising robustness to bulk motion effects. The comparisons with MOTSA provide insight as to the additions required to create a comparable scan.