MoDL: Model-Based Deep Learning Architecture for Inverse Problems

DGEDDGAN: A dual-domain generator and edge-enhanced dual discriminator generative adversarial network for MRI reconstruction

Magnetic resonance imaging (MRI) as a critical clinical tool in medical imaging, requires a long scan time for producing high-quality MRI images. To accelerate the speed of MRI while reconstructing high-quality images with sharper edges and fewer aliases, a novel dual-domain generator and edge-enhancement dual discriminator generative adversarial network structure named DGEDDGAN for MRI reconstruction is proposed, in which one discriminator is responsible for holistic image reconstruction, whereas the other is adopted to enhance the edge preservation. A dual-domain U-Net structure that cascades the frequency domain and image domain is designed for the generator. The densely connected residual block is used to replace the traditional U-Net convolution block to improve the feature reuse capability while overcoming the gradient vanishing problem. The coordinate attention mechanism in each skip connection is employed to effectively reduce the loss of spatial information and enforce the feature selection capability. Extensive experiments on two publicly available datasets i.e., IXI dataset and CC-359, demonstrate that the proposed method can reconstruct the high-quality MRI images with more edge details and fewer artifacts, outperforming several state-of-the-art methods under various sampling rates and masks. The time of single-image reconstruction is below 13 ms, which meets the demand of faster processing.

MHWT: Wide-range attention modeling using window transformer for multi-modal MRI reconstruction

The Swin Transformer, with its window-based attention mechanism, demonstrates strong feature modeling capabilities. However, it struggles with high-resolution feature maps due to its fixed window size, particularly when capturing long-range dependencies in magnetic resonance image reconstruction tasks. To overcome this, we propose a novel multi-modal hybrid window attention Transformer (MHWT) that introduces a retractable attention mechanism combined with shape-alternating window design. This approach expands attention coverage while maintaining computational efficiency. Additionally, we employ a variable and shifted window attention strategy to model both local and global dependencies more flexibly. Improvements to the Transformer encoder, including adjustments to normalization and attention score computation, enhance training stability and reconstruction performance. Experimental results on multiple public datasets show that our method outperforms state-of-the-art approaches in both single-modal and multi-modal scenarios, demonstrating superior image reconstruction ability and adaptability. The code is publicly available at https://github.com/EnieHan/MHWT.

An extragradient and noise-tuning adaptive iterative network for diffusion MRI-based microstructural estimation

Diffusion MRI (dMRI) is a powerful technique for investigating tissue microstructure properties. However, advanced dMRI models are typically complex and nonlinear, requiring a large number of acquisitions in the q-space. Deep learning techniques, specifically optimization-based networks, have been proposed to improve the model fitting with limited q-space data. Previous optimization procedures relied on the empirical selection of iteration block numbers and the network structures were based on the iterative hard thresholding (IHT) algorithm, which may suffer from instability during sparse reconstruction. In this study, we introduced an extragradient and noise-tuning adaptive iterative network, a generic network for estimating dMRI model parameters. We proposed an adaptive mechanism that flexibly adjusts the sparse representation process, depending on specific dMRI models, datasets, and downsampling strategies, avoiding manual selection and accelerating inference. In addition, we proposed a noise-tuning module to assist the network in escaping from local minimum/saddle points. The network also included an additional projection of the extragradient to ensure its convergence. We evaluated the performance of the proposed network on the neurite orientation dispersion and density imaging (NODDI) model and diffusion basis spectrum imaging (DBSI) model on two 3T Human Connectome Project (HCP) datasets and a 7T HCP dataset with six different downsampling strategies. The proposed framework demonstrated superior accuracy and generalizability compared to other state-of-the-art microstructural estimation algorithms.

Joint coil sensitivity and motion correction in parallel MRI with a self-calibrating score-based diffusion model

Magnetic Resonance Imaging (MRI) stands as a powerful modality in clinical diagnosis. However, it faces challenges such as long acquisition time and vulnerability to motion-induced artifacts. While many existing motion correction algorithms have shown success, most fail to account for the impact of motion artifacts on coil sensitivity map (CSM) estimation during fast MRI reconstruction. This oversight can lead to significant performance degradation, as errors in the estimated CSMs can propagate and compromise motion correction. In this work, we propose JSMoCo, a novel method for jointly estimating motion parameters and time-varying coil sensitivity maps for under-sampled MRI reconstruction. The joint estimation presents a highly ill-posed inverse problem due to the increased number of unknowns. To address this challenge, we leverage score-based diffusion models as powerful priors and apply MRI physical principles to effectively constrain the solution space. Specifically, we parameterize rigid motion with trainable variables and model CSMs as polynomial functions. A Gibbs sampler is employed to ensure system consistency between the sensitivity maps and the reconstructed images, effectively preventing error propagation from pre-estimated sensitivity maps to the final reconstructed images. We evaluate JSMoCo through 2D and 3D motion correction experiments on simulated motion-corrupted fastMRI dataset and in-vivo real MRI brain scans. The results demonstrate that JSMoCo successfully reconstructs high-quality MRI images from under-sampled k-space data, achieving robust motion correction by accurately estimating time-varying coil sensitivities. The code is available at https://github.com/MeijiTian/JSMoCo.

JotlasNet: Joint tensor low-rank and attention-based sparse unrolling network for accelerating dynamic MRI

Joint low-rank and sparse unrolling networks have shown superior performance in dynamic MRI reconstruction. However, existing works mainly utilized matrix low-rank priors, neglecting the tensor characteristics of dynamic MRI images, and only a global threshold is applied for the sparse constraint to the multi-channel data, limiting the flexibility of the network. Additionally, most of them have inherently complex network structure, with intricate interactions among variables. In this paper, we propose a novel deep unrolling network, JotlasNet, for dynamic MRI reconstruction by jointly utilizing tensor low-rank and attention-based sparse priors. Specifically, we utilize tensor low-rank prior to exploit the structural correlations in high-dimensional data. Convolutional neural networks are used to adaptively learn the low-rank and sparse transform domains. A novel attention-based soft thresholding operator is proposed to assign a unique learnable threshold to each channel of the data in the CNN-learned sparse domain. The network is unrolled from the elaborately designed composite splitting algorithm and thus features a simple yet efficient parallel structure. Extensive experiments on two datasets (OCMR, CMRxRecon) demonstrate the superior performance of JotlasNet in dynamic MRI reconstruction.

AutoDPS: An unsupervised diffusion model based method for multiple degradation removal in MRI

BACKGROUND AND OBJECTIVE

Diffusion models have demonstrated their ability in image generation and solving inverse problems like restoration. Unlike most existing deep-learning based image restoration techniques which rely on unpaired or paired data for degradation awareness, diffusion models offer an unsupervised degradation independent alternative. This is well-suited in the context of restoring artifact-corrupted Magnetic Resonance Images (MRI), where it is impractical to exactly model the degradations apriori. In MRI, multiple corruptions arise, for instance, from patient movement compounded by undersampling artifacts from the acquisition settings.

METHODS

To tackle this scenario, we propose AutoDPS, an unsupervised method for corruption removal in brain MRI based on Diffusion Posterior Sampling. Our method (i) performs motion-related corruption parameter estimation using a blind iterative solver, and (ii) utilizes the knowledge of the undersampling pattern when the corruption consists of both motion and undersampling artifacts. We incorporate this corruption operation during sampling to guide the generation in recovering high-quality images.

RESULTS

Despite being trained to denoise and tested on completely unseen corruptions, our method AutoDPS has shown ∼ 1.63 dB of improvement in PSNR over baselines for realistic 3D motion restoration and ∼ 0.5 dB of improvement for random motion with undersampling. Additionally, our experiments demonstrate AutoDPS's resilience to noise and its generalization capability under domain shift, showcasing its robustness and adaptability.

CONCLUSION

In this paper, we propose an unsupervised method that removes multiple corruptions, mainly motion with undersampling, in MRI images which are essential for accurate diagnosis. The experiments show promising results on realistic and composite artifacts with higher improvement margins as compared to other methods. Our code is available at https://github.com/arunima101/AutoDPS/tree/master.

End-to-end deep learning-based motion correction and reconstruction for accelerated whole-heart joint T1/T2 mapping

PURPOSE

To accelerate 3D whole-heart joint T1/T2 mapping for myocardial tissue characterization using an end-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction of multi-contrast undersampled data.

METHODS

A free-breathing high-resolution motion-compensated 3D joint T1/T2 water/fat sequence is employed. The sequence consists of the acquisition of four interleaved volumes with 2-echo encoding, resulting in eight volumes with different contrasts. An end-to-end non-rigid motion-corrected reconstruction network is used to estimate high quality motion-corrected reconstructions from the eight multi-contrast undersampled data for subsequent joint T1/T2 mapping. Reconstruction with the proposed approach was compared against state-of-the-art motion-corrected HD-PROST reconstruction.

RESULTS

The proposed approach yields images with good visual agreement compared to the reference reconstructions. The comparison of the quantitative values in the T1 and T2 maps showed the absence of systematic errors, and a small bias of -6.35 ms and -1.8 ms, respectively. The proposed reconstruction time was 24 seconds in comparison to 2.5 hours with motion-corrected HD-PROST, resulting in a reconstruction speed-up of over 370 times.

CONCLUSION

In conclusion, this study presents a promising method for efficient whole-heart myocardial tissue characterization. Specifically, the research highlights the potential of the multi-contrast end-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction of multi-contrast undersampled data. The findings underscore its ability to compute T1 and T2 values with good agreement when compared to the reference motion-corrected HD-PROST method, while substantially reducing reconstruction time.

One for multiple: Physics-informed synthetic data boosts generalizable deep learning for fast MRI reconstruction

Magnetic resonance imaging (MRI) is a widely used radiological modality renowned for its radiation-free, comprehensive insights into the human body, facilitating medical diagnoses. However, the drawback of prolonged scan times hinders its accessibility. The k-space undersampling offers a solution, yet the resultant artifacts necessitate meticulous removal during image reconstruction. Although deep learning (DL) has proven effective for fast MRI image reconstruction, its broader applicability across various imaging scenarios has been constrained. Challenges include the high cost and privacy restrictions associated with acquiring large-scale, diverse training data, coupled with the inherent difficulty of addressing mismatches between training and target data in existing DL methodologies. Here, we present a novel Physics-Informed Synthetic data learning Framework for fast MRI, called PISF. PISF marks a breakthrough by enabling generalizable DL for multi-scenario MRI reconstruction through a single trained model. Our approach separates the reconstruction of a 2D image into many 1D basic problems, commencing with 1D data synthesis to facilitate generalization. We demonstrate that training DL models on synthetic data, coupled with enhanced learning techniques, yields in vivo MRI reconstructions comparable to or surpassing those of models trained on matched realistic datasets, reducing the reliance on real-world MRI data by up to 96 %. With a single trained model, our PISF supports the high-quality reconstruction under 4 sampling patterns, 5 anatomies, 6 contrasts, 5 vendors, and 7 centers, exhibiting remarkable generalizability. Its adaptability to 2 neuro and 2 cardiovascular patient populations has been validated through evaluations by 10 experienced medical professionals. In summary, PISF presents a feasible and cost-effective way to significantly boost the widespread adoption of DL in various fast MRI applications.

A critical assessment of artificial intelligence in magnetic resonance imaging of cancer

Given the enormous output and pace of development of artificial intelligence (AI) methods in medical imaging, it can be challenging to identify the true success stories to determine the state-of-the-art of the field. This report seeks to provide the magnetic resonance imaging (MRI) community with an initial guide into the major areas in which the methods of AI are contributing to MRI in oncology. After a general introduction to artificial intelligence, we proceed to discuss the successes and current limitations of AI in MRI when used for image acquisition, reconstruction, registration, and segmentation, as well as its utility for assisting in diagnostic and prognostic settings. Within each section, we attempt to present a balanced summary by first presenting common techniques, state of readiness, current clinical needs, and barriers to practical deployment in the clinical setting. We conclude by presenting areas in which new advances must be realized to address questions regarding generalizability, quality assurance and control, and uncertainty quantification when applying MRI to cancer to maintain patient safety and practical utility.

Deep learning initialized compressed sensing (Deli-CS) in volumetric spatio-temporal subspace reconstruction

OBJECT

Spatio-temporal MRI methods offer rapid whole-brain multi-parametric mapping, yet they are often hindered by prolonged reconstruction times or prohibitively burdensome hardware requirements. The aim of this project is to reduce reconstruction time using deep learning.

MATERIALS AND METHODS

This study focuses on accelerating the reconstruction of volumetric multi-axis spiral projection MRF, aiming for whole-brain T1 and T2 mapping, while ensuring a streamlined approach compatible with clinical requirements. To optimize reconstruction time, the traditional method is first revamped with a memory-efficient GPU implementation. Deep Learning Initialized Compressed Sensing (Deli-CS) is then introduced, which initiates iterative reconstruction with a DL-generated seed point, reducing the number of iterations needed for convergence.

RESULTS

The full reconstruction process for volumetric multi-axis spiral projection MRF is completed in just 20 min compared to over 2 h for the previously published implementation. Comparative analysis demonstrates Deli-CS's efficiency in expediting iterative reconstruction while maintaining high-quality results.

DISCUSSION

By offering a rapid warm start to the iterative reconstruction algorithm, this method substantially reduces processing time while preserving reconstruction quality. Its successful implementation paves the way for advanced spatio-temporal MRI techniques, addressing the challenge of extensive reconstruction times and ensuring efficient, high-quality imaging in a streamlined manner.

Rosette Trajectory MRI Reconstruction with Vision Transformers

INTRODUCTION

An efficient pipeline for rosette trajectory magnetic resonance imaging reconstruction is proposed, combining the inverse Fourier transform with a vision transformer (ViT) network enhanced with a convolutional layer. This method addresses the challenges of reconstructing high-quality images from non-Cartesian data by leveraging the ViT's ability to handle complex spatial dependencies without extensive preprocessing.

MATERIALS AND METHODS

The inverse fast Fourier transform provides a robust initial approximation, which is refined by the ViT network to produce high-fidelity images.

RESULTS AND DISCUSSION

This approach outperforms established deep learning techniques for normalized root mean squared error, peak signal-to-noise ratio, and entropy-based image quality scores; offers better runtime performance; and remains competitive with respect to other metrics.

FDuDoCLNet: Fully dual-domain contrastive learning network for parallel MRI reconstruction

Magnetic resonance imaging (MRI) is a non-invasive medical imaging technique that is widely used for high-resolution imaging of soft tissues and organs. However, the slow speed of MRI imaging, especially in high-resolution or dynamic scans, makes MRI reconstruction an important research topic. Currently, MRI reconstruction methods based on deep learning (DL) have garnered significant attention, and they improve the reconstruction quality by learning complex image features. However, DL-based MR image reconstruction methods exhibit certain limitations. First, the existing reconstruction networks seldom account for the diverse frequency features in the wavelet domain. Second, existing dual-domain reconstruction methods may pay too much attention to the features of a single domain (such as the global information in the image domain or the local details in the wavelet domain), resulting in the loss of either critical global structures or fine details in certain regions of the reconstructed image. In this work, inspired by the lifting scheme in wavelet theory, we propose a novel Fully Dual-Domain Contrastive Learning Network (FDuDoCLNet) based on variational networks (VarNet) for accelerating PI in both the image and wavelet domains. It is composed of several cascaded dual-domain regularization units and data consistency (DC) layers, in which a novel dual-domain contrastive loss is introduced to optimize the reconstruction performance effectively. The proposed FDuDoCLNet was evaluated on the publicly available fastMRI multi-coil knee dataset under a 6× acceleration factor, achieving a PSNR of 34.439 dB and a SSIM of 0.895.

Deep-ER: Deep Learning ECCENTRIC Reconstruction for fast high-resolution neurometabolic imaging

INTRODUCTION

Altered neurometabolism is an important pathological mechanism in many neurological diseases and brain cancer, which can be mapped non-invasively by Magnetic Resonance Spectroscopic Imaging (MRSI). Advanced MRSI using non-cartesian compressed-sense acquisition enables fast high-resolution metabolic imaging but has lengthy reconstruction times that limits throughput and needs expert user interaction. Here, we present a robust and efficient Deep Learning reconstruction embedded in a physical model within an end-to-end automated processing pipeline to obtain high-quality metabolic maps.

METHODS

Fast high-resolution whole-brain metabolic imaging was performed at 3.4 mm3 isotropic resolution with acquisition times between 4:11-9:21 min:s using ECCENTRIC pulse sequence on a 7T MRI scanner. Data were acquired in a high-resolution phantom and 27 human participants, including 22 healthy volunteers and 5 glioma patients. A deep neural network using recurring interlaced convolutional layers with joint dual-space feature representation was developed for deep learning ECCENTRIC reconstruction (Deep-ER). 21 subjects were used for training and 6 subjects for testing. Deep-ER performance was compared to iterative compressed sensing Total Generalized Variation reconstruction using image and spectral quality metrics.

RESULTS

Deep-ER demonstrated 600-fold faster reconstruction than conventional methods, providing improved spatial-spectral quality and metabolite quantification with 12%-45% (P<0.05) higher signal-to-noise and 8%-50% (P<0.05) smaller Cramer-Rao lower bounds. Metabolic images clearly visualize glioma tumor heterogeneity and boundary. Deep-ER generalizes reliably to unseen data.

CONCLUSION

Deep-ER provides efficient and robust reconstruction for sparse-sampled MRSI. The accelerated acquisition-reconstruction MRSI is compatible with high-throughput imaging workflow. It is expected that such improved performance will facilitate basic and clinical MRSI applications for neuroscience and precision medicine.

Spatial-frequency aware zero-centric residual unfolding network for MRI reconstruction

Magnetic Resonance Imaging is a cornerstone of medical diagnostics, providing high-quality soft tissue contrast through non-invasive methods. However, MRI technology faces critical limitations in imaging speed and resolution. Prolonged scan times not only increase patient discomfort but also contribute to motion artifacts, further compromising image quality. Compressed Sensing (CS) theory has enabled the acquisition of partial k-space data, which can then be effectively reconstructed to recover the original image using advanced reconstruction algorithms. Recently, deep learning has been widely applied to MRI reconstruction, aiming to reduce the artifacts in the image domain caused by undersampling in k-space and enhance image quality. As deep learning continues to evolve, the undersampling factors in k-space have gradually increased in recent years. However, these layers are limited in compensating for reconstruction errors in the unsampled areas, impeding further performance improvements. To address this, we propose a learnable spatial-frequency difference-aware module that complements the learnable data consistency layer, mapping k-space domain differences to the spatial image domain for perceptual compensation. Additionally, inspired by wavelet decomposition, we introduce explicit priors by decomposing images into mean and residual components, enforcing a refined zero-mean constraint on the residuals while maintaining computational efficiency. Comparative experiments on the FastMRI and Calgary-Campinas datasets demonstrate that our method achieves superior reconstruction performance against seven state-of-the-art techniques. Ablation studies further confirm the efficacy of our model's architecture, establishing a new pathway for enhanced MRI reconstruction.

Optimizing bladder magnetic resonance imaging: accelerating scan time and improving image quality through deep learning

PURPOSE

To investigate the value of deep learning (DL) in T2-weighted imaging (T2DL) of the bladder regarding acquisition time (TA), image quality, and diagnostic confidence compared to standard T2-weighted turbo-spin-echo (TSE) imaging (T2S).

METHODS

We prospectively enrolled a total of 28 consecutive patients for the evaluation of bladder cancer. T2S and T2DL sequences in three planes were performed for each participant, and acquisition time was compared between the two acquisition protocols. The image evaluation was conducted independently by two radiologists using a 5-point Likert scale for artifacts, noise, overall image quality, and diagnostic confidence, with 5 indicating the best quality. Additionally, T2 scoring based on Vesical Imaging-Reporting and Data System (VI-RADS) was performed by two readers.

RESULTS

Compared to T2S, the acquisition time of T2DL was reduced by 49.4% in the axial and by 43.8% in the coronal and sagittal orientations. The severity and impact of artifacts and noise levels were superior in T2DL versus T2S (both p < 0.05). The overall image quality in T2DL was demonstrated to be higher compared to that in T2S in axial and sagittal imaging (both p < 0.05). The diagnostic confidence and T2 scoring of both sequences in all planes did not differ (p > 0.05).

CONCLUSIONS

Our study preliminarily demonstrated the feasibility of T2-weighted imaging with DL reconstruction of bladder MR in clinical practice. T2DL achieved a reduction in acquisition time, superior lesion detectability, and overall image quality with similar diagnostic confidence and T2 score compared to the standard T2 TSE sequence.

Ensemble and low-frequency mixing with diffusion models for accelerated MRI reconstruction

Magnetic resonance imaging (MRI) is an important imaging modality in medical diagnosis, providing comprehensive anatomical information with detailed tissue structures. However, the long scan time required to acquire high-quality MR images is a major challenge, especially in urgent clinical scenarios. Although diffusion models have achieved remarkable performance in accelerated MRI, there are several challenges. In particular, they struggle with the long inference time due to the high number of iterations in the reverse process of diffusion models. Additionally, they occasionally create artifacts or 'hallucinate' tissues that do not exist in the original anatomy. To address these problems, we propose ensemble and adaptive low-frequency mixing on the diffusion model, namely ELF-Diff for accelerated MRI. The proposed method consists of three key components in the reverse diffusion step: (1) optimization based on unified data consistency; (2) low-frequency mixing; and (3) aggregation of multiple perturbations of the predicted images for the ensemble in each step. We evaluate ELF-Diff on two MRI datasets, FastMRI and SKM-TEA. ELF-Diff surpasses other existing diffusion models for MRI reconstruction. Furthermore, extensive experiments, including a subtask of pathology detection, further demonstrate the superior anatomical precision of our method. ELF-Diff outperforms the existing state-of-the-art MRI reconstruction methods without being limited to specific undersampling patterns.

The state-of-the-art in cardiac MRI reconstruction: Results of the CMRxRecon challenge in MICCAI 2023

Cardiac magnetic resonance imaging (MRI) provides detailed and quantitative evaluation of the heart's structure, function, and tissue characteristics with high-resolution spatial-temporal imaging. However, its slow imaging speed and motion artifacts are notable limitations. Undersampling reconstruction, especially data-driven algorithms, has emerged as a promising solution to accelerate scans and enhance imaging performance using highly under-sampled data. Nevertheless, the scarcity of publicly available cardiac k-space datasets and evaluation platform hinder the development of data-driven reconstruction algorithms. To address this issue, we organized the Cardiac MRI Reconstruction Challenge (CMRxRecon) in 2023, in collaboration with the 26th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI). CMRxRecon presented an extensive k-space dataset comprising cine and mapping raw data, accompanied by detailed annotations of cardiac anatomical structures. With overwhelming participation, the challenge attracted more than 285 teams and over 600 participants. Among them, 22 teams successfully submitted Docker containers for the testing phase, with 7 teams submitted for both cine and mapping tasks. All teams use deep learning based approaches, indicating that deep learning has predominately become a promising solution for the problem. The first-place winner of both tasks utilizes the E2E-VarNet architecture as backbones. In contrast, U-Net is still the most popular backbone for both multi-coil and single-coil reconstructions. This paper provides a comprehensive overview of the challenge design, presents a summary of the submitted results, reviews the employed methods, and offers an in-depth discussion that aims to inspire future advancements in cardiac MRI reconstruction models. The summary emphasizes the effective strategies observed in Cardiac MRI reconstruction, including backbone architecture, loss function, pre-processing techniques, physical modeling, and model complexity, thereby providing valuable insights for further developments in this field.

Deep Learning-Based Accelerated MR Cholangiopancreatography Without Fully-Sampled Data

The purpose of this study was to accelerate MR cholangiopancreatography (MRCP) acquisitions using deep learning-based (DL) reconstruction at 3 and 0.55 T. A total of 35 healthy volunteers underwent conventional twofold accelerated MRCP scans at field strengths of 3 and 0.55 T. We trained DL reconstructions using two different training strategies, supervised (SV) and self-supervised (SSV), with retrospectively sixfold undersampled data obtained at 3 T. We then evaluated the DL reconstructions against standard techniques, parallel imaging (PI) and compressed sensing (CS), focusing on peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) as metrics. We also tested DL reconstructions with prospectively accelerated acquisitions and evaluated their robustness when changing fields strengths from 3 to 0.55 T. DL reconstructions demonstrated a reduction in average acquisition time from 599/542 to 255/180 s for MRCP at 3 T/0.55 T. In both retrospective and prospective undersampling, PSNR and SSIM of DL reconstructions were higher than those of PI and CS. At the same time, DL reconstructions preserved the image quality of undersampled data, including sharpness and the visibility of hepatobiliary ducts. In addition, both DL approaches produced high-quality reconstructions at 0.55 T. In summary, DL reconstructions trained for highly accelerated MRCP enabled a reduction in acquisition time by a factor of 2.4/3.0 at 3 T/0.55 T while maintaining the image quality of conventional acquisitions.

Attention incorporated network for sharing low-rank, image and k-space information during MR image reconstruction to achieve single breath-hold cardiac Cine imaging

Cardiac Cine Magnetic Resonance Imaging (MRI) provides an accurate assessment of heart morphology and function in clinical practice. However, MRI requires long acquisition times, with recent deep learning-based methods showing great promise to accelerate imaging and enhance reconstruction quality. Existing networks exhibit some common limitations that constrain further acceleration possibilities, including single-domain learning, reliance on a single regularization term, and equal feature contribution. To address these limitations, we propose to embed information from multiple domains, including low-rank, image, and k-space, in a novel deep learning network for MRI reconstruction, which we denote as A-LIKNet. A-LIKNet adopts a parallel-branch structure, enabling independent learning in the k-space and image domain. Coupled information sharing layers realize the information exchange between domains. Furthermore, we introduce attention mechanisms into the network to assign greater weights to more critical coils or important temporal frames. Training and testing were conducted on an in-house dataset, including 91 cardiovascular patients and 38 healthy subjects scanned with 2D cardiac Cine using retrospective undersampling. Additionally, we evaluated A-LIKNet on the real-time prospectively undersampled data from the OCMR dataset. The results demonstrate that our proposed A-LIKNet outperforms existing methods and provides high-quality reconstructions. The network can effectively reconstruct highly retrospectively undersampled dynamic MR images up to 24× accelerations, indicating its potential for single breath-hold imaging.

DONet: Dual-Octave Network for Fast MR Image Reconstruction

Magnetic resonance (MR) image acquisition is an inherently prolonged process, whose acceleration has long been the subject of research. This is commonly achieved by obtaining multiple undersampled images, simultaneously, through parallel imaging. In this article, we propose the dual-octave network (DONet), which is capable of learning multiscale spatial-frequency features from both the real and imaginary components of MR data, for parallel fast MR image reconstruction. More specifically, our DONet consists of a series of dual-octave convolutions (Dual-OctConvs), which are connected in a dense manner for better reuse of features. In each Dual-OctConv, the input feature maps and convolutional kernels are first split into two components (i.e., real and imaginary) and then divided into four groups according to their spatial frequencies. Then, our Dual-OctConv conducts intragroup information updating and intergroup information exchange to aggregate the contextual information across different groups. Our framework provides three appealing benefits: 1) it encourages information interaction and fusion between the real and imaginary components at various spatial frequencies to achieve richer representational capacity; 2) the dense connections between the real and imaginary groups in each Dual-OctConv make the propagation of features more efficient by feature reuse; and 3) DONet enlarges the receptive field by learning multiple spatial-frequency features of both the real and imaginary components. Extensive experiments on two popular datasets (i.e., clinical knee and fastMRI), under different undersampling patterns and acceleration factors, demonstrate the superiority of our model in accelerated parallel MR image reconstruction.

Deep Equilibrium Unfolding Learning for Noise Estimation and Removal in Optical Molecular Imaging

In clinical optical molecular imaging, the need for real-time high frame rates and low excitation doses to ensure patient safety inherently increases susceptibility to detection noise. Faced with the challenge of image degradation caused by severe noise, image denoising is essential for mitigating the trade-off between acquisition cost and image quality. However, prevailing deep learning methods exhibit uncontrollable and suboptimal performance with limited interpretability, primarily due to neglecting underlying physical model and frequency information. In this work, we introduce an end-to-end model-driven Deep Equilibrium Unfolding Mamba (DEQ-UMamba) that integrates proximal gradient descent technique and learnt spatial-frequency characteristics to decouple complex noise structures into statistical distributions, enabling effective noise estimation and suppression in fluorescent images. Moreover, to address the computational limitations of unfolding networks, DEQ-UMamba trains an implicit mapping by directly differentiating the equilibrium point of the convergent solution, thereby ensuring stability and avoiding non-convergent behavior. With each network module aligned to a corresponding operation in the iterative optimization process, the proposed method achieves clear structural interpretability and strong performance. Comprehensive experiments conducted on both clinical and in vivo datasets demonstrate that DEQ-UMamba outperforms current state-of-the-art alternatives while utilizing fewer parameters, facilitating the advancement of cost-effective and high-quality clinical molecular imaging.

Accelerated cardiac cine with spatio-coil regularized deep learning reconstruction

PURPOSE

To develop an iterative deep learning (DL) reconstruction with spatio-coil regularization and multichannel k-space data consistency for accelerated cine imaging.

METHODS

This study proposes a Spatio-Coil Regularized DL (SCR-DL) approach for iterative deep learning reconstruction incorporating multicoil information in data consistency and regularizer. SCR-DL uses shift-invariant convolutional kernels to interpolate missing k-space lines and reconstruct individual coil images, followed by a regularizer that operates simultaneously across spatial and coil dimensions using learned image priors. At 8-fold acceleration, SCR-DL was compared with Generalized Autocalibrating Partially Parallel Acquisition (GRAPPA), sensitivity encoding (SENSE)-based DL and spatio-temporal regularized (STR)-DL reconstruction. In the retrospective undersampled cine, images were quantitatively evaluated using normalized mean square error (NMSE) and structural similarity index measure (SSIM). Additionally, agreement for left-ventricular ejection fraction and left-ventricular mass were assessed using prospectively accelerated cine images at 2-fold and 8-fold accelerations.

RESULTS

The SCR-DL algorithm successfully reconstructed highly accelerated cine images. SCR-DL had significant improvements in NMSE (0.03 ± 0.02) and SSIM (91.4% ± 2.7%) compared with GRAPPA (NMSE: 0.09 ± 0.04, SSIM: 69.9% ± 11.1%; p < 0.001), SENSE-DL (NMSE: 0.07 ± 0.04, SSIM: 86.9% ± 3.2%; p < 0.001), and STR-DL (NMSE: 0.04 ± 0.03, SSIM: 90.0% ± 2.5%; p < 0.001) with retrospective undersampled cine. Despite the 3-fold reduction in scan time, there was no difference between left-ventricular ejection fraction (59.8 ± 4.5 vs. 60.8 ± 4.8, p = 0.46) or left-ventricular mass (73.6 ± 19.4 g vs. 73.2 ± 19.7 g, p = 0.95) between R = 2 and R = 8 prospectively accelerated cine images.

CONCLUSIONS

SCR-DL enabled highly accelerated cardiac cine imaging, significantly reducing breath-hold time. Compared with GRAPPA or SENSE-DL, images reconstructed with SCR-DL showed superior NMSE and SSIM.

Deep learning-based Intraoperative MRI reconstruction

BACKGROUND

We retrospectively evaluated the quality of deep learning (DL) reconstructions of on-scanner accelerated intraoperative MRI (iMRI) during respective brain tumor surgery.

METHODS

Accelerated iMRI was performed using dual surface coils positioned around the area of resection. A DL model was trained on the fastMRI neuro dataset to mimic the data from the iMRI protocol. The evaluation was performed on imaging material from 40 patients imaged from Nov 1, 2021, to June 1, 2023, who underwent iMRI during tumor resection surgery. A comparative analysis was conducted between the conventional compressed sense (CS) method and the trained DL reconstruction method. Blinded evaluation of multiple image quality metrics was performed by two neuroradiologists and one neurosurgeon using a 1-to-5 Likert scale (1, nondiagnostic; 2, poor; 3, acceptable; 4, good; and 5, excellent), and the favored reconstruction variant.

RESULTS

The DL reconstruction was strongly favored or favored over the CS reconstruction for 33/40, 39/40, and 8/40 of cases for readers 1, 2, and 3, respectively. For the evaluation metrics, the DL reconstructions had a higher score than their respective CS counterparts for 72%, 72%, and 14% of the cases for readers 1, 2, and 3, respectively. Still, the DL reconstructions exhibited shortcomings such as a striping artifact and reduced signal.

CONCLUSION

DL shows promise in allowing for high-quality reconstructions of iMRI. The neuroradiologists noted an improvement in the perceived spatial resolution, signal-to-noise ratio, diagnostic confidence, diagnostic conspicuity, and spatial resolution compared to CS, while the neurosurgeon preferred the CS reconstructions across all metrics.

RELEVANCE STATEMENT

DL shows promise to allow for high-quality reconstructions of iMRI, however, due to the challenging setting of iMRI, further optimization is needed.

KEY POINTS

iMRI is a surgical tool with a challenging image setting. DL allowed for high-quality reconstructions of iMRI. Additional optimization is needed due to the challenging intraoperative setting.

GroupCDL: Interpretable Denoising and Compressed Sensing MRI via Learned Group-Sparsity and Circulant Attention

Nonlocal self-similarity within images has become an increasingly popular prior in deep-learning models. Despite their successful image restoration performance, such models remain largely uninterpretable due to their black-box construction. Our previous studies have shown that interpretable construction of a fully convolutional denoiser (CDLNet), with performance on par with state-of-the-art black-box counterparts, is achievable by unrolling a convolutional dictionary learning algorithm. In this manuscript, we seek an interpretable construction of a convolutional network with a nonlocal self-similarity prior that performs on par with black-box nonlocal models. We show that such an architecture can be effectively achieved by up-grading theℓ 1 sparsity prior (soft-thresholding) of CDLNet to an image-adaptive group-sparsity prior (group-thresholding). The proposed learned group-thresholding makes use of nonlocal attention to perform spatially varying soft-thresholding on the latent representation. To enable effective training and inference on large images with global artifacts, we propose a novel circulant-sparse attention. We achieve competitive natural-image denoising performance compared to black-box nonlocal DNNs and transformers. The interpretable construction of our network allows for a straightforward extension to Compressed Sensing MRI (CS-MRI), yielding state-of-the-art performance. Lastly, we show robustness to noise-level mismatches between training and inference for denoising and CS-MRI reconstruction.

Advancing MRI Reconstruction: A Systematic Review of Deep Learning and Compressed Sensing Integration

Magnetic resonance imaging (MRI) is a non-invasive imaging modality and provides comprehensive anatomical and functional insights into the human body. However, its long acquisition times can lead to patient discomfort, motion artifacts, and limiting real-time applications. To address these challenges, strategies such as parallel imaging have been applied, which utilize multiple receiver coils to speed up the data acquisition process. Additionally, compressed sensing (CS) is a method that facilitates image reconstruction from sparse data, significantly reducing image acquisition time by minimizing the amount of data collection needed. Recently, deep learning (DL) has emerged as a powerful tool for improving MRI reconstruction. It has been integrated with parallel imaging and CS principles to achieve faster and more accurate MRI reconstructions. This review comprehensively examines DL-based techniques for MRI reconstruction. We categorize and discuss various DL-based methods, including end-to-end approaches, unrolled optimization, and federated learning, highlighting their potential benefits. Our systematic review highlights significant contributions and underscores the potential of DL in MRI reconstruction. Additionally, we summarize key results and trends in DL-based MRI reconstruction, including quantitative metrics, the dataset, acceleration factors, and the progress of and research interest in DL techniques over time. Finally, we discuss potential future directions and the importance of DL-based MRI reconstruction in advancing medical imaging. To facilitate further research in this area, we provide a GitHub repository that includes up-to-date DL-based MRI reconstruction publications and public datasets-https://github.com/mosaf/Awesome-DL-based-CS-MRI.

SPIRiT-Diffusion: Self-Consistency Driven Diffusion Model for Accelerated MRI

Diffusion models have emerged as a leading methodology for image generation and have proven successful in the realm of magnetic resonance imaging (MRI) reconstruction. However, existing reconstruction methods based on diffusion models are primarily formulated in the image domain, making the reconstruction quality susceptible to inaccuracies in coil sensitivity maps (CSMs). k-space interpolation methods can effectively address this issue but conventional diffusion models are not readily applicable in k-space interpolation. To overcome this challenge, we introduce a novel approach called SPIRiT-Diffusion, which is a diffusion model for k-space interpolation inspired by the iterative self-consistent SPIRiT method. Specifically, we utilize the iterative solver of the self-consistent term (i.e., k-space physical prior) in SPIRiT to formulate a novel stochastic differential equation (SDE) governing the diffusion process. Subsequently, k-space data can be interpolated by executing the diffusion process. This innovative approach highlights the optimization model's role in designing the SDE in diffusion models, enabling the diffusion process to align closely with the physics inherent in the optimization model-a concept referred to as model-driven diffusion. We evaluated the proposed SPIRiT-Diffusion method using a 3D joint intracranial and carotid vessel wall imaging dataset. The results convincingly demonstrate its superiority over image-domain reconstruction methods, achieving high reconstruction quality even at a substantial acceleration rate of 10. Our code are available at https://github.com/zhyjSIAT/SPIRiT-Diffusion.

MRI recovery with self-calibrated denoisers without fully-sampled data

OBJECTIVE

Acquiring fully sampled training data is challenging for many MRI applications. We present a self-supervised image reconstruction method, termed ReSiDe, capable of recovering images solely from undersampled data.

MATERIALS AND METHODS

ReSiDe is inspired by plug-and-play (PnP) methods, but unlike traditional PnP approaches that utilize pre-trained denoisers, ReSiDe iteratively trains the denoiser on the image or images that are being reconstructed. We introduce two variations of our method: ReSiDe-S and ReSiDe-M. ReSiDe-S is scan-specific and works with a single set of undersampled measurements, while ReSiDe-M operates on multiple sets of undersampled measurements and provides faster inference. Studies I, II, and III compare ReSiDe-S and ReSiDe-M against other self-supervised or unsupervised methods using data from T1- and T2-weighted brain MRI, MRXCAT digital perfusion phantom, and first-pass cardiac perfusion, respectively.

RESULTS

ReSiDe-S and ReSiDe-M outperform other methods in terms of peak signal-to-noise ratio and structural similarity index measure for Studies I and II, and in terms of expert scoring for Study III.

DISCUSSION

We present a self-supervised image reconstruction method and validate it in both static and dynamic MRI applications. These developments can benefit MRI applications where the availability of fully sampled training data is limited.

Accelerating multi-coil MR image reconstruction using weak supervision

Deep-learning-based MR image reconstruction in settings where large fully sampled dataset collection is infeasible requires methods that effectively use both under-sampled and fully sampled datasets. This paper evaluates a weakly supervised, multi-coil, physics-guided approach to MR image reconstruction, leveraging both dataset types, to improve both the quality and robustness of reconstruction. A physics-guided end-to-end variational network (VarNet) is pretrained in a self-supervised manner using a 4 × under-sampled dataset following the self-supervised learning via data undersampling (SSDU) methodology. The pre-trained weights are transferred to another VarNet, which is fine-tuned using a smaller, fully sampled dataset by optimizing multi-scale structural similarity (MS-SSIM) loss in image space. The proposed methodology is compared with fully self-supervised and fully supervised training. Reconstruction quality improvements in SSIM, PSNR, and NRMSE when abundant training data is available (the high-data regime), and enhanced robustness when training data is scarce (the low-data regime) are demonstrated using weak supervision for knee and brain MR image reconstructions at 8 × and 10 × acceleration, respectively. Multi-coil physics-guided MR image reconstruction using both under-sampled and fully sampled datasets is achievable with transfer learning and fine-tuning. This methodology can provide improved reconstruction quality in the high-data regime and improved robustness in the low-data regime at high acceleration rates.

Highly accelerated MRI via implicit neural representation guided posterior sampling of diffusion models

Reconstructing high-fidelity magnetic resonance (MR) images from under-sampled k-space is a commonly used strategy to reduce scan time. The posterior sampling of diffusion models based on the real measurement data holds significant promise of improved reconstruction accuracy. However, traditional posterior sampling methods often lack effective data consistency guidance, leading to inaccurate and unstable reconstructions. Implicit neural representation (INR) has emerged as a powerful paradigm for solving inverse problems by modeling a signal's attributes as a continuous function of spatial coordinates. In this study, we present a novel posterior sampler for diffusion models using INR, named DiffINR. The INR-based component incorporates both the diffusion prior distribution and the MRI physical model to ensure high data fidelity. DiffINR demonstrates superior performance on in-distribution datasets with remarkable accuracy, even under high acceleration factors (up to R = 12 in single-channel reconstruction). Furthermore, DiffINR exhibits excellent generalizability across various tissue contrasts and anatomical structures with low uncertainty. Overall, DiffINR significantly improves MRI reconstruction in terms of accuracy, generalizability and stability, paving the way for further accelerating MRI acquisition. Notably, our proposed framework can be a generalizable framework to solve inverse problems in other medical imaging tasks.

Deep learning for efficient reconstruction of highly accelerated 3D FLAIR MRI in neurological deficits

OBJECTIVE

To compare compressed sensing (CS) and the Cascades of Independently Recurrent Inference Machines (CIRIM) with respect to image quality and reconstruction times when 12-fold accelerated scans of patients with neurological deficits are reconstructed.

MATERIALS AND METHODS

Twelve-fold accelerated 3D T2-FLAIR images were obtained from a cohort of 62 patients with neurological deficits on 3 T MRI. Images were reconstructed offline via CS and the CIRIM. Image quality was assessed in a blinded and randomized manner by two experienced interventional neuroradiologists and one experienced pediatric neuroradiologist on imaging artifacts, perceived spatial resolution (sharpness), anatomic conspicuity, diagnostic confidence, and contrast. The methods were also compared in terms of self-referenced quality metrics, image resolution, patient groups and reconstruction time. In ten scans, the contrast ratio (CR) was determined between lesions and white matter. The effect of acceleration factor was assessed in a publicly available fully sampled dataset, since ground truth data are not available in prospectively accelerated clinical scans. Specifically, 451 FLAIR scans, including scans with white matter lesions, were adopted from the FastMRI database to evaluate structural similarity (SSIM) and the CR of lesions and white matter on ranging acceleration factors from four-fold up to 12-fold.

RESULTS

Interventional neuroradiologists significantly preferred the CIRIM for imaging artifacts, anatomic conspicuity, and contrast. One rater significantly preferred the CIRIM in terms of sharpness and diagnostic confidence. The pediatric neuroradiologist preferred CS for imaging artifacts and sharpness. Compared to CS, the CIRIM reconstructions significantly improved in terms of imaging artifacts and anatomic conspicuity (p < 0.01) for higher resolution scans while yielding a 28% higher SNR (p = 0.001) and a 5.8% lower CR (p = 0.04). There were no differences between patient groups. Additionally, CIRIM was five times faster than CS was. An increasing acceleration factor did not lead to changes in CR (p = 0.92), but led to lower SSIM (p = 0.002).

DISCUSSION

Patients with neurological deficits can undergo MRI at a range of moderate to high acceleration. DL reconstruction outperforms CS in terms of image resolution, efficient denoising with a modest reduction in contrast and reduced reconstruction times.

Data- and Physics-Driven Deep Learning Based Reconstruction for Fast MRI: Fundamentals and Methodologies

Magnetic Resonance Imaging (MRI) is a pivotal clinical diagnostic tool, yet its extended scanning times often compromise patient comfort and image quality, especially in volumetric, temporal and quantitative scans. This review elucidates recent advances in MRI acceleration via data and physics-driven models, leveraging techniques from algorithm unrolling models, enhancement-based methods, and plug-and-play models to the emerging full spectrum of generative model-based methods. We also explore the synergistic integration of data models with physics-based insights, encompassing the advancements in multi-coil hardware accelerations like parallel imaging and simultaneous multi-slice imaging, and the optimization of sampling patterns. We then focus on domain-specific challenges and opportunities, including image redundancy exploitation, image integrity, evaluation metrics, data heterogeneity, and model generalization. This work also discusses potential solutions and future research directions, with an emphasis on the role of data harmonization and federated learning for further improving the general applicability and performance of these methods in MRI reconstruction.

QRAGE-Simultaneous multiparametric quantitative MRI of water content, T1, T2*, and magnetic susceptibility at ultrahigh field strength

PURPOSE

To introduce quantitative rapid gradient-echo (QRAGE), a novel approach for the simultaneous mapping of multiple quantitative MRI parameters, including water content, T1, T2*, and magnetic susceptibility at ultrahigh field strength.

METHODS

QRAGE leverages a newly developed multi-echo MPnRAGE sequence, facilitating the acquisition of 171 distinct contrast images across a range of TI and TE points. To maintain a short acquisition time, we introduce MIRAGE2, a novel model-based reconstruction method that exploits prior knowledge of temporal signal evolution, represented as damped complex exponentials. MIRAGE2 minimizes local Block-Hankel and Casorati matrices. Parameter maps are derived from the reconstructed contrast images through postprocessing steps. We validate QRAGE through extensive simulations, phantom studies, and in vivo experiments, demonstrating its capability for high-precision imaging.

RESULTS

In vivo brain measurements show the promising performance of QRAGE, with test-retest SDs and deviations from reference methods of < 0.8% for water content, < 17 ms for T1, and < 0.7 ms for T2*. QRAGE achieves whole-brain coverage at a 1-mm isotropic resolution in just 7 min and 15 s, comparable to the acquisition time of an MP2RAGE scan. In addition, QRAGE generates a contrast image akin to the UNI image produced by MP2RAGE.

CONCLUSION

QRAGE is a new, successful approach for simultaneously mapping multiple MR parameters at ultrahigh field.

Deep unfolding network with spatial alignment for multi-modal MRI reconstruction

Multi-modal Magnetic Resonance Imaging (MRI) offers complementary diagnostic information, but some modalities are limited by the long scanning time. To accelerate the whole acquisition process, MRI reconstruction of one modality from highly under-sampled k-space data with another fully-sampled reference modality is an efficient solution. However, the misalignment between modalities, which is common in clinic practice, can negatively affect reconstruction quality. Existing deep learning-based methods that account for inter-modality misalignment perform better, but still share two main common limitations: (1) The spatial alignment task is not adaptively integrated with the reconstruction process, resulting in insufficient complementarity between the two tasks; (2) the entire framework has weak interpretability. In this paper, we construct a novel Deep Unfolding Network with Spatial Alignment, termed DUN-SA, to appropriately embed the spatial alignment task into the reconstruction process. Concretely, we derive a novel joint alignment-reconstruction model with a specially designed aligned cross-modal prior term. By relaxing the model into cross-modal spatial alignment and multi-modal reconstruction tasks, we propose an effective algorithm to solve this model alternatively. Then, we unfold the iterative stages of the proposed algorithm and design corresponding network modules to build DUN-SA with interpretability. Through end-to-end training, we effectively compensate for spatial misalignment using only reconstruction loss, and utilize the progressively aligned reference modality to provide inter-modality prior to improve the reconstruction of the target modality. Comprehensive experiments on four real datasets demonstrate that our method exhibits superior reconstruction performance compared to state-of-the-art methods.

Fast and Reliable Score-Based Generative Model for Parallel MRI

The score-based generative model (SGM) can generate high-quality samples, which have been successfully adopted for magnetic resonance imaging (MRI) reconstruction. However, the recent SGMs may take thousands of steps to generate a high-quality image. Besides, SGMs neglect to exploit the redundancy in space. To overcome the above two drawbacks, in this article, we propose a fast and reliable SGM (FRSGM). First, we propose deep ensemble denoisers (DEDs) consisting of SGM and the deep denoiser, which are used to solve the proximal problem of the implicit regularization term. Second, we propose a spatially adaptive self-consistency (SASC) term as the regularization term of the -space data. We use the alternating direction method of multipliers (ADMM) algorithm to solve the minimization model of compressed sensing (CS)-MRI incorporating the image prior term and the SASC term, which is significantly faster than the related works based on SGM. Meanwhile, we can prove that the iterating sequence of the proposed algorithm has a unique fixed point. In addition, the DED and the SASC term can significantly improve the generalization ability of the algorithm. The features mentioned above make our algorithm reliable, including the fixed-point convergence guarantee, the exploitation of the space, and the powerful generalization ability.

Generalizable Reconstruction for Accelerating MR Imaging via Federated Learning With Neural Architecture Search

Heterogeneous data captured by different scanning devices and imaging protocols can affect the generalization performance of the deep learning magnetic resonance (MR) reconstruction model. While a centralized training model is effective in mitigating this problem, it raises concerns about privacy protection. Federated learning is a distributed training paradigm that can utilize multi-institutional data for collaborative training without sharing data. However, existing federated learning MR image reconstruction methods rely on models designed manually by experts, which are complex and computationally expensive, suffering from performance degradation when facing heterogeneous data distributions. In addition, these methods give inadequate consideration to fairness issues, namely ensuring that the model's training does not introduce bias towards any specific dataset's distribution. To this end, this paper proposes a generalizable federated neural architecture search framework for accelerating MR imaging (GAutoMRI). Specifically, automatic neural architecture search is investigated for effective and efficient neural network representation learning of MR images from different centers. Furthermore, we design a fairness adjustment approach that can enable the model to learn features fairly from inconsistent distributions of different devices and centers, and thus facilitate the model to generalize well to the unseen center. Extensive experiments show that our proposed GAutoMRI has better performances and generalization ability compared with seven state-of-the-art federated learning methods. Moreover, the GAutoMRI model is significantly more lightweight, making it an efficient choice for MR image reconstruction tasks. The code will be made available at https://github.com/ternencewu123/GAutoMRI.

Fourier Convolution Block with global receptive field for MRI reconstruction

Reconstructing images from under-sampled Magnetic Resonance Imaging (MRI) signals significantly reduces scan time and improves clinical practice. However, Convolutional Neural Network (CNN)-based methods, while demonstrating great performance in MRI reconstruction, may face limitations due to their restricted receptive field (RF), hindering the capture of global features. This is particularly crucial for reconstruction, as aliasing artifacts are distributed globally. Recent advancements in Vision Transformers have further emphasized the significance of a large RF. In this study, we proposed a novel global Fourier Convolution Block (FCB) with whole image RF and low computational complexity by transforming the regular spatial domain convolutions into frequency domain. Visualizations of the effective RF and trained kernels demonstrated that FCB improves the RF of reconstruction models in practice. The proposed FCB was evaluated on four popular CNN architectures using brain and knee MRI datasets. Models with FCB achieved superior PSNR and SSIM than baseline models and exhibited more details and texture recovery. The code is publicly available at https://github.com/Haozhoong/FCB.

An improved low-rank plus sparse unrolling network method for dynamic magnetic resonance imaging

BACKGROUND

Recent advances in deep learning have sparked new research interests in dynamic magnetic resonance imaging (MRI) reconstruction. However, existing deep learning-based approaches suffer from insufficient reconstruction efficiency and accuracy due to the lack of time correlation modeling during the reconstruction procedure.

PURPOSE

Inappropriate tensor processing steps and deep learning models may lead to not only a lack of modeling in the time dimension but also an increase in the overall size of the network. Therefore, this study aims to find suitable tensor processing methods and deep learning models to achieve better reconstruction results and a smaller network size.

METHODS

We propose a novel unrolling network method that enhances the reconstruction quality and reduces the parameter redundancy by introducing time correlation modeling into MRI reconstruction with low-rank core matrix and convolutional long short-term memory (ConvLSTM) unit.

RESULTS

We conduct extensive experiments on AMRG Cardiac MRI dataset to evaluate our proposed approach. The results demonstrate that compared to other state-of-the-art approaches, our approach achieves higher peak signal-to-noise ratios and structural similarity indices at different accelerator factors with significantly fewer parameters.

CONCLUSIONS

The improved reconstruction performance demonstrates that our proposed time correlation modeling is simple and effective for accelerating MRI reconstruction. We hope our approach can serve as a reference for future research in dynamic MRI reconstruction.

AutoSamp: Autoencoding k-Space Sampling via Variational Information Maximization for 3D MRI

Accelerated MRI protocols routinely involve a predefined sampling pattern that undersamples the k-space. Finding an optimal pattern can enhance the reconstruction quality, however this optimization is a challenging task. To address this challenge, we introduce a novel deep learning framework, AutoSamp, based on variational information maximization that enables joint optimization of sampling pattern and reconstruction of MRI scans. We represent the encoder as a non-uniform Fast Fourier Transform that allows continuous optimization of k-space sample locations on a non-Cartesian plane, and the decoder as a deep reconstruction network. Experiments on public 3D acquired MRI datasets show improved reconstruction quality of the proposed AutoSamp method over the prevailing variable density and variable density Poisson disc sampling for both compressed sensing and deep learning reconstructions. We demonstrate that our data-driven sampling optimization method achieves 4.4dB, 2.0dB, 0.75dB, 0.7dB PSNR improvements over reconstruction with Poisson Disc masks for acceleration factors of R =5, 10, 15, 25, respectively. Prospectively accelerated acquisitions with 3D FSE sequences using our optimized sampling patterns exhibit improved image quality and sharpness. Furthermore, we analyze the characteristics of the learned sampling patterns with respect to changes in acceleration factor, measurement noise, underlying anatomy, and coil sensitivities. We show that all these factors contribute to the optimization result by affecting the sampling density, k-space coverage and point spread functions of the learned sampling patterns.

CineVN: Variational network reconstruction for rapid functional cardiac cine MRI

PURPOSE

To develop a reconstruction method for highly accelerated cardiac cine MRI with high spatiotemporal resolution and low temporal blurring, and to demonstrate accurate estimation of ventricular volumes and myocardial strain in healthy subjects and in patients.

METHODS

The proposed method, called CineVN, employs a spatiotemporal Variational Network combined with conjugate gradient descent for optimized data consistency and improved image quality. The method is first evaluated on retrospectively undersampled cine MRI data in terms of image quality. Then, prospectively accelerated data are acquired in 18 healthy subjects both segmented over two heartbeats per slice as well as in real time with 1.6 mm isotropic resolution. Ventricular volumes and strain parameters are computed and compared to a compressed sensing reconstruction and to a conventional reference cine MRI acquisition. Lastly, the method is demonstrated in 46 patients and ventricular volumes and strain parameters are evaluated.

RESULTS

CineVN outperformed compressed sensing in image quality metrics on retrospectively undersampled data. Functional parameters and myocardial strain were the most accurate for CineVN compared to two state-of-the-art compressed sensing methods.

CONCLUSION

Deep learning-based reconstruction using our proposed method enables accurate evaluation of cardiac function in real-time cine MRI with high spatiotemporal resolution. This has the potential to improve cardiac imaging particularly for patients with arrhythmia or impaired breath-hold capability.

Systems Engineering Approach Towards Sensitive Cellular Fluorine-19 MRI

In vivo fluorine-19 MRI using F-based tracer media has shown utility and versatility for a wide range of biomedical uses, particularly immune and stem cell detection, as well as biosensing. As with many advanced MRI acquisition techniques, the sensitivity and limit of detection (LOD) in vivo is a key consideration for a successful study outcome. In this review, we analyze the primary factors that limit cell LOD. The achievable sensitivity is strongly dependent on the specific composition of tracer, cell type of interest, cell activity, data acquisition and reconstruction methods, and MRI hardware design. Recent innovations in molecular 19F tracer design and image acquisition-reconstruction methods have achieved significant leaps in 19F MRI sensitivity, and integration of these new materials and methods into studies can result in > 10-fold improvement in LOD. These developments will help unlock the full potential of clinical 19F MRI for biomedical applications.

Spherical echo-planar time-resolved imaging (sEPTI) for rapid 3D quantitative T 2 * and susceptibility imaging

PURPOSE

To develop a 3D spherical EPTI (sEPTI) acquisition and a comprehensive reconstruction pipeline for rapid high-quality whole-brain submillimeterT 2 * $$ {\mathrm{T}}_2^{\ast } $$ and QSM quantification.

METHODS

For the sEPTI acquisition, spherical k-space coverage is utilized with variable echo-spacing and maximum kx ramp-sampling to improve efficiency and signal incoherency compared to existing EPTI approaches. For reconstruction, an iterative rank-shrinking B0 estimation and odd-even high-order phase correction algorithms were incorporated into the reconstruction to better mitigate artifacts from field imperfections. A physics-informed unrolled network was utilized to boost the SNR, where 1-mm and 0.75-mm isotropic whole-brain imaging were performed in 45 and 90 s at 3 T, respectively. These protocols were validated through simulations, phantom, and in vivo experiments. Ten healthy subjects were recruited to provide sufficient data for the unrolled network. The entire pipeline was validated on additional five healthy subjects where different EPTI sampling approaches were compared. Two additional pediatric patients with epilepsy were recruited to demonstrate the generalizability of the unrolled reconstruction.

RESULTS

sEPTI achieved 1.4× $$ \times $$ faster imaging with improved image quality and quantitative map precision compared to existing EPTI approaches. The B0 update and the phase correction provide improved reconstruction performance with lower artifacts. The unrolled network boosted the SNR, achieving high-qualityT 2 * $$ {\mathrm{T}}_2^{\ast } $$ and QSM quantification with single average data. High-quality reconstruction was also obtained in the pediatric patients using this network.

CONCLUSION

sEPTI achieved whole-brain distortion-free multi-echo imaging andT 2 * $$ {\mathrm{T}}_2^{\ast } $$ and QSM quantification at 0.75 mm in 90 s which has the potential to be useful for wide clinical applications.

Incorporating Radiologist Knowledge Into MRI Quality Metrics for Machine Learning Using Rank-Based Ratings

BACKGROUND

Deep learning (DL) often requires an image quality metric; however, widely used metrics are not designed for medical images.

PURPOSE

To develop an image quality metric that is specific to MRI using radiologists image rankings and DL models.

STUDY TYPE

Retrospective.

POPULATION

A total of 19,344 rankings on 2916 unique image pairs from the NYU fastMRI Initiative neuro database was used for the neural network-based image quality metrics training with an 80%/20% training/validation split and fivefold cross-validation.

FIELD STRENGTH/SEQUENCE

1.5 T and 3 T T1, T1 postcontrast, T2, and FLuid Attenuated Inversion Recovery (FLAIR).

ASSESSMENT

Synthetically corrupted image pairs were ranked by radiologists (N = 7), with a subset also scoring images using a Likert scale (N = 2). DL models were trained to match rankings using two architectures (EfficientNet and IQ-Net) with and without reference image subtraction and compared to ranking based on mean squared error (MSE) and structural similarity (SSIM). Image quality assessing DL models were evaluated as alternatives to MSE and SSIM as optimization targets for DL denoising and reconstruction.

STATISTICAL TESTS

Radiologists' agreement was assessed by a percentage metric and quadratic weighted Cohen's kappa. Ranking accuracies were compared using repeated measurements analysis of variance. Reconstruction models trained with IQ-Net score, MSE and SSIM were compared by paired t test. P < 0.05 was considered significant.

RESULTS

Compared to direct Likert scoring, ranking produced a higher level of agreement between radiologists (70.4% vs. 25%). Image ranking was subjective with a high level of intraobserver agreement (94.9 % ± 2.4 % $$ 94.9\%\pm 2.4\% $$ ) and lower interobserver agreement (61.47 % ± 5.51 % $$ 61.47\%\pm 5.51\% $$ ). IQ-Net and EfficientNet accurately predicted rankings with a reference image (75.2 % ± 1.3 % $$ 75.2\%\pm 1.3\% $$ and79.2 % ± 1.7 % $$ 79.2\%\pm 1.7\% $$ ). However, EfficientNet resulted in images with artifacts and high MSE when used in denoising tasks while IQ-Net optimized networks performed well for both denoising and reconstruction tasks.

DATA CONCLUSION

Image quality networks can be trained from image ranking and used to optimize DL tasks.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 1.

Free-Breathing Ungated Radial Simultaneous Multi-Slice Cardiac T1 Mapping

BACKGROUND

Modified Look-Locker imaging (MOLLI) T1 mapping sequences are acquired during breath-holding and require ECG gating with consistent R-R intervals, which is problematic for patients with atrial fibrillation (AF). Consequently, there is a need for a free-breathing and ungated framework for cardiac T1 mapping.

PURPOSE

To develop and evaluate a free-breathing ungated radial simultaneous multi-slice (SMS) cardiac T1 mapping (FURST) framework.

STUDY TYPE

Retrospective, nonconsecutive cohort study.

POPULATION

Twenty-four datasets from 17 canine and 7 human subjects (4 males,51 ± 22 $$ 51\pm 22 $$ years; 3 females,56 ± 19 $$ 56\pm 19 $$ years). Canines were from studies involving AF induction and ablation treatment. The human population included separate subjects with suspected microvascular disease, acute coronary syndrome with persistent AF, and transthyretin amyloidosis with persistent AF. The remaining human subjects were healthy volunteers.

FIELD STRENGTH/SEQUENCE

Pre- and post-contrast T1 mapping with the free-breathing and ungated SMS inversion recovery sequence with gradient echo readout and with conventional MOLLI sequences at 1.5 T and 3.0 T.

ASSESSMENT

MOLLI and FURST were acquired in all subjects, and American Heart Association (AHA) segmentation was used for segment-wise analysis. Pre-contrast T1, post-contrast T1, and ECV were analyzed using correlation and Bland-Altman plots in 13 canines and 7 human subjects. T1 difference box plots for repeated acquisitions in four canine subjects were used to assess reproducibility. The PIQUE image quality metric was used to evaluate the perceptual quality of T1 maps.

STATISTICAL TESTS

Paired t-tests were used for all comparisons between FURST and MOLLI, withP < 0.05 $$ P<0.05 $$ indicating statistical significance.

RESULTS

There were no significant differences between FURST and MOLLI pre-contrast T1 reproducibility (25 ± 18 $$ 25\pm 18 $$ and19 ± 16 msec $$ 19\pm 16\ \mathrm{msec} $$ ,P = 0.19 $$ P=0.19 $$ ), FURST and MOLLI ECV (29 % ± 11 % $$ 29\%\pm 11\% $$ and28 % ± 11 % $$ 28\%\pm 11\% $$ ,P = 0.05 $$ P=0.05 $$ ), or FURST and MOLLI PIQUE scores (52 ± 8 $$ 52\pm 8 $$ and53 ± 10 $$ 53\pm 10 $$ ,P = 0.18 $$ P=0.18 $$ ). The ECV mean difference was0.48 $$ 0.48 $$ with95 % CI : 6.0 × 10 - 4 , 0.96 $$ 95\%\mathrm{CI}:\left(6.0\times {10}^{-4},0.96\right) $$ .

CONCLUSIONS

FURST had similar quality pre-contrast T1, post-contrast T1, and ECV maps and similar reproducibility compared to MOLLI.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: 1.

Motion-Guided Deep Image Prior for Cardiac MRI

Cardiovascular magnetic resonance imaging is a powerful diagnostic tool for assessing cardiac structure and function. Traditional breath-held imaging protocols, however, pose challenges for patients with arrhythmias or limited breath-holding capacity. We introduce Motion-Guided Deep Image prior (M-DIP), a novel unsupervised reconstruction framework for accelerated real-time cardiac MRI. M-DIP employs a spatial dictionary to synthesize a time-dependent template image, which is further refined using time-dependent deformation fields that model cardiac and respiratory motion. Unlike prior DIP-based methods, M-DIP simultaneously captures physiological motion and frame-to-frame content variations, making it applicable to a wide range of dynamic applications. We validate M-DIP using simulated MRXCAT cine phantom data as well as free-breathing real-time cine and single-shot late gadolinium enhancement data from clinical patients. Comparative analyses against state-of-the-art supervised and unsupervised approaches demonstrate M-DIP's performance and versatility. M-DIP achieved better image quality metrics on phantom data, as well as higher reader scores for in-vivo patient data.

PixCUE: Joint Uncertainty Estimation and Image Reconstruction in MRI using Deep Pixel Classification

Deep learning (DL) models are effective in leveraging latent representations from MR data, emerging as state-of-the-art solutions for accelerated MRI reconstruction. However, challenges arise due to the inherent uncertainties associated with undersampling in k-space, coupled with the over- or under-parameterized and opaque nature of DL models. Addressing uncertainty has thus become a critical issue in DL MRI reconstruction. Monte Carlo (MC) inference techniques are commonly employed to estimate uncertainty, involving multiple reconstructions of the same scan to compute variance as a measure of uncertainty. Nevertheless, these methods entail significant computational expenses, requiring multiple inferences through the DL model. In this context, we propose a novel approach to uncertainty estimation during MRI reconstruction using a pixel classification framework. Our method, PixCUE (Pixel Classification Uncertainty Estimation), generates both the reconstructed image and an uncertainty map in a single forward pass through the DL model. We validate the efficacy of this approach by demonstrating that PixCUE-generated uncertainty maps exhibit a strong correlation with reconstruction errors across various MR imaging sequences and under diverse adversarial conditions. We present an empirical relationship between uncertainty estimations using PixCUE and established reconstruction metrics such as NMSE, PSNR, and SSIM. Furthermore, we establish a correlation between the estimated uncertainties from PixCUE and the conventional MC method. Our findings affirm that PixCUE reliably estimates uncertainty in MRI reconstruction with minimal additional computational cost.

Ferumoxytol-Enhanced Cardiac Cine MRI Reconstruction Using a Variable-Splitting Spatiotemporal Network

BACKGROUND

Balanced steady-state free precession (bSSFP) imaging is commonly used in cardiac cine MRI but prone to image artifacts. Ferumoxytol-enhanced (FE) gradient echo (GRE) has been proposed as an alternative. Utilizing the abundance of bSSFP images to develop a computationally efficient network that is applicable to FE GRE cine would benefit future network development.

PURPOSE

To develop a variable-splitting spatiotemporal network (VSNet) for image reconstruction, trained on bSSFP cine images and applicable to FE GRE cine images.

STUDY TYPE

Retrospective and prospective.

SUBJECTS

41 patients (26 female, 53 ± 19 y/o) for network training, 31 patients (19 female, 49 ± 17 y/o) and 5 healthy subjects (5 female, 30 ± 7 y/o) for testing.

FIELD STRENGTH/SEQUENCE

1.5T and 3T, bSSFP and GRE.

ASSESSMENT

VSNet was compared to VSNet with total variation loss, compressed sensing and low rank methods for 14× accelerated data. The GRAPPA×2/×3 images served as the reference. Peak signal-to-noise-ratio (PSNR), structural similarity index (SSIM), left ventricular (LV) and right ventricular (RV) end-diastolic volume (EDV), end-systolic volume (ESV), and ejection fraction (EF) were measured. Qualitative image ranking and scoring were independently performed by three readers. Latent scores were calculated based on scores of each method relative to the reference.

STATISTICS

Linear mixed-effects regression, Tukey method, Fleiss' Kappa, Bland-Altman analysis, and Bayesian categorical cumulative probit model. A P-value <0.05 was considered statistically significant.

RESULTS

VSNet achieved significantly higher PSNR (32.7 ± 0.2), SSIM (0.880 ± 0.004), rank (2.14 ± 0.06), and latent scores (-1.72 ± 0.22) compared to other methods (rank >2.90, latent score < -2.63). Fleiss' Kappa was 0.52 for scoring and 0.61 for ranking. VSNet showed no significantly different LV and RV ESV (P = 0.938) and EF (P = 0.143) measurements, but statistically significant different (2.62 mL) EDV measurements compared to the reference.

CONCLUSION

VSNet produced the highest image quality and the most accurate functional measurements for FE GRE cine images among the tested 14× accelerated reconstruction methods.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 1.

Self-supervised learning for improved calibrationless radial MRI with NLINV-Net

PURPOSE

To develop a neural network architecture for improved calibrationless reconstruction of radial data when no ground truth is available for training.

METHODS

NLINV-Net is a model-based neural network architecture that directly estimates images and coil sensitivities from (radial) k-space data via nonlinear inversion (NLINV). Combined with a training strategy using self-supervision via data undersampling (SSDU), it can be used for imaging problems where no ground truth reconstructions are available. We validated the method for (1) real-time cardiac imaging and (2) single-shot subspace-based quantitative T1 mapping. Furthermore, region-optimized virtual (ROVir) coils were used to suppress artifacts stemming from outside the field of view and to focus the k-space-based SSDU loss on the region of interest. NLINV-Net-based reconstructions were compared with conventional NLINV and PI-CS (parallel imaging + compressed sensing) reconstruction and the effect of the region-optimized virtual coils and the type of training loss was evaluated qualitatively.

RESULTS

NLINV-Net-based reconstructions contain significantly less noise than the NLINV-based counterpart. ROVir coils effectively suppress streakings which are not suppressed by the neural networks while the ROVir-based focused loss leads to visually sharper time series for the movement of the myocardial wall in cardiac real-time imaging. For quantitative imaging, T1-maps reconstructed using NLINV-Net show similar quality as PI-CS reconstructions, but NLINV-Net does not require slice-specific tuning of the regularization parameter.

CONCLUSION

NLINV-Net is a versatile tool for calibrationless imaging which can be used in challenging imaging scenarios where a ground truth is not available.

ACGRHA-Net: Accelerated multi-contrast MR imaging with adjacency complementary graph assisted residual hybrid attention network

Multi-contrast magnetic resonance (MR) imaging is an advanced technology used in medical diagnosis, but the long acquisition process can lead to patient discomfort and limit its broader application. Shortening acquisition time by undersampling k-space data introduces noticeable aliasing artifacts. To address this, we propose a method that reconstructs multi-contrast MR images from zero-filled data by utilizing a fully-sampled auxiliary contrast MR image as a prior to learn an adjacency complementary graph. This graph is then combined with a residual hybrid attention network, forming the adjacency complementary graph assisted residual hybrid attention network (ACGRHA-Net) for multi-contrast MR image reconstruction. Specifically, the optimal structural similarity is represented by a graph learned from the fully sampled auxiliary image, where the node features and adjacency matrices are designed to precisely capture structural information among different contrast images. This structural similarity enables effective fusion with the target image, improving the detail reconstruction. Additionally, a residual hybrid attention module is designed in parallel with the graph convolution network, allowing it to effectively capture key features and adaptively emphasize these important features in target contrast MR images. This strategy prioritizes crucial information while preserving shallow features, thereby achieving comprehensive feature fusion at deeper levels to enhance multi-contrast MR image reconstruction. Extensive experiments on the different datasets, using various sampling patterns and accelerated factors demonstrate that the proposed method outperforms the current state-of-the-art reconstruction methods.

Using High-Pass Filter to Enhance Scan Specific Learning for MRI Reconstruction without Any Extra Training Data

In accelerated MRI, the robust artificial-neural-network for k-space interpolation (RAKI) method is an attractive learning-based reconstruction that does not require additional training data. This study was focused on obtaining high quality MR images from regular under-sampled multi-coil k-space data using a high-pass filtered RAKI (HP-RAKI) reconstruction without any extra training data. MRI scan from human subjects was under-sampled with a regular pattern using skipped phase encoding and a fully sampled k-space center. A high-pass (HP) filter was applied in k-space to reduce image support to facilitate linear prediction. The HP filtered k-space center was used to train the RAKI network without any extra training data. The unacquired k-space data can be predicted from a trained RAKI network with optimized parameters. Final reconstruction was obtained after performing an inverse HP filtering for the predicted k-space data. This HP-RAKI method can be extended to corresponding residual structure (HP-rRAKI). HP-RAKI was compared with GRAPPA, HP-GRAPPA, RAKI and MW-RAKI algorithms, and HP-rRAKI was compared with corresponding residual extensions, including rRAKI and MW-rRAKI, all qualitatively and quantitatively using visual inspection and such metrics as SSIM and PSNR. HP-RAKI and HP-rRAKI were found to be effective in reconstructing MR images even at high acceleration factors. HP-RAKI and HP-rRAKI compared favorably with other algorithms. Using high-pass filtered central k-space data for training, HP-RAKI offers higher reconstruction quality for regularly under-sampled multi-coil k-space data without any extra training data. It has shown promising capabilities for fast MRI applications, especially those lacking fully sampled training data.

Accelerated High-Resolution Deep Learning Reconstruction Turbo Spin Echo MRI of the Knee at 7 T

OBJECTIVES

The aim of this study was to compare the image quality of 7 T turbo spin echo (TSE) knee images acquired with varying factors of parallel-imaging acceleration reconstructed with deep learning (DL)-based and conventional algorithms.

MATERIALS AND METHODS

This was a prospective single-center study. Twenty-three healthy volunteers underwent 7 T knee magnetic resonance imaging. Two-, 3-, and 4-fold accelerated high-resolution fat-signal-suppressing proton density (PD-fs) and T1-weighted coronal 2D TSE acquisitions with an encoded voxel volume of 0.31 × 0.31 × 1.5 mm 3 were acquired. Each set of raw data was reconstructed with a DL-based and a conventional Generalized Autocalibrating Partially Parallel Acquisition (GRAPPA) algorithm. Three readers rated image contrast, sharpness, artifacts, noise, and overall quality. Friedman analysis of variance and the Wilcoxon signed rank test were used for comparison of image quality criteria.

RESULTS

The mean age of the participants was 32.0 ± 8.1 years (15 male, 8 female). Acquisition times at 4-fold acceleration were 4 minutes 15 seconds (PD-fs, Supplemental Video is available at http://links.lww.com/RLI/A938 ) and 3 minutes 9 seconds (T1, Supplemental Video available at http://links.lww.com/RLI/A939 ). At 4-fold acceleration, image contrast, sharpness, noise, and overall quality of images reconstructed with the DL-based algorithm were significantly better rated than the corresponding GRAPPA reconstructions ( P < 0.001). Four-fold accelerated DL-reconstructed images scored significantly better than 2- to 3-fold GRAPPA-reconstructed images with regards to image contrast, sharpness, noise, and overall quality ( P ≤ 0.031). Image contrast of PD-fs images at 2-fold acceleration ( P = 0.087), image noise of T1-weighted images at 2-fold acceleration ( P = 0.180), and image artifacts for both sequences at 2- and 3-fold acceleration ( P ≥ 0.102) of GRAPPA reconstructions were not rated differently than those of 4-fold accelerated DL-reconstructed images. Furthermore, no significant difference was observed for all image quality measures among 2-fold, 3-fold, and 4-fold accelerated DL reconstructions ( P ≥ 0.082).

CONCLUSIONS

This study explored the technical potential of DL-based image reconstruction in accelerated 2D TSE acquisitions of the knee at 7 T. DL reconstruction significantly improved a variety of image quality measures of high-resolution TSE images acquired with a 4-fold parallel-imaging acceleration compared with a conventional reconstruction algorithm.