Undersampling Pattern Research Articles

Fast Multi-Contrast MRI Acquisition by Optimal Sampling of Information Complementary to Pre-acquired MRI Contrast.

Recent studies on multi-contrast MRI reconstruction have demonstrated the potential of further accelerating MRI acquisition by exploiting correlation between contrasts. Most of the state-of-the-art approaches have achieved improvement through the development of network architectures for fixed under-sampling patterns, without considering inter-contrast correlation in the under-sampling pattern design. On the other hand, sampling pattern learning methods have shown better reconstruction performance than those with fixed under-sampling patterns. However, most under-sampling pattern learning algorithms are designed for single contrast MRI without exploiting complementary information between contrasts. To this end, we propose a framework to optimize the under-sampling pattern of a target MRI contrast which complements the acquired fully-sampled reference contrast. Specifically, a novel image synthesis network is introduced to extract the redundant information contained in the reference contrast, which is exploited in the subsequent joint pattern optimization and reconstruction network. We have demonstrated superior performance of our learned under-sampling patterns on both public and in-house datasets, compared to the commonly used under-sampling patterns and state-of-the-art methods that jointly optimize the reconstruction network and the under-sampling patterns, up to 8-fold under-sampling factor.

An analysis of reconstruction noise from undersampled 4D flow MRI

Novel Magnetic Resonance (MR) imaging modalities can quantify hemodynamics but require long acquisition times, precluding its widespread use for early diagnosis of cardiovascular disease. To reduce acquisition times, flow reconstruction from undersampled data is routinely performed.Reconstructed anatomical and hemodynamic images may present visual artifacts. While some artifacts are reconstruction errors, and a consequence of undersampling, others are due to measurement noise or the random choice of samples. A reconstructed image becomes thus a random variable: its bias leads to systematic reconstruction errors, whereas its fluctuations may induce spatial correlations that may be misconstrued for visual information or that may carry to quantities of interest computed from the image. Although the former has been studied in the literature, the latter has not received as much attention.In this study, we investigate the theoretical properties of the random perturbations arising from the reconstruction process. To our knowledge, this is the first study on this topic. We perform numerical experiments on simulated flow, on aortic phantom flow, and on aortic flow. These show that the correlation length remains limited to two to three pixels when a Gaussian undersampling pattern is combined with ℓ1-norm minimization methods. The correlation length may increase significantly for other undersampling patterns, higher undersampling factors (i.e., higher than 8× compression), and other reconstruction methods. Our findings suggest that the reconstruction method has a large impact on the correlation. As reconstruction methods are routinely used in practice, the impact of these random perturbations in practical applications merits further study.

Deep Learning MRI Reconstruction for Accelerating Turbo Spin Echo Hand and Wrist Imaging: A Comparison of Image Quality, Visualization of Anatomy, and Detection of Common Pathologies with Standard Imaging

Magnetic resonance imaging (MRI) of the hand and wrist is a routine MRI examination and takes about 15-20 minutes, which can lead to problems resulting from the relatively long scan time, such as decreased image quality due to motion artifacts and lower patient throughput. The objective of this study was to evaluate a deep learning (DL) reconstruction for turbo spin echo (TSE) sequences of the hand and wrist regarding image quality, visualization of anatomy, and diagnostic performance concerning common pathologies. Twenty-one patients (mean age: 43 ± 19 [19-85] years, 10 men, 11 female) were prospectively enrolled in this study between October 2020 and June 2021. Each participant underwent two MRI protocols: first, standard fully sampled TSE sequences reconstructed with a standard GRAPPA reconstruction (TSES) and second, prospectively undersampled TSE sequences using a conventional parallel imaging undersampling pattern reconstructed with a DL reconstruction (TSEDL). Both protocols were acquired consecutively in one examination. Two experienced MSK-imaging radiologists qualitatively evaluated the images concerning image quality, noise, edge sharpness, artifacts, and diagnostic confidence, as well as the delineation of anatomical structures (triangular fibrocartilage complex, tendon of the extensor carpi ulnaris muscle, extrinsic and intrinsic ligaments, median nerve, cartilage) using a five-point Likert scale and assessed common pathologies. Wilcoxon signed-rank test and kappa statistics were performed to compare the sequences. Overall image quality, artifacts, delineation of anatomical structures, and diagnostic confidence of TSEDL were rated to be comparable to TSES (p > 0.05). Additionally, TSEDL showed decreased image noise (4.90, median 5, IQR 5-5) compared to TSES (4.52, median 5, IQR 4-5, p < 0.05) and improved edge sharpness (TSEDL: 4.10, median 4, IQR 3.5-5; TSES: 3.57, median 4, IQR 3-4; p < 0.05). Inter- and intrareader agreement was substantial to almost perfect (κ=0.632-1.000) for the detection of common pathologies. Time of acquisition could be reduced by more than 60% with the protocol using TSEDL. Compared to TSES, TSEDL provided decreased noise and increased edge sharpness, equal image quality, delineation of anatomical structures, detection of pathologies, and diagnostic confidence. Therefore, TSEDL may be clinically relevant for hand and wrist imaging, as it reduces examination time by more than 60%, thus increasing patient comfort and patient throughput.

A Genetic Algorithm Optimized Undersampling Method for Seismic Sparse Acquisition and Reconstruction

The irregular observation region poses challenges to seismic acquisition systems design. The commonly-used parallel-type acquisition system requires that the geophones are located at equally spaced positions and therefore is hard to implement in an irregular observation region. An acquisition system which allows implementation of sparse and irregular observation (e.g., the node-type geometry) followed by a reconstruction procedure is a solution. It can not only fit in irregular observation regions but also make a significant reduction in seismic acquisition costs. The seismic sparse acquisition can be mathematically modeled as a undersampling operator in the seismic reconstruction problem. A suboptimal undersampling pattern will lead to an inferior reconstruction result. To optimize the seismic sparse acquisition, I propose a undersampling method based on bionic intelligence in this study. In the proposed method, a Shannon entropy maximum model is proposed to improve the observed ergodicity and reduce the undersampling artifacts. To solve the maximum problem, an improved version of genetic algorithm is presented. The proposed method is applicable to irregular observation regions and can optimize the subsequent reconstruction performance. I provide the detailed algorithm framework and discuss the undersampling artifacts of different undersampling methods. The application to synthetic and field seismic data validates the effectiveness of the proposed method.

De-Aliasing and Accelerated Sparse Magnetic Resonance Image Reconstruction Using Fully Dense CNN with Attention Gates.

When sparsely sampled data are used to accelerate magnetic resonance imaging (MRI), conventional reconstruction approaches produce significant artifacts that obscure the content of the image. To remove aliasing artifacts, we propose an advanced convolutional neural network (CNN) called fully dense attention CNN (FDA-CNN). We updated the Unet model with the fully dense connectivity and attention mechanism for MRI reconstruction. The main benefit of FDA-CNN is that an attention gate in each decoder layer increases the learning process by focusing on the relevant image features and provides a better generalization of the network by reducing irrelevant activations. Moreover, densely interconnected convolutional layers reuse the feature maps and prevent the vanishing gradient problem. Additionally, we also implement a new, proficient under-sampling pattern in the phase direction that takes low and high frequencies from the k-space both randomly and non-randomly. The performance of FDA-CNN was evaluated quantitatively and qualitatively with three different sub-sampling masks and datasets. Compared with five current deep learning-based and two compressed sensing MRI reconstruction techniques, the proposed method performed better as it reconstructed smoother and brighter images. Furthermore, FDA-CNN improved the mean PSNR by 2 dB, SSIM by 0.35, and VIFP by 0.37 compared with Unet for the acceleration factor of 5.

Mitigating undersampling errors in MR fingerprinting bysequence optimization.

To develop a method for MR Fingerprinting (MRF) sequence optimization that takes both the applied undersampling pattern and a realistic reference map into account. A predictive model for the undersampling error leveraging on perturbation theory was exploited to optimize the MRF flip angle sequence for improved robustness against undersampling artifacts. In this framework parameter maps from a previously acquired MRF scan were used as reference. Sequences were optimized for different sequence lengths, smoothness constraints and undersampling factors. Numerical simulations and in vivo measurements in eight healthy subjects were performed to assess the effect of the performed optimization. The optimized MRF sequences were compared to a conventionally shaped flip angle pattern and an optimized pattern based on the Cramér-Rao lower bound (CRB). Numerical simulations and in vivo results demonstrate that the undersampling errors can be suppressed by flip angle optimization. Analysis of the in vivo results show that a sequence optimized for improved robustness against undersampling with a flip angle train of length 400 yielded significantly lower median absolute errors in : and : compared to the conventional ( : , : ) and CRB-based ( : , : ) sequences. The proposed method is able to optimize the MRF flip angle pattern such that significant mitigation of the artifacts from strong k-space undersampling in MRF is achieved.

Assessment of pulmonary morphometry using hyperpolarized 129 Xe diffusion-weighted MRI with variable-sampling-ratio compressed sensing patterns.

Hyperpolarized (HP) 129 Xe multiple b-values diffusion-weighted magnetic resonance imaging (DW-MRI) has been widely used for quantifying pulmonary microstructural morphometry. However, the technique requires long acquisition times, making it hard to apply in patients with severe pulmonary diseases, who cannot sustain long breath holds. To develop and evaluate the technique of variable-sampling-ratio compressed sensing (VCS) patterns for accelerating HP 129 Xe multiple b-values DW-MRI in humans. Optimal variable sampling ratios and corresponding k-space undersampling patterns for each b-value were obtained by retrospective simulations based on the fully sampled (FS) DW-MRI dataset acquired from six young healthy volunteers. Then, the FS datasets were retrospectively undersampled using both VCS patterns and conventional compressed sensing (CS) pattern with a similar average acceleration factor. The quality of reconstructed images with retrospective VCS (rVCS) and CS (rCS) datasets were quantified using mean absolute error (MAE) and structural similarity (SSIM). Pulmonary morphometric parameters were also evaluated between rVCS and FS datasets. In addition, prospective VCS multiple b-values 129 Xe DW-MRI datasets were acquired from 14 cigarette smokers and 13 age-matched healthy volunteers. The differences of lung morphological parameters obtained with the proposed method were compared between the groups using independent samples t-test. Pearson correlation coefficient was also utilized for evaluating the correlation of the pulmonary physiological parameters obtained with VCS DW-MRI and pulmonary function tests. Lower MAE and higher SSIM values were found in the reconstructed images with rVCS measurement when compared to those using conventional rCS measurement. The details and quality of the images obtained with rVCS and FS measurements were found to be comparable. The mean values of the morphological parameters derived from rVCS and FS datasets showed no significant differences (p>0.05), and the mean differences of measured acinar duct radius, mean linear intercept, surface-to-volume ratio, and apparent diffusion coefficient with cylinder model were -0.87%, -2.42%, 2.04%, and -0.50%, respectively. By using the VCS technique, significant differences were delineated between the pulmonary morphometric parameters of healthy volunteers and cigarette smokers (p<0.001), while the acquisition time was reduced by four times. A fourfold reduction in acquisition time was achieved using the proposed VCS method while preserving good image quality. Our preliminary results demonstrated that the proposed method can be used for evaluating pulmonary injuries caused by cigarette smoking and may prove to be helpful in diagnosing lung diseases in clinical practice.

Simultaneously optimizing sampling pattern for joint acceleration of multi-contrast MRI using model-based deep learning.

Acceleration of MR imaging (MRI) is a popular research area, and usage of deep learning for acceleration has become highly widespread in the MR community. Joint acceleration of multiple-acquisition MRI was proven to be effective over a single-acquisition approach. Also, optimization in the sampling pattern demonstrated its advantage over conventional undersampling pattern. However, optimizing the sampling patterns for joint acceleration of multiple-acquisition MRI has not been investigated well. To develop a model-based deep learning scheme to optimize sampling patterns for a joint acceleration of multi-contrast MRI. The proposed scheme combines sampling pattern optimization and multi-contrast MRI reconstruction. It was extended from the physics-guided method of the joint model-based deep learning (J-MoDL) scheme to optimize the separate sampling pattern for each of multiple contrasts simultaneously for their joint reconstruction. Tests were performed with three contrasts of T2-weighted, FLAIR, and T1-weighted images. The proposed multi-contrast method was compared to (i) single-contrast method with sampling optimization (baseline J-MoDL), (ii) multi-contrast method without sampling optimization, and (iii) multi-contrast method with single common sampling optimization for all contrasts. The optimized sampling patterns were analyzed for sampling location overlap across contrasts. The scheme was also tested in a data-driven scenario, where the inversion between input and label was learned from the under-sampled data directly and tested on knee datasets for generalization test. The proposed scheme demonstrated a quantitative and qualitative advantage over the single-contrast scheme with sampling pattern optimization and the multi-contrast scheme without sampling pattern optimization. Optimizing the separate sampling pattern for each of the multi-contrasts was superior to optimizing only one common sampling pattern for all contrasts. The proposed scheme showed less overlap in sampling locations than the single-contrast scheme. The main hypothesis was also held in the data-driven situation as well. The brain-trained model worked well on the knee images, demonstrating its generalizability. Our study introduced an effective scheme that combines the sampling optimization and the multi-contrast acceleration. The seamless combination resulted in superior performance over the other existing methods.

Assessment of data consistency through cascades of independently recurrent inference machines for fast and robust accelerated MRI reconstruction

Objective. Machine Learning methods can learn how to reconstruct magnetic resonance images (MRI) and thereby accelerate acquisition, which is of paramount importance to the clinical workflow. Physics-informed networks incorporate the forward model of accelerated MRI reconstruction in the learning process. With increasing network complexity, robustness is not ensured when reconstructing data unseen during training. We aim to embed data consistency (DC) in deep networks while balancing the degree of network complexity. While doing so, we will assess whether either explicit or implicit enforcement of DC in varying network architectures is preferred to optimize performance. Approach. We propose a scheme called Cascades of Independently Recurrent Inference Machines (CIRIM) to assess DC through unrolled optimization. Herein we assess DC both implicitly by gradient descent and explicitly by a designed term. Extensive comparison of the CIRIM to compressed sensing as well as other Machine Learning methods is performed: the End-to-End Variational Network (E2EVN), CascadeNet, KIKINet, LPDNet, RIM, IRIM, and UNet. Models were trained and evaluated on T1-weighted and FLAIR contrast brain data, and T2-weighted knee data. Both 1D and 2D undersampling patterns were evaluated. Robustness was tested by reconstructing 7.5× prospectively undersampled 3D FLAIR MRI data of multiple sclerosis (MS) patients with white matter lesions. Main results. The CIRIM performed best when implicitly enforcing DC, while the E2EVN required an explicit DC formulation. Through its cascades, the CIRIM was able to score higher on structural similarity and PSNR compared to other methods, in particular under heterogeneous imaging conditions. In reconstructing MS patient data, prospectively acquired with a sampling pattern unseen during model training, the CIRIM maintained lesion contrast while efficiently denoising the images. Significance. The CIRIM showed highly promising generalization capabilities maintaining a very fair trade-off between reconstructed image quality and fast reconstruction times, which is crucial in the clinical workflow.

A Deep Learning Method for Optimal Undersampling Patterns and Image Recovery for MRI Exploiting Losses and Projections

Compressed Sensing was recently proposed to reduce the long acquisition time of Magnetic Resonance Imaging by undersampling the signal frequency content and then algorithmically reconstructing the original image. We propose a way to significantly improve the above method by exploiting a deep neural network to tackle both problems of frequency sub-sampling and image reconstruction simultaneously, thanks to the introduction of a new loss function to drive the training and the addition of a post-processing non-neural stage. Furthermore, we highlight how some of the quantities along the processing chain can be used as a proxy of the quality of the recovered image, thus allowing a self-assessment of the whole technique. All improvements hinge on the possibility of identifying constraints to which the final image must obey and suitably enforce them. The effectiveness of our approach is tested on real-world MRI acquisitions from the fastMRI public database and achieves an appreciable improvement in Peak Signal-to-Noise Ratio with respect to the original CS-based proposal with speed-up factors 4 and 8.

Time efficiency analysis for undersampled quantitative MRI acquisitions.

To realize Quantitative MRI (QMRI) with clinically acceptable scan time, acceleration factors achieved by conventional parallel imaging techniques are often inadequate. Further acceleration is possible using model-based reconstruction. We propose a theoretical metric called TEUSQA: Time Efficiency for UnderSampled QMRI Acquisitions to inform sequence design and sample pattern optimisation. TEUSQA is designed for a particular class of reconstruction techniques that directly estimate tissue parameters, possibly using prior information to regularize the estimation. TEUSQA can be used to evaluate undersampling patterns for multi-contrast QMRI sequences targeting any tissue parameter. To verify the time efficiency predicted by TEUSQA, we performed Monte Carlo simulations and an accelerated parameter mapping with two sequences (Inversion prepared fast spin echo for T1 and T2 mapping and 3D GRASE for T2 and B0 inhomogeneity mapping). Using TEUSQA, we assessed several ways to generate undersampling patterns in silico, providing insight into the relation between sample distribution and time efficiency for different acceleration factors. The time efficiency predicted by TEUSQA was within 15% of that observed in the Monte Carlo simulations and the prospective acquisition experiment. The assessment of undersampling patterns showed that a class of good patterns could be obtained by low-discrepancy sampling. We believe that TEUSQA offers a valuable instrument for developers of novel QMRI sequences pushing the boundaries of acceleration to achieve clinically feasible protocols. Finally, we applied a time-efficient undersampling pattern selected using TEUSQA for a 32-fold accelerated scan to map T1 & T2 mapping of a healthy volunteer.

Coverage of In Situ Climatological Observations in the World's Mountains

Many mountainous environments and ecosystems around the world are responding rapidly to ongoing climate change. Long-term climatological time-series from such regions are crucial for developing improving understanding of the mechanisms driving such changes and ultimately delivering more reliable future impact projections to environmental managers and other decision makers. Whilst it is already established that high elevation regions tend to be comparatively under-sampled, detailed spatial and other patterns in the coverage of mountain climatological data have not yet been comprehensively assessed on a global basis. To begin to address this deficiency, we analyse the coverage of mountainous records from the Global Historical Climatological Network-Daily (GHCNd) inventory with respect to space, time, and elevation. Three key climate-related variables—air temperature, precipitation, and snow depth—are considered across 292 named mountain ranges. Several additional datasets are also introduced to characterize data coverage relative to topographic, hydrological, and socio-economic factors. Spatial mountain data coverage is found to be highly uneven, with station densities in several “Water Tower Units” that were previously identified as having great hydrological importance to society being especially low. Several mountainous regions whose elevational distribution is severely undersampled by GHCNd stations are identified, and mountain station density is shown to be only weakly related to the human population or economic output of the corresponding downstream catchments. Finally, we demonstrate the capabilities of a script (which is provided in the Supplementary Material) to produce detailed assessments of individual records' temporal coverage and measurement quality information. Overall, our contribution should help international authorities and regional stakeholders identify areas, variables, and other monitoring-related considerations that should be prioritized for infrastructure and capacity investment. Finally, the transparent and reproducible approach taken will enable the analysis to be rapidly repeated for subsequent versions of GHCNd, and could act as a basis for similar analyses using other spatial reporting boundaries and/or environmental monitoring station networks.

2D probabilistic undersampling pattern optimization for MR image reconstruction.

With 3D magnetic resonance imaging (MRI), a tradeoff exists between higher image quality and shorter scan time. One way to solve this problem is to reconstruct high-quality MRI images from undersampled k-space. There have been many recent studies exploring effective k-space undersampling patterns and designing MRI reconstruction methods from undersampled k-space, which are two necessary steps. Most studies separately considered these two steps, although in theory, their performance is dependent on each other. In this study, we propose a joint optimization model, trained end-to-end, to simultaneously optimize the undersampling pattern in the Fourier domain and the reconstruction model in the image domain. A 2D probabilistic undersampling layer was designed to optimize the undersampling pattern and probability distribution in a differentiable manner. A 2D inverse Fourier transform layer was implemented to connect the Fourier domain and the image domain during the forward and back propagation. Finally, we discovered an optimized relationship between the probability distribution of the undersampling pattern and its corresponding sampling rate. Further testing was performed using 3D T1-weighted MR images of the brain from the MICCAI 2013 Grand Challenge on Multi-Atlas Labeling dataset and locally acquired brain 3D T1-weighted MR images of healthy volunteers and contrast-enhanced 3D T1-weighted MR images of high-grade glioma patients. The results showed that the recovered MR images using our 2D probabilistic undersampling pattern (with or without the reconstruction network) significantly outperformed those using the existing start-of-the-art undersampling strategies for both qualitative and quantitative comparison, suggesting the advantages and some extent of the generalization of our proposed method.

Enhanced detection of paramagnetic fluorine-19 magnetic resonance imaging agents using zero echo time sequence and compressed sensing.

Fluorine-19 (19 F) magnetic resonance imaging (MRI) is an emerging technique offering specific detection of labeled cells in vivo. Lengthy acquisition times and modest signal-to-noise ratio (SNR) makes three-dimensional spin-density-weighted 19 F imaging challenging. Recent advances in tracer paramagnetic metallo-perfluorocarbon (MPFC) nanoemulsion probes have shown multifold SNR improvements due to an accelerated 19 F T1 relaxation rate and a commensurate gain in imaging speed and averages. However, 19 F T2 -reduction and increased linewidth limit the amount of metal additive in MPFC probes, thus constraining the ultimate SNR. To overcome these barriers, we describe a compressed sampling (CS) scheme, implemented using a "zero" echo time (ZTE) sequence, with data reconstructed via a sparsity-promoting algorithm. Our CS-ZTE scheme acquires k-space data using an undersampled spherical radial pattern and signal averaging. Image reconstruction employs off-the-shelf sparse solvers to solve a joint total variation and -norm regularized least square problem. To evaluate CS-ZTE, we performed simulations and acquired 19 F MRI data at 11.7T in phantoms and mice receiving MPFC-labeled dendritic cells. For MPFC-labeled cells in vivo, we show SNR gains of ~6.3×with 8-fold undersampling. We show that this enhancement is due to three mechanisms including undersampling and commensurate increase in signal averaging in a fixed scan time, denoising attributes from the CS algorithm, and paramagnetic reduction of T1 . Importantly, 19 F image intensity analyses yield accurate estimates of absolute quantification of 19 F spins. Overall, the CS-ZTE method using MPFC probes achieves ultrafast imaging, a substantial boost in detection sensitivity, accurate 19 F spin quantification, and minimal image artifacts.

ReconResNet: Regularised residual learning for MR image reconstruction of Undersampled Cartesian and Radial data

MRI is an inherently slow process, which leads to long scan time for high-resolution imaging. The speed of acquisition can be increased by ignoring parts of the data (undersampling). Consequently, this leads to the degradation of image quality, such as loss of resolution or introduction of image artefacts. This work aims to reconstruct highly undersampled Cartesian or radial MR acquisitions, with better resolution and with less to no artefact compared to conventional techniques like compressed sensing. In recent times, deep learning has emerged as a very important area of research and has shown immense potential in solving inverse problems, e.g. MR image reconstruction. In this paper, a deep learning based MR image reconstruction framework is proposed, which includes a modified regularised version of ResNet as the network backbone to remove artefacts from the undersampled image, followed by data consistency steps that fusions the network output with the data already available from undersampled k-space in order to further improve reconstruction quality. The performance of this framework for various undersampling patterns has also been tested, and it has been observed that the framework is robust to deal with various sampling patterns, even when mixed together while training, and results in very high quality reconstruction, in terms of high SSIM (highest being 0.990 ± 0.006 for acceleration factor of 3.5), while being compared with the fully sampled reconstruction. It has been shown that the proposed framework can successfully reconstruct even for an acceleration factor of 20 for Cartesian (0.968 ± 0.005) and 17 for radially (0.962 ± 0.012) sampled data. Furthermore, it has been shown that the framework preserves brain pathology during reconstruction while being trained on healthy subjects.

Single-Pass Object-Adaptive Data Undersampling and Reconstruction for MRI

There is much recent interest in techniques to accelerate the data acquisition process in MRI by acquiring limited measurements. Often sophisticated reconstruction algorithms are deployed to maintain high image quality in such settings. In this work, we propose a data-driven sampler using a convolutional neural network, MNet, to provide object-specific sampling patterns adaptive to each scanned object. The network observes very limited low-frequency k-space data for each object and rapidly predicts the desired undersampling pattern in one go that achieves high image reconstruction quality. We propose an accompanying alternating-type training framework with a mask-backward procedure that efficiently generates training labels for the sampler network and jointly trains an image reconstruction network. Experimental results on the fastMRI knee dataset demonstrate the ability of the proposed learned undersampling network to generate object-specific masks at fourfold and eightfold acceleration that achieve superior image reconstruction performance than several existing schemes. The source code for the proposed joint sampling and reconstruction learning framework is available at https://github.com/zhishenhuang/mri.

Learned Low-Rank Priors in Dynamic MR Imaging.

Deep learning methods have achieved attractive performance in dynamic MR cine imaging. However, most of these methods are driven only by the sparse prior of MR images, while the important low-rank (LR) prior of dynamic MR cine images is not explored, which may limit further improvements in dynamic MR reconstruction. In this paper, a learned singular value thresholding (Learned-SVT) operator is proposed to explore low-rank priors in dynamic MR imaging to obtain improved reconstruction results. In particular, we put forward a model-based unrolling sparse and low-rank network for dynamic MR imaging, dubbed as SLR-Net. SLR-Net is defined over a deep network flow graph, which is unrolled from the iterative procedures in the iterative shrinkage-thresholding algorithm (ISTA) for optimizing a sparse and LR-based dynamic MRI model. Experimental results on a single-coil scenario show that the proposed SLR-Net can further improve the state-of-the-art compressed sensing (CS) methods and sparsity-driven deep learning-based methods with strong robustness to different undersampling patterns, both qualitatively and quantitatively. Besides, SLR-Net has been extended to a multi-coil scenario, and achieved excellent reconstruction results compared with a sparsity-driven multi-coil deep learning-based method under a high acceleration. Prospective reconstruction results on an open real-time dataset further demonstrate the capability and flexibility of the proposed method on real-time scenarios.

Multi-band- and in-plane-accelerated diffusion MRI enabled by model-based deep learning in q-space and its extension to learning in the spherical harmonic domain.

To propose a new method for the recovery of combined in-plane- and multi-band (MB)-accelerated diffusion MRI data. Combining MB acceleration with in-plane acceleration is crucial to improve the time efficiency of high (angular and spatial) resolution diffusion scans. However, as the MB factor and in-plane acceleration increase, the reconstruction becomes challenging due to the heavy aliasing. The new reconstruction utilizes an additional q-space prior to constrain the recovery, which is derived from the previously proposed qModeL framework. Specifically, the qModeL prior provides a pre-learned representation of the diffusion signal space to which the measured data belongs. We show that the pre-learned q-space prior along with a model-based iterative reconstruction that accommodate multi-band unaliasing, can efficiently reconstruct the in-plane- and MB-accelerated data. The power of joint reconstruction is maximally utilized by using an incoherent under-sampling pattern in the k-q domain. We tested the proposed method on single- and multi-shell data, with high/low angular resolution, high/low spatial resolution, healthy/abnormal tissues, and 3T/7T field strengths. Furthermore, the learning is extended to the spherical harmonic basis, to provide a rotational invariant learning framework. The qModeL joint reconstruction is shown to simultaneously unalias and jointly recover DWIs with reasonable accuracy in all the cases studied. The reconstruction error from 18-fold accelerated multi-shell datasets was <3%. The microstructural maps derived from the accelerated acquisitions also exhibit reasonable accuracy for both healthy and abnormal tissues. The deep learning (DL)-enabled reconstructions are comparable to those derived using traditional methods. qModeL enables the joint recovery of combined in-plane- and MB-accelerated dMRI utilizing DL.

An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset

This work aims at developing a generalizable Magnetic Resonance Imaging (MRI) reconstruction method in the meta-learning framework. Specifically, we develop a deep reconstruction network induced by a learnable optimization algorithm (LOA) to solve the nonconvex nonsmooth variational model of MRI image reconstruction. In this model, the nonconvex nonsmooth regularization term is parameterized as a structured deep network where the network parameters can be learned from data. We partition these network parameters into two parts: a task-invariant part for the common feature encoder component of the regularization, and a task-specific part to account for the variations in the heterogeneous training and testing data. We train the regularization parameters in a bilevel optimization framework which significantly improves the robustness of the training process and the generalization ability of the network. We conduct a series of numerical experiments using heterogeneous MRI data sets with various undersampling patterns, ratios, and acquisition settings. The experimental results show that our network yields greatly improved reconstruction quality over existing methods and can generalize well to new reconstruction problems whose undersampling patterns/trajectories are not present during training.

Time domain principal component analysis for rapid, real-time 2D MRI reconstruction from undersampled data.

A rapid real-time 2D accelerated method was developed for magnetic resonance imaging (MRI) using principal component analysis (PCA) in the temporal domain. This method employs a moving window of previous dynamic frames to reconstruct the current, real-time frame within this window. This technique could be particularly useful in real-time tracking applications such as in MR-guided radiotherapy, where low latency real-time reconstructions are essential. The method was tested retrospectively on 15 fully-sampled data sets of lung patient data acquired on a 3T Philips Achieva system. High frequency data are incoherently undersampled, while the central low-frequency data are always acquired to characterize the temporal fluctuations through PCA. The undersampling pattern is derived in such a way that all of k-space is acquired within a pre-determined number of frames. The missing data in the current frame are then filled in by fitting the temporal characterizations to the acquired undersampled data, using a pre-determined number of PCs. A subset of six patients was used to test the contour ability of the images. Various accelerations between 3x and 8x were tested along with the optimal number of PCs for fitting. A comparison was also performed with previous work from our group proposed by Dietz etal. as well as with a standard low resolution acquisition. In order to determine how the method would perform at lower signal to noise ratio (SNR), noise levels of 2×, 4×, and 6× were added to the 3T data. Metrics such as normalised mean square error and Dice coefficient were used to measure the reconstruction image quality and contour ability. The proposed method demonstrated good temporal robustness as consistent metrics were detected for the duration of the imaging session. It was found that the optimal number of PCs for temporal fitting was dependent on the acceleration rate. For the data tested, five PCs were found to be optimal at the acceleration rates of 3× and 4×. This number decreases to three at accelerations of 5× and 6× and further decreases to two at an acceleration rate of 8×, likely due to greater instability with fewer acquired data points. The use of too many PCs for fitting increased the chances of noisy reconstruction which affected contourability. The proposed 2D real-time MR acceleration method demonstrated greater robustness in the metrics over time when compared with previous real-time PCA methods using metrics such as normalised mean squared error, peak SNR and structural similarity up to an acceleration of 8x. Improved temporal robustness of image structure contourability and accurate definition was also demonstrated using several metrics including the Dice coefficient. Reconstruction of raw acquired data can be performed at approximately 50ms per frame using an Intel core i5 CPU. The method has the advantage of being very flexible in terms of hardware requirements as it can operate successfully on a single coil channel and does not require specialized computing power to implement in real-time.