Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction

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

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

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

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

Deep learning-based denoising image reconstruction of body magnetic resonance imaging in children

BACKGROUND

Radial k-space sampling is widely employed in paediatric magnetic resonance imaging (MRI) to mitigate motion and aliasing artefacts. Artificial intelligence (AI)-based image reconstruction has been developed to enhance image quality and accelerate acquisition time.

OBJECTIVE

To assess image quality of deep learning (DL)-based denoising image reconstruction of body MRI in children.

MATERIALS AND METHODS

Children who underwent thoraco-abdominal MRI employing radial k-space filling technique (PROPELLER) with conventional and DL-based image reconstruction between April 2022 and January 2023 were eligible for this retrospective study. Only cases with previous MRI including comparable PROPELLER sequences with conventional image reconstruction were selected. Image quality was compared between DL-reconstructed axial T1-weighted and T2-weighted images and conventionally reconstructed images from the same PROPELLER acquisition. Quantitative image quality was assessed by signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the liver and spleen. Qualitative image quality was evaluated by three observers using a 4-point Likert scale and included presence of noise, motion artefact, depiction of peripheral lung vessels and subsegmental bronchi at the lung bases, sharpness of abdominal organ borders, and visibility of liver and spleen vessels. Image quality was compared with the Wilcoxon signed-rank test. Scan time length was compared to prior MRI obtained with conventional image reconstruction.

RESULTS

In 21 children (median age 7 years, range 1.5 years to 15.8 years) included, the SNR and CNR of the liver and spleen on T1-weighted and T2-weighted images were significantly higher with DL-reconstruction (P<0.001) than with conventional reconstruction. The DL-reconstructed images showed higher overall image quality, with improved delineation of the peripheral vessels and the subsegmental bronchi in the lung bases, sharper abdominal organ margins and increased visibility of the peripheral vessels in the liver and spleen. Not respiratory-gated DL-reconstructed T1-weighted images demonstrated more pronounced respiratory motion artefacts in comparison to conventional reconstruction (P=0.015), while there was no difference for the respiratory-gated T2-weighted images. The median scan time per slice was reduced from 6.3 s (interquartile range, 4.2 - 7.0 s) to 4.8 s (interquartile range, 4.4 - 4.9 s) for the T1-weighted images and from 5.6 s (interquartile range, 5.4 - 5.9 s) to 4.2 s (interquartile range, 3.9 - 4.8 s) for the T2-weighted images.

CONCLUSION

DL-based denoising image reconstruction of paediatric body MRI sequences employing radial k-space sampling allowed for improved overall image quality at shorter scan times. Respiratory motion artefacts were more pronounced on ungated T1-weighted images.

Accelerated intracranial time-of-flight MR angiography with image-based deep learning image enhancement reduces scan times and improves image quality at 3-T and 1.5-T

PURPOSE

Three-dimensional time-of-flight magnetic resonance angiography (TOF-MRA) is effective for cerebrovascular disease assessment, but clinical application is limited by long scan times and low spatial resolution. Recent advances in deep learning-based reconstruction have shown the potential to improve image quality and reduce scan times. This study aimed to evaluate the effectiveness of accelerated intracranial TOF-MRA using deep learning-based image enhancement (TOF-DL) compared to conventional TOF-MRA (TOF-Con) at both 3-T and 1.5-T.

MATERIALS AND METHODS

In this retrospective study, patients who underwent both conventional and 40% accelerated TOF-MRA protocols on 1.5-T or 3-T scanners from July 2022 to March 2023 were included. A commercially available DL-based image enhancement algorithm was applied to the accelerated MRA. Quantitative image quality assessments included signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), contrast ratio (CR), and vessel sharpness (VS), while qualitative assessments were conducted using a five-point Likert scale. Cohen's d was used to compare the quantitative image metrics, and a cumulative link mixed regression model analyzed the readers' scores.

RESULTS

A total of 129 patients (mean age, 64 years ± 12 [SD], 99 at 3-T and 30 at 1.5-T) were included. TOF-DL showed significantly higher SNR, CNR, CR, and VS compared to TOF-Con (CNR = 183.89 vs. 45.58; CR = 0.63 vs. 0.59; VS = 0.73 vs. 0.61; all p < 0.001). The improvement in VS was more pronounced at 1.5-T (Cohen's d = 2.39) compared to 3-T HR and routine (Cohen's d = 0.83 and 0.75, respectively). TOF-DL also outperformed TOF-Con in qualitative image parameters, enhancing the visibility of small- and medium-sized vessels, regardless of the degree of resolution and field strength. TOF-DL showed comparable diagnostic accuracy (AUC: 0.77-0.85) to TOF-Con (AUC: 0.79-0.87) but had higher specificity for steno-occlusive lesions.

CONCLUSIONS

Accelerated intracranial MRA with deep learning-based reconstruction reduces scan times by 40% and significantly enhances image quality over conventional TOF-MRA at both 3-T and 1.5-T.

CAMP-Net: Consistency-Aware Multi-Prior Network for Accelerated MRI Reconstruction

Undersampling -space data in magnetic resonance imaging (MRI) reduces scan time but pose challenges in image reconstruction. Considerable progress has been made in reconstructing accelerated MRI. However, restoration of high-frequency image details in highly undersampled data remains challenging. To address this issue, we propose CAMP-Net, an unrolling-based Consistency-Aware Multi-Prior Network for accelerated MRI reconstruction. CAMP-Net leverages complementary multi-prior knowledge and multi-slice information from various domains to enhance reconstruction quality. Specifically, CAMP-Net comprises three interleaved modules for image enhancement, -space restoration, and calibration consistency, respectively. These modules jointly learn priors from data in image domain, -domain, and calibration region, respectively, in data-driven manner during each unrolled iteration. Notably, the encoded calibration prior knowledge extracted from auto-calibrating signals implicitly guides the learning of consistency-aware -space correlation for reliable interpolation of missing -space data. To maximize the benefits of image domain and -domain prior knowledge, the reconstructions are aggregated in a frequency fusion module, exploiting their complementary properties to optimize the trade-off between artifact removal and fine detail preservation. Additionally, we incorporate a surface data fidelity layer during the learning of -domain and calibration domain priors to prevent degradation of the reconstruction caused by padding-induced data imperfections. We evaluate the generalizability and robustness of our method on three large public datasets with varying acceleration factors and sampling patterns. The experimental results demonstrate that our method outperforms state-of-the-art approaches in terms of both reconstruction quality and mapping estimation, particularly in scenarios with high acceleration factors.

Accelerated cardiac cine with spatio-coil regularized deep learning reconstruction

PURPOSE

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

METHODS

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

RESULTS

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

CONCLUSIONS

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

Deep learning-based Intraoperative MRI reconstruction

BACKGROUND

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

METHODS

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

RESULTS

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

CONCLUSION

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

RELEVANCE STATEMENT

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

KEY POINTS

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

The scientific evidence of commercial AI products for MRI acceleration: a systematic review

OBJECTIVES

This study explores the methods employed by commercially available AI products to accelerate MRI protocols and investigates the strength of their diagnostic image quality assessment.

MATERIALS AND METHODS

All commercial AI products for MRI acceleration were identified from the exhibitors presented at the RSNA 2023 and ECR 2024 annual meetings. Peer-reviewed scientific articles describing validation of clinical performance were searched for each product. Information was extracted regarding the MRI acceleration technique, achieved acceleration, diagnostic performance metrics, test cohort, and hallucinatory artifacts. The strength of the diagnostic image quality was assessed using scientific evidence levels ranging from "product's technical feasibility for clinical purposes" to "product's economic impact on society".

RESULTS

Out of 1046 companies, 14 products of 14 companies were included. No scientific articles were found for four products (29%). For the remaining ten products (71%), 21 articles were retrieved. Four acceleration methods were identified: noise reduction, raw data reconstruction, personalized scanning protocols, and synthetic image generation. Only a limited number of articles prospectively demonstrated impact on patient outcomes (n = 4, 19%), and no articles discussed an evaluation in a prospective cohort of > 100 patients or performed an economic analysis. None of the articles performed an analysis of hallucinatory artifacts.

CONCLUSION

Currently, commercially available AI products for MRI acceleration can be categorized into four main methods. The acceleration methods lack prospective scientific evidence on clinical performance in large cohorts and economic analysis, which would help to get a better insight into their diagnostic performance and enable safe and effective clinical implementation.

KEY POINTS

Question There is a growing interest in AI products that reduce MRI scan time, but an overview of these methods and their scientific evidence is missing. Findings Only a limited number of articles (n = 4, 19%) prospectively demonstrated the impact of the software for accelerating MRI on diagnostic performance metrics. Clinical relevance Although various commercially available products shorten MRI acquisition time, more studies in large cohorts are needed to get a better insight into the diagnostic performance of AI-constructed MRI.

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

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

Unsupervised reconstruction of accelerated cardiac cine MRI using neural fields

BACKGROUND

Cardiac cine MRI is the gold standard for cardiac functional assessment, but the inherently slow acquisition process creates the necessity of reconstruction approaches for accelerated undersampled acquisitions. Several regularization approaches that exploit spatial-temporal redundancy have been proposed to reconstruct undersampled cardiac cine MRI. More recently, methods based on supervised deep learning have been also proposed to further accelerate acquisition and reconstruction. However, these techniques rely on usually large dataset for training, which are not always available and might introduce biases.

METHODS

In this work we propose NF-cMRI, an unsupervised approach based on implicit neural field representations for cardiac cine MRI. We evaluate our method in in-vivo undersampled golden-angle radial multi-coil acquisitions for undersampling factors of 13x, 17x and 26x.

RESULTS

The proposed method achieves excellent scores in sharpness and robustness to artifacts and comparable or improved spatial-temporal depiction than state-of-the-art conventional and unsupervised deep learning reconstruction techniques.

CONCLUSIONS

We have demonstrated NF-cMRI potential for cardiac cine MRI reconstruction with highly undersampled data.

Accelerating multi-coil MR image reconstruction using weak supervision

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

Evaluation of Image Quality and Scan Time Efficiency in Accelerated 3D T1-Weighted Pediatric Brain MRI Using Deep Learning-Based Reconstruction

OBJECTIVE

This study evaluated the effect of an accelerated three-dimensional (3D) T1-weighted pediatric brain MRI protocol using a deep learning (DL)-based reconstruction algorithm on scan time and image quality.

MATERIALS AND METHODS

This retrospective study included 46 pediatric patients who underwent conventional and accelerated, pre- and post-contrast, 3D T1-weighted brain MRI using a 3T scanner (SIGNA Premier; GE HealthCare) at a single tertiary referral center between March 1, 2023, and April 30, 2023. Conventional scans were reconstructed using intensity Filter A (Conv), whereas accelerated scans were reconstructed using intensity Filter A (Fast_A) and a DL-based algorithm (Fast_DL). Image quality was assessed quantitatively based on the coefficient of variation, relative contrast, apparent signal-to-noise ratio (aSNR), and apparent contrast-to-noise ratio (aCNR) and qualitatively according to radiologists' ratings of overall image quality, artifacts, noisiness, gray-white matter differentiation, and lesion conspicuity.

RESULTS

The acquisition times for the pre- and post-contrast scans were 191 and 135 seconds, respectively, for the conventional scan. With the accelerated protocol, these were reduced to 135 and 80 seconds, achieving time reductions of 29.3% and 40.7%, respectively. DL-based reconstruction significantly reduced the coefficient of variation, improved the aSNR, aCNR, and overall image quality, and reduced the number of artifacts compared with the conventional acquisition method (all P < 0.05). However, the lesion conspicuity remained similar between the two protocols.

CONCLUSION

Utilizing a DL-based reconstruction algorithm in accelerated 3D T1-weighted pediatric brain MRI can significantly shorten the acquisition time, enhance image quality, and reduce artifacts, making it a viable option for pediatric imaging.

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.

Artificial Intelligence-Empowered Radiology-Current Status and Critical Review

Humanity stands at a pivotal moment of technological revolution, with artificial intelligence (AI) reshaping fields traditionally reliant on human cognitive abilities. This transition, driven by advancements in artificial neural networks, has transformed data processing and evaluation, creating opportunities for addressing complex and time-consuming tasks with AI solutions. Convolutional networks (CNNs) and the adoption of GPU technology have already revolutionized image recognition by enhancing computational efficiency and accuracy. In radiology, AI applications are particularly valuable for tasks involving pattern detection and classification; for example, AI tools have enhanced diagnostic accuracy and efficiency in detecting abnormalities across imaging modalities through automated feature extraction. Our analysis reveals that neuroimaging and chest imaging, as well as CT and MRI modalities, are the primary focus areas for AI products, reflecting their high clinical demand and complexity. AI tools are also used to target high-prevalence diseases, such as lung cancer, stroke, and breast cancer, underscoring AI's alignment with impactful diagnostic needs. The regulatory landscape is a critical factor in AI product development, with the majority of products certified under the Medical Device Directive (MDD) and Medical Device Regulation (MDR) in Class IIa or Class I categories, indicating compliance with moderate-risk standards. A rapid increase in AI product development from 2017 to 2020, peaking in 2020 and followed by recent stabilization and saturation, was identified. In this work, the authors review the advancements in AI-based imaging applications, underscoring AI's transformative potential for enhanced diagnostic support and focusing on the critical role of CNNs, regulatory challenges, and potential threats to human labor in the field of diagnostic imaging.

A Flow-based Truncated Denoising Diffusion Model for super-resolution Magnetic Resonance Spectroscopic Imaging

Magnetic Resonance Spectroscopic Imaging (MRSI) is a non-invasive imaging technique for studying metabolism and has become a crucial tool for understanding neurological diseases, cancers and diabetes. High spatial resolution MRSI is needed to characterize lesions, but in practice MRSI is acquired at low resolution due to time and sensitivity restrictions caused by the low metabolite concentrations. Therefore, there is an imperative need for a post-processing approach to generate high-resolution MRSI from low-resolution data that can be acquired fast and with high sensitivity. Deep learning-based super-resolution methods provided promising results for improving the spatial resolution of MRSI, but they still have limited capability to generate accurate and high-quality images. Recently, diffusion models have demonstrated superior learning capability than other generative models in various tasks, but sampling from diffusion models requires iterating through a large number of diffusion steps, which is time-consuming. This work introduces a Flow-based Truncated Denoising Diffusion Model (FTDDM) for super-resolution MRSI, which shortens the diffusion process by truncating the diffusion chain, and the truncated steps are estimated using a normalizing flow-based network. The network is conditioned on upscaling factors to enable multi-scale super-resolution. To train and evaluate the deep learning models, we developed a 1H-MRSI dataset acquired from 25 high-grade glioma patients. We demonstrate that FTDDM outperforms existing generative models while speeding up the sampling process by over 9-fold compared to the baseline diffusion model. Neuroradiologists' evaluations confirmed the clinical advantages of our method, which also supports uncertainty estimation and sharpness adjustment, extending its potential clinical applications.

Beyond the Conventional Structural MRI: Clinical Application of Deep Learning Image Reconstruction and Synthetic MRI of the Brain

ABSTRACT

Recent technological advancements have revolutionized routine brain magnetic resonance imaging (MRI) sequences, offering enhanced diagnostic capabilities in intracranial disease evaluation. This review explores 2 pivotal breakthrough areas: deep learning reconstruction (DLR) and quantitative MRI techniques beyond conventional structural imaging. DLR using deep neural networks facilitates accelerated imaging with improved signal-to-noise ratio and spatial resolution, enhancing image quality with short scan times. DLR focuses on supervised learning applied to clinical implementation and applications. Quantitative MRI techniques, exemplified by 2D multidynamic multiecho, 3D quantification using interleaved Look-Locker acquisition sequences with T2 preparation pulses, and magnetic resonance fingerprinting, enable precise calculation of brain-tissue parameters and further advance diagnostic accuracy and efficiency. Potential DLR instabilities and quantification and bias limitations will be discussed. This review underscores the synergistic potential of DLR and quantitative MRI, offering prospects for improved brain imaging beyond conventional methods.

Agreement of Image Quality Metrics with Radiological Evaluation in the Presence of Motion Artifacts

Purpose

Reliable image quality assessment is crucial for evaluating new motion correction methods for magnetic resonance imaging. In this work, we compare the performance of commonly used reference-based and reference-free image quality metrics on a unique dataset with real motion artifacts. We further analyze the image quality metrics' robustness to typical pre-processing techniques.

Methods

We compared five reference-based and five reference-free image quality metrics on data acquired with and without intentional motion (2D and 3D sequences). The metrics were recalculated seven times with varying pre-processing steps. The anonymized images were rated by radiologists and radiographers on a 1-5 Likert scale. Spearman correlation coefficients were computed to assess the relationship between image quality metrics and observer scores.

Results

All reference-based image quality metrics showed strong correlation with observer assessments, with minor performance variations across sequences. Among reference-free metrics, Average Edge Strength offers the most promising results, as it consistently displayed stronger correlations across all sequences compared to the other reference-free metrics. Overall, the strongest correlation was achieved with percentile normalization and restricting the metric values to the skull-stripped brain region. In contrast, correlations were weaker when not applying any brain mask and using min-max or no normalization.

Conclusion

Reference-based metrics reliably correlate with radiological evaluation across different sequences and datasets. Pre-processing steps, particularly normalization and brain masking, significantly influence the correlation values. Future research should focus on refining pre-processing techniques and exploring machine learning approaches for automated image quality evaluation.

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

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

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.

Improved Resolution and Image Quality of Musculoskeletal Magnetic Resonance Imaging using Deep Learning-based Denoising Reconstruction: A Prospective Clinical Study

OBJECTIVE

To prospectively evaluate a deep learning-based denoising reconstruction (DLR) for improved resolution and image quality in musculoskeletal (MSK) magnetic resonance imaging (MRI).

METHODS

Images from 137 contrast-weighted sequences in 40 MSK patients were evaluated. Each sequence was performed twice, first with the routine parameters and reconstructed with a routine reconstruction filter (REF), then with higher resolution and reconstructed with DLR, and with three conventional reconstruction filters (NL2, GA43, GA53). The five reconstructions (REF, DLR, NL2, GA43, and GA53) were de-identified, randomized, and blindly reviewed by three MSK radiologists using eight scoring criteria and a forced ranking. Quantitative SNR, CNR, and structure's full width at half maximum (FWHM) for resolution assessment were measured and compared. To account for repeated measures, Generalized Estimating Equations (GEE) with Bonferroni adjustment was used to compare the reader's scores, SNR, CNR, and FWHM between DLR vs. NL2, GA43, GA53, and REF.

RESULTS

Compared to the routine REF images, the resolution was improved by 47.61% with DLR from 0.39 ± 0.15 mm2 to 0.20 ± 0.06 mm2 (p < 0.001). Per-sequence average scan time was shortened by 7.93% with DLR from 165.58 ± 21.86 s to 152.45 ± 25.65 s (p < 0.001). Based on the average scores, DLR images were rated significantly higher in all image quality criteria and the forced ranking (p < 0.001).

CONCLUSION

This prospective clinical evaluation demonstrated that DLR allows approximately two times finer resolution and improved image quality compared to the standard-of-care images.

Deep-learning-based image reconstruction with limited data: generating synthetic raw data using deep learning

OBJECT

Deep learning has shown great promise for fast reconstruction of accelerated MRI acquisitions by learning from large amounts of raw data. However, raw data is not always available in sufficient quantities. This study investigates synthetic data generation to complement small datasets and improve reconstruction quality.

MATERIALS AND METHODS

An adversarial auto-encoder was trained to generate phase and coil sensitivity maps from magnitude images, which were combined into synthetic raw data. On a fourfold accelerated MR reconstruction task, deep-learning-based reconstruction networks were trained with varying amounts of training data (20 to 160 scans). Test set performance was compared between baseline experiments and experiments that incorporated synthetic training data.

RESULTS

Training with synthetic raw data showed decreasing reconstruction errors with increasing amounts of training data, but importantly this was magnitude-only data, rather than real raw data. For small training sets, training with synthetic data decreased the mean absolute error (MAE) by up to 7.5%, whereas for larger training sets the MAE increased by up to 2.6%.

DISCUSSION

Synthetic raw data generation improved reconstruction quality in scenarios with limited training data. A major advantage of synthetic data generation is that it allows for the reuse of magnitude-only datasets, which are more readily available than raw datasets.

Spectro-ViT: A vision transformer model for GABA-edited MEGA-PRESS reconstruction using spectrograms

This study investigated the use of a Vision Transformer (ViT) for reconstructing GABA-edited Magnetic Resonance Spectroscopy (MRS) data from a reduced number of transients. Transients refer to the samples collected during an MRS acquisition by repeating the experiment to generate a signal of sufficient quality. Specifically, 80 transients were used instead of the typical 320 transients, aiming to reduce scan time. The 80 transients were pre-processed and converted into a spectrogram image representation using the Short-Time Fourier Transform (STFT). A pre-trained ViT, named Spectro-ViT, was fine-tuned and then tested using in-vivo GABA-edited MEGA-PRESS data. Its performance was compared against other pipelines in the literature using quantitative quality metrics and estimated metabolite concentration values, with the typical 320-transient scans serving as the reference for comparison. The Spectro-ViT model exhibited the best overall quality metrics among all other pipelines against which it was compared. The metabolite concentrations from Spectro-ViT's reconstructions for GABA+ achieved the best average R2 value of 0.67 and the best average Mean Absolute Percentage Error (MAPE) value of 9.68%, with no significant statistical differences found compared to the 320-transient reference. The code to reproduce this research is available at https://github.com/MICLab-Unicamp/Spectro-ViT.

An unrolled neural network for accelerated dynamic MRI based on second-order half-quadratic splitting model

The reconstruction of dynamic magnetic resonance images from incomplete k-space data has sparked significant research interest due to its potential to reduce scan time. However, traditional iterative optimization algorithms fail to faithfully reconstruct images at higher acceleration factors and incur long reconstruction time. Furthermore, end-to-end deep learning-based reconstruction algorithms suffer from large model parameters and lack robustness in the reconstruction results. Recently, unrolled deep learning models, have shown immense potential in algorithm stability and applicability flexibility. In this paper, we propose an unrolled deep learning network based on a second-order Half-Quadratic Splitting(HQS) algorithm, where the forward propagation process of this framework strictly follows the computational flow of the HQS algorithm. In particular, we propose a degradation-sense module by associating random sampling patterns with intermediate variables to guide the iterative process. We introduce the Information Fusion Transformer(IFT) to extract both local and non-local prior information from image sequences, thereby removing aliasing artifacts resulting from random undersampling. Finally, we impose low-rank constraints within the HQS algorithm to further enhance the reconstruction results. The experiments demonstrate that each component module of our proposed model contributes to the improvement of the reconstruction task. Our proposed method achieves comparably satisfying performance to the state-of-the-art methods and it exhibits excellent generalization capabilities across different sampling masks. At the low acceleration factor, there is a 0.7% enhancement in the PSNR. Furthermore, when the acceleration factor reached 8 and 12, the PSNR achieves an improvement of 3.4% and 5.8% respectively.

Assessing deep learning reconstruction for faster prostate MRI: visual vs. diagnostic performance metrics

OBJECTIVE

Deep learning (DL) MRI reconstruction enables fast scan acquisition with good visual quality, but the diagnostic impact is often not assessed because of large reader study requirements. This study used existing diagnostic DL to assess the diagnostic quality of reconstructed images.

MATERIALS AND METHODS

A retrospective multisite study of 1535 patients assessed biparametric prostate MRI between 2016 and 2020. Likely clinically significant prostate cancer (csPCa) lesions (PI-RADS ≥ 4) were delineated by expert radiologists. T2-weighted scans were retrospectively undersampled, simulating accelerated protocols. DL reconstruction (DLRecon) and diagnostic DL detection (DLDetect) were developed. The effect on the partial area under (pAUC), the Free-Response Operating Characteristic (FROC) curve, and the structural similarity (SSIM) were compared as metrics for diagnostic and visual quality, respectively. DLDetect was validated with a reader concordance analysis. Statistical analysis included Wilcoxon, permutation, and Cohen's kappa tests for visual quality, diagnostic performance, and reader concordance.

RESULTS

DLRecon improved visual quality at 4- and 8-fold (R4, R8) subsampling rates, with SSIM (range: -1 to 1) improved to 0.78 ± 0.02 (p < 0.001) and 0.67 ± 0.03 (p < 0.001) from 0.68 ± 0.03 and 0.51 ± 0.03, respectively. However, diagnostic performance at R4 showed a pAUC FROC of 1.33 (CI 1.28-1.39) for DL and 1.29 (CI 1.23-1.35) for naive reconstructions, both significantly lower than fully sampled pAUC of 1.58 (DL: p = 0.024, naïve: p = 0.02). Similar trends were noted for R8.

CONCLUSION

DL reconstruction produces visually appealing images but may reduce diagnostic accuracy. Incorporating diagnostic AI into the assessment framework offers a clinically relevant metric essential for adopting reconstruction models into clinical practice.

CLINICAL RELEVANCE STATEMENT

In clinical settings, caution is warranted when using DL reconstruction for MRI scans. While it recovered visual quality, it failed to match the prostate cancer detection rates observed in scans not subjected to acceleration and DL reconstruction.

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.

vSHARP: Variable Splitting Half-quadratic ADMM algorithm for reconstruction of inverse-problems

Medical Imaging (MI) tasks, such as accelerated parallel Magnetic Resonance Imaging (MRI), often involve reconstructing an image from noisy or incomplete measurements. This amounts to solving ill-posed inverse problems, where a satisfactory closed-form analytical solution is not available. Traditional methods such as Compressed Sensing (CS) in MRI reconstruction can be time-consuming or prone to obtaining low-fidelity images. Recently, a plethora of Deep Learning (DL) approaches have demonstrated superior performance in inverse-problem solving, surpassing conventional methods. In this study, we propose vSHARP (variable Splitting Half-quadratic ADMM algorithm for Reconstruction of inverse Problems), a novel DL-based method for solving ill-posed inverse problems arising in MI. vSHARP utilizes the Half-Quadratic Variable Splitting method and employs the Alternating Direction Method of Multipliers (ADMM) to unroll the optimization process. For data consistency, vSHARP unrolls a differentiable gradient descent process in the image domain, while a DL-based denoiser, such as a U-Net architecture, is applied to enhance image quality. vSHARP also employs a dilated-convolution DL-based model to predict the Lagrange multipliers for the ADMM initialization. We evaluate vSHARP on tasks of accelerated parallel MRI Reconstruction using two distinct datasets and on accelerated parallel dynamic MRI Reconstruction using another dataset. Our comparative analysis with state-of-the-art methods demonstrates the superior performance of vSHARP in these applications.

ISRSL0: compressed sensing MRI with image smoothness regularized-smoothed [Formula: see text] norm

The reconstruction of MR images has always been a challenging inverse problem in medical imaging. Acceleration of MR scanning is of great importance for clinical research and cutting-edge applications. One of the primary efforts to achieve this is using compressed sensing (CS) theory. The CS aims to reconstruct MR images using a small number of sampled data in k-space. The CS-MRI techniques face challenges, including the potential loss of fine structure and increased computational complexity. We introduce a novel framework based on a regularized sparse recovery problem and a sharpening step to improve the CS-MRI approaches regarding fine structure loss under high acceleration factors. This problem is solved via the Half Quadratic Splitting (HQS) approach. The inverse problem for reconstructing MR images is converted into two distinct sub-problems, each of which can be solved separately. One key feature of the proposed approach is the replacement of one sub-problem with a denoiser. This regularization assists the optimization of the Smoothed [Formula: see text] (SL0) norm in escaping local minimums and enhances its precision. The proposed method consists of smoothing, feature modification, and Smoothed [Formula: see text] cost function optimization. The proposed approach improves the SL0 algorithm for MRI reconstruction without complicating it. The convergence of the proposed approach is illustrated analytically. The experimental results show an acceptable performance of the proposed method compared to the network-based approaches.

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.

Evaluation of 3D T1-weighted spoiled gradient echo MR image quality using artificial intelligence image reconstruction techniques in the pediatric brain

PURPOSE

To assess image quality and diagnostic confidence of 3D T1-weighted spoiled gradient echo (SPGR) MRI using artificial intelligence (AI) reconstruction.

MATERIALS AND METHODS

This prospective, IRB-approved study enrolled 50 pediatric patients (mean age = 11.8 ± 3.1 years) undergoing clinical brain MRI. In addition to standard of care (SOC) compressed SENSE (CS = 2.5), 3D T1-weighted SPGR images were obtained with higher CS acceleration factors (5 and 8) to evaluate the ability of AI reconstruction to improve image quality and reduce scan time. Images were reviewed independently on dedicated research PACS workstations by two neuroradiologists. Quantitative analysis of signal intensities to calculate apparent grey and white matter signal to noise (aSNR) and grey-white matter apparent contrast to noise ratios (aCNR) was performed.

RESULTS

AI improved overall image quality compared to standard CS reconstruction in 35% (35/100) of evaluations in CS = 2.5 (average scan time = 221 ± 6.9 s), 100% (46/46) of CS = 5 (average scan time = 113.3 ± 4.6 s) and 94% (47/50) of CS = 8 (average scan time = 74.1 ± 0.01 s). Quantitative analysis revealed significantly higher grey matter aSNR, white matter aSNR and grey-white matter aCNR with AI reconstruction compared to standard reconstruction for CS 5 and 8 (all p-values < 0.001), however not for CS 2.5.

CONCLUSIONS

AI reconstruction improved overall image quality and gray-white matter qualitative and quantitative aSNR and aCNR in highly accelerated (CS = 5 and 8) 3D T1W SPGR images in the majority of pediatric patients.

Frequency and phase correction of GABA-edited magnetic resonance spectroscopy using complex-valued convolutional neural networks

PURPOSE

To determine the significance of complex-valued inputs and complex-valued convolutions compared to real-valued inputs and real-valued convolutions in convolutional neural networks (CNNs) for frequency and phase correction (FPC) of GABA-edited magnetic resonance spectroscopy (MRS) data.

METHODS

An ablation study using simulated data was performed to determine the most effective input (real or complex) and convolution type (real or complex) to predict frequency and phase shifts in GABA-edited MEGA-PRESS data using CNNs. The best CNN model was subsequently compared using both simulated and in vivo data to two recently proposed deep learning (DL) methods for FPC of GABA-edited MRS. All methods were trained using the same experimental setup and evaluated using the signal-to-noise ratio (SNR) and linewidth of the GABA peak, choline artifact, and by visually assessing the reconstructed final difference spectrum. Statistical significance was assessed using the Wilcoxon signed rank test.

RESULTS

The ablation study showed that using complex values for the input represented by real and imaginary channels in our model input tensor, with complex convolutions was most effective for FPC. Overall, in the comparative study using simulated data, our CC-CNN model (that received complex-valued inputs with complex convolutions) outperformed the other models as evaluated by the mean absolute error.

CONCLUSION

Our results indicate that the optimal CNN configuration for GABA-edited MRS FPC uses a complex-valued input and complex convolutions. Overall, this model outperformed existing DL models.

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.

Learning a stable approximation of an existing but unknown inverse mapping: application to the half-time circular Radon transform

Supervised deep learning-based methods have inspired a new wave of image reconstruction methods that implicitly learn effective regularization strategies from a set of training data. While they hold potential for improving image quality, they have also raised concerns regarding their robustness. Instabilities can manifest when learned methods are applied to find approximate solutions to ill-posed image reconstruction problems for which a unique and stable inverse mapping does not exist, which is a typical use case. In this study, we investigate the performance of supervised deep learning-based image reconstruction in an alternate use case in which a stable inverse mapping is known to exist but is not yet analytically available in closed form. For such problems, a deep learning-based method can learn a stable approximation of the unknown inverse mapping that generalizes well to data that differ significantly from the training set. The learned approximation of the inverse mapping eliminates the need to employ an implicit (optimization-based) reconstruction method and can potentially yield insights into the unknown analytic inverse formula. The specific problem addressed is image reconstruction from a particular case of radially truncated circular Radon transform (CRT) data, referred to as 'half-time' measurement data. For the half-time image reconstruction problem, we develop and investigate a learned filtered backprojection method that employs a convolutional neural network to approximate the unknown filtering operation. We demonstrate that this method behaves stably and readily generalizes to data that differ significantly from training data. The developed method may find application to wave-based imaging modalities that include photoacoustic computed tomography.

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.

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.

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.

Evaluation of T2W FLAIR MR image quality using artificial intelligence image reconstruction techniques in the pediatric brain

BACKGROUND

Artificial intelligence (AI) reconstruction techniques have the potential to improve image quality and decrease imaging time. However, these techniques must be assessed for safe and effective use in clinical practice.

OBJECTIVE

To assess image quality and diagnostic confidence of AI reconstruction in the pediatric brain on fluid-attenuated inversion recovery (FLAIR) imaging.

MATERIALS AND METHODS

This prospective, institutional review board (IRB)-approved study enrolled 50 pediatric patients (median age=12 years, Q1=10 years, Q3=14 years) undergoing clinical brain MRI. T2-weighted (T2W) FLAIR images were reconstructed by both standard clinical and AI reconstruction algorithms (strong denoising). Images were independently rated by two neuroradiologists on a dedicated research picture archiving and communication system (PACS) to indicate whether AI increased, decreased, or had no effect on image quality compared to standard reconstruction. Quantitative analysis of signal intensities was also performed to calculate apparent signal to noise (aSNR) and apparent contrast to noise (aCNR) ratios.

RESULTS

AI reconstruction was better than standard in 99% (reader 1, 49/50; reader 2, 50/50) for overall image quality, 99% (reader 1, 49/50; reader 2, 50/50) for subjective SNR, and 98% (reader 1, 49/50; reader 2, 49/50) for diagnostic preference. Quantitative analysis revealed significantly higher gray matter aSNR (30.6±6.5), white matter aSNR (21.4±5.6), and gray-white matter aCNR (7.1±1.6) in AI-reconstructed images compared to standard reconstruction (18±2.7, 14.2±2.8, 4.4±0.8, p<0.001) respectively.

CONCLUSION

We conclude that AI reconstruction improved T2W FLAIR image quality in most patients when compared with standard reconstruction in pediatric patients.

Federated learning for medical image analysis: A survey

Machine learning in medical imaging often faces a fundamental dilemma, namely, the small sample size problem. Many recent studies suggest using multi-domain data pooled from different acquisition sites/centers to improve statistical power. However, medical images from different sites cannot be easily shared to build large datasets for model training due to privacy protection reasons. As a promising solution, federated learning, which enables collaborative training of machine learning models based on data from different sites without cross-site data sharing, has attracted considerable attention recently. In this paper, we conduct a comprehensive survey of the recent development of federated learning methods in medical image analysis. We have systematically gathered research papers on federated learning and its applications in medical image analysis published between 2017 and 2023. Our search and compilation were conducted using databases from IEEE Xplore, ACM Digital Library, Science Direct, Springer Link, Web of Science, Google Scholar, and PubMed. In this survey, we first introduce the background of federated learning for dealing with privacy protection and collaborative learning issues. We then present a comprehensive review of recent advances in federated learning methods for medical image analysis. Specifically, existing methods are categorized based on three critical aspects of a federated learning system, including client end, server end, and communication techniques. In each category, we summarize the existing federated learning methods according to specific research problems in medical image analysis and also provide insights into the motivations of different approaches. In addition, we provide a review of existing benchmark medical imaging datasets and software platforms for current federated learning research. We also conduct an experimental study to empirically evaluate typical federated learning methods for medical image analysis. This survey can help to better understand the current research status, challenges, and potential research opportunities in this promising research field.

Integrating data distribution prior via Langevin dynamics for end-to-end MR reconstruction

PURPOSE

To develop a novel deep learning-based method inheriting the advantages of data distribution prior and end-to-end training for accelerating MRI.

METHODS

Langevin dynamics is used to formulate image reconstruction with data distribution before facilitate image reconstruction. The data distribution prior is learned implicitly through the end-to-end adversarial training to mitigate the hyper-parameter selection and shorten the testing time compared to traditional probabilistic reconstruction. By seamlessly integrating the deep equilibrium model, the iteration of Langevin dynamics culminates in convergence to a fix-point, ensuring the stability of the learned distribution.

RESULTS

The feasibility of the proposed method is evaluated on the brain and knee datasets. Retrospective results with uniform and random masks show that the proposed method demonstrates superior performance both quantitatively and qualitatively than the state-of-the-art.

CONCLUSION

The proposed method incorporating Langevin dynamics with end-to-end adversarial training facilitates efficient and robust reconstruction for MRI. Empirical evaluations conducted on brain and knee datasets compellingly demonstrate the superior performance of the proposed method in terms of artifact removing and detail preserving.

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.

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.

The beating heart: artificial intelligence for cardiovascular application in the clinic

Artificial intelligence (AI) integration in cardiac magnetic resonance imaging presents new and exciting avenues for advancing patient care, automating post-processing tasks, and enhancing diagnostic precision and outcomes. The use of AI significantly streamlines the examination workflow through the reduction of acquisition and postprocessing durations, coupled with the automation of scan planning and acquisition parameters selection. This has led to a notable improvement in examination workflow efficiency, a reduction in operator variability, and an enhancement in overall image quality. Importantly, AI unlocks new possibilities to achieve spatial resolutions that were previously unattainable in patients. Furthermore, the potential for low-dose and contrast-agent-free imaging represents a stride toward safer and more patient-friendly diagnostic procedures. Beyond these benefits, AI facilitates precise risk stratification and prognosis evaluation by adeptly analysing extensive datasets. This comprehensive review article explores recent applications of AI in the realm of cardiac magnetic resonance imaging, offering insights into its transformative potential in the field.

Results of the 2023 ISBI challenge to reduce GABA-edited MRS acquisition time

PURPOSE

Use a conference challenge format to compare machine learning-based gamma-aminobutyric acid (GABA)-edited magnetic resonance spectroscopy (MRS) reconstruction models using one-quarter of the transients typically acquired during a complete scan.

METHODS

There were three tracks: Track 1: simulated data, Track 2: identical acquisition parameters with in vivo data, and Track 3: different acquisition parameters with in vivo data. The mean squared error, signal-to-noise ratio, linewidth, and a proposed shape score metric were used to quantify model performance. Challenge organizers provided open access to a baseline model, simulated noise-free data, guides for adding synthetic noise, and in vivo data.

RESULTS

Three submissions were compared. A covariance matrix convolutional neural network model was most successful for Track 1. A vision transformer model operating on a spectrogram data representation was most successful for Tracks 2 and 3. Deep learning (DL) reconstructions with 80 transients achieved equivalent or better SNR, linewidth and fit error compared to conventional 320 transient reconstructions. However, some DL models optimized linewidth and SNR without actually improving overall spectral quality, indicating a need for more robust metrics.

CONCLUSION

DL-based reconstruction pipelines have the promise to reduce the number of transients required for GABA-edited MRS.

Evaluating Machine Learning-Based MRI Reconstruction Using Digital Image Quality Phantoms

Quantitative and objective evaluation tools are essential for assessing the performance of machine learning (ML)-based magnetic resonance imaging (MRI) reconstruction methods. However, the commonly used fidelity metrics, such as mean squared error (MSE), structural similarity (SSIM), and peak signal-to-noise ratio (PSNR), often fail to capture fundamental and clinically relevant MR image quality aspects. To address this, we propose evaluation of ML-based MRI reconstruction using digital image quality phantoms and automated evaluation methods. Our phantoms are based upon the American College of Radiology (ACR) large physical phantom but created in k-space to simulate their MR images, and they can vary in object size, signal-to-noise ratio, resolution, and image contrast. Our evaluation pipeline incorporates evaluation metrics of geometric accuracy, intensity uniformity, percentage ghosting, sharpness, signal-to-noise ratio, resolution, and low-contrast detectability. We demonstrate the utility of our proposed pipeline by assessing an example ML-based reconstruction model across various training and testing scenarios. The performance results indicate that training data acquired with a lower undersampling factor and coils of larger anatomical coverage yield a better performing model. The comprehensive and standardized pipeline introduced in this study can help to facilitate a better understanding of the performance and guide future development and advancement of ML-based reconstruction algorithms.

Artificial intelligence solution to accelerate the acquisition of MRI images: Impact on the therapeutic care in oncology in radiology and radiotherapy departments

PURPOSE

MRI is essential in the management of brain tumours. However, long waiting times reduce patient accessibility. Reducing acquisition time could improve access but at the cost of spatial resolution and diagnostic quality. A commercially available artificial intelligence (AI) solution, SubtleMR™, can increase the resolution of acquired images. The objective of this prospective study was to evaluate the impact of this algorithm that halves the acquisition time on the detectability of brain lesions in radiology and radiotherapy.

MATERIAL AND METHODS

The T1/T2 MRI of 33 patients with brain metastases or meningiomas were analysed. Images acquired quickly have a matrix divided by two which halves the acquisition time. The visual quality and lesion detectability of the AI images were evaluated by radiologists and radiation oncologist as well as pixel intensity and lesions size.

RESULTS

The subjective quality of the image is lower for the AI images compared to the reference images. However, the analysis of lesion detectability shows a specificity of 1 and a sensitivity of 0.92 and 0.77 for radiology and radiotherapy respectively. Undetected lesions on the IA image are lesions with a diameter less than 4mm and statistically low average gadolinium-enhancement contrast.

CONCLUSION

It is possible to reduce MRI acquisition times by half using the commercial algorithm to restore the characteristics of the image and obtain good specificity and sensitivity for lesions with a diameter greater than 4mm.

MRI reconstruction with enhanced self-similarity using graph convolutional network

BACKGROUND

Recent Convolutional Neural Networks (CNNs) perform low-error reconstruction in fast Magnetic Resonance Imaging (MRI). Most of them convolve the image with kernels and successfully explore the local information. Nonetheless, the non-local image information, which is embedded among image patches relatively far from each other, may be lost due to the limitation of the receptive field of the convolution kernel. We aim to incorporate a graph to represent non-local information and improve the reconstructed images by using the Graph Convolutional Enhanced Self-Similarity (GCESS) network.

METHODS

First, the image is reconstructed into the graph to extract the non-local self-similarity in the image. Second, GCESS uses spatial convolution and graph convolution to process the information in the image, so that local and non-local information can be effectively utilized. The network strengthens the non-local similarity between similar image patches while reconstructing images, making the reconstruction of structure more reliable.

RESULTS

Experimental results on in vivo knee and brain data demonstrate that the proposed method achieves better artifact suppression and detail preservation than state-of-the-art methods, both visually and quantitatively. Under 1D Cartesian sampling with 4 × acceleration (AF = 4), the PSNR of knee data reached 34.19 dB, 1.05 dB higher than that of the compared methods; the SSIM achieved 0.8994, 2% higher than the compared methods. Similar results were obtained for the reconstructed images under other sampling templates as demonstrated in our experiment.

CONCLUSIONS

The proposed method successfully constructs a hybrid graph convolution and spatial convolution network to reconstruct images. This method, through its training process, amplifies the non-local self-similarities, significantly benefiting the structural integrity of the reconstructed images. Experiments demonstrate that the proposed method outperforms the state-of-the-art reconstruction method in suppressing artifacts, as well as in preserving image details.

Accelerated MRI reconstructions via variational network and feature domain learning

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

AI-based motion artifact severity estimation in undersampled MRI allowing for selection of appropriate reconstruction models

BACKGROUND

Magnetic Resonance acquisition is a time consuming process, making it susceptible to patient motion during scanning. Even motion in the order of a millimeter can introduce severe blurring and ghosting artifacts, potentially necessitating re-acquisition. Magnetic Resonance Imaging (MRI) can be accelerated by acquiring only a fraction of k-space, combined with advanced reconstruction techniques leveraging coil sensitivity profiles and prior knowledge. Artificial intelligence (AI)-based reconstruction techniques have recently been popularized, but generally assume an ideal setting without intra-scan motion.

PURPOSE

To retrospectively detect and quantify the severity of motion artifacts in undersampled MRI data. This may prove valuable as a safety mechanism for AI-based approaches, provide useful information to the reconstruction method, or prompt for re-acquisition while the patient is still in the scanner.

METHODS

We developed a deep learning approach that detects and quantifies motion artifacts in undersampled brain MRI. We demonstrate that synthetically motion-corrupted data can be leveraged to train the convolutional neural network (CNN)-based motion artifact estimator, generalizing well to real-world data. Additionally, we leverage the motion artifact estimator by using it as a selector for a motion-robust reconstruction model in case a considerable amount of motion was detected, and a high data consistency model otherwise.

RESULTS

Training and validation were performed on 4387 and 1304 synthetically motion-corrupted images and their uncorrupted counterparts, respectively. Testing was performed on undersampled in vivo motion-corrupted data from 28 volunteers, where our model distinguished head motion from motion-free scans with 91% and 96% accuracy when trained on synthetic and on real data, respectively. It predicted a manually defined quality label ('Good', 'Medium' or 'Bad' quality) correctly in 76% and 85% of the time when trained on synthetic and real data, respectively. When used as a selector it selected the appropriate reconstruction network 93% of the time, achieving near optimal SSIM values.

CONCLUSIONS

The proposed method quantified motion artifact severity in undersampled MRI data with high accuracy, enabling real-time motion artifact detection that can help improve the safety and quality of AI-based reconstructions.