Keywords: AI/ML Image Reconstruction, Machine Learning/Artificial Intelligence, Diffusion tensor imaging, convolutional neural network, curve fitting, mean diffusivity, fractional anisotropy

Motivation: Diffusion-MRI faced limitations due to extended scan times and scanner/protocol variations.

Goal(s): This study aims to assess its ability to accelerate imaging procedures and unify data from diverse sources.

Approach: A convolutional neural network was employed to reconstruct diffusion-weighted images into diffusion tensor images. The effectiveness of reconstructed model was evaluated by normalized mean-square error (NMSE) and structural similarity index (SSIM).

Results: The CNN showed significantly better SSIM and lower NMSE in FA and MD (p < 0.001) compared to conventional methods. Moreover, the CNN model maintained strong performance when applied to other Scanners for FA and MD.

Impact: Through convolutional neural networks, images might be acquired fast and easily be harmonized across platforms . Subsequent research will further utilize deep/machine learning tools to investigate the impact of reconstructed image-segmented brain regions on the performance of classification models.

Keywords: AI/ML Image Reconstruction, Diffusion Tensor Imaging

Motivation: Diffusion Tensor Cardiovascular Magnetic Resonance (DT-CMR) is hindered by low resolution and long acquisitions. Spiral trajectories could be efficient with effective removal of artefacts from undersampled images.

Goal(s): To remove artefacts from highly accelerated spiral in-vivo DT-CMR acquisitions using a novel deep learning method.

Approach: We proposed a Residual U-Net based Complex-valued Edge Attention Network (CEAN) to remove undersampling artefacts. Training with and without transfer learning were explored.

Results: CEAN with transfer learning outperformed other networks, achieving the lowest Mean Absolute Error (MAE) for DT-CMR parameters and preserving diffusion encoding information, suggesting future potentials in accelerating clinical DT-CMR studies.

Impact: This work will allow the acquisition and reconstruction of highly accelerated STEAM spiral DT-CMR, aided by the proposed deep Complex-valued Edge Attention Network. Further developments will allow increases in spatial resolution to facilitate clinical translation of DT-CMR.

Keywords: AI/ML Image Reconstruction, Diffusion/other diffusion imaging techniques, implicit neural representation, q-space undersampling, spherical harmonics

Motivation: Most diffusion MRI techniques require extensive sampling of q-space to effectively resolve fiber structures at a fine detail. The scan times become impractically long, especially for clinical settings.

Goal(s): Our goal is to arbitrarily interpolate the q-space data to enable downsampling of q-space, while maintaining high fidelity diffusion metrics.

Approach: We propose QUCCI, a subject-specific unsupervised implicit network model that utilizes both implicit and physics-driven explicit regularization to encode diffusion MRI signals with angular continuity.

Results: QUCCI achieves superior q-space interpolation, outperforming traditional and deep learning methods.

Impact: QUCCI provides high-fidelity diffusion MRI metrics via improving the angular interpolation of diffusion MRI signals under highly undersampled q-space cases, which may especially be beneficial in the clinical settings where excessively long scan times are impractical.

Keywords: Acquisition Methods, Low-Field MRI

Motivation: Adequate image signal-to-noise ratio (SNR) and resolution within a reasonable scan time is challenging for low-field diffusion quantitative MRI.

Goal(s): To present a PROPELLER-acquisition and ADC mapping joint learning neural network to accelerate DWI with improved image SNR and resolution.

Approach: Considering the similar anatomical structure between diffusion-weighted MR images, this work achieved DWI PROPELLER-acquisition optimization and reconstructed high quality ADC maps from data acquired by optimized acquisition using U-net.

Results: In vivo and simulation results demonstrate that our proposed method can generate high SNR and resolution ADC maps within 2 minutes scan time under 0.23T human scanner.

Impact: Joint learning acquisition and quantitative reconstruction based on PROPELLER sampling trajectory using neural network has successfully achieved fast ADC mapping, offering great possibility for quantitative analysis in low-field diffusion MRI.

Keywords: AI/ML Image Reconstruction, Machine Learning/Artificial Intelligence, super-resolution, turbo-spin-echo, joint-optimization

Motivation: TSE flip angle trains can have a strong influence on the actual resolution of the acquired image and have consequently a considerable impact on the performance of a super-resolution task.

Goal(s): We demonstrate the advantage of end-to-end optimization of sequence and neural network parameter compared to pure network training approaches.

Approach: This MR-physics-informed training procedure jointly optimizes radiofrequency pulse trains of a PD- and T2-weighted TSE and subsequently applied CNN to predict corresponding PDw and T2w super-resolution TSE images.

Results: The method generalizes from simulation-based optimization to in vivo measurements and acquired super-resolution images show higher accuracy compared to pure network training approaches.

Impact: Acquired super-resolution image may improve evaluation of the data. Reduction of acquisition time compared to direct high-resolution acquisition leads to increase in patient comfort and minimization of motion artifacts.

Keywords: AI/ML Image Reconstruction, Data Processing

Motivation: For magnetic resonance imaging (MRI) applications, rapid imaging and automatic segmentation of target tissues are critical. However, most existing methods barely consider MR image segmentation in fast imaging scenarios.

Goal(s): Our goal is to simultaneously achieve high scanning acceleration and accurate multi-class tissue segmentation results under a unified framework.

Approach: We propose a novel multi-task method  with a novel interaction module to reconstruct undersampled MR images based on modified ISTA-Net  and simultaneously segment tissues based on lightweight U-Net.

Results: Experiments on cardiac and knee datasets demonstrate that our method outperforms existing state-of-the-art multi-task approaches for joint MR image reconstruction and segmentation.

Impact: The proposed multi-task interaction method can effectively achieve high scanning acceleration and accurate segmentation results simultaneously, which can further expand the application of MR in clinical disease diagnosis.

Keywords: AI/ML Image Reconstruction, Heart

Motivation: The currently limited quality of accelerated cardiac cine reconstruction may potentially be improved by the emerging diffusion models, but the clinically unacceptable long processing time poses a challenge.

Goal(s): To develop a clinically feasible diffusion-model-based reconstruction pipeline to improve the image quality of cine MRI.

Approach: A multi-in-multi-out diffusion enhancement model together with fast inference strategies were developed to be used in conjunction with a reconstruction model.

Results: The diffusion reconstruction reduced spatial and temporal blurring in prospectively undersampled clinical data, as validated by experts’ inspection. The 1.5s/video processing time enabled the approach to be applied in clinical scenarios.

Impact: The proposed diffusion reconstruction pipeline provides a practical solution to cardiac cine reconstruction with enhanced quality for clinical usage. This pipeline may be transfered to the clinical application of other diffusion-based methods.

Keywords: AI/ML Image Reconstruction, Cardiovascular, Motion-compensated reconstruction

Motivation: Cardiac Cine MRI is commonly used for assessing cardiac function. However, extended acquisition times may cause patient discomfort or can result in respiratory motion artifacts and slice misalignments due to multiple breath-holds.

Goal(s): We aim to accelerate data acquisition into a single breath-hold ($$$\sim$$$24×) with spatial-temporal sharing along the cardiac cycle for accurate morphological and functional reconstruction.

Approach: We integrated inter-frame motion field estimations with a deep learning-based reconstruction. The motion-compensated A-LIKNet was trained on 115 subjects and tested on 14 subjects.

Results: The proposed method reconstructs high-quality images, especially improving morphological accuracy, and thus enables cardiac Cine imaging in a single breath-hold.

Impact: The proposed deep learning-based motion-compensated A-LIKNet can efficiently reconstruct highly undersampled cardiac Cine MRI for up to 24× accelerated acquisitions of a single breath-hold. Results demonstrate higher morphological authenticity, sharper details, and reduced artifacts compared to other methods. 

Keywords: AI/ML Image Reconstruction, Image Reconstruction

Motivation: GRASP allows for free-breathing DCE-MRI with high spatial and temporal resolution. However, the current 4D iterative reconstruction is slow and still suffers from streaking artifacts, limiting clinical use.

Goal(s): Develop a DL solution that significantly reduces the reconstruction time and improves image quality.

Approach: A model-assisted DL reconstruction combining a sparsity model with an efficient 3D spatiotemporal network for fast and robust reconstruction of accelerated scans with high resolution.

Results: A sparsity-constrained DL-based can provide robust and fast reconstructions with improved image quality, evidenced by the superior quantitative metrics and the qualitative analysis of cases under-represented in the training data. 

Impact: GRASP offers high-resolution 4D free-breathing DCE-MRI; however, it still suffers from under-sampling artifacts and long reconstruction times. A model-assisted DL reconstruction can reduce the reconstruction time, improve image quality, and increase system robustness—essential in translating to clinical practice.

Keywords: AI/ML Image Reconstruction, Cardiovascular

Motivation: Deep learning reconstruction algorithms offer significant advantages for accelerating 4D-Flow MRI acquisition. However, a large high-quality fully-sampled dataset is usually unavailable for network training.

Goal(s): To propose an unsupervised algorithm for 4D-Flow MRI reconstruction, without the need for any fully-sampled data.

Approach: We use branched CNNs and a Graph-Convolution-Network as the generator. Additionally, we devise an ADMM algorithm to alternately optimize the images and the network parameters. Experiments are conducted on aortic and intracranial 4D-Flow data.

Results: The proposed algorithm demonstrates superior reconstruction results, outperforming even supervised deep-learning method. Moreover, it exhibits good generalization capability when applied to another imaging target.

Impact: The proposed method is a promising algorithm for accelerating MR blood-flow imaging, owing to its exceptional performance and generalization capacity. Furthermore, the algorithm introduces a new model for 4D-flow MRI reconstruction which is valuable for further research.

Keywords: AI/ML Image Reconstruction, Image Reconstruction, Non-cartesian Reconstruction, Memory Efficient Reconstruction, Deep Learning, AI

Motivation: Exploring the stability of various memory efficient deep equilibrium architectures for non-cartesian stack of stars reconstruction.

Goal(s): Implementing memory efficient inversion-based and inversion free deep equilibrium models and assessing performance against unrolled network baselines.

Approach: Applying an implicit layer consisting of data-consistency and a 2D-UNet denoiser to find a fixed point and iteratively solving the optimization problem using proximal gradient descent with inversion free and inversion-based equilibrium backpropagation. Compare the efficacy of spectral normalization and Jacobian regularization on stability.

Results: The inversion-free deep equilibrium model exhibited performance similar to unrolled networks, achieving a significant 50% reduction in GPU memory usage.

Impact: Stable and memory-efficient training advances AI-based reconstructions by enhancing their robustness and efficiency, leading to more accurate diagnoses and treatments, ultimately improving patient care. Moreover, it renders these advanced AI solutions more accessible to resource-constrained systems.

Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, coronary magnetic resonance angiography, compressed sensing, deep learning

Motivation: While classical non-learning reconstruction methods for 3D coronary magnetic resonance angiography (CMRA) lack a task-adaptive image prior, 3D deep unrolling suffers from a low memory efficiency, causing a reduced number of iterations and a compromised image quality.

Goal(s): We aim to combine compressed sensing and deep learning regularization by using a trained de-aliasing network as the sparsifying transform.

Approach: We compared the method with PROST, Plug-and-Play, DAGAN, and MoDL for accelerating CMRA in 20 healthy subjects.

Results: Visual inspections and quantitative comparisons both found a substantially improved reconstruction quality from DARCS relative to the other methods.

Impact: The proposed method overcomes an important limitation of 3D unrolling while maintaining its core advantage of task-adaptive regularization. The method not only can accelerate 3D CMRA, but also has the potential for general 3D image reconstructions.

Keywords: AI/ML Image Reconstruction, Machine Learning/Artificial Intelligence

Motivation: Federated MR image reconstruction can make full use of data from multiple institutions while protecting patient privacy, but the images obtained by existing methods still need improvement in terms of fine structures.

Goal(s): Our goal is to improve the quality of clinical diagnosis by achieving accurate MR image reconstruction.

Approach: A Laplacian attention mechanism is proposed to capture fine structure and details for accurate MR image reconstruction from undersampled data.

Results: Qualitative and quantitative experimental results on an in-house and two public datasets validate the effectiveness of our method.

Impact: Federated MR image reconstruction promotes collaboration across multiple institutions and effectively leverages data from different organizations to enhance model performance, while mitigating privacy concerns.

Keywords: AI/ML Image Reconstruction, Precision & Accuracy, image quality assessment, digital phantoms

Motivation: Quantitative image quality evaluation tools are needed for machine learning-based MR reconstruction. 

Goal(s): To introduce digital image quality phantoms and evaluation metrics tailored for machine learning-based MR reconstruction, scalable to form large test sets, and flexible to simulate various object size, image contrast, signal-to-noise-ratio, resolution etc.

Approach: We created 2D disks, resolution arrays, and low-contrast phantoms resembling MR ACR phantom properties. The evaluation includes geometric accuracy, intensity uniformity, resolution, and low-contrast detectability. We evaluated the AUTOMAP reconstruction model trained on the M4Raw and FastMRI datasets with these phantoms. 

Results: The study provides a tool for evaluating machine learning-based MRI reconstruction.

Impact: This research establishes digital phantoms and quantitative metrics for evaluating machine learning-based MRI reconstruction. These tools enable accurate assessment of fundamental image quality and generalizability over scan conditions, offering valuable feedback for improving machine learning-based methods development.

Keywords: AI/ML Image Reconstruction, Image Reconstruction

Motivation: Parallel imaging coil constraints can make it difficult to design comfortable coil arrays.

Goal(s): 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.

Approach: We synthesized an eight-channel head coil configuration and gradually increased coil overlap making the coils less ideal for parallel imaging. A DL reconstruction method was compared to a traditional non-DL method.

Results: As coil overlap increased, a smaller decrease in reconstruction performance was seen when using a DL method versus a non-DL method.

Impact: Our works suggests parallel imaging geometric coil constraints may be relaxed when using a deep learning reconstruction method. This flexibility would lead to an increased range of coil configurations that allow for improved patient comfort while decreasing scan times.

Keywords: AI/ML Image Reconstruction, Machine Learning/Artificial Intelligence, MR Physics Model

Motivation: Deep learning (DL) is powerful for fast MRI reconstruction, but remains largely untapped in multiple clinical imaging scenarios.

Goal(s): To provide a feasible and cost-effective way to markedly boost the widespread usage of DL in various fast MRI applications.

Approach: In this work, we present a Physics-Informed Synthetic data learning framework for Fast MRI, called PISF, which is the first to enable generalizable DL for multi-scenario MRI reconstruction using solely one trained model.

Results: PISF trained on synthetic data enables high-quality, ultra-fast, and robust MRI reconstruction from different 4contrasts, 5 anatomies, 5 vendors and centers, and 2 pathologies, without further re-training.

Impact: Physics-informed synthetic data learning (DL) provides a feasible and cost-effective way to markedly boost the widespread usage of DL in various fast MRI applications, while freeing from the intractable ethical and practical considerations of in vivo human data acquisitions.

Keywords: AI Diffusion Models, Data Processing, Phase error correct, Diffusion models.

Motivation: To address EPI phase error correction caused by the problem of inconsistent positive and negative phases.

Goal(s): We introduce an image prior-based method termed Phase Error Correction Diffusion-based Reconstruction with Echo Apart Magnitude-Consistency(PEC-DREAM).

Approach: The method was trained on structural imaging data, and it performs robustly the inference on EPI phase error correction task without specific model finetune. Here, we introduce novel data consistency including k-space and magnitude consistency to enhance the performance of the SGM during reverse diffusion.

Results: Experiments demonstrate the versatility of our approach across various scenarios, including human and rodent EPI, accelerated and non-accelerated imaging and SMS sampling.

Impact: The method we have proposed effectively addresses EPI phase error correction. Prospective experiments demonstrate the versatility of our innovative approach across various scenarios, and our method holds promise as a potent tool.

Keywords: AI/ML Image Reconstruction, Machine Learning/Artificial Intelligence, k-space, motion, artifacts, quality metric

Motivation: Motion correction in MRI predominantly relies on image-based methods and continues to be a challenge. Innovative approaches could harness better motion information latent in k-space (i.e., the measurement space).

Goal(s): Developing a reference-less motion correction pipeline in k-space using deep learning.

Approach: Our k-space motion correction pipeline combines deep learning for motion parameter estimation with model-based image reconstruction. Large datasets were generated through physics-based simulations on 2D brain MRI acquisitions to enhance model training and performance.

Results: Our deep-learning model performs well in motion parameter estimation, even for successive motion events, effectively removing substantial motion artifacts when combined with model-based reconstruction.

Impact: SISMIK, our deep learning model successfully estimates motion parameters in the acquisition space of multi-slice 2D brain MRI. It allows substantial motion artifact removal through a model-based reconstruction approach, which is, by design, free of hallucination artifacts.

Keywords: AI/ML Image Reconstruction, Spectroscopy, Lipid-Suppression, Brain, High-Field MR, Deep Learning

Motivation: Magnetic Resonance Spectroscopic Imaging (MRSI) offers non-invasive metabolic concentration mapping, aiding early pathology detection like brain tumors. However, extracranial lipid signals can compromise neurochemical data.

Goal(s): The potential of supervised neural networks remains unexplored, despite their success in other artifact removal and metabolite quantification tasks. We introduce a deep learning method for robust lipid removal.

Approach: Our approach is compared to a state-of-the-art L2-lipid-regularization using simulated and in-vivo whole-brain MRSI data.

Results: Our supervised deep learning method showed improved performance to the L2-lipid-regularization method by eliminating more lipid signal while preserving metabolic signals and spectral baseline.

Impact: The LIPCON method achieves accurate lipid suppression across whole-brain MRSI datasets without the need for parameter tuning and within a few seconds. This should mark a step in enhancing the reproducibility and efficiency of MRSI pipelines.

Keywords: Analysis/Processing, Machine Learning/Artificial Intelligence

Motivation: Complex-valued deep neural network has not been fully investigated in MRS reconstruction and preprocessing.

Goal(s): We aim to solve the spectroscopy inverse problems in domain transform from FIDs to spectra, especially for accelerated MRS reconstruction.

Approach: A complex-valued deep learning framework artificial Fourier transform network (AFT-Net) is proposed to directly reconstruct and process the complex-valued raw data in the sensor domain.

Results: Evaluation of different acceleration rates was performed on the in vivo dataset. AFT-Net demonstrated the ability to reconstruct the data under up to 80 times acceleration rate. The proposed AFT-Net is an efficient and accurate approach for MEGA-PRESS MRS accelerated reconstruction.

Impact: MRS reconstruction and preprocessing with AFT-Net should be able to determine the domain-manifold mapping and process FID data directly, which shows superior performance compared with numerical method and can be served as an efficient and accurate approach for MRS acquisition.