Learning a variational network for reconstruction of accelerated MRI data

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.

Recovering high-quality fiber orientation distributions from a reduced number of diffusion-weighted images using a model-driven deep learning architecture

PURPOSE

The aim of this study was to develop a model-based deep learning architecture to accurately reconstruct fiber orientation distributions (FODs) from a reduced number of diffusion-weighted images (DWIs), facilitating accurate analysis with reduced acquisition times.

METHODS

Our proposed architecture, Spherical Deconvolution Network (SDNet), performed FOD reconstruction by mapping 30 DWIs to fully sampled FODs, which have been fit to 288 DWIs. SDNet included DWI-consistency blocks within the network architecture, and a fixel-classification penalty within the loss function. SDNet was trained on a subset of the Human Connectome Project, and its performance compared with FOD-Net, and multishell multitissue constrained spherical deconvolution.

RESULTS

SDNet achieved the strongest results with respect to angular correlation coefficient and sum of squared errors. When the impact of the fixel-classification penalty was increased, we observed an improvement in performance metrics reliant on segmenting the FODs into the correct number of fixels.

CONCLUSION

Inclusion of DWI-consistency blocks improved reconstruction performance, and the fixel-classification penalty term offered increased control over the angular separation of fixels in the reconstructed FODs.

ATOMMIC: An Advanced Toolbox for Multitask Medical Imaging Consistency to facilitate Artificial Intelligence applications from acquisition to analysis in Magnetic Resonance Imaging

BACKGROUND AND OBJECTIVES

Artificial intelligence (AI) is revolutionizing Magnetic Resonance Imaging (MRI) along the acquisition and processing chain. Advanced AI frameworks have been applied in various successive tasks, such as image reconstruction, quantitative parameter map estimation, and image segmentation. However, existing frameworks are often designed to perform tasks independently of each other or are focused on specific models or single datasets, limiting generalization. This work introduces the Advanced Toolbox for Multitask Medical Imaging Consistency (ATOMMIC), a novel open-source toolbox that streamlines AI applications for accelerated MRI reconstruction and analysis. ATOMMIC implements several tasks using deep learning (DL) models and enables MultiTask Learning (MTL) to perform related tasks in an integrated manner, targeting generalization in the MRI domain.

METHODS

We conducted a comprehensive literature review and analyzed 12,479 GitHub repositories to assess the current landscape of AI frameworks for MRI. Subsequently, we demonstrate how ATOMMIC standardizes workflows and improves data interoperability, enabling effective benchmarking of various DL models across MRI tasks and datasets. To showcase ATOMMIC's capabilities, we evaluated twenty-five DL models on eight publicly available datasets, focusing on accelerated MRI reconstruction, segmentation, quantitative parameter map estimation, and joint accelerated MRI reconstruction and segmentation using MTL.

RESULTS

ATOMMIC's high-performance training and testing capabilities, utilizing multiple GPUs and mixed precision support, enable efficient benchmarking of multiple models across various tasks. The framework's modular architecture implements each task through a collection of data loaders, models, loss functions, evaluation metrics, and pre-processing transformations, facilitating seamless integration of new tasks, datasets, and models. Our findings demonstrate that ATOMMIC supports MTL for multiple MRI tasks with harmonized complex-valued and real-valued data support while maintaining active development and documentation. Task-specific evaluations demonstrate that physics-based models outperform other approaches in reconstructing highly accelerated acquisitions. These high-quality reconstruction models also show superior accuracy in estimating quantitative parameter maps. Furthermore, when combining high-performing reconstruction models with robust segmentation networks through MTL, performance is improved in both tasks.

CONCLUSIONS

ATOMMIC advances MRI reconstruction and analysis by leveraging MTL and ensuring consistency across tasks, models, and datasets. This comprehensive framework serves as a versatile platform for researchers to use existing AI methods and develop new approaches in medical imaging.

Enhancing gadoxetic acid-enhanced liver MRI: a synergistic approach with deep learning CAIPIRINHA-VIBE and optimized fat suppression techniques

OBJECTIVE

To investigate whether a deep learning (DL) controlled aliasing in parallel imaging results in higher acceleration (CAIPIRINHA)-volumetric interpolated breath-hold examination (VIBE) technique can improve image quality, lesion conspicuity, and lesion detection compared to a standard CAIPIRINHA-VIBE technique in gadoxetic acid-enhanced liver MRI.

METHODS

This retrospective single-center study included 168 patients who underwent gadoxetic acid-enhanced liver MRI at 3 T using both standard CAIPIRINHA-VIBE and DL CAIPIRINHA-VIBE techniques on pre-contrast and hepatobiliary phase (HBP) images. Additionally, high-resolution (HR) DL CAIPIRINHA-VIBE was obtained with 1-mm slice thickness on the HBP. Three abdominal radiologists independently assessed the image quality and lesion conspicuity of pre-contrast and HBP images. Statistical analyses involved the Wilcoxon signed-rank test for image quality assessment and the generalized estimation equation for lesion conspicuity and detection evaluation.

RESULTS

DL and HR-DL CAIPIRINHA-VIBE demonstrated significantly improved overall image quality and reduced artifacts on pre-contrast and HBP images compared to standard CAIPIRINHA-VIBE (p < 0.001), with a shorter acquisition time (DL vs standard, 11 s vs 17 s). However, the former presented a more synthetic appearance (both p < 0.05). HR-DL CAIPIRINHA-VIBE showed superior lesion conspicuity to standard and DL CAIPIRINHA-VIBE on HBP images (p < 0.001). Moreover, HR-DL CAIPIRINHA-VIBE exhibited a significantly higher detection rate of small (< 2 cm) solid focal liver lesions (FLLs) on HBP images compared to standard CAIPIRINHA-VIBE (92.5% vs 87.4%; odds ratio = 1.83; p = 0.036).

CONCLUSION

DL and HR-DL CAIPIRINHA-VIBE achieved superior image quality compared to standard CAIPIRINHA-VIBE. Additionally, HR-DL CAIPIRINHA-VIBE improved the lesion conspicuity and detection of small solid FLLs. DL and HR-DL CAIPIRINHA-VIBE hold the potential clinical utility for gadoxetic acid-enhanced liver MRI.

CLINICAL RELEVANCE STATEMENT

DL and HR-DL CAIPIRINHA-VIBE hold promise as potential alternatives to standard CAIPIRINHA-VIBE in routine clinical liver MRI, improving the image quality and lesion conspicuity, enhancing the detection of small (< 2 cm) solid focal liver lesions, and reducing the acquisition time.

KEY POINTS

• DL and HR-DL CAIPIRINHA-VIBE demonstrated improved overall image quality and reduced artifacts on pre-contrast and HBP images compared to standard CAIPIRINHA-VIBE, in addition to a shorter acquisition time. • DL and HR-DL CAIPIRINHA-VIBE yielded a more synthetic appearance than standard CAIPIRINHA-VIBE. • HR-DL CAIPIRINHA-VIBE showed improved lesion conspicuity than standard CAIPIRINHA-VIBE on HBP images, with a higher detection of small (< 2 cm) solid focal liver lesions.

Effect of MR head coil geometry on deep-learning-based MR image reconstruction

PURPOSE

To investigate whether parallel imaging-imposed geometric coil constraints can be relaxed when using a deep learning (DL)-based image reconstruction method as opposed to a traditional non-DL method.

THEORY AND METHODS

Traditional and DL-based MR image reconstruction approaches operate in fundamentally different ways: Traditional methods solve a system of equations derived from the image data whereas DL methods use data/target pairs to learn a generalizable reconstruction model. Two sets of head coil profiles were evaluated: (1) 8-channel and (2) 32-channel geometries. A DL model was compared to conjugate gradient SENSE (CG-SENSE) and L1-wavelet compressed sensing (CS) through quantitative metrics and visual assessment as coil overlap was increased.

RESULTS

Results were generally consistent between experiments. As coil overlap increased, there was a significant (p < 0.001) decrease in performance in most cases for all methods. The decrease was most pronounced for CG-SENSE, and the DL models significantly outperformed (p < 0.001) their non-DL counterparts in all scenarios. CS showed improved robustness to coil overlap and signal-to-noise ratio (SNR) versus CG-SENSE, but had quantitatively and visually poorer reconstructions characterized by blurriness as compared to DL. DL showed virtually no change in performance across SNR and very small changes across coil overlap.

CONCLUSION

The DL image reconstruction method produced images that were robust to coil overlap and of higher quality than CG-SENSE and CS. This suggests that geometric coil design constraints can be relaxed when using DL reconstruction methods.

Numerical and Clinical Evaluation of the Robustness of Open-source Networks for Parallel MR Imaging Reconstruction

PURPOSE

Deep neural networks (DNNs) for MRI reconstruction often require large datasets for training. Still, in clinical settings, the domains of datasets are diverse, and how robust DNNs are to domain differences between training and testing datasets has been an open question. Here, we numerically and clinically evaluate the generalization of the reconstruction networks across various domains under clinically practical conditions and provide practical guidance on what points to consider when selecting models for clinical application.

METHODS

We compare the reconstruction performance between four network models: U-Net, the deep cascade of convolutional neural networks (DC-CNNs), Hybrid Cascade, and variational network (VarNet). We used the public multicoil dataset fastMRI for training and testing and performed a single-domain test, where the domains of the dataset used for training and testing were the same, and cross-domain tests, where the source and target domains were different. We conducted a single-domain test (Experiment 1) and cross-domain tests (Experiments 2-4), focusing on six factors (the number of images, sampling pattern, acceleration factor, noise level, contrast, and anatomical structure) both numerically and clinically.

RESULTS

U-Net had lower performance than the three model-based networks and was less robust to domain shifts between training and testing datasets. VarNet had the highest performance and robustness among the three model-based networks, followed by Hybrid Cascade and DC-CNN. Especially, VarNet showed high performance even with a limited number of training images (200 images/10 cases). U-Net was more robust to domain shifts concerning noise level than the other model-based networks. Hybrid Cascade showed slightly better performance and robustness than DC-CNN, except for robustness to noise-level domain shifts. The results of the clinical evaluations generally agreed with the results of the quantitative metrics.

CONCLUSION

In this study, we numerically and clinically evaluated the robustness of the publicly available networks using the multicoil data. Therefore, this study provided practical guidance for clinical applications.

Learning to reconstruct accelerated MRI through K-space cold diffusion without noise

Deep learning-based MRI reconstruction models have achieved superior performance these days. Most recently, diffusion models have shown remarkable performance in image generation, in-painting, super-resolution, image editing and more. As a generalized diffusion model, cold diffusion further broadens the scope and considers models built around arbitrary image transformations such as blurring, down-sampling, etc. In this paper, we propose a k-space cold diffusion model that performs image degradation and restoration in k-space without the need for Gaussian noise. We provide comparisons with multiple deep learning-based MRI reconstruction models and perform tests on a well-known large open-source MRI dataset. Our results show that this novel way of performing degradation can generate high-quality reconstruction images for accelerated MRI.

Nonlinear metamaterials enhanced surface coil array for parallel magnetic resonance imaging

Magnetic resonance imaging (MRI) is a crucial medical imaging modality, with parallel MRI accelerating scans but often reducing the signal-to-noise ratio (SNR). Recent advances in metamaterials have shown considerable potential in enhancing the SNR of MRI and consequently improving the quality of parallel MRI. In this study, we present a nonlinear metamaterial comprising nonlinear meta-atoms designed to selectively enhance the radio-frequency reception field in MRI, while minimizing interference with the radio-frequency transmission field. We develop an electromagnetic field-circuit joint simulation process for analyzing and optimizing the nonlinear response. Experimental validation confirms that nonlinear metamaterial integration in a surface coil array delivers a 3-fold SNR increase for parallel MRI compared to using the surface coil array alone. This research advances our understanding of such metamaterials and demonstrates their potential for practical utilization in MRI and clinical settings, thereby promising substantial enhancements in imaging capabilities.

Deep Learning k-Space-to-Image Reconstruction Facilitates High Spatial Resolution and Scan Time Reduction in Diffusion-Weighted Imaging Breast MRI

BACKGROUND

For time-consuming diffusion-weighted imaging (DWI) of the breast, deep learning-based imaging acceleration appears particularly promising.

PURPOSE

To investigate a combined k-space-to-image reconstruction approach for scan time reduction and improved spatial resolution in breast DWI.

STUDY TYPE

Retrospective.

POPULATION

133 women (age 49.7 ± 12.1 years) underwent multiparametric breast MRI.

FIELD STRENGTH/SEQUENCE

3.0T/T2 turbo spin echo, T1 3D gradient echo, DWI (800 and 1600 sec/mm2).

ASSESSMENT

DWI data were retrospectively processed using deep learning-based k-space-to-image reconstruction (DL-DWI) and an additional super-resolution algorithm (SRDL-DWI). In addition to signal-to-noise ratio and apparent diffusion coefficient (ADC) comparisons among standard, DL- and SRDL-DWI, a range of quantitative similarity (e.g., structural similarity index [SSIM]) and error metrics (e.g., normalized root mean square error [NRMSE], symmetric mean absolute percent error [SMAPE], log accuracy error [LOGAC]) was calculated to analyze structural variations. Subjective image evaluation was performed independently by three radiologists on a seven-point rating scale.

STATISTICAL TESTS

Friedman's rank-based analysis of variance with Bonferroni-corrected pairwise post-hoc tests. P < 0.05 was considered significant.

RESULTS

Both DL- and SRDL-DWI allowed for a 39% reduction in simulated scan time over standard DWI (5 vs. 3 minutes). The highest image quality ratings were assigned to SRDL-DWI with good interreader agreement (ICC 0.834; 95% confidence interval 0.818-0.848). Irrespective of b-value, both standard and DL-DWI produced superior SNR compared to SRDL-DWI. ADC values were slightly higher in SRDL-DWI (+0.5%) and DL-DWI (+3.4%) than in standard DWI. Structural similarity was excellent between DL-/SRDL-DWI and standard DWI for either b value (SSIM ≥ 0.86). Calculation of error metrics (NRMSE ≤ 0.05, SMAPE ≤ 0.02, and LOGAC ≤ 0.04) supported the assumption of low voxel-wise error.

DATA CONCLUSION

Deep learning-based k-space-to-image reconstruction reduces simulated scan time of breast DWI by 39% without influencing structural similarity. Additionally, super-resolution interpolation allows for substantial improvement of subjective image quality.

EVIDENCE LEVEL

4 TECHNICAL EFFICACY: Stage 1.

SPICER: Self-supervised learning for MRI with automatic coil sensitivity estimation and reconstruction

PURPOSE

To introduce a novel deep model-based architecture (DMBA), SPICER, that uses pairs of noisy and undersampled k-space measurements of the same object to jointly train a model for MRI reconstruction and automatic coil sensitivity estimation.

METHODS

SPICER consists of two modules to simultaneously reconstructs accurate MR images and estimates high-quality coil sensitivity maps (CSMs). The first module, CSM estimation module, uses a convolutional neural network (CNN) to estimate CSMs from the raw measurements. The second module, DMBA-based MRI reconstruction module, forms reconstructed images from the input measurements and the estimated CSMs using both the physical measurement model and learned CNN prior. With the benefit of our self-supervised learning strategy, SPICER can be efficiently trained without any fully sampled reference data.

RESULTS

We validate SPICER on both open-access datasets and experimentally collected data, showing that it can achieve state-of-the-art performance in highly accelerated data acquisition settings (up to10 × $$ 10\times $$ ). Our results also highlight the importance of different modules of SPICER-including the DMBA, the CSM estimation, and the SPICER training loss-on the final performance of the method. Moreover, SPICER can estimate better CSMs than pre-estimation methods especially when the ACS data is limited.

CONCLUSION

Despite being trained on noisy undersampled data, SPICER can reconstruct high-quality images and CSMs in highly undersampled settings, which outperforms other self-supervised learning methods and matches the performance of the well-known E2E-VarNet trained on fully sampled ground-truth data.

Deep learning reconstructed T2-weighted Dixon imaging of the spine: Impact on acquisition time and image quality

PURPOSE

To assess the image quality and impact on acquisition time of a novel deep learning based T2 Dixon sequence (T2DL) of the spine.

METHODS

This prospective, single center study included n = 44 consecutive patients with a clinical indication for lumbar MRI at our university radiology department between September 2022 and March 2023. MRI examinations were performed on 1.5-T and 3-T scanners (MAGNETOM Aera and Vida; Siemens Healthineers, Erlangen, Germany) using dedicated spine coils. The MR study protocol consisted of our standard clinical protocol, including a T2 weighted standard Dixon sequence (T2std) and an additional T2DL acquisition. The latter used a conventional sampling pattern with a higher parallel acceleration factor. The individual contrasts acquired for Dixon water-fat separation were then reconstructed using a dedicated research application. After reconstruction of the contrast images from k-space data, a conventional water-fat separation was performed to provide derived water images. Two readers with 6 and 4 years of experience in interpreting MSK imaging, respectively, analyzed the images in a randomized fashion. Regarding overall image quality, banding artifacts, artifacts, sharpness, noise, and diagnostic confidence were analyzed using a 5-point Likert scale (from 1 = non-diagnostic to 5 = excellent image quality). Statistical analyses included the Wilcoxon signed-rank test and weighted Cohen's kappa statistics.

RESULTS

Forty-four patients (mean age 53 years (±18), male sex: 39 %) were prospectively included. Thirty-one examinations were performed on 1.5 T and 13 examinations on 3 T scanners. A sequence was successfully acquired in all patients. The total acquisition time of T2DL was 93 s at 1.5-T and 86 s at 3-T, compared to 235 s, and 257 s, respectively for T2std (reduction of acquisition time: 60.4 % at 1.5-T, and 66.5 % at 3-T; p < 0.01). Overall image quality was rated equal for both sequences (median T2DL: 5[3 -5], and median T2std: 5 [2 -5]; p = 0.57). T2DL showed significantly reduced noise levels compared to T2std (5 [4 -5] versus 4 [3 -4]; p < 0.001). In addition, sharpness was rated to be significantly higher in T2DL (5 [4 -5] versus 4 [3 -5]; p < 0.001). Although T2DL displayed significantly more banding artifacts (5 [2 -5] versus 5 [4 -5]; p < 0.001), no significant impact on readers diagnostic confidence between sequences was noted (T2std: 5 [2 -5], and T2DL: 5 [3 -5]; p = 0.61). Substantial inter-reader and intrareader agreement was observed for T2DL overall image quality (κ: 0.77, and κ: 0.8, respectively).

CONCLUSION

T2DL is feasible, yields an image quality comparable to the reference standard while substantially reducing the acquisition time.

Using a deep learning prior for accelerating hyperpolarized 13C MRSI on synthetic cancer datasets

PURPOSE

We aimed to incorporate a deep learning prior with k-space data fidelity for accelerating hyperpolarized carbon-13 MRSI, demonstrated on synthetic cancer datasets.

METHODS

A two-site exchange model, derived from the Bloch equation of MR signal evolution, was firstly used in simulating training and testing data, that is, synthetic phantom datasets. Five singular maps generated from each simulated dataset were used to train a deep learning prior, which was then employed with the fidelity term to reconstruct the undersampled MRI k-space data. The proposed method was assessed on synthetic human brain tumor images (N = 33), prostate cancer images (N = 72), and mouse tumor images (N = 58) for three undersampling factors and 2.5% additive Gaussian noise. Furthermore, varied levels of Gaussian noise with SDs of 2.5%, 5%, and 10% were added on synthetic prostate cancer data, and corresponding reconstruction results were evaluated.

RESULTS

For quantitative evaluation, peak SNRs were approximately 32 dB, and the accuracy was generally improved for 5 to 8 dB compared with those from compressed sensing with L1-norm regularization or total variation regularization. Reasonable normalized RMS error were obtained. Our method also worked robustly against noise, even on a data with noise SD of 10%.

CONCLUSION

The proposed singular value decomposition + iterative deep learning model could be considered as a general framework that extended the application of deep learning MRI reconstruction to metabolic imaging. The morphology of tumors and metabolic images could be measured robustly in six times acceleration using our method.

MLMFNet: A multi-level modality fusion network for multi-modal accelerated MRI reconstruction

Magnetic resonance imaging produces detailed anatomical and physiological images of the human body that can be used in the clinical diagnosis and treatment of diseases. However, MRI suffers its comparatively longer acquisition time than other imaging methods and is thus vulnerable to motion artifacts, which ultimately lead to likely failed or even wrong diagnosis. In order to perform faster reconstruction, deep learning-based methods along with traditional strategies such as parallel imaging and compressed sensing come into play in recent years in this field. Meanwhile, in order to better analyze the diseases, it is also often necessary to acquire images in the same region of interest under different modalities, which yield images with different contrast levels. However, most of these aforementioned methods tend to use single-modal images for reconstruction, neglecting the correlation and redundancy information embedded in MR images acquired with different modalities. While there are works on multi-modal reconstruction, the information is yet to be efficiently explored. In this paper, we propose an end-to-end neural network called MLMFNet, which helps the reconstruction of the target modality by using information from the auxiliary modality across feature channels and layers. Specifically, this is highlighted by three components: (I) An encoder based on UNet with a single-stream strategy that fuses auxiliary and target modalities; (II) a decoder that tends to multi-level features from all layers of the encoder, and (III) a channel attention module. Quantitative and qualitative analyses are performed on a public brain dataset and knee brain dataset, which show that the proposed method achieves satisfying results in MRI reconstruction within the multi-modal context, and also demonstrate its effectiveness and potential to be used in clinical practice.

Recent advances in artificial intelligent strategies for tissue engineering and regenerative medicine

BACKGROUND

Tissue engineering and regenerative medicine (TERM) aim to repair or replace damaged or lost tissues or organs due to accidents, diseases, or aging, by applying different sciences. For this purpose, an essential part of TERM is the designing, manufacturing, and evaluating of scaffolds, cells, tissues, and organs. Artificial intelligence (AI) or the intelligence of machines or software can be effective in all areas where computers play a role.

METHODS

The "artificial intelligence," "machine learning," "tissue engineering," "clinical evaluation," and "scaffold" keywords used for searching in various databases and articles published from 2000 to 2024 were evaluated.

RESULTS

The combination of tissue engineering and AI has created a new generation of technological advancement in the biomedical industry. Experience in TERM has been refined using advanced design and manufacturing techniques. Advances in AI, particularly deep learning, offer an opportunity to improve scientific understanding and clinical outcomes in TERM.

CONCLUSION

The findings of this research show the high potential of AI, machine learning, and robots in the selection, design, and fabrication of scaffolds, cells, tissues, or organs, and their analysis, characterization, and evaluation after their implantation. AI can be a tool to accelerate the introduction of tissue engineering products to the bedside.

HIGHLIGHTS

The capabilities of artificial intelligence (AI) can be used in different ways in all the different stages of TERM and not only solve the existing limitations, but also accelerate the processes, increase efficiency and precision, reduce costs, and complications after transplantation. ML predicts which technologies have the most efficient and easiest path to enter the market and clinic. The use of AI along with these imaging techniques can lead to the improvement of diagnostic information, the reduction of operator errors when reading images, and the improvement of image analysis (such as classification, localization, regression, and segmentation).

Deep learning-based rapid image reconstruction and motion correction for high-resolution cartesian first-pass myocardial perfusion imaging at 3T

PURPOSE

To develop and evaluate a deep learning (DL) -based rapid image reconstruction and motion correction technique for high-resolution Cartesian first-pass myocardial perfusion imaging at 3T with whole-heart coverage for both single-slice (SS) and simultaneous multi-slice (SMS) acquisitions.

METHODS

3D physics-driven unrolled network architectures were utilized for the reconstruction of high-resolution Cartesian perfusion imaging. The SS and SMS multiband (MB) = 2 networks were trained from 135 slices from 20 subjects. Structural similarity index (SSIM), peak SNR (PSNR), and normalized RMS error (NRMSE) were assessed, and prospective images were blindly graded by two experienced cardiologists (5, excellent; 1, poor). For respiratory motion correction, a 2D U-Net based motion corrected network was proposed, and the temporal fidelity and second-order derivative were calculated to assess the performance of the motion correction.

RESULTS

Excellent performance was demonstrated in the proposed technique with high SSIM and PSNR, and low NRMSE. Image quality scores were (4.3 [4.3, 4.4], 4.5 [4.4, 4.6], 4.3 [4.3, 4.4], and 4.5 [4.3, 4.5]) for SS DL and SS L1-SENSE, MB = 2 DL and MB = 2 SMS-L1-SENSE, respectively, showing no statistically significant difference (p > 0.05 for SS and SMS) between (SMS)-L1-SENSE and the proposed DL technique. The network inference time was around 4 s per dynamic perfusion series with 40 frames while the time of (SMS)-L1-SENSE with GPU acceleration was approximately 30 min.

CONCLUSION

The proposed DL-based image reconstruction and motion correction technique enabled rapid and high-quality reconstruction for SS and SMS MB = 2 high-resolution Cartesian first-pass perfusion imaging at 3T.

DIRECTION: Deep cascaded reconstruction residual-based feature modulation network for fast MRI reconstruction

Deep cascaded networks have been extensively studied and applied to accelerate Magnetic Resonance Imaging (MRI) and have shown promising results. Most existing works employ a large cascading number for the sake of superior performances. However, due to the lack of proper guidance, the reconstruction performance can easily reach a plateau and even face degradation if simply increasing the cascading number. In this paper, we aim to boost the reconstruction performance from a novel perspective by proposing a parallel architecture called DIRECTION that fully exploits the guiding value of the reconstruction residual of each subnetwork. Specifically, we introduce a novel Reconstruction Residual-Based Feature Modulation Mechanism (RRFMM) which utilizes the reconstruction residual of the previous subnetwork to guide the next subnetwork at the feature level. To achieve this, a Residual Attention Modulation Block (RAMB) is proposed to generate attention maps using multi-scale residual features to modulate the image features of the corresponding scales. Equipped with this strategy, each subnetwork within the cascaded network possesses its unique optimization objective and emphasis rather than blindly updating its parameters. To further boost the performance, we introduce the Cross-Stage Feature Reuse Connection (CSFRC) and the Reconstruction Dense Connection (RDC), which can reduce information loss and enhance representative ability. We conduct sufficient experiments and evaluate our method on the fastMRI knee dataset using multiple subsampling masks. Comprehensive experimental results show that our method can markedly boost the performance of cascaded networks and significantly outperforms other compared state-of-the-art methods quantitatively and qualitatively.

Model-based federated learning for accurate MR image reconstruction from undersampled k-space data

Deep learning-based methods have achieved encouraging performances in the field of Magnetic Resonance (MR) image reconstruction. Nevertheless, building powerful and robust deep learning models requires collecting large and diverse datasets from multiple centers. This raises concerns about ethics and data privacy. Recently, federated learning has emerged as a promising solution, enabling the utilization of multi-center data without the need for data transfer between institutions. Despite its potential, existing federated learning methods face challenges due to the high heterogeneity of data from different centers. Aggregation methods based on simple averaging, which are commonly used to combine the client's information, have shown limited reconstruction and generalization capabilities. In this paper, we propose a Model-based Federated learning framework (ModFed) to address these challenges. ModFed has three major contributions: (1) Different from existing data-driven federated learning methods, ModFed designs attention-assisted model-based neural networks that can alleviate the need for large amounts of data on each client; (2) To address the data heterogeneity issue, ModFed proposes an adaptive dynamic aggregation scheme, which can improve the generalization capability and robustness of the trained neural network models; (3) ModFed incorporates a spatial Laplacian attention mechanism and a personalized client-side loss regularization to capture the detailed information for accurate image reconstruction. The effectiveness of the proposed ModFed is evaluated on three in-vivo datasets. Experimental results show that when compared to six existing state-of-the-art federated learning approaches, ModFed achieves better MR image reconstruction performance with increased generalization capability. Codes will be made available at https://github.com/ternencewu123/ModFed.

Modern acceleration in musculoskeletal MRI: applications, implications, and challenges

Magnetic resonance imaging (MRI) is crucial for accurately diagnosing a wide spectrum of musculoskeletal conditions due to its superior soft tissue contrast resolution. However, the long acquisition times of traditional two-dimensional (2D) and three-dimensional (3D) fast and turbo spin-echo (TSE) pulse sequences can limit patient access and comfort. Recent technical advancements have introduced acceleration techniques that significantly reduce MRI times for musculoskeletal examinations. Key acceleration methods include parallel imaging (PI), simultaneous multi-slice acquisition (SMS), and compressed sensing (CS), enabling up to eightfold faster scans while maintaining image quality, resolution, and safety standards. These innovations now allow for 3- to 6-fold accelerated clinical musculoskeletal MRI exams, reducing scan times to 4 to 6 min for joints and spine imaging. Evolving deep learning-based image reconstruction promises even faster scans without compromising quality. Current research indicates that combining acceleration techniques, deep learning image reconstruction, and superresolution algorithms will eventually facilitate tenfold accelerated musculoskeletal MRI in routine clinical practice. Such rapid MRI protocols can drastically reduce scan times by 80-90% compared to conventional methods. Implementing these rapid imaging protocols does impact workflow, indirect costs, and workload for MRI technologists and radiologists, which requires careful management. However, the shift from conventional to accelerated, deep learning-based MRI enhances the value of musculoskeletal MRI by improving patient access and comfort and promoting sustainable imaging practices. This article offers a comprehensive overview of the technical aspects, benefits, and challenges of modern accelerated musculoskeletal MRI, guiding radiologists and researchers in this evolving field.

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.

Real-time fMRI neurofeedback of the anterior insula using arterial spin labelling

Arterial spin labelling (ASL) is the only non-invasive technique that allows absolute quantification of perfusion and is increasingly used in brain activation studies. Contrary to the blood oxygen level-dependent (BOLD) effect ASL measures the cerebral blood flow (CBF) directly. However, the ASL signal has a lower signal-to-noise ratio (SNR), than the BOLD signal, which constrains its utilization in neurofeedback studies. If successful, ASL neurofeedback can be used to aid in the rehabilitation of health conditions with impaired blood flow, for example, stroke. We provide the first ASL-based neurofeedback study incorporating a double-blind, sham-controlled design. A pseudo-continuous ASL (pCASL) approach with background suppression and 3D GRASE readout was combined with a real-time post-processing pipeline. The real-time pipeline allows to monitor the ASL signal and provides real-time feedback on the neural activity to the subject. In total 41 healthy adults (19-56 years) divided into three groups underwent a neurofeedback-based emotion imagery training of the left anterior insula. Two groups differing only in the explicitness level of instruction received real training and a third group received sham feedback. Only those participants receiving real feedback with explicit instruction showed significantly higher absolute CBF values in the trained region during neurofeedback than participants receiving sham feedback. However, responder analyses of percent signal change values show no differences in activation between the three groups. Persisting limitations, such as the lower SNR, confounding effects of arterial transit time and partial volume effects still impact negatively the implementation of ASL neurofeedback.

Accelerated T2-weighted MRI of the bowel at 3T using a single-shot technique with deep learning-based image reconstruction: impact on image quality and disease detection

RATIONALE AND OBJECTIVE

A single-shot T2-weighted deep-learning-based image reconstruction (DL-HASTE) has been recently developed allowing for shorter acquisition time than conventional half-Fourier acquisition single-shot turbo-spin echo (HASTE). The purpose of this study was to compare image quality of conventional 6 mm HASTE with DL-HASTE at 4 mm and 6 mm slice thickness.

MATERIALS AND METHODS

91 patients (51 female; mean±SD age: 44±10years) who underwent 3T MR enterography from 5/15/2023-7/15/2023 including pelvic conventional HASTE and DL-HASTE were included. Patients either had 4 mm-DL-HASTE or 6 mm-DL-HASTE. Four abdominal radiologists, blinded to sequence type, independently evaluated overall image quality, artifacts over bowel, bowel wall sharpness, and confidence for the presence/absence of bowel abnormalities on 5-point Likert scales. Readers recorded the presence/absence of ileal wall thickening, ileal inflammation, stricture, and penetrating disease on each sequence. Wilcoxon signed-rank test with continuity correction was used for paired comparisons and Wilcoxon rank sum test was used for unpaired ordinal comparisons. A p < .05 indicated statistical significance.

RESULTS

Acquisition times for 6 mm HASTE, 4 mm-DL-HASTE, and 6 mm-DL-HASTE were 64 s, 51 s, and 49 s, respectively. Overall image quality and bowel sharpness were significantly improved for 4 mm-DL-HASTE versus HASTE for 3/4 readers (all p < .05) and similar for the 4th reader (p > .05). Diagnostic confidence was similar for all readers (p > .05). 6 mm-DL-HASTE was similar to HASTE for bowel sharpness, image quality, and confidence for 3/4 readers (all p > .05). The presence of ileal thickening, ileal inflammation, stricture, and penetrating disease were similar for all readers for HASTE, 4 mm-DL-HASTE, and 6 mm-DL-HASTE (all p > .05).

CONCLUSION

4 mm-DL-HASTE had superior image quality than conventional HASTE at shorter acquisition time.

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.

Deep Learning-Based Reconstruction of 3D T1 SPACE Vessel Wall Imaging Provides Improved Image Quality with Reduced Scan Times: A Preliminary Study

BACKGROUND AND PURPOSE

Intracranial vessel wall imaging is technically challenging to implement, given the simultaneous requirements of high spatial resolution, excellent blood and CSF signal suppression, and clinically acceptable gradient times. Herein, we present our preliminary findings on the evaluation of a deep learning-optimized sequence using T1-weighted imaging.

MATERIALS AND METHODS

Clinical and optimized deep learning-based image reconstruction T1 3D Sampling Perfection with Application optimized Contrast using different flip angle Evolution (SPACE) were evaluated, comparing noncontrast sequences in 10 healthy controls and postcontrast sequences in 5 consecutive patients. Images were reviewed on a Likert-like scale by 4 fellowship-trained neuroradiologists. Scores (range, 1-4) were separately assigned for 11 vessel segments in terms of vessel wall and lumen delineation. Additionally, images were evaluated in terms of overall background noise, image sharpness, and homogeneous CSF signal. Segment-wise scores were compared using paired samples t tests.

RESULTS

The scan time for the clinical and deep learning-based image reconstruction sequences were 7:26 minutes and 5:23 minutes respectively. Deep learning-based image reconstruction images showed consistently higher wall signal and lumen visualization scores, with the differences being statistically significant in most vessel segments on both pre- and postcontrast images. Deep learning-based image reconstruction had lower background noise, higher image sharpness, and uniform CSF signal. Depiction of intracranial pathologies was better or similar on the deep learning-based image reconstruction.

CONCLUSIONS

Our preliminary findings suggest that deep learning-based image reconstruction-optimized intracranial vessel wall imaging sequences may be helpful in achieving shorter gradient times with improved vessel wall visualization and overall image quality. These improvements may help with wider adoption of intracranial vessel wall imaging in clinical practice and should be further validated on a larger cohort.

Artificial Intelligence, the Digital Surgeon: Unravelling Its Emerging Footprint in Healthcare - The Narrative Review

Background

Artificial Intelligence (AI) holds transformative potential for the healthcare industry, offering innovative solutions for diagnosis, treatment planning, and improving patient outcomes. As AI continues to be integrated into healthcare systems, it promises advancements across various domains. This review explores the diverse applications of AI in healthcare, along with the challenges and limitations that need to be addressed. The aim is to provide a comprehensive overview of AI's impact on healthcare and to identify areas for further development and focus.

Main Applications

The review discusses the broad range of AI applications in healthcare. In medical imaging and diagnostics, AI enhances the accuracy and efficiency of diagnostic processes, aiding in early disease detection. AI-powered clinical decision support systems assist healthcare professionals in patient management and decision-making. Predictive analytics using AI enables the prediction of patient outcomes and identification of potential health risks. AI-driven robotic systems have revolutionized surgical procedures, improving precision and outcomes. Virtual assistants and chatbots enhance patient interaction and support, providing timely information and assistance. In the pharmaceutical industry, AI accelerates drug discovery and development by identifying potential drug candidates and predicting their efficacy. Additionally, AI improves administrative efficiency and operational workflows in healthcare, streamlining processes and reducing costs. AI-powered remote monitoring and telehealth solutions expand access to healthcare, particularly in underserved areas.

Challenges and Limitations

Despite the significant promise of AI in healthcare, several challenges persist. Ensuring the reliability and consistency of AI-driven outcomes is crucial. Privacy and security concerns must be navigated carefully, particularly in handling sensitive patient data. Ethical considerations, including bias and fairness in AI algorithms, need to be addressed to prevent unintended consequences. Overcoming these challenges is critical for the ethical and successful integration of AI in healthcare.

Conclusion

The integration of AI into healthcare is advancing rapidly, offering substantial benefits in improving patient care and operational efficiency. However, addressing the associated challenges is essential to fully realize the transformative potential of AI in healthcare. Future efforts should focus on enhancing the reliability, transparency, and ethical standards of AI technologies to ensure they contribute positively to global health outcomes.

Optimizing Image Quality with High-Resolution, Deep-Learning-Based Diffusion-Weighted Imaging in Breast Cancer Patients at 1.5 T

The objective of this study was to evaluate a high-resolution deep-learning (DL)-based diffusion-weighted imaging (DWI) sequence for breast magnetic resonance imaging (MRI) in comparison to a standard DWI sequence (DWIStd) at 1.5 T. It is a prospective study of 38 breast cancer patients, who were scanned with DWIStd and DWIDL. Both DWI sequences were scored for image quality, sharpness, artifacts, contrast, noise, and diagnostic confidence with a Likert-scale from 1 (non-diagnostic) to 5 (excellent). The lesion diameter was evaluated on b 800 DWI, apparent diffusion coefficient (ADC), and the second subtraction (SUB) of the contrast-enhanced T1 VIBE. SNR was also calculated. Statistics included correlation analyses and paired t-tests. High-resolution DWIDL offered significantly superior image quality, sharpness, noise, contrast, and diagnostic confidence (each p < 0.02)). Artifacts were significantly higher in DWIDL by one reader (M = 4.62 vs. 4.36 Likert scale, p < 0.01) without affecting the diagnostic confidence. SNR was higher in DWIDL for b 50 and ADC maps (each p = 0.07). Acquisition time was reduced by 22% in DWIDL. The lesion diameters in DWI b 800DL and Std and ADCDL and Std were respectively 6% lower compared to the 2nd SUB. A DL-based diffusion sequence at 1.5 T in breast MRI offers a higher resolution and a faster acquisition, including only minimally more artefacts without affecting the diagnostic confidence.

Attention-Based MultiOffset Deep Learning Reconstruction of Chemical Exchange Saturation Transfer (AMO-CEST) MRI

One challenge of chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI) is the long scan time due to multiple acquisitions of images at different saturation frequency offsets. k-space under-sampling strategy is commonly used to accelerate MRI acquisition, while this could introduce artifacts and reduce signal-to-noise ratio (SNR). To accelerate CEST-MRI acquisition while maintaining suitable image quality, we proposed an attention-based multioffset deep learning reconstruction network (AMO-CEST) with a multiple radial k-space sampling strategy for CEST-MRI. The AMO-CEST also contains dilated convolution to enlarge the receptive field and data consistency module to preserve the sampled k-space data. We evaluated the proposed method on a mouse brain dataset containing 5760 CEST images acquired at a pre-clinical 3 T MRI scanner. Quantitative results demonstrated that AMO-CEST showed obvious improvement over zero-filling method with a PSNR enhancement of 11 dB, a SSIM enhancement of 0.15, and a NMSE decrease of [Formula: see text] in three acquisition orientations. Compared with other deep learning-based models, AMO-CEST showed visual and quantitative improvements in images from three different orientations. We also extracted molecular contrast maps, including the amide proton transfer (APT) and the relayed nuclear Overhauser enhancement (rNOE). The results demonstrated that the CEST contrast maps derived from the CEST images of AMO-CEST were comparable to those derived from the original high-resolution CEST images. The proposed AMO-CEST can efficiently reconstruct high-quality CEST images from under-sampled k-space data and thus has the potential to accelerate CEST-MRI acquisition.

Inflammatory Knee Synovitis: Evaluation of an Accelerated FLAIR Sequence Compared With Standard Contrast-Enhanced Imaging

OBJECTIVES

The aim of this study was to assess the diagnostic value and accuracy of a deep learning (DL)-accelerated fluid attenuated inversion recovery (FLAIR) sequence with fat saturation (FS) in patients with inflammatory synovitis of the knee.

MATERIALS AND METHODS

Patients with suspected knee synovitis were retrospectively included between January and September 2023. All patients underwent a 3 T knee magnetic resonance imaging including a DL-accelerated noncontrast FLAIR FS sequence (acquisition time: 1 minute 38 seconds) and a contrast-enhanced (CE) T1-weighted FS sequence (acquisition time: 4 minutes 50 seconds), which served as reference standard. All knees were scored by 2 radiologists using the semiquantitative modified knee synovitis score, effusion synovitis score, and Hoffa inflammation score. Diagnostic confidence, image quality, and image artifacts were rated on separate Likert scales. Wilcoxon signed rank test was used to compare the semiquantitative scores. Interreader and intrareader reproducibility were calculated using Cohen κ.

RESULTS

Fifty-five patients (mean age, 52 ± 17 years; 28 females) were included in the study. Twenty-seven patients (49%) had mild to moderate synovitis (synovitis score 6-13), and 17 patients (31%) had severe synovitis (synovitis score >14). No signs of synovitis were detected in 11 patients (20%) (synovitis score <5). Semiquantitative assessment of the whole knee synovitis score showed no significant difference between the DL-accelerated FLAIR sequence and the CE T1-weighted sequence (mean FLAIR score: 10.69 ± 8.83, T1 turbo spin-echo FS: 10.74 ± 10.32; P = 0.521). Both interreader and intrareader reproducibility were excellent (range Cohen κ [0.82-0.96]).

CONCLUSIONS

Assessment of inflammatory knee synovitis using a DL-accelerated noncontrast FLAIR FS sequence was feasible and equivalent to CE T1-weighted FS imaging.

Deep learning-accelerated T2WI: image quality, efficiency, and staging performance against BLADE T2WI for gastric cancer

PURPOSE

The purpose of our study is to investigate image quality, efficiency, and diagnostic performance of a deep learning-accelerated single-shot breath-hold (DLSB) against BLADE for T2-weighted MR imaging (T2WI) for gastric cancer (GC).

METHODS

112 patients with GCs undergoing gastric MRI were prospectively enrolled between Aug 2022 and Dec 2022. Axial DLSB-T2WI and BLADE-T2WI of stomach were scanned with same spatial resolution. Three radiologists independently evaluated the image qualities using a 5-scale Likert scales (IQS) in terms of lesion delineation, gastric wall boundary conspicuity, and overall image quality. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated in measurable lesions. T staging was conducted based on the results of both sequences for GC patients with gastrectomy. Pairwise comparisons between DLSB-T2WI and BLADE-T2WI were performed using the Wilcoxon signed-rank test, paired t-test, and chi-squared test. Kendall's W, Fleiss' Kappa, and intraclass correlation coefficient values were used to determine inter-reader reliability.

RESULTS

Against BLADE, DLSB reduced total acquisition time of T2WI from 495 min (mean 4:42 per patient) to 33.6 min (18 s per patient), with better overall image quality that produced 9.43-fold, 8.00-fold, and 18.31-fold IQS upgrading against BALDE, respectively, in three readers. In 69 measurable lesions, DLSB-T2WI had higher mean SNR and higher CNR than BLADE-T2WI. Among 71 patients with gastrectomy, DLSB-T2WI resulted in comparable accuracy to BLADE-T2WI in staging GCs (P > 0.05).

CONCLUSIONS

DLSB-T2WI demonstrated shorter acquisition time, better image quality, and comparable staging accuracy, which could be an alternative to BLADE-T2WI for gastric cancer imaging.

Knowledge-driven deep learning for fast MR imaging: Undersampled MR image reconstruction from supervised to un-supervised learning

Deep learning (DL) has emerged as a leading approach in accelerating MRI. It employs deep neural networks to extract knowledge from available datasets and then applies the trained networks to reconstruct accurate images from limited measurements. Unlike natural image restoration problems, MRI involves physics-based imaging processes, unique data properties, and diverse imaging tasks. This domain knowledge needs to be integrated with data-driven approaches. Our review will introduce the significant challenges faced by such knowledge-driven DL approaches in the context of fast MRI along with several notable solutions, which include learning neural networks and addressing different imaging application scenarios. The traits and trends of these techniques have also been given which have shifted from supervised learning to semi-supervised learning, and finally, to unsupervised learning methods. In addition, MR vendors' choices of DL reconstruction have been provided along with some discussions on open questions and future directions, which are critical for the reliable imaging systems.

Deep Learning Reconstruction of Accelerated MRI: False-Positive Cartilage Delamination Inserted in MRI Arthrography Under Traction

OBJECTIVES

The radiological imaging industry is developing and starting to offer a range of novel artificial intelligence software solutions for clinical radiology. Deep learning reconstruction of magnetic resonance imaging data seems to allow for the acceleration and undersampling of imaging data. Resulting reduced acquisition times would lead to greater machine utility and to greater cost-efficiency of machine operations.

MATERIALS AND METHODS

Our case shows images from magnetic resonance arthrography under traction of the right hip joint from a 30-year-old, otherwise healthy, male patient.

RESULTS

The undersampled image data when reconstructed by a deep learning tool can contain false-positive cartilage delamination and false-positive diffuse cartilage defects.

CONCLUSIONS

In the future, precision of this novel technology will have to be put to thorough testing. Bias of systems, in particular created by the choice of training data, will have to be part of those assessments.

Accelerated motion correction with deep generative diffusion models

PURPOSE

The aim of this work is to develop a method to solve the ill-posed inverse problem of accelerated image reconstruction while correcting forward model imperfections in the context of subject motion during MRI examinations.

METHODS

The proposed solution uses a Bayesian framework based on deep generative diffusion models to jointly estimate a motion-free image and rigid motion estimates from subsampled and motion-corrupt two-dimensional (2D) k-space data.

RESULTS

We demonstrate the ability to reconstruct motion-free images from accelerated two-dimensional (2D) Cartesian and non-Cartesian scans without any external reference signal. We show that our method improves over existing correction techniques on both simulated and prospectively accelerated data.

CONCLUSION

We propose a flexible framework for retrospective motion correction of accelerated MRI based on deep generative diffusion models, with potential application to other forward model corruptions.

Feature Fusion for Multi-Coil Compressed MR Image Reconstruction

Magnetic resonance imaging (MRI) occupies a pivotal position within contemporary diagnostic imaging modalities, offering non-invasive and radiation-free scanning. Despite its significance, MRI's principal limitation is the protracted data acquisition time, which hampers broader practical application. Promising deep learning (DL) methods for undersampled magnetic resonance (MR) image reconstruction outperform the traditional approaches in terms of speed and image quality. However, the intricate inter-coil correlations have been insufficiently addressed, leading to an underexploitation of the rich information inherent in multi-coil acquisitions. In this article, we proposed a method called "Multi-coil Feature Fusion Variation Network" (MFFVN), which introduces an encoder to extract the feature from multi-coil MR image directly and explicitly, followed by a feature fusion operation. Coil reshaping enables the 2D network to achieve satisfactory reconstruction results, while avoiding the introduction of a significant number of parameters and preserving inter-coil information. Compared with VN, MFFVN yields an improvement in the average PSNR and SSIM of the test set, registering enhancements of 0.2622 dB and 0.0021 dB respectively. This uplift can be attributed to the integration of feature extraction and fusion stages into the network's architecture, thereby effectively leveraging and combining the multi-coil information for enhanced image reconstruction quality. The proposed method outperforms the state-of-the-art methods on fastMRI dataset of multi-coil brains under a fourfold acceleration factor without incurring substantial computation overhead.

Highly-accelerated CEST MRI using frequency-offset-dependent k-space sampling and deep-learning reconstruction

PURPOSE

To develop a highly accelerated CEST Z-spectral acquisition method using a specifically-designed k-space sampling pattern and corresponding deep-learning-based reconstruction.

METHODS

For k-space down-sampling, a customized pattern was proposed for CEST, with the randomized probability following a frequency-offset-dependent (FOD) function in the direction of saturation offset. For reconstruction, the convolution network (CNN) was enhanced with a Partially Separable (PS) function to optimize the spatial domain and frequency domain separately. Retrospective experiments on a self-acquired human brain dataset (13 healthy adults and 15 brain tumor patients) were conducted using k-space resampling. The prospective performance was also assessed on six healthy subjects.

RESULTS

In retrospective experiments, the combination of FOD sampling and PS network (FOD + PSN) showed the best quantitative metrics for reconstruction, outperforming three other combinations of conventional sampling with varying density and a regular CNN (nMSE and SSIM, p < 0.001 for healthy subjects). Across all acceleration factors from 4 to 14, the FOD + PSN approach consistently outperformed the comparative methods in four contrast maps including MTRasym, MTRrex, as well as the Lorentzian Difference maps of amide and nuclear Overhauser effect (NOE). In the subspace replacement experiment, the error distribution demonstrated the denoising benefits achieved in the spatial subspace. Finally, our prospective results obtained from healthy adults and brain tumor patients (14×) exhibited the initial feasibility of our method, albeit with less accurate reconstruction than retrospective ones.

CONCLUSION

The combination of FOD sampling and PSN reconstruction enabled highly accelerated CEST MRI acquisition, which may facilitate CEST metabolic MRI for brain tumor patients.

A Feasibility Study of AI-Assisted Compressed Sensing in Prostate T2-Weighted Imaging

RATIONALE AND OBJECTIVES

To evaluate the image quality and PI-RADS scoring performance of prostate T2-weighted imaging (T2WI) based on AI-assisted compressed sensing (ACS).

MATERIALS AND METHODS

In this prospective study, adult male urological outpatients or inpatients underwent prostate MRI, including T2WI, diffusion-weighted imaging and apparent diffusion coefficient maps. Three accelerated scanning protocols using parallel imaging (PI) and ACS: T2WIPI, T2WIACS1 and T2WIACS2 were evaluated through comparative analysis. Quantitative analysis included signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), slope profile, and edge rise distance (ERD). Image quality was qualitatively assessed using a five-point Likert scale (ranging from 1 = non-diagnostic to 5 = excellent). PI-RADS scores were determined for the largest or most suspicious lesions in each patient. The Friedman test and one-way ANOVA with post hoc tests were utilized for group comparisons, with statistical significance set at P < 0.05.

RESULTS

This study included 40 participants. Compared to PI, ACS reduced acquisition time by over 50%, significantly enhancing the CNR of sagittal and axial T2WI (P < 0.05), significantly improving the image quality of sagittal and axial T2WI (P < 0.05). No significant differences were observed in slope profile, ERD, and PI-RADS scores between groups (P > 0.05).

CONCLUSION

ACS reduced prostate T2WI acquisition time by half while improving image quality without affecting PI-RADS scores.

Deep Learning Reconstruction of Prospectively Accelerated MRI of the Pancreas: Clinical Evaluation of Shortened Breath-Hold Examinations With Dixon Fat Suppression

OBJECTIVE

Deep learning (DL)-enabled magnetic resonance imaging (MRI) reconstructions can enable shortening of breath-hold examinations and improve image quality by reducing motion artifacts. Prospective studies with DL reconstructions of accelerated MRI of the upper abdomen in the context of pancreatic pathologies are lacking. In a clinical setting, the purpose of this study is to investigate the performance of a novel DL-based reconstruction algorithm in T1-weighted volumetric interpolated breath-hold examinations with partial Fourier sampling and Dixon fat suppression (hereafter, VIBE-DixonDL). The objective is to analyze its impact on acquisition time, image sharpness and quality, diagnostic confidence, pancreatic lesion conspicuity, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR).

METHODS

This prospective single-center study included participants with various pancreatic pathologies who gave written consent from January 2023 to September 2023. During the same session, each participant underwent 2 MRI acquisitions using a 1.5 T scanner: conventional precontrast and postcontrast T1-weighted VIBE acquisitions with Dixon fat suppression (VIBE-Dixon, reference standard) using 4-fold parallel imaging acceleration and 6-fold accelerated VIBE-Dixon acquisitions with partial Fourier sampling utilizing a novel DL reconstruction tailored to the acquisition. A qualitative image analysis was performed by 4 readers. Acquisition time, image sharpness, overall image quality, image noise and artifacts, diagnostic confidence, as well as pancreatic lesion conspicuity and size were compared. Furthermore, a quantitative analysis of SNR and CNR was performed.

RESULTS

Thirty-two participants were evaluated (mean age ± SD, 62 ± 19 years; 20 men). The VIBE-DixonDL method enabled up to 52% reduction in average breath-hold time (7 seconds for VIBE-DixonDL vs 15 seconds for VIBE-Dixon, P < 0.001). A significant improvement of image sharpness, overall image quality, diagnostic confidence, and pancreatic lesion conspicuity was observed in the images recorded using VIBE-DixonDL (P < 0.001). Furthermore, a significant reduction of image noise and motion artifacts was noted in the images recorded using the VIBE-DixonDL technique (P < 0.001). In addition, for all readers, there was no evidence of a difference in lesion size measurement between VIBE-Dixon and VIBE-DixonDL. Interreader agreement between VIBE-Dixon and VIBE-DixonDL regarding lesion size was excellent (intraclass correlation coefficient, >90). Finally, a statistically significant increase of pancreatic SNR in VIBE-DIXONDL was observed in both the precontrast (P = 0.025) and postcontrast images (P < 0.001). Also, an increase of splenic SNR in VIBE-DIXONDL was observed in both the precontrast and postcontrast images, but only reaching statistical significance in the postcontrast images (P = 0.34 and P = 0.003, respectively). Similarly, an increase of pancreas CNR in VIBE-DIXONDL was observed in both the precontrast and postcontrast images, but only reaching statistical significance in the postcontrast images (P = 0.557 and P = 0.026, respectively).

CONCLUSIONS

The prospectively accelerated, DL-enhanced VIBE with Dixon fat suppression was clinically feasible. It enabled a 52% reduction in breath-hold time and provided superior image quality, diagnostic confidence, and pancreatic lesion conspicuity. This technique might be especially useful for patients with limited breath-hold capacity.

Prospective and multi-reader evaluation of deep learning reconstruction-based accelerated rectal MRI: image quality, diagnostic performance, and reading time

OBJECTIVES

To evaluate deep learning reconstruction (DLR)-based accelerated rectal magnetic resonance imaging (MRI) compared with standard MRI.

MATERIALS AND METHODS

Patients with biopsy-confirmed rectal adenocarcinoma between November/2022 and May/2023 in a single centre were prospectively enrolled for an intra-individual comparison between standard fast spin-echo (FSEstandard) and DLR-based FSE (FSEDL) sequences. Quantitative and qualitative image quality metrics of the pre-therapeutic MRIs were evaluated in all patients; diagnostic performance and evaluating time for T-staging, N-staging, extramural vascular invasion (EMVI), and mesorectal fascia (MRF) status was further analysed in patients undergoing curative surgery, with histopathologic results as the diagnostic gold standard.

RESULTS

A total of 117 patients were enrolled, with 60 patients undergoing curative surgery. FSEDL reduced the acquisition time by 65% than FSEstandard. FSEDL exhibited higher signal-to-noise ratios, contrast-to-noise ratio, and subjective scores (noise, tumour margin clarity, visualisation of bowel wall layering and MRF, overall image quality, and diagnostic confidence) than FSEstandard (p < 0.001). Reduced artefacts were observed in FSEDL for patients without spasmolytics (p < 0.05). FSEDL provided higher T-staging accuracy by junior readers than FSEstandard (reader 1, 58.33% vs 70.00%, p = 0.016; reader 3, 60.00% vs 76.67%, p = 0.021), with similar N-staging, EMVI, and MRF performance. No significant difference was observed for senior readers. FSEDL exhibited shorter diagnostic time in all readers' T-staging and overall evaluation, and junior readers' EMVI and MRF (p < 0.05).

CONCLUSION

FSEDL provided improved image quality, reading time, and junior radiologists' T-staging accuracy than FSEstandard, while reducing the acquisition time by 65%.

CLINICAL RELEVANCE STATEMENT

DLR is clinically applicable for rectal MRI, providing improved image quality with shorter scanning time, which may ease the examination burden. It is beneficial for diagnostic optimisation in improving junior radiologists' T-staging accuracy and reading time.

KEY POINTS

The rising incidence of rectal cancer has demanded enhanced efficiency and quality in imaging examinations. FSEDL demonstrated superior image quality and had a 65% reduced acquisition time. FSEDL can improve the diagnostic accuracy of T-staging and reduce the reading time for assessing rectal cancer.

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 mm3 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.

Accelerated cardiac cine magnetic resonance imaging using deep low-rank plus sparse network: validation in patients

Background

Accurate and reproducible assessment of left ventricular (LV) volumes is important in managing various cardiac conditions. However, patients are required to hold their breath multiple times during data acquisition, which may result in discomfort and restrict cardiac motion, potentially compromising the accuracy of the detected results. Accelerated imaging techniques can help reduce the number of breath holds needed, potentially improving patient comfort and the reliability of the LV assessment. This study aimed to prospectively evaluate the feasibility and accuracy of LV assessment with a model-based low-rank plus sparse network (L+S-Net) for accelerated magnetic resonance (MR) cine imaging.

Methods

Fourty-one patients with different cardiac conditions were recruited in this study. Both accelerated MR cine imaging with L+S-Net and traditional electrocardiogram (ECG)-gated segmented cine were performed for each patient. Subjective image quality (IQ) score and quantitative LV volume function parameters were measured and compared between L+S-Net and traditional standards. The IQ score and LV volume measurements of cardiovascular magnetic resonance (CMR) images reconstructed by L+S-Net and standard cine were compared by paired t-test. The acquisition time of the two methods was also calculated.

Results

In a quantitative analysis, L+S-Net and standard cine yielded similar measurements for all parameters of LV function (ejection fraction: 35±22 for standard vs. 33±23 for L+S-Net), although L+S-Net had slightly lower IQ scores than standard cine CMR (4.2±0.5 for L+S-Net vs. 4.8±0.4 for standard cine; P<0.001). The mean acquisition time of L+S-Net and standard cine was 0.83±0.08 vs. 6.35±0.78 s per slice (P<0.001).

Conclusions

Assessment of LV function with L+S-Net at 3.0 T yields comparable results to the reference standard, albeit with a reduced acquisition time. This feature enhances the clinical applicability of the L+S-Net approach, helping alleviate patient discomfort and motion artifacts that may arise due to prolonged acquisition time.

Rapid 2D 23Na MRI of the calf using a denoising convolutional neural network

PURPOSE

23Na MRI can be used to quantify in-vivo tissue sodium concentration (TSC), but the inherently low 23Na signal leads to long scan times and/or noisy or low-resolution images. Reconstruction algorithms such as compressed sensing (CS) have been proposed to mitigate low signal-to-noise ratio (SNR); although, these can result in unnatural images, suboptimal denoising and long processing times. Recently, machine learning has been increasingly used to denoise 1H MRI acquisitions; however, this approach typically requires large volumes of high-quality training data, which is not readily available for 23Na MRI. Here, we propose using 1H data to train a denoising convolutional neural network (CNN), which we subsequently demonstrate on prospective 23Na images of the calf.

METHODS

1893 1H fat-saturated transverse slices of the knee from the open-source fastMRI dataset were used to train denoising CNNs for different levels of noise. Synthetic low SNR images were generated by adding gaussian noise to the high-quality 1H k-space data before reconstruction to create paired training data. For prospective testing, 23Na images of the calf were acquired in 10 healthy volunteers with a total of 150 averages over ten minutes, which were used as a reference throughout the study. From this data, images with fewer averages were retrospectively reconstructed using a non-uniform fast Fourier transform (NUFFT) as well as CS, with the NUFFT images subsequently denoised using the trained CNN.

RESULTS

CNNs were successfully applied to 23Na images reconstructed with 50, 40 and 30 averages. Muscle and skin apparent TSC quantification from CNN-denoised images were equivalent to those from CS images, with <0.9 mM bias compared to reference values. Estimated SNR was significantly higher in CNN-denoised images compared to NUFFT, CS and reference images. Quantitative edge sharpness was equivalent for all images. For subjective image quality ranking, CNN-denoised images ranked equally best with reference images and significantly better than NUFFT and CS images.

CONCLUSION

Denoising CNNs trained on 1H data can be successfully applied to 23Na images of the calf; thus, allowing scan time to be reduced from ten minutes to two minutes with little impact on image quality or apparent TSC quantification accuracy.

Task-based assessment for neural networks: evaluating undersampled MRI reconstructions based on human observer signal detection

Purpose

Recent research explores using neural networks to reconstruct undersampled magnetic resonance imaging. Because of the complexity of the artifacts in the reconstructed images, there is a need to develop task-based approaches to image quality. We compared conventional global quantitative metrics to evaluate image quality in undersampled images generated by a neural network with human observer performance in a detection task. The purpose is to study which acceleration (2×, 3×, 4×, 5×) would be chosen with the conventional metrics and compare it to the acceleration chosen by human observer performance.

Approach

We used common global metrics for evaluating image quality: the normalized root mean squared error (NRMSE) and structural similarity (SSIM). These metrics are compared with a measure of image quality that incorporates a subtle signal for a specific task to allow for image quality assessment that locally evaluates the effect of undersampling on a signal. We used a U-Net to reconstruct under-sampled images with 2×, 3×, 4×, and 5× one-dimensional undersampling rates. Cross-validation was performed for a 500- and a 4000-image training set with both SSIM and MSE losses. A two-alternative forced choice (2-AFC) observer study was carried out for detecting a subtle signal (small blurred disk) from images with the 4000-image training set.

Results

We found that for both loss functions, the human observer performance on the 2-AFC studies led to a choice of a 2× undersampling, but the SSIM and NRMSE led to a choice of a 3× undersampling.

Conclusions

For this detection task using a subtle small signal at the edge of detectability, SSIM and NRMSE led to an overestimate of the achievable undersampling using a U-Net before a steep loss of image quality between 2×, 3×, 4×, 5× undersampling rates when compared to the performance of human observers in the detection task.

Prospective Deployment of Deep Learning Reconstruction Facilitates Highly Accelerated Upper Abdominal MRI

RATIONALE AND OBJECTIVES

To compare a conventional T1 volumetric interpolated breath-hold examination (VIBE) with SPectral Attenuated Inversion Recovery (SPAIR) fat saturation and a deep learning (DL)-reconstructed accelerated VIBE sequence with SPAIR fat saturation achieving a 50 % reduction in breath-hold duration (hereafter, VIBE-SPAIRDL) in terms of image quality and diagnostic confidence.

MATERIALS AND METHODS

This prospective study enrolled consecutive patients referred for upper abdominal MRI from November 2023 to December 2023 at a single tertiary center. Patients underwent upper abdominal MRI with acquisition of non-contrast and gadobutrol-enhanced conventional VIBE-SPAIR (fourfold acceleration, acquisition time 16 s) and VIBE-SPAIRDL (sixfold acceleration, acquisition time 8 s) on a 1.5 T scanner. Image analysis was performed by four readers, evaluating homogeneity of fat suppression, perceived signal-to-noise ratio (SNR), edge sharpness, artifact level, lesion detectability and diagnostic confidence. A statistical power analysis for patient sample size estimation was performed. Image quality parameters were compared by a repeated measures analysis of variance, and interreader agreement was assessed using Fleiss' κ.

RESULTS

Among 450 consecutive patients, 45 patients were evaluated (mean age, 60 years ± 15 [SD]; 27 men, 18 women). VIBE-SPAIRDL acquisition demonstrated superior SNR (P < 0.001), edge sharpness (P < 0.001), and reduced artifacts (P < 0.001) with substantial to almost perfect interreader agreement for non-contrast (κ: 0.70-0.91) and gadobutrol-enhanced MRI (κ: 0.68-0.87). No evidence of a difference was found between conventional VIBE-SPAIR and VIBE-SPAIRDL regarding homogeneity of fat suppression, lesion detectability, or diagnostic confidence (all P > 0.05).

CONCLUSION

Deep learning reconstruction of VIBE-SPAIR facilitated a reduction of breath-hold duration by half, while reducing artifacts and improving image quality.

SUMMARY

Deep learning reconstruction of prospectively accelerated T1 volumetric interpolated breath-hold examination for upper abdominal MRI enabled a 50 % reduction in breath-hold time with superior image quality.

KEY RESULTS

1) In a prospective analysis of 45 patients referred for upper abdominal MRI, accelerated deep learning (DL)-reconstructed VIBE images with spectral fat saturation (SPAIR) showed better overall image quality, with better perceived signal-to-noise ratio and less artifacts (all P < 0.001), despite a 50 % reduction in acquisition time compared to conventional VIBE. 2) No evidence of a difference was found between conventional VIBE-SPAIR and accelerated VIBE-SPAIRDL regarding lesion detectability or diagnostic confidence.

Dual-domain faster Fourier convolution based network for MR image reconstruction

Deep learning methods for fast MRI have shown promise in reconstructing high-quality images from undersampled multi-coil k-space data, leading to reduced scan duration. However, existing methods encounter challenges related to limited receptive fields in dual-domain (k-space and image domains) reconstruction networks, rigid data consistency operations, and suboptimal refinement structures, which collectively restrict overall reconstruction performance. This study introduces a comprehensive framework that addresses these challenges and enhances MR image reconstruction quality. Firstly, we propose Faster Inverse Fourier Convolution (FasterIFC), a frequency domain convolutional operator that significantly expands the receptive field of k-space domain reconstruction networks. Expanding the information extraction range to the entire frequency spectrum according to the spectral convolution theorem in Fourier theory enables the network to easily utilize richer redundant long-range information from adjacent, symmetrical, and diagonal locations of multi-coil k-space data. Secondly, we introduce a novel softer Data Consistency (softerDC) layer, which achieves an enhanced balance between data consistency and smoothness. This layer facilitates the implementation of diverse data consistency strategies across distinct frequency positions, addressing the inflexibility observed in current methods. Finally, we present the Dual-Domain Faster Fourier Convolution Based Network (D2F2), which features a centrosymmetric dual-domain parallel structure based on FasterIFC. This architecture optimally leverages dual-domain data characteristics while substantially expanding the receptive field in both domains. Coupled with the softerDC layer, D2F2 demonstrates superior performance on the NYU fastMRI dataset at multiple acceleration factors, surpassing state-of-the-art methods in both quantitative and qualitative evaluations.

Deep learning for accelerated and robust MRI reconstruction

Deep learning (DL) has recently emerged as a pivotal technology for enhancing magnetic resonance imaging (MRI), a critical tool in diagnostic radiology. This review paper provides a comprehensive overview of recent advances in DL for MRI reconstruction, and focuses on various DL approaches and architectures designed to improve image quality, accelerate scans, and address data-related challenges. It explores end-to-end neural networks, pre-trained and generative models, and self-supervised methods, and highlights their contributions to overcoming traditional MRI limitations. It also discusses the role of DL in optimizing acquisition protocols, enhancing robustness against distribution shifts, and tackling biases. Drawing on the extensive literature and practical insights, it outlines current successes, limitations, and future directions for leveraging DL in MRI reconstruction, while emphasizing the potential of DL to significantly impact clinical imaging practices.

Large-scale 3D non-Cartesian coronary MRI reconstruction using distributed memory-efficient physics-guided deep learning with limited training data

OBJECT

To enable high-quality physics-guided deep learning (PG-DL) reconstruction of large-scale 3D non-Cartesian coronary MRI by overcoming challenges of hardware limitations and limited training data availability.

MATERIALS AND METHODS

While PG-DL has emerged as a powerful image reconstruction method, its application to large-scale 3D non-Cartesian MRI is hindered by hardware limitations and limited availability of training data. We combine several recent advances in deep learning and MRI reconstruction to tackle the former challenge, and we further propose a 2.5D reconstruction using 2D convolutional neural networks, which treat 3D volumes as batches of 2D images to train the network with a limited amount of training data. Both 3D and 2.5D variants of the PG-DL networks were compared to conventional methods for high-resolution 3D kooshball coronary MRI.

RESULTS

Proposed PG-DL reconstructions of 3D non-Cartesian coronary MRI with 3D and 2.5D processing outperformed all conventional methods both quantitatively and qualitatively in terms of image assessment by an experienced cardiologist. The 2.5D variant further improved vessel sharpness compared to 3D processing, and scored higher in terms of qualitative image quality.

DISCUSSION

PG-DL reconstruction of large-scale 3D non-Cartesian MRI without compromising image size or network complexity is achieved, and the proposed 2.5D processing enables high-quality reconstruction with limited training data.

Deep Learning-based Image Enhancement Techniques for Fast MRI in Neuroimaging

Despite its superior soft tissue contrast and non-invasive nature, MRI requires long scan times due to its intrinsic signal acquisition principles, a main drawback which technological advancements in MRI have been focused on. In particular, scan time reduction is a natural requirement in neuroimaging due to detailed structures requiring high resolution imaging and often volumetric (3D) acquisitions, and numerous studies have recently attempted to harness deep learning (DL) technology in enabling scan time reduction and image quality improvement. Various DL-based image reconstruction products allow for additional scan time reduction on top of existing accelerated acquisition methods without compromising the image quality.

Stop moving: MR motion correction as an opportunity for artificial intelligence

Subject motion is a long-standing problem of magnetic resonance imaging (MRI), which can seriously deteriorate the image quality. Various prospective and retrospective methods have been proposed for MRI motion correction, among which deep learning approaches have achieved state-of-the-art motion correction performance. This survey paper aims to provide a comprehensive review of deep learning-based MRI motion correction methods. Neural networks used for motion artifacts reduction and motion estimation in the image domain or frequency domain are detailed. Furthermore, besides motion-corrected MRI reconstruction, how estimated motion is applied in other downstream tasks is briefly introduced, aiming to strengthen the interaction between different research areas. Finally, we identify current limitations and point out future directions of deep learning-based MRI motion correction.

Improving quantitative MRI using self-supervised deep learning with model reinforcement: Demonstration for rapid T1 mapping

PURPOSE

This paper proposes a novel self-supervised learning framework that uses model reinforcement, REference-free LAtent map eXtraction with MOdel REinforcement (RELAX-MORE), for accelerated quantitative MRI (qMRI) reconstruction. The proposed method uses an optimization algorithm to unroll an iterative model-based qMRI reconstruction into a deep learning framework, enabling accelerated MR parameter maps that are highly accurate and robust.

METHODS

Unlike conventional deep learning methods which require large amounts of training data, RELAX-MORE is a subject-specific method that can be trained on single-subject data through self-supervised learning, making it accessible and practically applicable to many qMRI studies. Using quantitativeT 1 $$ {\mathrm{T}}_1 $$ mapping as an example, the proposed method was applied to the brain, knee and phantom data.

RESULTS

The proposed method generates high-quality MR parameter maps that correct for image artifacts, removes noise, and recovers image features in regions of imperfect image conditions. Compared with other state-of-the-art conventional and deep learning methods, RELAX-MORE significantly improves efficiency, accuracy, robustness, and generalizability for rapid MR parameter mapping.

CONCLUSION

This work demonstrates the feasibility of a new self-supervised learning method for rapid MR parameter mapping, that is readily adaptable to the clinical translation of qMRI.

Learning-Assisted Fast Determination of Regularization Parameter in Constrained Image Reconstruction

OBJECTIVE

To leverage machine learning (ML) for fast selection of optimal regularization parameter in constrained image reconstruction.

METHODS

Constrained image reconstruction is often formulated as a regularization problem and selecting a good regularization parameter value is an essential step. We solved this problem using an ML-based approach by leveraging the finding that for a specific constrained reconstruction problem defined for a fixed class of image functions, the optimal regularization parameter value is weakly subject-dependent and the dependence can be captured using few experimental data. The proposed method has four key steps: a) solution of a given constrained reconstruction problem for a few (say, 3) pre-selected regularization parameter values, b) extraction of multiple approximated quality metrics from the initial reconstructions, c) predicting the true quality metrics values from the approximated values using pre-trained neural networks, and d) determination of the optimal regularization parameter by fusing the predicted quality metrics.

RESULTS

The effectiveness of the proposed method was demonstrated in two constrained reconstruction problems. Compared with L-curve-based method, the proposed method determined the regularization parameters much faster and produced substantially improved reconstructions. Our method also outperformed state-of-the-art learning-based methods when trained with limited experimental data.

CONCLUSION

This paper demonstrates the feasibility and improved reconstruction quality by using machine learning to determine the regularization parameter in constrained reconstruction.

SIGNIFICANCE

The proposed method substantially reduces the computational burden of the traditional methods (e.g., L-curve) or relaxes the requirement of large training data by modern learning-based methods, thus enhancing the practical utility of constrained reconstruction.

CMRxRecon: A publicly available k-space dataset and benchmark to advance deep learning for cardiac MRI

Cardiac magnetic resonance imaging (CMR) has emerged as a valuable diagnostic tool for cardiac diseases. However, a significant drawback of CMR is its slow imaging speed, resulting in low patient throughput and compromised clinical diagnostic quality. The limited temporal resolution also causes patient discomfort and introduces artifacts in the images, further diminishing their overall quality and diagnostic value. There has been growing interest in deep learning-based CMR imaging algorithms that can reconstruct high-quality images from highly under-sampled k-space data. However, the development of deep learning methods requires large training datasets, which have so far not been made publicly available for CMR. To address this gap, we released a dataset that includes multi-contrast, multi-view, multi-slice and multi-coil CMR imaging data from 300 subjects. Imaging studies include cardiac cine and mapping sequences. The 'CMRxRecon' dataset contains raw k-space data and auto-calibration lines. Our aim is to facilitate the advancement of state-of-the-art CMR image reconstruction by introducing standardized evaluation criteria and making the dataset freely accessible to the research community.

Evaluation of deep-learning TSE images in clinical musculoskeletal imaging

In this study, we compared the fat-saturated (FS) and non-FS turbo spin echo (TSE) magnetic resonance imaging knee sequences reconstructed conventionally (conventional-TSE) against a deep learning-based reconstruction of accelerated TSE (DL-TSE) scans. A total of 232 conventional-TSE and DL-TSE image pairs were acquired for comparison. For each consenting patient, one of the clinically acquired conventional-TSE proton density-weighted sequences in the sagittal or coronal planes (FS and non-FS), or in the axial plane (non-FS), was repeated using a research DL-TSE sequence. The DL-TSE reconstruction resulted in an image resolution that increased by at least 45% and scan times that were up to 52% faster compared to the conventional TSE. All images were acquired on a MAGNETOM Vida 3T scanner (Siemens Healthineers AG, Erlangen, Germany). The reporting radiologists, blinded to the acquisition time, were requested to qualitatively compare the DL-TSE against the conventional-TSE reconstructions. Despite having a faster acquisition time, the DL-TSE was rated to depict smaller structures better for 139/232 (60%) cases, equivalent for 72/232 (31%) cases and worse for 21/232 (9%) cases compared to the conventional-TSE. Overall, the radiologists preferred the DL-TSE reconstruction in 124/232 (53%) cases and stated no preference, implying equivalence, for 65/232 (28%) cases. DL-TSE reconstructions enabled faster acquisition times while enhancing spatial resolution and preserving the image contrast. From these results, the DL-TSE provided added or comparable clinical value and utility in less time. DL-TSE offers the opportunity to further reduce the overall examination time and improve patient comfort with no loss in diagnostic accuracy.