Acquiring images of the same anatomy with multiple different contrasts increases the diversity of diagnostic information available in an MR exam. Yet, scan time limitations may prohibit acquisition of certain contrasts, and some contrasts may be corrupted by noise and artifacts. In such cases, the ability to synthesize unacquired or corrupted contrasts can improve diagnostic utility. For multi-contrast synthesis, current methods learn a nonlinear intensity transformation between the source and target images, either via nonlinear regression or deterministic neural networks. These methods can in turn suffer from the loss of structural details in synthesized images. Here, we propose a new approach for multi-contrast MRI synthesis based on conditional generative adversarial networks. The proposed approach preserves intermediate-to-high frequency details via an adversarial loss; and it offers enhanced synthesis performance via pixel-wise and perceptual losses for registered multi-contrast images and a cycle-consistency loss for unregistered images. Information from neighboring cross-sections are utilized to further improve synthesis quality. Demonstrations on T1- and T2- weighted images from healthy subjects and patients clearly indicate the superior performance of the proposed approach compared to previous state-of-the-art methods. Our synthesis approach can help improve quality and versatility of multicontrast MRI exams without the need for prolonged or repeated examinations.

Cardiac magnetic resonance (CMR) is an essential clinical tool for the assessment of cardiovascular disease. Deep learning (DL) has recently revolutionized the field through image reconstruction techniques that allow unprecedented data undersampling rates. These fast acquisitions have the potential to considerably impact the diagnosis and treatment of cardiovascular disease. Herein, we provide a comprehensive review of DL-based reconstruction methods for CMR. We place special emphasis on state-of-the-art unrolled networks, which are heavily based on a conventional image reconstruction framework. We review the main DL-based methods and connect them to the relevant conventional reconstruction theory. Next, we review several methods developed to tackle specific challenges that arise from the characteristics of CMR data. Then, we focus on DL-based methods developed for specific CMR applications, including flow imaging, late gadolinium enhancement, and quantitative tissue characterization. Finally, we discuss the pitfalls and future outlook of DL-based reconstructions in CMR, focusing on the robustness, interpretability, clinical deployment, and potential for new methods.

Purpose: To develop a framework that jointly estimates rigid motion and polarizing magnetic field (B₀) perturbations () for brain MRI using a single navigator of a few milliseconds in duration, and to additionally allow for navigator acquisition at arbitrary timings within any type of sequence to obtain high‐temporal resolution estimates.Theory and Methods: Methods exist that match navigator data to a low‐resolution single‐contrast image (scout) to estimate either motion or . In this work, called QUEEN (QUantitatively Enhanced parameter Estimation from Navigators), we propose combined motion and estimation from a fast, tailored trajectory with arbitrary‐contrast navigator data. To this end, the concept of a quantitative scout (Q‐Scout) acquisition is proposed from which contrast‐matched scout data is predicted for each navigator. Finally, navigator trajectories, contrast‐matched scout, and are integrated into a motion‐informed parallel‐imaging framework.Results: Simulations and in vivo experiments show the need to model to obtain accurate motion parameters estimated in the presence of strong . Simulations confirm that tailored navigator trajectories are needed to robustly estimate both motion and . Furthermore, experiments show that a contrast‐matched scout is needed for parameter estimation from multicontrast navigator data. A retrospective, in vivo reconstruction experiment shows improved image quality when using the proposed Q‐Scout and QUEEN estimation.Conclusions: We developed a framework to jointly estimate rigid motion parameters and from navigators. Combing a contrast‐matched scout with the proposed trajectory allows for navigator deployment in almost any sequence and/or timing, which allows for higher temporal‐resolution motion and estimates.

Magnetic resonance imaging (MRI) is a common and life-saving medical imaging technique. However, acquiring high signal-to-noise ratio MRI scans requires long scan times, resulting in increased costs and patient discomfort, and decreased throughput. Thus, there is great interest in denoising MRI scans, especially for the subtype of diffusion MRI scans that are severely SNR-limited. While most prior MRI denoising methods are supervised in nature, acquiring supervised training datasets for the multitude of anatomies, MRI scanners, and scan parameters proves impractical. Here, we propose Denoising Diffusion Models for Denoising Diffusion MRI (DDM$^2$), a self-supervised denoising method for MRI denoising using diffusion denoising generative models. Our three-stage framework integrates statistic-based denoising theory into diffusion models and performs denoising through conditional generation. During inference, we represent input noisy measurements as a sample from an intermediate posterior distribution within the diffusion Markov chain. We conduct experiments on 4 real-world in-vivo diffusion MRI datasets and show that our DDM$^2$ demonstrates superior denoising performances ascertained with clinically-relevant visual qualitative and quantitative metrics.

Introduction Spatio-temporal MRI methods enable whole-brain multi-parametric mapping at ultra-fast acquisition times through efficient k-space encoding, but can have very long reconstruction times, which limit their integration into clinical practice. Deep learning (DL) is a promising approach to accelerate reconstruction, but can be computationally intensive to train and deploy due to the large dimensionality of spatio-temporal MRI. DL methods also need large training data sets and can produce results that don’t match the acquired data if data consistency is not enforced. The aim of this project is to reduce reconstruction time using DL whilst simultaneously limiting the risk of deep learning induced hallucinations, all with modest hardware requirements. Methods Deep Learning Initialized Compressed Sensing (Deli-CS) is proposed to reduce the reconstruction time of iterative reconstructions by “kick-starting” the iterative reconstruction with a DL generated starting point. The proposed framework is applied to volumetric multi-axis spiral projection MRF that achieves whole-brain T1 and T2 mapping at 1-mm isotropic resolution for a 2-minute acquisition. First, the traditional reconstruction is optimized from over two hours to less than 40 minutes while using more than 90% less RAM and only 4.7 GB GPU memory, by using a memory-efficient GPU implementation. The Deli-CS framework is then implemented and evaluated against the above reconstruction. Results Deli-CS achieves comparable reconstruction quality with 50% fewer iterations bringing the full reconstruction time to 20 minutes. Conclusion Deli-CS reduces the reconstruction time of subspace reconstruction of volumetric spatio-temporal acquisitions by providing a warm start to the iterative reconstruction algorithm.

Purpose: To facilitate ease of data compilation across diverse populations for training models to synthesize clinical contrast-weighted images from magnetic resonance fingerprinting. Methods: We leverage a semi-supervised training framework using highly accelerated acquisitions of the target contrasts used as ground truths. We utilize complementary randomized data sampling masks across training subjects and contrasts for homogeneous learning in k-space, together with multi-task learning. Results: Our experiments indicate that the proposed method achieves high-quality synthesis with networks trained on retrospectively and prospectively undersampled data of the contrast-weighted images, enabling undersampling up to 12–16\(\times \). Conclusions: The proposed method enables semi-supervised learning for synthesis from MRF with an end-to-end, ultra-fast training data acquisition protocol that is easier to obtain across a large population in clinical settings.