MoDL-MUSSELS: Model-Based Deep Learning for Multishot Sensitivity-Encoded Diffusion MRI

Navigator-free multi-shot diffusion MRI via non-local low-rank reconstruction

PURPOSE

To develop a non-local low-rank (NLLR) reconstruction method for multi-shot EPI (ms-EPI) in DWI, addressing phase inconsistencies and noise issues while maintaining high spatial resolution in clinically feasible scan times.

THEORY AND METHODS

Single-shot EPI (ss-EPI) is widely used for DWI but suffers from geometric distortions and T2* blurring. ms-EPI improves spatial resolution but introduces shot-to-shot phase variations requiring correction strategies. Traditional navigator-based approaches may increase acquisition time. Recent low-rank regularization reconstruction techniques, such as locally low-rank (LLR) methods, can estimate the phase errors but rely strictly on local neighborhood information along the shot dimension. The proposed NLLR method extends this framework by leveraging non-local patch matching by grouping similar image patches across spatially distant image locations, enhancing non-local redundancy exploitation for improved phase estimation and correction as well as noise suppression. The method was validated in simulations and in vivo experiments and compared to existing post-processing denoising and navigator-free approaches.

RESULTS

In simulation experiments, compared to post-processing denoising algorithms, NLLR demonstrated superior noise suppression and structural preservation across all metrics, even when reconstructing from a single diffusion direction. In the in-vivo experiments, NLLR outperformed conventional navigator-free approaches particularly regarding noise suppression. Fractional anisotropy maps reconstructed using NLLR exhibited improved visualization of fine structures with improved SNR, with performance differences becoming more pronounced at higher resolutions.

CONCLUSION

The proposed NLLR approach provides an efficient and good solution for high-resolution DWI reconstruction, improving image quality while mitigating phase variations and noise.

Fast and ultra-high shot diffusion MRI image reconstruction with self-adaptive Hankel subspace

Multi-shot interleaved echo planar imaging is widely employed for acquiring high-resolution and low-distortion diffusion weighted images (DWI). These DWI images, however, are easily affected by motion artifacts induced by inter-shot phase variations which could be removed by enforcing the low-rankness of a huge 2D block Hankel matrix of the k-space. Successful applications have been evidenced on 4∼8 shots DWI but failure was observed on ultra-high shots, e.g. 10∼12 shots, limiting the extension to higher-resolution DWI. Moreover, the 2D Hankel matrix reconstruction is very time-consuming. Here, we propose to accelerate the reconstruction through decomposing this huge 2D matrix into small 1D lOw-raNk HAnkel (DONA) matrices from every k-space readout line. This extension encounters another problem of variant low-rankness across the k-space. To address this issue, we propose to separate signal subspaces of 1D Hankel matrices into the strong and uncertain ones. The former is pre-estimated from an initial image to reduce the degree of freedom in reconstruction. The latter protects image details in reconstruction by avoiding the overshadowing on small singular values. This method is called DONA with self-adapTive subspacE estimation (DONATE). In vivo results show that DONATE can not only accomplish 4-shot reconstruction in 10 s but also the reconstruction of 12 shots with 10 times faster computation. Besides, DONATE shows superiority on low-distortion spine DWI reconstruction and subjective image quality evaluation in terms of blind scoring by 4 radiologists.

3D distortion-free, reduced FOV diffusion-prepared gradient echo at 3 T

PURPOSE

To develop a 3D distortion-free reduced-FOV diffusion-prepared gradient-echo sequence and demonstrate its application in vivo for diffusion imaging of the spinal cord in healthy volunteers.

METHODS

A 3D multi-shot reduced-FOV diffusion-prepared gradient-echo acquisition is achieved using a slice-selective tip-down pulse in the phase-encoding direction in the diffusion preparation, combined with magnitude stabilizers, centric k-space encoding, and 2D phase navigators to correct for intershot phase errors. The accuracy of the ADC values obtained using the proposed approach was evaluated in a diffusion phantom and compared to the tabulated reference ADC values and to the ADC values obtained using a standard spin echo diffusion-weighted single-shot EPI sequence (DW-SS-EPI). Five healthy volunteers were scanned at 3 T using the proposed sequence, DW-SS-EPI, and a clinical diffusion-weighted multi-shot readout-segmented EPI sequence (RESOLVE) for cervical spinal cord imaging. Image quality, perceived SNR, and image distortion were assessed by two expert radiologists. ADC maps were calculated, and ADC values obtained with the proposed sequence were compared to those obtained using DW-SS-EPI and RESOLVE.

RESULTS

Consistent ADC estimates were measured in the diffusion phantom with the proposed sequence and the conventional DW-SS-EPI sequence, and the ADC values were in close agreement with the reference values provided by the manufacturer of the phantom. In vivo, the proposed sequence demonstrated improved image quality, improved perceived SNR, and reduced perceived distortion compared to DW-SS-EPI, whereas all measures were comparable against RESOLVE. There were no significant differences in ADC values estimated in vivo for each of the sequences.

CONCLUSION

3D distortion-free diffusion-prepared imaging can be achieved using the proposed sequence.

Self-Navigated 3D Diffusion MRI Using an Optimized CAIPI Sampling and Structured Low-Rank Reconstruction Estimated Navigator

3D multi-slab acquisitions are an appealing approach for diffusion MRI because they are compatible with the imaging regime delivering optimal SNR efficiency. In conventional 3D multi-slab imaging, shot-to-shot phase variations caused by motion pose challenges due to the use of multi-shot k-space acquisition. Navigator acquisition after each imaging echo is typically employed to correct phase variations, which prolongs scan time and increases the specific absorption rate (SAR). The aim of this study is to develop a highly efficient, self-navigated method to correct for phase variations in 3D multi-slab diffusion MRI without explicitly acquiring navigators. The sampling of each shot is carefully designed to intersect with the central kz=0 plane of each slab, and the multi-shot sampling is optimized for self-navigation performance while retaining decent reconstruction quality. The kz=0 intersections from all shots are jointly used to reconstruct a 2D phase map for each shot using a structured low-rank constrained reconstruction that leverages the redundancy in shot and coil dimensions. The phase maps are used to eliminate the shot-to-shot phase inconsistency in the final 3D multi-shot reconstruction. We demonstrate the method's efficacy using retrospective simulations and prospectively acquired in-vivo experiments at 1.22 mm and 1.09 mm isotropic resolutions. Compared to conventional navigated 3D multi-slab imaging, the proposed self-navigated method achieves comparable image quality while shortening the scan time by 31.7% and improving the SNR efficiency by 15.5%. The proposed method produces comparable quality of DTI and white matter tractography to conventional navigated 3D multi-slab acquisition with a much shorter scan time.

DIMOND: DIffusion Model OptimizatioN with Deep Learning

Diffusion magnetic resonance imaging is an important tool for mapping tissue microstructure and structural connectivity non-invasively in the in vivo human brain. Numerous diffusion signal models are proposed to quantify microstructural properties. Nonetheless, accurate estimation of model parameters is computationally expensive and impeded by image noise. Supervised deep learning-based estimation approaches exhibit efficiency and superior performance but require additional training data and may be not generalizable. A new DIffusion Model OptimizatioN framework using physics-informed and self-supervised Deep learning entitled "DIMOND" is proposed to address this problem. DIMOND employs a neural network to map input image data to model parameters and optimizes the network by minimizing the difference between the input acquired data and synthetic data generated via the diffusion model parametrized by network outputs. DIMOND produces accurate diffusion tensor imaging results and is generalizable across subjects and datasets. Moreover, DIMOND outperforms conventional methods for fitting sophisticated microstructural models including the kurtosis and NODDI model. Importantly, DIMOND reduces NODDI model fitting time from hours to minutes, or seconds by leveraging transfer learning. In summary, the self-supervised manner, high efficacy, and efficiency of DIMOND increase the practical feasibility and adoption of microstructure and connectivity mapping in clinical and neuroscientific applications.

FD-Net: An unsupervised deep forward-distortion model for susceptibility artifact correction in EPI

PURPOSE

To introduce an unsupervised deep-learning method for fast and effective correction of susceptibility artifacts in reversed phase-encode (PE) image pairs acquired with echo planar imaging (EPI).

METHODS

Recent learning-based correction approaches in EPI estimate a displacement field, unwarp the reversed-PE image pair with the estimated field, and average the unwarped pair to yield a corrected image. Unsupervised learning in these unwarping-based methods is commonly attained via a similarity constraint between the unwarped images in reversed-PE directions, neglecting consistency to the acquired EPI images. This work introduces a novel unsupervised deep Forward-Distortion Network (FD-Net) that predicts both the susceptibility-induced displacement field and the underlying anatomically correct image. Unlike previous methods, FD-Net enforces the forward-distortions of the correct image in both PE directions to be consistent with the acquired reversed-PE image pair. FD-Net further leverages a multiresolution architecture to maintain high local and global performance.

RESULTS

FD-Net performs competitively with a gold-standard reference method (TOPUP) in image quality, while enabling a leap in computational efficiency. Furthermore, FD-Net outperforms recent unwarping-based methods for unsupervised correction in terms of both image and field quality.

CONCLUSION

The unsupervised FD-Net method introduces a deep forward-distortion approach to enable fast, high-fidelity correction of susceptibility artifacts in EPI by maintaining consistency to measured data. Therefore, it holds great promise for improving the anatomical accuracy of EPI imaging.

Validation of deep learning techniques for quality augmentation in diffusion MRI for clinical studies

The objective of this study is to evaluate the efficacy of deep learning (DL) techniques in improving the quality of diffusion MRI (dMRI) data in clinical applications. The study aims to determine whether the use of artificial intelligence (AI) methods in medical images may result in the loss of critical clinical information and/or the appearance of false information. To assess this, the focus was on the angular resolution of dMRI and a clinical trial was conducted on migraine, specifically between episodic and chronic migraine patients. The number of gradient directions had an impact on white matter analysis results, with statistically significant differences between groups being drastically reduced when using 21 gradient directions instead of the original 61. Fourteen teams from different institutions were tasked to use DL to enhance three diffusion metrics (FA, AD and MD) calculated from data acquired with 21 gradient directions and a b-value of 1000 s/mm2. The goal was to produce results that were comparable to those calculated from 61 gradient directions. The results were evaluated using both standard image quality metrics and Tract-Based Spatial Statistics (TBSS) to compare episodic and chronic migraine patients. The study results suggest that while most DL techniques improved the ability to detect statistical differences between groups, they also led to an increase in false positive. The results showed that there was a constant growth rate of false positives linearly proportional to the new true positives, which highlights the risk of generalization of AI-based tasks when assessing diverse clinical cohorts and training using data from a single group. The methods also showed divergent performance when replicating the original distribution of the data and some exhibited significant bias. In conclusion, extreme caution should be exercised when using AI methods for harmonization or synthesis in clinical studies when processing heterogeneous data in clinical studies, as important information may be altered, even when global metrics such as structural similarity or peak signal-to-noise ratio appear to suggest otherwise.

High-fidelity mesoscale in-vivo diffusion MRI through gSlider-BUDA and circular EPI with S-LORAKS reconstruction

PURPOSE

To develop a high-fidelity diffusion MRI acquisition and reconstruction framework with reduced echo-train-length for less T2* image blurring compared to typical highly accelerated echo-planar imaging (EPI) acquisitions at sub-millimeter isotropic resolution.

METHODS

We first proposed a circular-EPI trajectory with partial Fourier sampling on both the readout and phase-encoding directions to minimize the echo-train-length and echo time. We then utilized this trajectory in an interleaved two-shot EPI acquisition with reversed phase-encoding polarity, to aid in the correction of off-resonance-induced image distortions and provide complementary k-space coverage in the missing partial Fourier regions. Using model-based reconstruction with structured low-rank constraint and smooth phase prior, we corrected the shot-to-shot phase variations across the two shots and recover the missing k-space data. Finally, we combined the proposed acquisition/reconstruction framework with an SNR-efficient RF-encoded simultaneous multi-slab technique, termed gSlider, to achieve high-fidelity 720 µm and 500 µm isotropic resolution in-vivo diffusion MRI.

RESULTS

Both simulation and in-vivo results demonstrate the effectiveness of the proposed acquisition and reconstruction framework to provide distortion-corrected diffusion imaging at the mesoscale with markedly reduced T2*-blurring. The in-vivo results of 720 µm and 500 µm datasets show high-fidelity diffusion images with reduced image blurring and echo time using the proposed approaches.

CONCLUSIONS

The proposed method provides high-quality distortion-corrected diffusion-weighted images with ∼40% reduction in the echo-train-length and T2* blurring at 500µm-isotropic-resolution compared to standard multi-shot EPI.

3D-EPI blip-up/down acquisition (BUDA) with CAIPI and joint Hankel structured low-rank reconstruction for rapid distortion-free high-resolution T 2 * mapping

PURPOSE

This work aims to develop a novel distortion-free 3D-EPI acquisition and image reconstruction technique for fast and robust, high-resolution, whole-brain imaging as well as quantitativeT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping.

METHODS

3D Blip-up and -down acquisition (3D-BUDA) sequence is designed for both single- and multi-echo 3D gradient recalled echo (GRE)-EPI imaging using multiple shots with blip-up and -down readouts to encode B0 field map information. Complementary k-space coverage is achieved using controlled aliasing in parallel imaging (CAIPI) sampling across the shots. For image reconstruction, an iterative hard-thresholding algorithm is employed to minimize the cost function that combines field map information informed parallel imaging with the structured low-rank constraint for multi-shot 3D-BUDA data. Extending 3D-BUDA to multi-echo imaging permitsT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping. For this, we propose constructing a joint Hankel matrix along both echo and shot dimensions to improve the reconstruction.

RESULTS

Experimental results on in vivo multi-echo data demonstrate that, by performing joint reconstruction along with both echo and shot dimensions, reconstruction accuracy is improved compared to standard 3D-BUDA reconstruction. CAIPI sampling is further shown to enhance image quality. ForT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping, parameter values from 3D-Joint-CAIPI-BUDA and reference multi-echo GRE are within limits of agreement as quantified by Bland-Altman analysis.

CONCLUSIONS

The proposed technique enables rapid 3D distortion-free high-resolution imaging andT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping. Specifically, 3D-BUDA enables 1-mm isotropic whole-brain imaging in 22 s at 3T and 9 s on a 7T scanner. The combination of multi-echo 3D-BUDA with CAIPI acquisition and joint reconstruction enables distortion-free whole-brainT 2 * $$ {\mathrm{T}}_2^{\ast } $$ mapping in 47 s at 1.1 × 1.1 × 1.0 mm3 resolution.

Deep learning prediction of diffusion MRI data with microstructure-sensitive loss functions

Deep learning prediction of diffusion MRI (DMRI) data relies on the utilization of effective loss functions. Existing losses typically measure the signal-wise differences between the predicted and target DMRI data without considering the quality of derived diffusion scalars that are eventually utilized for quantification of tissue microstructure. Here, we propose two novel loss functions, called microstructural loss and spherical variance loss, to explicitly consider the quality of both the predicted DMRI data and derived diffusion scalars. We apply these loss functions to the prediction of multi-shell data and enhancement of angular resolution. Evaluation based on infant and adult DMRI data indicates that both microstructural loss and spherical variance loss improve the quality of derived diffusion scalars.

ENSURE: A General Approach for Unsupervised Training of Deep Image Reconstruction Algorithms

Image reconstruction using deep learning algorithms offers improved reconstruction quality and lower reconstruction time than classical compressed sensing and model-based algorithms. Unfortunately, clean and fully sampled ground-truth data to train the deep networks is often unavailable in several applications, restricting the applicability of the above methods. We introduce a novel metric termed the ENsemble Stein's Unbiased Risk Estimate (ENSURE) framework, which can be used to train deep image reconstruction algorithms without fully sampled and noise-free images. The proposed framework is the generalization of the classical SURE and GSURE formulation to the setting where the images are sampled by different measurement operators, chosen randomly from a set. We evaluate the expectation of the GSURE loss functions over the sampling patterns to obtain the ENSURE loss function. We show that this loss is an unbiased estimate for the true mean-square error, which offers a better alternative to GSURE, which only offers an unbiased estimate for the projected error. Our experiments show that the networks trained with this loss function can offer reconstructions comparable to the supervised setting. While we demonstrate this framework in the context of MR image recovery, the ENSURE framework is generally applicable to arbitrary inverse problems.

Motion-robust, blood-suppressed, reduced-distortion diffusion MRI of the liver

PURPOSE

To evaluate feasibility and reproducibility of liver diffusion-weighted (DW) MRI using cardiac-motion-robust, blood-suppressed, reduced-distortion techniques.

METHODS

DW-MRI data were acquired at 3T in an anatomically accurate liver phantom including controlled pulsatile motion, in eight healthy volunteers and four patients with known or suspected liver metastases. Standard monopolar and motion-robust (M1-nulled, and M1-optimized) DW gradient waveforms were each acquired with single-shot echo-planar imaging (ssEPI) and multishot EPI (msEPI). In the motion phantom, apparent diffusion coefficient (ADC) was measured in the motion-affected volume. In healthy volunteers, ADC was measured in the left and right liver lobes separately to evaluate ADC reproducibility between the two lobes. Image distortions were quantified using the normalized cross-correlation coefficient, with an undistorted T2-weighted reference.

RESULTS

In the motion phantom, ADC mean and SD in motion-affected volumes substantially increased with increasing motion for monopolar waveforms. ADC remained stable in the presence of increasing motion when using motion-robust waveforms. M1-optimized waveforms suppressed slow flow signal present with M1-nulled waveforms. In healthy volunteers, monopolar waveforms generated significantly different ADC measurements between left and right liver lobes ( p = 0 . 0078 $$ p=0.0078 $$ , reproducibility coefficients (RPC) =  470 × 1 0 - 6 $$ 470\times 1{0}^{-6} $$ mm 2 $$ {}^2 $$ /s for monopolar-msEPI), while M1-optimized waveforms showed more reproducible ADC values ( p = 0 . 29 $$ p=0.29 $$ , RPC = 220 × 1 0 - 6 $$ \mathrm{RPC}=220\times 1{0}^{-6} $$ mm 2 $$ {}^2 $$ /s for M1-optimized-msEPI). In phantom and healthy volunteer studies, motion-robust acquisitions with msEPI showed significantly reduced image distortion ( p < 0 . 001 $$ p<0.001 $$ ) compared to ssEPI. Patient scans showed reduction of wormhole artifacts when combining M1-optimized waveforms with msEPI.

CONCLUSION

Synergistic effects of combined M1-optimized diffusion waveforms and msEPI acquisitions enable reproducible liver DWI with motion robustness, blood signal suppression, and reduced distortion.

High-resolution multi-shot diffusion-weighted MRI combining markerless prospective motion correction and locally low-rank constrained reconstruction

PURPOSE

Subject head motion is a major challenge in DWI, leading to image blurring, signal losses, and biases in the estimated diffusion parameters. Here, we investigate a combined application of prospective motion correction and spatial-angular locally low-rank constrained reconstruction to obtain robust, multi-shot, high-resolution diffusion-weighted MRI under substantial motion.

METHODS

Single-shot EPI with retrospective motion correction can mitigate motion artifacts and resolve any mismatching of gradient encoding orientations; however, it is limited by low spatial resolution and image distortions. Multi-shot acquisition strategies could achieve higher resolution and image fidelity but increase the vulnerability to motion artifacts and phase variations related to cardiac pulsations from shot to shot. We use prospective motion correction with optical markerless motion tracking to remove artifacts and reduce image blurring due to bulk motion, combined with locally low-rank regularization to correct for remaining artifacts due to shot-to-shot phase variations.

RESULTS

The approach was evaluated on healthy adult volunteers at 3 Tesla under different motion patterns. In multi-shot DWI, image blurring due to motion with 20 mm translations and 30° rotations was successfully removed by prospective motion correction, and aliasing artifacts caused by shot-to-shot phase variations were addressed by locally low-rank regularization. The ability of prospective motion correction to preserve the orientational information in DTI without requiring a reorientation of the b-matrix is highlighted.

CONCLUSION

The described technique is proved to hold valuable potential for mapping brain diffusivity and connectivity at high resolution for studies in subjects/cohorts where motion is common, including neonates, pediatrics, and patients with neurological disorders.

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.

Wave-Encoded Model-Based Deep Learning for Highly Accelerated Imaging with Joint Reconstruction

A recently introduced model-based deep learning (MoDL) technique successfully incorporates convolutional neural network (CNN)-based regularizers into physics-based parallel imaging reconstruction using a small number of network parameters. Wave-controlled aliasing in parallel imaging (CAIPI) is an emerging parallel imaging method that accelerates imaging acquisition by employing sinusoidal gradients in the phase- and slice/partition-encoding directions during the readout to take better advantage of 3D coil sensitivity profiles. We propose wave-encoded MoDL (wave-MoDL) combining the wave-encoding strategy with unrolled network constraints for highly accelerated 3D imaging while enforcing data consistency. We extend wave-MoDL to reconstruct multicontrast data with CAIPI sampling patterns to leverage similarity between multiple images to improve the reconstruction quality. We further exploit this to enable rapid quantitative imaging using an interleaved look-locker acquisition sequence with T₂ preparation pulse (3D-QALAS). Wave-MoDL enables a 40 s MPRAGE acquisition at 1 mm resolution at 16-fold acceleration. For quantitative imaging, wave-MoDL permits a 1:50 min acquisition for T₁, T₂, and proton density mapping at 1 mm resolution at 12-fold acceleration, from which contrast-weighted images can be synthesized as well. In conclusion, wave-MoDL allows rapid MR acquisition and high-fidelity image reconstruction and may facilitate clinical and neuroscientific applications by incorporating unrolled neural networks into wave-CAIPI reconstruction.

Highly accelerated EPI with wave encoding and multi-shot simultaneous multislice imaging

PURPOSE

To introduce wave-encoded acquisition and reconstruction techniques for highly accelerated EPI with reduced g-factor penalty and image artifacts.

THEORY AND METHODS

Wave-EPI involves application of sinusoidal gradients during the EPI readout, which spreads the aliasing in all spatial directions, thereby taking better advantage of 3D coil sensitivity profiles. The amount of voxel spreading that can be achieved by the wave gradients during the short EPI readout period is constrained by the slew rate of the gradient coils and peripheral nerve stimulation monitor. We propose to use a "half-cycle" sinusoidal gradient to increase the amount of voxel spreading that can be achieved while respecting the slew and stimulation constraints. Extending wave-EPI to multi-shot acquisition minimizes geometric distortion and voxel blurring at high in-plane resolutions, while structured low-rank regularization mitigates shot-to-shot phase variations. To address gradient imperfections, we propose to use different point spread functions for the k-space lines with positive and negative polarities, which are calibrated with a FLEET-based reference scan.

RESULTS

Wave-EPI enabled whole-brain single-shot gradient-echo (GE) and multi-shot spin-echo (SE) EPI acquisitions at high acceleration factors at 3T and was combined with g-Slider encoding to boost the SNR level in 1 mm isotropic diffusion imaging. Relative to blipped-CAIPI, wave-EPI reduced average and maximum g-factors by up to 1.21- and 1.37-fold at R_in  × R_sms  = 3 × 3, respectively.

CONCLUSION

Wave-EPI allows highly accelerated single- and multi-shot EPI with reduced g-factor and artifacts and may facilitate clinical and neuroscientific applications of EPI by improving the spatial and temporal resolution in functional and diffusion imaging.

Multiple B-Value Model-Based Residual Network (MORN) for Accelerated High-Resolution Diffusion-Weighted Imaging

Single-Shot Echo Planar Imaging (SSEPI) based Diffusion Weighted Imaging (DWI) has shortcomings such as low resolution and severe distortions. In contrast, Multi-Shot EPI (MSEPI) provides optimal spatial resolution but increases scan time. This study proposed a Multiple b-value mOdel-based Residual Network (MORN) model to reconstruct multiple b-value high-resolution DWI from undersampled k-space data simultaneously. We incorporated Parallel Imaging (PI) into a residual U-net to reconstruct multiple b-value multi-coil data with the supervision of MUltiplexed Sensitivity-Encoding (MUSE) reconstructed Multi-Shot DWI (MSDWI). Moreover, asymmetric concatenations among different b-values and the combined loss to back propagate helped the feature transfer. After training and validation of the MORN in a dataset of 32 healthy cases, additional assessments were performed on 6 patients with different tumor types. The experimental results demonstrated that the MORN model outperformed conventional PI reconstruction (i.e. SENSE) and two state-of-the-art deep learning methods (SENSE-GAN and VSNet) in terms of PSNR (Peak Signal-to-Noise Ratio), SSIM (Structual SIMilarity) and apparent diffusion coefficient maps. In addition, using the pre-trained model under DWI, the MORN achieved consistent fractional anisotrophy and mean diffusivity reconstructed from multiple diffusion directions. Hence, the proposed method shows potential in clinical application according to the observations on tumor patients as well as images of multiple diffusion directions.

Accelerating multi-echo chemical shift encoded water-fat MRI using model-guided deep learning

PURPOSE

To accelerate chemical shift encoded (CSE) water-fat imaging by applying a model-guided deep learning water-fat separation (MGDL-WF) framework to the undersampled k-space data.

METHODS

A model-guided deep learning water-fat separation framework is proposed for the acceleration using Cartesian/radial undersampling data. The proposed MGDL-WF combines the power of CSE water-fat imaging model and data-driven deep learning by jointly using a multi-peak fat model and a modified residual U-net network. The model is used to guide the image reconstruction, and the network is used to capture the artifacts induced by the undersampling. A data consistency layer is used in MGDL-WF to ensure the output images to be consistent with the k-space measurements. A Gauss-Newton iteration algorithm is adapted for the gradient updating of the networks.

RESULTS

Compared with the compressed sensing water-fat separation (CS-WF) algorithm/2-step procedure algorithm, the MGDL-WF increased peak signal-to-noise ratio (PSNR) by 5.31/5.23, 6.11/4.54, and 4.75 dB/1.88 dB with Cartesian sampling, and by 4.13/6.53, 2.90/4.68, and 1.68 dB/3.48 dB with radial sampling, at acceleration rates (R) of 4, 6, and 8, respectively. By using MGDL-WF, radial sampling increased the PSNR by 2.07 dB at R = 8, compared with Cartesian sampling.

CONCLUSIONS

The proposed MGDL-WF enables exploiting features of the water images and fat images from the undersampled multi-echo data, leading to improved performance in the accelerated CSE water-fat imaging. By using MGDL-WF, radial sampling can further improve the image quality with comparable scan time in comparison with Cartesian sampling.

Blip up-down acquisition for spin- and gradient-echo imaging (BUDA-SAGE) with self-supervised denoising enables efficient T2 , T2 *, para- and dia-magnetic susceptibility mapping

PURPOSE

To rapidly obtain high resolution T₂ , T₂ *, and quantitative susceptibility mapping (QSM) source separation maps with whole-brain coverage and high geometric fidelity.

METHODS

We propose Blip Up-Down Acquisition for Spin And Gradient Echo imaging (BUDA-SAGE), an efficient EPI sequence for quantitative mapping. The acquisition includes multiple T₂ *-, T₂ '-, and T₂ -weighted contrasts. We alternate the phase-encoding polarities across the interleaved shots in this multi-shot navigator-free acquisition. A field map estimated from interim reconstructions was incorporated into the joint multi-shot EPI reconstruction with a structured low rank constraint to eliminate distortion. A self-supervised neural network (NN), MR-Self2Self (MR-S2S), was used to perform denoising to boost SNR. Using Slider encoding allowed us to reach 1 mm isotropic resolution by performing super-resolution reconstruction on volumes acquired with 2 mm slice thickness. Quantitative T₂ (=1/R₂ ) and T₂ * (=1/R₂ *) maps were obtained using Bloch dictionary matching on the reconstructed echoes. QSM was estimated using nonlinear dipole inversion on the gradient echoes. Starting from the estimated R₂ /R₂ * maps, R₂ ' information was derived and used in source separation QSM reconstruction, which provided additional para- and dia-magnetic susceptibility maps.

RESULTS

In vivo results demonstrate the ability of BUDA-SAGE to provide whole-brain, distortion-free, high-resolution, multi-contrast images and quantitative T₂ /T₂ * maps, as well as yielding para- and dia-magnetic susceptibility maps. Estimated quantitative maps showed comparable values to conventional mapping methods in phantom and in vivo measurements.

CONCLUSION

BUDA-SAGE acquisition with self-supervised denoising and Slider encoding enables rapid, distortion-free, whole-brain T₂ /T₂ * mapping at 1 mm isotropic resolution under 90 s.

Improved unsupervised physics-informed deep learning for intravoxel incoherent motion modeling and evaluation in pancreatic cancer patients

PURPOSE

Earlier work showed that IVIM-NETorig , an unsupervised physics-informed deep neural network, was faster and more accurate than other state-of-the-art intravoxel-incoherent motion (IVIM) fitting approaches to diffusion-weighted imaging (DWI). This study presents a substantially improved version, IVIM-NEToptim , and characterizes its superior performance in pancreatic cancer patients.

METHOD

In simulations (signal-to-noise ratio [SNR] = 20), the accuracy, independence, and consistency of IVIM-NET were evaluated for combinations of hyperparameters (fit S0, constraints, network architecture, number of hidden layers, dropout, batch normalization, learning rate), by calculating the normalized root-mean-square error (NRMSE), Spearman's ρ, and the coefficient of variation (CVNET ), respectively. The best performing network, IVIM-NEToptim was compared to least squares (LS) and a Bayesian approach at different SNRs. IVIM-NEToptim 's performance was evaluated in an independent dataset of 23 patients with pancreatic ductal adenocarcinoma. Fourteen of the patients received no treatment between two repeated scan sessions and nine received chemoradiotherapy between the repeated sessions. Intersession within-subject standard deviations (wSD) and treatment-induced changes were assessed.

RESULTS

In simulations (SNR = 20), IVIM-NEToptim outperformed IVIM-NETorig in accuracy (NRMSE(D) = 0.177 vs 0.196; NMRSE(f) = 0.220 vs 0.267; NMRSE(D*) = 0.386 vs 0.393), independence (ρ(D*, f) = 0.22 vs 0.74), and consistency (CVNET (D) = 0.013 vs 0.104; CVNET (f) = 0.020 vs 0.054; CVNET (D*) = 0.036 vs 0.110). IVIM-NEToptim showed superior performance to the LS and Bayesian approaches at SNRs < 50. In vivo, IVIM-NEToptim showed significantly less noisy parameter maps with lower wSD for D and f than the alternatives. In the treated cohort, IVIM-NEToptim detected the most individual patients with significant parameter changes compared to day-to-day variations.

CONCLUSION

IVIM-NEToptim is recommended for accurate, informative, and consistent IVIM fitting to DWI data.

Axillary lymph node metastasis prediction by contrast-enhanced computed tomography images for breast cancer patients based on deep learning

When doctors use contrast-enhanced computed tomography (CECT) images to predict the metastasis of axillary lymph nodes (ALN) for breast cancer patients, the prediction performance could be degraded by subjective factors such as experience, psychological factors, and degree of fatigue. This study aims to exploit efficient deep learning schemes to predict the metastasis of ALN automatically via CECT images. A new construction called deformable sampling module (DSM) was meticulously designed as a plug-and-play sampling module in the proposed deformable attention VGG19 (DA-VGG19). A dataset of 800 samples labeled from 800 CECT images of 401 breast cancer patients retrospectively enrolled in the last three years was adopted to train, validate, and test the deep convolutional neural network models. By comparing the accuracy, positive predictive value, negative predictive value, sensitivity and specificity indices, the performance of the proposed model is analyzed in detail. The best-performing DA-VGG19 model achieved an accuracy of 0.9088, which is higher than that of other classification neural networks. As such, the proposed intelligent diagnosis algorithm can provide doctors with daily diagnostic assistance and advice and reduce the workload of doctors. The source code mentioned in this article will be released later.

Deep learning in magnetic resonance image reconstruction

Magnetic resonance (MR) imaging visualises soft tissue contrast in exquisite detail without harmful ionising radiation. In this work, we provide a state-of-the-art review on the use of deep learning in MR image reconstruction from different image acquisition types involving compressed sensing techniques, parallel image acquisition and multi-contrast imaging. Publications with deep learning-based image reconstruction for MR imaging were identified from the literature (PubMed and Google Scholar), and a comprehensive description of each of the works was provided. A detailed comparison that highlights the differences, the data used and the performance of each of these works were also made. A discussion of the potential use cases for each of these methods is provided. The sparse image reconstruction methods were found to be most popular in using deep learning for improved performance, accelerating acquisitions by around 4-8 times. Multi-contrast image reconstruction methods rely on at least one pre-acquired image, but can achieve 16-fold, and even up to 32- to 50-fold acceleration depending on the set-up. Parallel imaging provides frameworks to be integrated in many of these methods for additional speed-up potential. The successful use of compressed sensing techniques and multi-contrast imaging with deep learning and parallel acquisition methods could yield significant MR acquisition speed-ups within clinical routines in the near future.

Distortion-free, high-isotropic-resolution diffusion MRI with gSlider BUDA-EPI and multicoil dynamic B0 shimming

PURPOSE

We combine SNR-efficient acquisition and model-based reconstruction strategies with newly available hardware instrumentation to achieve distortion-free in vivo diffusion MRI of the brain at submillimeter-isotropic resolution with high fidelity and sensitivity on a clinical 3T scanner.

METHODS

We propose blip-up/down acquisition (BUDA) for multishot EPI using interleaved blip-up/blip-down phase encoding and incorporate B₀ forward-modeling into structured low-rank reconstruction to enable distortion-free and navigator-free diffusion MRI. We further combine BUDA-EPI with an SNR-efficient simultaneous multislab acquisition (generalized slice-dithered enhanced resolution ["gSlider"]), to achieve high-isotropic-resolution diffusion MRI. To validate gSlider BUDA-EPI, whole-brain diffusion data at 860-μm and 780-μm data sets were acquired. Finally, to improve the conditioning and minimize noise penalty in BUDA reconstruction at very high resolutions where B₀ inhomogeneity can have a detrimental effect, the level of B₀ inhomogeneity was reduced by incorporating slab-by-slab dynamic shimming with a 32-channel AC/DC coil into the acquisition. Whole-brain 600-μm diffusion data were then acquired with this combined approach of gSlider BUDA-EPI with dynamic shimming.

RESULTS

The results of 860-μm and 780-μm datasets show high geometry fidelity with gSlider BUDA-EPI. With dynamic shimming, the BUDA reconstruction's noise penalty was further alleviated. This enables whole-brain 600-μm isotropic resolution diffusion imaging with high image quality.

CONCLUSIONS

The gSlider BUDA-EPI method enables high-quality, distortion-free diffusion imaging across the whole brain at submillimeter resolution, where the use of multicoil dynamic B₀ shimming further improves reconstruction performance, which can be particularly useful at very high resolutions.

RUN-UP: Accelerated multishot diffusion-weighted MRI reconstruction using an unrolled network with U-Net as priors

PURPOSE

To accelerate and improve multishot diffusion-weighted MRI reconstruction using deep learning.

METHODS

An unrolled pipeline containing recurrences of model-based gradient updates and neural networks was introduced for accelerating multishot DWI reconstruction with shot-to-shot phase correction. The network was trained to predict results of jointly reconstructed multidirection data using single-direction data as input. In vivo brain and breast experiments were performed for evaluation.

RESULTS

The proposed method achieves a reconstruction time of 0.1 second per image, over 100-fold faster than a shot locally low-rank reconstruction. The resultant image quality is comparable to the target from the joint reconstruction with a peak signal-to-noise ratio of 35.3 dB, a normalized root-mean-square error of 0.0177, and a structural similarity index of 0.944. The proposed method also improves upon the locally low-rank reconstruction (2.9 dB higher peak signal-to-noise ratio, 29% lower normalized root-mean-square error, and 0.037 higher structural similarity index). With training data from the brain, this method also generalizes well to breast diffusion-weighted imaging, and fine-tuning further reduces aliasing artifacts.

CONCLUSION

A proposed data-driven approach enables almost real-time reconstruction with improved image quality, which improves the feasibility of multishot DWI in a wide range of clinical and neuroscientific studies.

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.

Accelerated Acquisition of High-resolution Diffusion-weighted Imaging of the Brain with a Multi-shot Echo-planar Sequence: Deep-learning-based Denoising

To accelerate high-resolution diffusion-weighted imaging with a multi-shot echo-planar sequence, we propose an approach based on reduced averaging and deep learning. Denoising convolutional neural networks can reduce amplified noise without requiring extensive averaging, enabling shorter scan times and high image quality. The preliminary experimental results demonstrate the superior performance of the proposed denoising method over state-of-the-art methods such as the widely used block-matching and 3D filtering.

Accelerated coronary MRI with sRAKI: A database-free self-consistent neural network k-space reconstruction for arbitrary undersampling

Purpose

To accelerate coronary MRI acquisitions with arbitrary undersampling patterns by using a novel reconstruction algorithm that applies coil self-consistency using subject-specific neural networks.

Methods

Self-consistent robust artificial-neural-networks for k-space interpolation (sRAKI) performs iterative parallel imaging reconstruction by enforcing self-consistency among coils. The approach bears similarity to SPIRiT, but extends the linear convolutions in SPIRiT to nonlinear interpolation using convolutional neural networks (CNNs). These CNNs are trained individually for each scan using the scan-specific autocalibrating signal (ACS) data. Reconstruction is performed by imposing the learned self-consistency and data-consistency, which enables sRAKI to support random undersampling patterns. Fully-sampled targeted right coronary artery MRI was acquired in six healthy subjects. The data were retrospectively undersampled, and reconstructed using SPIRiT, l₁-SPIRiT and sRAKI for acceleration rates of 2 to 5. Additionally, prospectively undersampled whole-heart coronary MRI was acquired to further evaluate reconstruction performance.

Results

sRAKI reduces noise amplification and blurring artifacts compared with SPIRiT and l₁-SPIRiT, especially at high acceleration rates in targeted coronary MRI. Quantitative analysis shows that sRAKI outperforms these techniques in terms of normalized mean-squared-error (~44% and ~21% over SPIRiT and l1-SPIRiT at rate 5) and vessel sharpness (~10% and ~20% over SPIRiT and l₁-SPIRiT at rate 5). Whole-heart data shows the sharpest coronary arteries when resolved using sRAKI, with 11% and 15% improvement in vessel sharpness over SPIRiT and l₁-SPIRiT, respectively.

Conclusion

sRAKI is a database-free neural network-based reconstruction technique that may further accelerate coronary MRI with arbitrary undersampling patterns, while improving noise resilience over linear parallel imaging and image sharpness over l₁ regularization techniques.