Structured Low-Rank Algorithms: Theory, Magnetic Resonance Applications, and Links to Machine Learning

Dynamic field mapping and distortion correction using single-shot blip-rewound EPI and joint multi-echo reconstruction

PURPOSE

To develop a method for dynamic∆ B 0 $$ \Delta {B}_0 $$ mapping and distortion correction.

METHODS

A blip-rewound EPI trajectory was developed to acquire multiple 2D EPI images in a single readout with an interleaved order, which allows a short TE difference. A joint multi-echo reconstruction was utilized to exploit the shared information between EPI images. The reconstructed images from each readout are combined to produce a final magnitude image. A∆ B 0 $$ \Delta {B}_0 $$ map is calculated from the phase of these images for distortion correction. The efficacy of the proposed method is assessed with phantom and in vivo experiments. The performance of the proposed method in the presence of subject motion is also investigated.

RESULTS

Compared to conventional multi-echo EPI, the proposed method allows dynamic∆ B 0 $$ \Delta {B}_0 $$ mapping at matched resolution with a much shorter TR. Phantom and in vivo results show that the proposed method can provide a comparable magnitude image as conventional single-shot EPI. The∆ B 0 $$ \Delta {B}_0 $$ maps calculated from the proposed method are consistent with conventional multi-echo EPI in the phantom experiment. For in vivo experiments, the proposed method provides a more accurate estimation of∆ B 0 $$ \Delta {B}_0 $$ than conventional multi-echo EPI, which is prone to phase wrapping problems due to the long TE difference. In-vivo scan with subject motion shows the proposed dynamic field mapping method can improve the temporal stability of EPI time series compared to gradient echo (GRE) based static field mapping.

CONCLUSION

The proposed method allows accurate dynamic∆ B 0 $$ \Delta {B}_0 $$ mapping for robust distortion correction without compromising spatial or temporal resolution.

High-Dimensional MR Reconstruction Integrating Subspace and Adaptive Generative Models

OBJECTIVE

To develop a new method that integrates subspace and generative image models for high-dimensional MR image reconstruction.

METHODS

We proposed a formulation that synergizes a low-dimensional subspace model of high-dimensional images, an adaptive generative image prior serving as spatial constraints on the sequence of "contrast-weighted" images or spatial coefficients of the subspace model, and a conventional sparsity regularization. A special pretraining plus subject-specific network adaptation strategy was proposed to construct an accurate generative-network-based representation for images with varying contrasts. An iterative algorithm was introduced to jointly update the subspace coefficients and the multi-resolution latent space of the generative image model that leveraged an recently proposed intermediate layer optimization technique for network inversion.

RESULTS

We evaluated the utility of the proposed method for two high-dimensional imaging applications: accelerated MR parameter mapping and high-resolution MR spectroscopic imaging. Improved performance over state-of-the-art subspace-based methods was demonstrated in both cases.

CONCLUSION

The proposed method provided a new way to address high-dimensional MR image reconstruction problems by incorporating an adaptive generative model as a data-driven spatial prior for constraining subspace reconstruction.

SIGNIFICANCE

Our work demonstrated the potential of integrating data-driven and adaptive generative priors with canonical low-dimensional modeling for high-dimensional imaging problems.

Self-calibrated subspace reconstruction for multidimensional MR fingerprinting for simultaneous relaxation and diffusion quantification

PURPOSE

To propose a new reconstruction method for multidimensional MR fingerprinting (mdMRF) to address shading artifacts caused by physiological motion-induced measurement errors without navigating or gating.

METHODS

The proposed method comprises two procedures: self-calibration and subspace reconstruction. The first procedure (self-calibration) applies temporally local matrix completion to reconstruct low-resolution images from a subset of under-sampled data extracted from the k-space center. The second procedure (subspace reconstruction) utilizes temporally global subspace reconstruction with pre-estimated temporal subspace from low-resolution images to reconstruct aliasing-free, high-resolution, and time-resolved images. After reconstruction, a customized outlier detection algorithm was employed to automatically detect and remove images corrupted by measurement errors. Feasibility, robustness, and scan efficiency were evaluated through in vivo human brain imaging experiments.

RESULTS

The proposed method successfully reconstructed aliasing-free, high-resolution, and time-resolved images, where the measurement errors were accurately represented. The corrupted images were automatically and robustly detected and removed. Artifact-free T1, T2, and ADC maps were generated simultaneously. The proposed reconstruction method demonstrated robustness across different scanners, parameter settings, and subjects. A high scan efficiency of less than 20 s per slice has been achieved.

CONCLUSION

The proposed reconstruction method can effectively alleviate shading artifacts caused by physiological motion-induced measurement errors. It enables simultaneous and artifact-free quantification of T1, T2, and ADC using mdMRF scans without prospective gating, with robustness and high scan efficiency.

Annihilation-Net: Learned annihilation relation for dynamic MR imaging

BACKGROUND

Deep learning methods driven by the low-rank regularization have achieved attractive performance in dynamic magnetic resonance (MR) imaging. The effectiveness of existing methods lies mainly in their ability to capture interframe relationships using network modules, which are lack interpretability.

PURPOSE

This study aims to design an interpretable methodology for modeling interframe relationships using convolutiona networks, namely Annihilation-Net and use it for accelerating dynamic MRI.

METHODS

Based on the equivalence between Hankel matrix product and convolution, we utilize convolutional networks to learn the null space transform for characterizing low-rankness. We employ low-rankness to represent interframe correlations in dynamic MR imaging, while combining with sparse constraints in the compressed sensing framework. The corresponding optimization problem is solved in an iterative form with the semi-quadratic splitting method (HQS). The iterative steps are unrolled into a network, dubbed Annihilation-Net. All the regularization parameters and null space transforms are set as learnable in the Annihilation-Net.

RESULTS

Experiments on the cardiac cine dataset show that the proposed model outperforms other competing methods both quantitatively and qualitatively. The training set and test set have 800 and 118 images, respectively.

CONCLUSIONS

The proposed Annihilation-Net improves the reconstruction quality of accelerated dynamic MRI with better interpretability.

New Theory and Faster Computations for Subspace-Based Sensitivity Map Estimation in Multichannel MRI

Sensitivity map estimation is important in many multichannel MRI applications. Subspace-based sensitivity map estimation methods like ESPIRiT are popular and perform well, though can be computationally expensive and their theoretical principles can be nontrivial to understand. In the first part of this work, we present a novel theoretical derivation of subspace-based sensitivity map estimation based on a linear-predictability/structured low-rank modeling perspective. This results in an estimation approach that is equivalent to ESPIRiT, but with distinct theory that may be more intuitive for some readers. In the second part of this work, we propose and evaluate a set of computational acceleration approaches (collectively known as PISCO) that can enable substantial improvements in computation time (up to  ∼ 100× in the examples we show) and memory for subspace-based sensitivity map estimation.

Locally low-rank denoising in transform domains

Purpose

To develop an extension to locally low rank (LLR) denoising techniques based on transform domain processing that reduces the number of images required in the MR image series for high-quality denoising.

Theory and Methods

LLR methods with random matrix theory-based thresholds are successfully used in the denoising of MR image series in a number of applications. The performance of these methods depend on how well the LLR assumption is satisfied, which deteriorates with few numbers of images, as is commonly encountered in quantitative MRI applications. We propose a transform-domain approach for denoising of MR image series to represent the underlying signal with higher fidelity when using a locally low rank approximation. The efficacy of the method is demonstrated for fully-sampled k-space, undersampled k-space, DICOM images, and complex-valued SENSE-1 images in quantitative MRI applications with as few as 4 images.

Results

For both MSK and brain applications, the transform domain denoising preserves local subtle variability, whereas the quantitative maps based on image domain LLR methods tend to be locally more homogeneous.

Conclusion

A transform domain extension to LLR denoising produces high quality images and is compatible with both raw k-space data and vendor reconstructed data. This allows for improved imaging and more accurate quantitative analyses and parameters obtained therefrom.

Complexities of deep learning-based undersampled MR image reconstruction

Artificial intelligence has opened a new path of innovation in magnetic resonance (MR) image reconstruction of undersampled k-space acquisitions. This review offers readers an analysis of the current deep learning-based MR image reconstruction methods. The literature in this field shows exponential growth, both in volume and complexity, as the capabilities of machine learning in solving inverse problems such as image reconstruction are explored. We review the latest developments, aiming to assist researchers and radiologists who are developing new methods or seeking to provide valuable feedback. We shed light on key concepts by exploring the technical intricacies of MR image reconstruction, highlighting the importance of raw datasets and the difficulty of evaluating diagnostic value using standard metrics.Relevance statement Increasingly complex algorithms output reconstructed images that are difficult to assess for robustness and diagnostic quality, necessitating high-quality datasets and collaboration with radiologists.Key points• Deep learning-based image reconstruction algorithms are increasing both in complexity and performance.• The evaluation of reconstructed images may mistake perceived image quality for diagnostic value.• Collaboration with radiologists is crucial for advancing deep learning technology.

Eddy current-induced artifacts correction in high gradient strength diffusion MRI with dynamic field monitoring: demonstration in ex vivo human brain imaging

Purpose

To demonstrate the advantages of spatiotemporal magnetic field monitoring to correct eddy current-induced artifacts (ghosting and geometric distortions) in high gradient strength diffusion MRI (dMRI).

Methods

A dynamic field camera with 16 NMR field probes was used to characterize eddy current fields induced from diffusion gradients for different gradients strengths (up to 300 mT/m), diffusion directions, and shots in a 3D multi-shot EPI sequence on a 3T Connectom scanner. The efficacy of dynamic field monitoring-based image reconstruction was demonstrated on high-resolution whole brain ex vivo dMRI. A 3D multi-shot image reconstruction framework was informed with the actual nonlinear phase evolution measured with the dynamic field camera, thereby accounting for high-order eddy currents fields on top of the image encoding gradients in the image formation model.

Results

Eddy current fields from diffusion gradients at high gradient strength in a 3T Connectom scanner are highly nonlinear in space and time, inducing high-order spatial phase modulations between odd/even echoes and shots that are not static during the readout. Superior reduction of ghosting and geometric distortion was achieved with dynamic field monitoring compared to ghosting approaches such as navigator- and structured low-rank-based methods or MUSE, followed by image-based distortion correction with eddy. Improved dMRI analysis is demonstrated with diffusion tensor imaging and high-angular resolution diffusion imaging.

Conclusion

Strong eddy current artifacts characteristic of high gradient strength dMRI can be well corrected with dynamic field monitoring-based image reconstruction, unlike the two-step approach consisting of ghosting correction followed by geometric distortion reduction with eddy.

Improving robustness of 3D multi-shot EPI by structured low-rank reconstruction of segmented CAIPI sampling for fMRI at 7T

Three-dimensional (3D) encoding methods are increasingly being explored as alternatives to two-dimensional (2D) multi-slice acquisitions in fMRI, particularly in cases where high isotropic resolution is needed. 3D multi-shot EPI acquisition, as the workhorse of 3D fMRI imaging, is susceptible to physiological fluctuations which can induce inter-shot phase variations, and thus reducing the achievable tSNR, negating some of the benefit of 3D encoding. This issue can be particularly problematic at ultra-high fields like 7T, which have more severe off-resonance effects. In this work, we aim to improve the temporal stability of 3D multi-shot EPI at 7T by improving its robustness to inter-shot phase variations. We presented a 3D segmented CAIPI sampling trajectory ("seg-CAIPI") and an improved reconstruction method based on Hankel structured low-rank matrix recovery. Simulation and in-vivo results demonstrate that the combination of the seg-CAIPI sampling scheme and the proposed structured low-rank reconstruction is a promising way to effectively reduce the unwanted temporal variance induced by inter-shot physiological fluctuations, and thus improve the robustness of 3D multi-shot EPI for fMRI.

Multi-band multi-shot diffusion MRI reconstruction with joint usage of structured low-rank constraints and explicit phase mapping

PURPOSE

To develop a joint reconstruction method for multi-band multi-shot diffusion MRI.

THEORY AND METHODS

Multi-band multi-shot EPI acquisition is an effective approach for high-resolution diffusion MRI, but requires specific algorithms to correct the inter-shot phase variations. The phase correction can be done by first estimating the explicit phase map and then feeding it into the k-space signal formulation model. Alternatively, the phase information can be used indirectly as structured low-rank constraints in k-space. The 2 methods differ in reconstruction accuracy and efficiency. We aim to combine the 2 different approaches for improved image quality and reconstruction efficiency simultaneously, termed "joint usage of structured low-rank constraints and explicit phase mapping" (JULEP). The proposed JULEP reconstruction is tested on both single-band and multi-band, multi-shot diffusion data, with different resolutions and b values. The results of JULEP are compared with conventional methods with explicit phase mapping (i.e., multiplexed sensitivity-encoding [MUSE]) and structured low-rank constraints (i.e., MUSSELS), and another joint reconstruction method (i.e., network estimated artifacts for tempered reconstruction [NEATR]).

RESULTS

JULEP improves the quality of the navigator and subsequently facilitates the reconstruction of final diffusion images. Compared with all 3 other methods (MUSE, MUSSELS, and NEATR), JULEP mitigates residual structural bias and improves temporal SNRs in the final diffusion image, particularly at high multi-band factors. Compared with MUSSELS, JULEP also improves computational efficiency.

CONCLUSION

The proposed JULEP method significantly improves the image quality and reconstruction efficiency of multi-band multi-shot diffusion MRI, which can promote a broader application of high-resolution diffusion MRI.

One-Dimensional Deep Low-Rank and Sparse Network for Accelerated MRI

Deep learning has shown astonishing performance in accelerated magnetic resonance imaging (MRI). Most state-of-the-art deep learning reconstructions adopt the powerful convolutional neural network and perform 2D convolution since many magnetic resonance images or their corresponding k-space are in 2D. In this work, we present a new approach that explores the 1D convolution, making the deep network much easier to be trained and generalized. We further integrate the 1D convolution into the proposed deep network, named as One-dimensional Deep Low-rank and Sparse network (ODLS), which unrolls the iteration procedure of a low-rank and sparse reconstruction model. Extensive results on in vivo knee and brain datasets demonstrate that, the proposed ODLS is very suitable for the case of limited training subjects and provides improved reconstruction performance than state-of-the-art methods both visually and quantitatively. Additionally, ODLS also shows nice robustness to different undersampling scenarios and some mismatches between the training and test data. In summary, our work demonstrates that the 1D deep learning scheme is memory-efficient and robust in fast MRI.

B-Spline Parameterized Joint Optimization of Reconstruction and K-Space Trajectories (BJORK) for Accelerated 2D MRI

Optimizing k-space sampling trajectories is a promising yet challenging topic for fast magnetic resonance imaging (MRI). This work proposes to optimize a reconstruction method and sampling trajectories jointly concerning image reconstruction quality in a supervised learning manner. We parameterize trajectories with quadratic B-spline kernels to reduce the number of parameters and apply multi-scale optimization, which may help to avoid sub-optimal local minima. The algorithm includes an efficient non-Cartesian unrolled neural network-based reconstruction and an accurate approximation for backpropagation through the non-uniform fast Fourier transform (NUFFT) operator to accurately reconstruct and back-propagate multi-coil non-Cartesian data. Penalties on slew rate and gradient amplitude enforce hardware constraints. Sampling and reconstruction are trained jointly using large public datasets. To correct for possible eddy-current effects introduced by the curved trajectory, we use a pencil-beam trajectory mapping technique. In both simulations and in- vivo experiments, the learned trajectory demonstrates significantly improved image quality compared to previous model-based and learning-based trajectory optimization methods for 10× acceleration factors. Though trained with neural network-based reconstruction, the proposed trajectory also leads to improved image quality with compressed sensing-based reconstruction.

Accelerated MRI Reconstruction With Separable and Enhanced Low-Rank Hankel Regularization

Magnetic resonance imaging serves as an essential tool for clinical diagnosis, however, suffers from a long acquisition time. Sparse sampling effectively saves this time but images need to be faithfully reconstructed from undersampled data. Among the existing reconstruction methods, the structured low-rank methods have advantages in robustness to the sampling patterns and lower error. However, the structured low-rank methods use the 2D or higher dimension k-space data to build a huge block Hankel matrix, leading to considerable time and memory consumption. To reduce the size of the Hankel matrix, we proposed to separably construct multiple small Hankel matrices from rows and columns of the k-space and then constrain the low-rankness on these small matrices. This separable model can significantly reduce the computational time but ignores the correlation existed in inter- and intra-row or column, resulting in increased reconstruction error. To improve the reconstructed image without obviously increasing the computation, we further introduced the self-consistency of k-space and virtual coil prior. Besides, the proposed separable model can be extended into other imaging scenarios which hold exponential characteristics in the parameter dimension. The in vivo experimental results demonstrated that the proposed method permits the lowest reconstruction error with a fast reconstruction. The proposed approach requires only 4% of the state-of-the-art STDLR-SPIRiT runtime for parallel imaging reconstruction, and achieves the fastest computational speed in parameter imaging reconstruction.

Calibrationless multi-slice Cartesian MRI via orthogonally alternating phase encoding direction and joint low-rank tensor completion

We propose a multi-slice acquisition with orthogonally alternating phase encoding (PE) direction and subsequent joint calibrationless reconstruction for accelerated multiple individual 2D slices or multi-slice 2D Cartesian MRI. Specifically, multi-slice multi-channel data are first acquired with random or uniform PE undersampling while orthogonally alternating PE direction between adjacent slices. They are then jointly reconstructed through a recently developed low-rank multi-slice Hankel tensor completion (MS-HTC) approach. The proposed acquisition and reconstruction strategy was evaluated with human brain MR data. It effectively suppressed aliasing artifacts even at high acceleration factor, outperforming the existing MS-HTC approach, where PE direction is the same between adjacent slices. More importantly, the new strategy worked robustly with uniform undersampling or random undersampling without any consecutive central k-space lines. In summary, our proposed multi-slice MRI strategy exploits both coil sensitivity and image content similarities across adjacent slices. Orthogonally alternating PE direction among slices substantially facilitates the low-rank completion process and improves image reconstruction quality. This new strategy is applicable to uniform and random PE undersampling. It can be easily implemented in practice for Cartesian parallel imaging of multiple individual 2D slices without any coil sensitivity calibration.

High-Quality MR Fingerprinting Reconstruction Using Structured Low-Rank Matrix Completion and Subspace Projection

Due to the capability of fast multiparametric quantitative imaging, magnetic resonance fingerprinting (MRF) is becoming a promising quantitative magnetic resonance imaging approach. However, the artifacts caused by the highly undersampled data acquisition lead to inaccurate estimation of the tissue parameter maps. Based on the assumption that the 3-D MRF data can be modeled as a piecewise smooth signal, with the discontinuities localized to the zero sets of a bandlimited function, we exploit the low-rank property of the structured Toeplitz matrix constructed from the Fourier measurements. In addition, we adopt the subspace projection scheme to improve the accuracy of parameter estimation. In order to efficiently solve the regularized problem, we propose an iterative two-stage algorithm, which alternately updates the k -space data and projects the space-time matrix into the dictionary space. Numerical experiments demonstrate that the proposed algorithm shows significant improvement in MRF time-series images reconstruction and can provide more accurate parameter maps over the state-of-the-art algorithms.

Reducing the Effects of Motion Artifacts in fMRI: A Structured Matrix Completion Approach

Functional MRI (fMRI) is widely used to study the functional organization of normal and pathological brains. However, the fMRI signal may be contaminated by subject motion artifacts that are only partially mitigated by motion correction strategies. These artifacts lead to distance-dependent biases in the inferred signal correlations. To mitigate these spurious effects, motion-corrupted volumes are censored from fMRI time series. Censoring can result in discontinuities in the fMRI signal, which may lead to substantial alterations in functional connectivity analysis. We propose a new approach to recover the missing entries from censoring based on structured low rank matrix completion. We formulated the artifact-reduction problem as the recovery of a super-resolved matrix from unprocessed fMRI measurements. We enforced a low rank prior on a large structured matrix, formed from the samples of the time series, to recover the missing entries. The recovered time series, in addition to being motion compensated, are also slice-time corrected at a fine temporal resolution. To achieve a fast and memory-efficient solution for our proposed optimization problem, we employed a variable splitting strategy. We validated the algorithm with simulations, data acquired under different motion conditions, and datasets from the ABCD study. Functional connectivity analysis showed that the proposed reconstruction resulted in connectivity matrices with lower errors in pair-wise correlation than non-censored and censored time series based on a standard processing pipeline. In addition, seed-based correlation analyses showed improved delineation of the default mode network. These demonstrate that the method can effectively reduce the adverse effects of motion in fMRI analysis.

Blind Primed Supervised (BLIPS) Learning for MR Image Reconstruction

This paper examines a combined supervised-unsupervised framework involving dictionary-based blind learning and deep supervised learning for MR image reconstruction from under-sampled k-space data. A major focus of the work is to investigate the possible synergy of learned features in traditional shallow reconstruction using adaptive sparsity-based priors and deep prior-based reconstruction. Specifically, we propose a framework that uses an unrolled network to refine a blind dictionary learning-based reconstruction. We compare the proposed method with strictly supervised deep learning-based reconstruction approaches on several datasets of varying sizes and anatomies. We also compare the proposed method to alternative approaches for combining dictionary-based methods with supervised learning in MR image reconstruction. The improvements yielded by the proposed framework suggest that the blind dictionary-based approach preserves fine image details that the supervised approach can iteratively refine, suggesting that the features learned using the two methods are complementary.

Fast data-driven learning of parallel MRI sampling patterns for large scale problems

In this study, a fast data-driven optimization approach, named bias-accelerated subset selection (BASS), is proposed for learning efficacious sampling patterns (SPs) with the purpose of reducing scan time in large-dimensional parallel MRI. BASS is applicable when Cartesian fully-sampled k-space measurements of specific anatomy are available for training and the reconstruction method for undersampled measurements is specified; such information is used to define the efficacy of any SP for recovering the values at the non-sampled k-space points. BASS produces a sequence of SPs with the aim of finding one of a specified size with (near) optimal efficacy. BASS was tested with five reconstruction methods for parallel MRI based on low-rankness and sparsity that allow a free choice of the SP. Three datasets were used for testing, two of high-resolution brain images ([Formula: see text]-weighted images and, respectively, [Formula: see text]-weighted images) and another of knee images for quantitative mapping of the cartilage. The proposed approach has low computational cost and fast convergence; in the tested cases it obtained SPs up to 50 times faster than the currently best greedy approach. Reconstruction quality increased by up to 45% over that provided by variable density and Poisson disk SPs, for the same scan time. Optionally, the scan time can be nearly halved without loss of reconstruction quality. Quantitative MRI and prospective accelerated MRI results show improvements. Compared with greedy approaches, BASS rapidly learns effective SPs for various reconstruction methods, using larger SPs and larger datasets; enabling better selection of sampling-reconstruction pairs for specific MRI problems.

Autoregression and Structured Low-Rank Modeling of Sinogram Neighborhoods

Sinograms are commonly used to represent the raw data from tomographic imaging experiments. Although it is already well-known that sinograms posess some amount of redundancy, in this work, we present novel theory suggesting that sinograms will often possess substantial additional redundancies that have not been explicitly exploited by previous methods. Specifically, we derive that sinograms will often satisfy multiple simple data-dependent autoregression relationships. This kind of autoregressive structure enables missing/degraded sinogram samples to be linearly predicted using a simple shift-invariant linear combination of neighboring samples. Our theory also further implies that if sinogram samples are assembled into a structured Hankel/Toeplitz matrix, then the matrix will be expected to have low-rank characteristics. As a result, sinogram restoration problems can be formulated as structured low-rank matrix recovery problems. Illustrations of this approach are provided using several different (real and simulated) X-ray imaging datasets, including comparisons against a state-of-the-art deep learning approach. Results suggest that structured low-rank matrix methods for sinogram recovery can have comparable performance to state-of-the-art approaches. Although our evaluation focuses on competitive comparisons against other approaches, we believe that autoregressive constraints are actually complementary to existing approaches with strong potential synergies.

High-dimensional fast convolutional framework (HICU) for calibrationless MRI

PURPOSE

To present a computational procedure for accelerated, calibrationless magnetic resonance image (Cl-MRI) reconstruction that is fast, memory efficient, and scales to high-dimensional imaging.

THEORY AND METHODS

Cl-MRI methods can enable high acceleration rates and flexible sampling patterns, but their clinical application is limited by computational complexity and large memory footprint. The proposed computational procedure, HIgh-dimensional fast convolutional framework (HICU), provides fast, memory-efficient recovery of unsampled k-space points. For demonstration, HICU is applied to 6 2D T2-weighted brain, 7 2D cardiac cine, 5 3D knee, and 1 multi-shot diffusion weighted imaging (MSDWI) datasets.

RESULTS

The 2D imaging results show that HICU can offer 1-2 orders of magnitude computation speedup compared to other Cl-MRI methods without sacrificing imaging quality. The 2D cine and 3D imaging results show that the computational acceleration techniques included in HICU yield computing time on par with SENSE-based compressed sensing methods with up to 3 dB improvement in signal-to-error ratio and better perceptual quality. The MSDWI results demonstrate the feasibility of HICU for a challenging multi-shot echo-planar imaging application.

CONCLUSIONS

The presented method, HICU, offers efficient computation and scalability as well as extendibility to a wide variety of MRI applications.

Deep Generalization of Structured Low-Rank Algorithms (Deep-SLR)

Structured low-rank (SLR) algorithms, which exploit annihilation relations between the Fourier samples of a signal resulting from different properties, is a powerful image reconstruction framework in several applications. This scheme relies on low-rank matrix completion to estimate the annihilation relations from the measurements. The main challenge with this strategy is the high computational complexity of matrix completion. We introduce a deep learning (DL) approach to significantly reduce the computational complexity. Specifically, we use a convolutional neural network (CNN)-based filterbank that is trained to estimate the annihilation relations from imperfect (under-sampled and noisy) k-space measurements of Magnetic Resonance Imaging (MRI). The main reason for the computational efficiency is the pre-learning of the parameters of the non-linear CNN from exemplar data, compared to SLR schemes that learn the linear filterbank parameters from the dataset itself. Experimental comparisons show that the proposed scheme can enable calibration-less parallel MRI; it can offer performance similar to SLR schemes while reducing the runtime by around three orders of magnitude. Unlike pre-calibrated and self-calibrated approaches, the proposed uncalibrated approach is insensitive to motion errors and affords higher acceleration. The proposed scheme also incorporates image domain priors that are complementary, thus significantly improving the performance over that of SLR schemes.

J-MoDL: Joint Model-Based Deep Learning for Optimized Sampling and Reconstruction

Modern MRI schemes, which rely on compressed sensing or deep learning algorithms to recover MRI data from undersampled multichannel Fourier measurements, are widely used to reduce the scan time. The image quality of these approaches is heavily dependent on the sampling pattern. We introduce a continuous strategy to optimize the sampling pattern and the network parameters jointly. We use a multichannel forward model, consisting of a non-uniform Fourier transform with continuously defined sampling locations, to realize the data consistency block within a model-based deep learning image reconstruction scheme. This approach facilitates the joint and continuous optimization of the sampling pattern and the CNN parameters to improve image quality. We observe that the joint optimization of the sampling patterns and the reconstruction module significantly improves the performance of most deep learning reconstruction algorithms. The source code is available at https://github.com/hkaggarwal/J-MoDL.

Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues

Following the success of deep learning in a wide range of applications, neural network-based machine learning techniques have received interest as a means of accelerating magnetic resonance imaging (MRI). A number of ideas inspired by deep learning techniques from computer vision and image processing have been successfully applied to non-linear image reconstruction in the spirit of compressed sensing for both low dose computed tomography and accelerated MRI. The additional integration of multi-coil information to recover missing k-space lines in the MRI reconstruction process, is still studied less frequently, even though it is the de-facto standard for currently used accelerated MR acquisitions. This manuscript provides an overview of the recent machine learning approaches that have been proposed specifically for improving parallel imaging. A general background introduction to parallel MRI is given that is structured around the classical view of image space and k-space based methods. Both linear and non-linear methods are covered, followed by a discussion of recent efforts to further improve parallel imaging using machine learning, and specifically using artificial neural networks. Image-domain based techniques that introduce improved regularizers are covered as well as k-space based methods, where the focus is on better interpolation strategies using neural networks. Issues and open problems are discussed as well as recent efforts for producing open datasets and benchmarks for the community.