Cardiac CINE magnetic resonance imaging is the gold-standard for assessment of cardiac function. Imaging accelerations have shown to enable 3D with left ventricular (LV) coverage in a single breath-hold. However, remains limited anisotropic resolution and long reconstruction times. Recently deep learning has promising results computationally efficient reconstructions highly accelerated 2D imaging. In this work, we propose novel 4D (3D + time) learning-based network, termed CINENet,...

To develop and evaluate a novel generalizable super-resolution (SR) deep-learning framework for motion-compensated isotropic 3D coronary MR angiography (CMRA), which allows free-breathing acquisitions in less than minute.Undersampled motion-corrected reconstructions have enabled CMRA ~5-10 min acquisition times. In this work, we propose deep-learning-based SR framework, combined with non-rigid respiratory motion compensation, to shorten the time 1 min. A generative adversarial network (GAN)...

Diffusion probabilistic models (DPMs) which employ explicit likelihood characterization and a gradual sampling process to synthesize data, have gained increasing research interest. Despite their huge computational burdens due the large number of steps involved during sampling, DPMs are widely appreciated in various medical imaging tasks for high-quality diversity generation. Magnetic resonance (MRI) is an important modality with excellent soft tissue contrast superb spatial resolution,...

Aneurysmal wall enhancement (AWE) has emerged as a new possible biomarker for depicting inflammation of the intracranial aneurysm (IA). However, relationships AWE with other risk factors are still unclear unruptured IA. The purpose this study was to investigate association between and metrics.Forty-eight patients saccular IAs diagnosed by digital subtraction angiography were recruited undergo magnetic resonance (MR) black-blood imaging. evaluated using pre- post-contrast MR images....

Purpose To develop a free‐running (free‐breathing, retrospective cardiac gating) 3D myocardial T 1 mapping with isotropic spatial resolution. Methods The sequence is inversion recovery (IR)‐prepared followed by continuous golden angle radial data acquisition. 1D respiratory motion signal extracted from the k‐space center of all spokes and used to bin into different states, enabling estimation correction translational motion, whereas recorded using electrocardiography synchronized compensated...

Purpose To develop a novel respiratory motion compensated three‐dimensional (3D) cardiac magnetic resonance fingerprinting (cMRF) approach for whole‐heart myocardial T 1 and 2 mapping from free‐breathing scan. Methods Two‐dimensional (2D) cMRF has been recently proposed simultaneous, co‐registered breath‐hold scan; however, coverage is limited. Here we propose 3D tissue characterization Variable inversion recovery preparation modules are used parametric encoding, bellows driven localized...

Non-rigid motion-corrected reconstruction has been proposed to account for the complex motion of heart in free-breathing 3D coronary magnetic resonance angiography (CMRA). This framework requires efficient and accurate estimation non-rigid fields from undersampled images at different respiratory positions (or bins). However, state-of-the-art registration methods can be time-consuming. article presents a novel unsupervised deep learning-based strategy fast inter-bin CMRA. The network...

To develop an end-to-end deep learning technique for nonrigid motion-corrected (MoCo) reconstruction of ninefold undersampled free-breathing whole-heart coronary MRA (CMRA).A novel framework was developed consisting a diffeomorphic registration network and motion-informed model-based (MoDL) network. The receives as input highly (~22×) respiratory-resolved images outputs 3D respiratory motion fields between the images. MoDL performs MoCo from data using predicted fields. whole framework,...

Physiological motion, such as cardiac and respiratory during Magnetic Resonance (MR) image acquisition can cause artifacts. Motion correction techniques have been proposed to compensate for these types of motion thoracic scans, relying on accurate estimation from undersampled motion-resolved reconstruction. A particular interest challenge lie in the derivation reliable non-rigid fields data. is usually formulated space via diffusion, parametric-spline, or optical flow methods. However,...

Cardiovascular magnetic resonance (CMR) T1ρ mapping can be used to detect ischemic or non-ischemic cardiomyopathy without the need of exogenous contrast agents. Current 2D myocardial requires multiple breath-holds and provides limited coverage. Respiratory gating by diaphragmatic navigation has recently been exploited enable free-breathing 3D mapping, which, however, low acquisition efficiency may result in unpredictable long scan times. This study aims develop a fast respiratory...

Cardiovascular magnetic resonance (CMR) is an important imaging modality for the assessment and management of adult patients with congenital heart disease (CHD). However, conventional techniques three-dimensional (3D) whole-heart acquisition involve long unpredictable scan times methods that accelerate scans via k-space undersampling often rely on iterative reconstructions. Deep-learning-based reconstruction have recently attracted much interest due to their capacity provide fast...

Purpose To develop a three-dimensional (3D) high-spatial-resolution time-efficient sequence for use in quantitative vessel wall T1 mapping. Materials and Methods A previously described sequence, simultaneous noncontrast angiography intraplaque hemorrhage (SNAP) imaging, was extended by introducing 3D golden angle radial k-space sampling (GOAL-SNAP). Sliding window reconstruction adopted to reconstruct images at different inversion delay times (different contrasts) voxelwise fitting. Phantom...

Purpose Develop a novel low‐rank motion‐corrected (LRMC) reconstruction for nonrigid MR fingerprinting (MRF). Methods Generalized (MC) reconstructions have been developed steady‐state imaging. Here we extend this framework to enable MC transient imaging applications with varying contrast, such as MRF. This is achieved by integrating dictionary‐based compression into the generalized model reconstruct singular images, reducing motion artifacts in resulting parametric maps. The proposed LRMC...

Purpose To introduce non‐rigid cardiac motion correction into a novel free‐running framework for the simultaneous acquisition of 3D whole‐heart myocardial and maps cine images, enabling 3‐min scan. Methods Data were acquired using golden‐angle radial readout interleaved with inversion recovery ‐preparation pulses. After translational respiratory motion, cardiac‐motion‐corrected reconstruction dictionary‐based low‐rank compression patch‐based regularization enabled mapping at any given phase...

Temporal correlation has been exploited for accelerated dynamic MRI reconstruction. Some methods have modeled inter-frame motion into the reconstruction process to produce temporally aligned image series and higher quality. However, traditional motion-compensated approaches requiring iterative optimization of registration are time-consuming, while most deep learning-based neglect in process. We propose an unrolled learning framework with each iteration consisting a groupwise diffeomorphic...

BACKGROUND: Late gadolinium enhancement (LGE) cardiac magnetic resonance (CMR) is a standard technique for diagnosing myocardial infarction (MI), which, however, poses risks due to contrast usage. Techniques enabling MI assessment based on contrast-free CMR are desirable overcome the limitations associated with enhancement. METHODS: We introduce novel deep generative learning method, termed cine-generated (CGE), which transforms cine into LGE-equivalent images assessment. CGE features...

Purpose To develop a reproducible and fast method to reconstruct MR fingerprinting arterial spin labeling (MRF‐ASL) perfusion maps using deep learning. Method A fully connected neural network, denoted as DeepMARS, was trained simulation data added Gaussian noise. Two MRF‐ASL models were used generate the data, specifically single‐compartment model with 4 unknowns parameters two‐compartment 7 unknown parameters. The DeepMARS evaluated from healthy subjects (N = 7) patients Moymoya disease 3)....

It has been proved that multi-contrast cardiovascular magnetic resonance (CMR) vessel wall imaging could be used to characterize carotid vulnerable plaque components according the signal intensity on different contrast images. The of is mainly dependent values T1 and T2 relaxation. mapping recently showed a potential in identifying but it not well validated by histology. This study aimed validate usefulness vivo assessing Thirty-four subjects (mean age, 64.0 ± 8.9 years; 26 males) with...

Purpose To propose a technique that can produce different T 1 and 2 contrasts in single scan for simultaneous mapping of the carotid plaque (SIMPLE). Methods An interleaved 3D golden angle radial trajectory was used conjunction with preparation variable duration (TE prep ) inversion recovery pulses. Sliding window reconstruction adopted to reconstruct images at delay time TE joint fitting. In fitting procedure, rapid B correction method presented. The accuracy SIMPLE investigated phantom...

In myocardial T1 mapping, undesirable motion poses significant challenges because uncorrected can affect estimation accuracy and cause incorrect diagnosis. this study, we propose evaluate a correction method for mapping using self-supervised deep learning based registration with contrast separation (SDRAP). A sparse coding was first proposed to separate the component from -weighted (T1w) images. Then, neural network cross-correlation (SDRAP-CC) or mutual information as similarity measurement...