ENLIVE: An Efficient Nonlinear Method for Calibrationless and Robust Parallel Imaging

Enhancing quality and speed in database-free neural network reconstructions of undersampled MRI with SCAMPI

PURPOSE

We present SCAMPI (Sparsity Constrained Application of deep Magnetic resonance Priors for Image reconstruction), an untrained deep Neural Network for MRI reconstruction without previous training on datasets. It expands the Deep Image Prior approach with a multidomain, sparsity-enforcing loss function to achieve higher image quality at a faster convergence speed than previously reported methods.

METHODS

Two-dimensional MRI data from the FastMRI dataset with Cartesian undersampling in phase-encoding direction were reconstructed for different acceleration rates for single coil and multicoil data.

RESULTS

The performance of our architecture was compared to state-of-the-art Compressed Sensing methods and ConvDecoder, another untrained Neural Network for two-dimensional MRI reconstruction. SCAMPI outperforms these by better reducing undersampling artifacts and yielding lower error metrics in multicoil imaging. In comparison to ConvDecoder, the U-Net architecture combined with an elaborated loss-function allows for much faster convergence at higher image quality. SCAMPI can reconstruct multicoil data without explicit knowledge of coil sensitivity profiles. Moreover, it is a novel tool for reconstructing undersampled single coil k-space data.

CONCLUSION

Our approach avoids overfitting to dataset features, that can occur in Neural Networks trained on databases, because the network parameters are tuned only on the reconstruction data. It allows better results and faster reconstruction than the baseline untrained Neural Network approach.

Scan-Specific Self-Supervised Bayesian Deep Non-Linear Inversion for Undersampled MRI Reconstruction

Magnetic resonance imaging is subject to slow acquisition times due to the inherent limitations in data sampling. Recently, supervised deep learning has emerged as a promising technique for reconstructing sub-sampled MRI. However, supervised deep learning requires a large dataset of fully-sampled data. Although unsupervised or self-supervised deep learning methods have emerged to address the limitations of supervised deep learning approaches, they still require a database of images. In contrast, scan-specific deep learning methods learn and reconstruct using only the sub-sampled data from a single scan. Here, we introduce Scan-Specific Self-Supervised Bayesian Deep Non-Linear Inversion (DNLINV) that does not require an auto calibration scan region. DNLINV utilizes a Deep Image Prior-type generative modeling approach and relies on approximate Bayesian inference to regularize the deep convolutional neural network. We demonstrate our approach on several anatomies, contrasts, and sampling patterns and show improved performance over existing approaches in scan-specific calibrationless parallel imaging and compressed sensing.

Accelerated MRI reconstructions via variational network and feature domain learning

We introduce three architecture modifications to enhance the performance of the end-to-end (E2E) variational network (VarNet) for undersampled MRI reconstructions. We first implemented the Feature VarNet, which propagates information throughout the cascades of the network in an N-channel feature-space instead of a 2-channel feature-space. Then, we add an attention layer that utilizes the spatial locations of Cartesian undersampling artifacts to further improve performance. Lastly, we combined the Feature and E2E VarNets into the Feature-Image (FI) VarNet, to facilitate cross-domain learning and boost accuracy. Reconstructions were evaluated on the fastMRI dataset using standard metrics and clinical scoring by three neuroradiologists. Feature and FI VarNets outperformed the E2E VarNet for 4 × , 5 × and 8 × Cartesian undersampling in all studied metrics. FI VarNet secured second place in the public fastMRI leaderboard for 4 × Cartesian undersampling, outperforming all open-source models in the leaderboard. Radiologists rated FI VarNet brain reconstructions with higher quality and sharpness than the E2E VarNet reconstructions. FI VarNet excelled in preserving anatomical details, including blood vessels, whereas E2E VarNet discarded or blurred them in some cases. The proposed FI VarNet enhances the reconstruction quality of undersampled MRI and could enable clinically acceptable reconstructions at higher acceleration factors than currently possible.

Coil sketching for computationally efficient MR iterative reconstruction

PURPOSE

Parallel imaging and compressed sensing reconstructions of large MRI datasets often have a prohibitive computational cost that bottlenecks clinical deployment, especially for three-dimensional (3D) non-Cartesian acquisitions. One common approach is to reduce the number of coil channels actively used during reconstruction as in coil compression. While effective for Cartesian imaging, coil compression inherently loses signal energy, producing shading artifacts that compromise image quality for 3D non-Cartesian imaging. We propose coil sketching, a general and versatile method for computationally-efficient iterative MR image reconstruction.

THEORY AND METHODS

We based our method on randomized sketching algorithms, a type of large-scale optimization algorithms well established in the fields of machine learning and big data analysis. We adapt the sketching theory to the MRI reconstruction problem via a structured sketching matrix that, similar to coil compression, considers high-energy virtual coils obtained from principal component analysis. But, unlike coil compression, it also considers random linear combinations of the remaining low-energy coils, effectively leveraging information from all coils.

RESULTS

First, we performed ablation experiments to validate the sketching matrix design on both Cartesian and non-Cartesian datasets. The resulting design yielded both improved computatioanal efficiency and preserved signal-to-noise ratio (SNR) as measured by the inverse g-factor. Then, we verified the efficacy of our approach on high-dimensional non-Cartesian 3D cones datasets, where coil sketching yielded up to three-fold faster reconstructions with equivalent image quality.

CONCLUSION

Coil sketching is a general and versatile reconstruction framework for computationally fast and memory-efficient reconstruction.

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.

Optimization in the space domain for density compensation with the nonuniform FFT

The non-uniform Discrete Fourier Transform algorithm has shown great utility for reconstructing images from non-uniformly spaced Fourier samples in several imaging modalities. Due to the non-uniform spacing, some correction for the variable density of the samples must be made. Common methods for generating density compensation values are either sub-optimal or only consider a finite set of points in the optimization. This manuscript presents an algorithm for generating density compensation values from a set of Fourier samples that takes into account the point spread function over an entire rectangular region in the image domain. We show that the reconstructed images using the density compensation values of this method are of superior quality when compared to other standard methods. Results are shown with a numerical phantom and with magnetic resonance images of the abdomen and the knee.

JSENSE-Pro: Joint sensitivity estimation and image reconstruction in parallel imaging using pre-learned subspaces of coil sensitivity functions

PURPOSE

To improve calibrationless parallel imaging using pre-learned subspaces of coil sensitivity functions.

THEORY AND METHODS

A subspace-based joint sensitivity estimation and image reconstruction method was developed for improved parallel imaging with no calibration data. Specifically, we proposed to use a probabilistic subspace model to capture prior information of the coil sensitivity functions from previous scans acquired using the same receiver system. Both the subspace basis and coefficient distributions were learned from a small set of training data. The learned subspace model was then incorporated into the regularized reconstruction formalism that includes a sparsity prior. The nonlinear optimization problem was solved using alternating minimization algorithm. Public fastMRI brain dataset was retrospectively undersampled by different schemes for performance evaluation of the proposed method.

RESULTS

With no calibration data, the proposed method consistently produced the most accurate coil sensitivity estimation and highest quality image reconstructions at all acceleration factors tested in comparison with state-of-the-art methods including JSENSE, DeepSENSE, P-LORAKS, and Sparse BLIP. Our results are comparable to or even better than those from SparseSENSE, which used calibration data for sensitivity estimation. The work also demonstrated that the probabilistic subspace model learned from T₂ w data can be generalized to aiding the reconstruction of FLAIR data acquired from the same receiver system.

CONCLUSION

A subspace-based method named JSENSE-Pro has been proposed for accelerated parallel imaging without the acquisition of companion calibration data. The method is expected to further enhance the practical utility of parallel imaging, especially in applications where calibration data acquisition is not desirable or limited.

Calibrationless reconstruction of uniformly‐undersampled multi‐channel MR data with deep learning estimated ESPIRiT maps

PURPOSE

To develop a truly calibrationless reconstruction method that derives An Eigenvalue Approach to Autocalibrating Parallel MRI (ESPIRiT) maps from uniformly-undersampled multi-channel MR data by deep learning.

METHODS

ESPIRiT, one commonly used parallel imaging reconstruction technique, forms the images from undersampled MR k-space data using ESPIRiT maps that effectively represents coil sensitivity information. Accurate ESPIRiT map estimation requires quality coil sensitivity calibration or autocalibration data. We present a U-Net based deep learning model to estimate the multi-channel ESPIRiT maps directly from uniformly-undersampled multi-channel multi-slice MR data. The model is trained using fully-sampled multi-slice axial brain datasets from the same MR receiving coil system. To utilize subject-coil geometric parameters available for each dataset, the training imposes a hybrid loss on ESPIRiT maps at the original locations as well as their corresponding locations within the standard reference multi-slice axial stack. The performance of the approach was evaluated using publicly available T1-weighed brain and cardiac data.

RESULTS

The proposed model robustly predicted multi-channel ESPIRiT maps from uniformly-undersampled k-space data. They were highly comparable to the reference ESPIRiT maps directly computed from 24 consecutive central k-space lines. Further, they led to excellent ESPIRiT reconstruction performance even at high acceleration, exhibiting a similar level of errors and artifacts to that by using reference ESPIRiT maps.

CONCLUSION

A new deep learning approach is developed to estimate ESPIRiT maps directly from uniformly-undersampled MR data. It presents a general strategy for calibrationless parallel imaging reconstruction through learning from the coil and protocol-specific data.

Deep, deep learning with BART

PURPOSE

To develop a deep-learning-based image reconstruction framework for reproducible research in MRI.

METHODS

The BART toolbox offers a rich set of implementations of calibration and reconstruction algorithms for parallel imaging and compressed sensing. In this work, BART was extended by a nonlinear operator framework that provides automatic differentiation to allow computation of gradients. Existing MRI-specific operators of BART, such as the nonuniform fast Fourier transform, are directly integrated into this framework and are complemented by common building blocks used in neural networks. To evaluate the use of the framework for advanced deep-learning-based reconstruction, two state-of-the-art unrolled reconstruction networks, namely the Variational Network and MoDL, were implemented.

RESULTS

State-of-the-art deep image-reconstruction networks can be constructed and trained using BART's gradient-based optimization algorithms. The BART implementation achieves a similar performance in terms of training time and reconstruction quality compared to the original implementations based on TensorFlow.

CONCLUSION

By integrating nonlinear operators and neural networks into BART, we provide a general framework for deep-learning-based reconstruction in MRI.

Pyramid Convolutional RNN for MRI Image Reconstruction

Fast and accurate MRI image reconstruction from undersampled data is crucial in clinical practice. Deep learning based reconstruction methods have shown promising advances in recent years. However, recovering fine details from undersampled data is still challenging. In this paper, we introduce a novel deep learning based method, Pyramid Convolutional RNN (PC-RNN), to reconstruct images from multiple scales. Based on the formulation of MRI reconstruction as an inverse problem, we design the PC-RNN model with three convolutional RNN (ConvRNN) modules to iteratively learn the features in multiple scales. Each ConvRNN module reconstructs images at different scales and the reconstructed images are combined by a final CNN module in a pyramid fashion. The multi-scale ConvRNN modules learn a coarse-to-fine image reconstruction. Unlike other common reconstruction methods for parallel imaging, PC-RNN does not employ coil sensitive maps for multi-coil data and directly model the multiple coils as multi-channel inputs. The coil compression technique is applied to standardize data with various coil numbers, leading to more efficient training. We evaluate our model on the fastMRI knee and brain datasets and the results show that the proposed model outperforms other methods and can recover more details. The proposed method is one of the winner solutions in the 2019 fastMRI competition.

Isotropic multichannel total variation framework for joint reconstruction of multicontrast parallel MRI

Purpose: To develop a synergistic image reconstruction framework that exploits multicontrast (MC), multicoil, and compressed sensing (CS) redundancies in magnetic resonance imaging (MRI). Approach: CS, MC acquisition, and parallel imaging (PI) have been individually well developed, but the combination of the three has not been equally well studied, much less the potential benefits of isotropy within such a setting. Inspired by total variation theory, we introduce an isotropic MC image regularizer and attain its full potential by integrating it into compressed MC multicoil MRI. A convex optimization problem is posed to model the new variational framework and a first-order algorithm is developed to solve the problem. Results: It turns out that the proposed isotropic regularizer outperforms many of the state-of-the-art reconstruction methods not only in terms of rotation-invariance preservation of symmetrical features, but also in suppressing noise or streaking artifacts, which are normally encountered in PI methods at aggressive undersampling rates. Moreover, the new framework significantly prevents intercontrast leakage of contrast-specific details, which seems to be a difficult situation to handle for some variational and low-rank MC reconstruction approaches. Conclusions: The new framework is a viable option for image reconstruction in fast protocols of MC parallel MRI, potentially reducing patient discomfort in otherwise long and time-consuming scans.

Aliasing-free reduced field-of-view parallel imaging

PURPOSE

To reconstruct aliasing-free full field-of-view (FOV) images for reduced FOV (rFOV) parallel imaging (PI) with Cartesian and Wave sampling, which suffers from aliasing artifacts using existing PI methods.

THEORY AND METHODS

The sensitivity encoding method (SENSE) was extended to the Soft-SENSE models supporting multiple-set coil sensitivity maps (CSM) and point spread functions (PSF) for Cartesian and Wave sampled rFOV PI, respectively. The multiple-set CSM and PSF were created from full FOV CSM and PSF according to the image folding process induced by rFOV sampling. The Soft-SENSE reconstructions could be solved by the same algorithms for the conventional full FOV SENSE reconstruction.

RESULTS

Soft-SENSE using multiple-set full FOV CSM and PSF successfully reconstruct aliasing-free full FOV image from rFOV PI data with Cartesian and Wave sampling. The proposed rFOV PI enables flexible control of the aliasing and achieves comparable geometry factors as the standard full FOV PI with the same net acceleration factor. Reduced FOV PI improves the computational efficiency of iterative compressed sensing (CS) and PI reconstruction, especially for high-resolution volumetric imaging, thanks to the reduced fast Fourier transform (FFT) size. Moreover, rFOV PI reconstruction provides a potential alternative to the phase oversampling for the FOV aliasing problem.

CONCLUSION

The proposed Soft-SENSE using full FOV CSM and PSF could reconstruct aliasing-free full FOV image for rFOV PI, and make it a viable solution enabling more flexible PI acceleration and effectively improving the computational efficiency of iterative CSPI reconstruction.

Improved Image Quality for Static BLADE Magnetic Resonance Imaging Using the Total-Variation Regularized Least Absolute Deviation Solver

In order to improve the image quality of BLADE magnetic resonance imaging (MRI) using the index tensor solvers and to evaluate MRI image quality in a clinical setting, we implemented BLADE MRI reconstructions using two tensor solvers (the least-squares solver and the L1 total-variation regularized least absolute deviation (L1TV-LAD) solver) on a graphics processing unit (GPU). The BLADE raw data were prospectively acquired and presented in random order before being assessed by two independent radiologists. Evaluation scores were examined for consistency and then by repeated measures analysis of variance (ANOVA) to identify the superior algorithm. The simulation showed the structural similarity index (SSIM) of various tensor solvers ranged between 0.995 and 0.999. Inter-reader reliability was high (Intraclass correlation coefficient (ICC) = 0.845, 95% confidence interval: 0.817, 0.87). The image score of L1TV-LAD was significantly higher than that of vendor-provided image and the least-squares method. The image score of the least-squares method was significantly lower than that of the vendor-provided image. No significance was identified in L1TV-LAD with a regularization strength of λ= 0.4–1.0. The L1TV-LAD with a regularization strength of λ= 0.4–0.7 was found consistently better than least-squares and vendor-provided reconstruction in BLADE MRI with a SENSitivity Encoding (SENSE) factor of 2. This warrants further development of the integrated computing system with the scanner.

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.

Accelerated calibrationless parallel transmit mapping using joint transmit and receive low-rank tensor completion

Purpose

To evaluate an algorithm for calibrationless parallel imaging to reconstruct undersampled parallel transmit field maps for the body and brain.

Methods

Using a combination of synthetic data and in vivo measurements from brain and body, 3 different approaches to a joint transmit and receive low-rank tensor completion algorithm are evaluated. These methods included: 1) virtual coils using the product of receive and transmit sensitivities, 2) joint-receiver coils that enforces a low rank structure across receive coils of all transmit modes, and 3) transmit low rank that uses a low rank structure for both receive and transmit modes simultaneously. The performance of each is investigated for different noise levels and different acceleration rates on an 8-channel parallel transmit 7 Tesla system.

Results

The virtual coils method broke down with increasing noise levels or acceleration rates greater than 2, producing normalized RMS error greater than 0.1. The joint receiver coils method worked well up to acceleration factors of 4, beyond which the normalized RMS error exceeded 0.1. Transmit low rank enabled an eightfold acceleration, with most normalized RMS errors remaining below 0.1.

Conclusion

This work demonstrates that undersampling factors of up to eightfold are feasible for transmit array mapping and can be reconstructed using calibrationless parallel imaging methods.

Maxwell parallel imaging

PURPOSE

To develop a general framework for parallel imaging (PI) with the use of Maxwell regularization for the estimation of the sensitivity maps (SMs) and constrained optimization for the parameter-free image reconstruction.

THEORY AND METHODS

Certain characteristics of both the SMs and the images are routinely used to regularize the otherwise ill-posed optimization-based joint reconstruction from highly accelerated PI data. In this paper, we rely on a fundamental property of SMs-they are solutions of Maxwell equations-we construct the subspace of all possible SM distributions supported in a given field-of-view, and we promote solutions of SMs that belong in this subspace. In addition, we propose a constrained optimization scheme for the image reconstruction, as a second step, once an accurate estimation of the SMs is available. The resulting method, dubbed Maxwell parallel imaging (MPI), works for both 2D and 3D, with Cartesian and radial trajectories, and minimal calibration signals.

RESULTS

The effectiveness of MPI is illustrated for various undersampling schemes, including radial, variable-density Poisson-disc, and Cartesian, and is compared against the state-of-the-art PI methods. Finally, we include some numerical experiments that demonstrate the memory footprint reduction of the constructed Maxwell basis with the help of tensor decomposition, thus allowing the use of MPI for full 3D image reconstructions.

CONCLUSION

The MPI framework provides a physics-inspired optimization method for the accurate and efficient image reconstruction from arbitrary accelerated scans.

Robust autocalibrated structured low-rank EPI ghost correction

PURPOSE

We propose and evaluate a new structured low-rank method for echo-planar imaging (EPI) ghost correction called Robust Autocalibrated LORAKS (RAC-LORAKS). The method can be used to suppress EPI ghosts arising from the differences between different readout gradient polarities and/or the differences between different shots. It does not require conventional EPI navigator signals, and is robust to imperfect autocalibration data.

METHODS

Autocalibrated LORAKS is a previous structured low-rank method for EPI ghost correction that uses GRAPPA-type autocalibration data to enable high-quality ghost correction. This method works well when the autocalibration data are pristine, but performance degrades substantially when the autocalibration information is imperfect. RAC-LORAKS generalizes Autocalibrated LORAKS in two ways. First, it does not completely trust the information from autocalibration data, and instead considers the autocalibration and EPI data simultaneously when estimating low-rank matrix structure. Second, it uses complementary information from the autocalibration data to improve EPI reconstruction in a multi-contrast joint reconstruction framework. RAC-LORAKS is evaluated using simulations and in vivo data, including comparisons to state-of-the-art methods.

RESULTS

RAC-LORAKS is demonstrated to have good ghost elimination performance compared to state-of-the-art methods in several complicated EPI acquisition scenarios (including gradient-echo brain imaging, diffusion-encoded brain imaging, and cardiac imaging).

CONCLUSIONS

RAC-LORAKS provides effective suppression of EPI ghosts and is robust to imperfect autocalibration data.