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.

The study compared the morphometric changes of the cervical spinal cord using dynamic magnetic resonance imaging (MRI) in patients with cervical spondylotic myelopathy (CSM) and assessed the correlation with kinematic changes, cord cross-sectional area (CSA), and high signal intensity (SI) on T2-weighted imaging (T2WI).
Patients with CSM were evaluated through dynamic MRI for sagittal and axial CSA changes of the cervical cord, cerebrospinal fluid (CSF) reserve ratio, degree of cord impingement, cord compression rate, range of motion (ROM), and severity of SI on T2WI. The degree of cord impingement was evaluated using the Muhle grading system. Clinical outcomes were assessed using Japanese Orthopaedic Association scoring and Nurick grade.
The study included 191 patients (113 males) with a mean age of 55.34 ± 12.09 years. The lowest sagittal CSF reserve ratio and cord occupation rate were observed during extension. Cord impingement and SI change were more prevalent in extension-positioned MRI. There was no difference between ROM on dynamic radiographs and dynamic MRI. Preoperative cervical ROM was greater in patients with intensely high SI change.
Dynamic MRI is useful for evaluating neck movement. Patients with high SI had greater ROM before surgery but worse outcomes after. Neck extension exacerbated cervical stenosis and cord compression compared to flexion, and cervical spinal motion contributed to the severity of CSM. Cervical spinal motion should be carefully evaluated, particularly in hyperextension, to prevent worsening of CSM.

To develop and assess a deep learning (DL) pipeline to learn dynamic MR image reconstruction from publicly available natural videos (Inter4K). Learning was performed for a range of DL architectures (VarNet, 3D UNet, FastDVDNet) and corresponding sampling patterns (Cartesian, radial, spiral) either from true multi-coil cardiac MR data (N = 692) or from synthetic MR data simulated from Inter4K natural videos (N = 588). Real-time undersampled dynamic MR images were reconstructed using DL networks trained with cardiac data and natural videos, and compressed sensing (CS). Differences were assessed in simulations (N = 104 datasets) in terms of MSE, PSNR, and SSIM and prospectively for cardiac cine (short axis, four chambers, N = 20) and speech cine (N = 10) data in terms of subjective image quality ranking, SNR and Edge sharpness. Friedman Chi Square tests with post-hoc Nemenyi analysis were performed to assess statistical significance. In simulated data, DL networks trained with cardiac data outperformed DL networks trained with natural videos, both of which outperformed CS (p < 0.05). However, in prospective experiments DL reconstructions using both training datasets were ranked similarly (and higher than CS) and presented no statistical differences in SNR and Edge Sharpness for most conditions.The developed pipeline enabled learning dynamic MR reconstruction from natural videos preserving DL reconstruction advantages such as high quality fast and ultra-fast reconstructions while overcoming some limitations (data scarcity or sharing). The natural video dataset, code and pre-trained networks are made readily available on github.

Lowery urinary tract symptoms (LUTS) affect a large majority of the aging population. 3D Dynamic MRI shows promise as a noninvasive diagnostic tool that can assess bladder anatomy and function (urodynamics) while overcoming challenges associated with current urodynamic assessment methods. However, validation of this technique remains an unmet need. In this study, an anatomically realistic, bladder-mimicking in vitro flow model was created and used to systematically benchmark 3D dynamic MRI performance using a highly controllable syringe pump. Time-resolved volumes of the synthetic bladder model were obtained during simulated filling and voiding events and used to calculate volumetric flowrate. During MRI acquisitions, pressure during each event was recorded and used to create PV loops for work assessment. Error between control and MRI-derived volume for voiding and filling events exhibited 3.36% and 4.66% differences, respectively. A slight increase in average error was observed for MRI-derived flowrate when compared to the control flowrate (4.90% and 7.67% for voiding and filling, respectively). Overall, average error in segmented volumes increased with decreasing volume flowrate. Pressure drops were observed during voiding. Pressure increased during filling. Enhanced validation of novel 3D MRI urodynamics is achieved by using high-resolution PIV for visualizing and quantifying velocity inside the bladder model, which is not currently possible with 3D Dynamic MRI.

BACKGROUND: Pelvic floor dysfunction (PFD) is frequently reported in both sexes. Dynamic magnetic resonance defecography (DMRD) is the preferred modality, mainly due to its superiority and complementary role in clinical examination. However, studies from the perspective of Indian patients are scarce and mostly restricted to females. Thus, we assessed the diagnostic performance of DMRD in patients with PFD and correlated the findings with those on clinical examination.
METHODS: This prospective, observational study involved 57 adult patients of either sex, presenting with pelvic floor symptoms (PFS) and diagnosed with PFD. Initially, the patients underwent clinical examination, and diagnosis was recorded. Subsequently, the patients were subjected to DMRD. The findings were correlated with the Pearson \"r\" correlation coefficient.
RESULTS: A significantly greater proportion of patients had involvement of multiple compartments (36 vs. 12, p<0.001), cystocele (23 vs. 8, p=0.002), and rectal prolapse (25 vs. 14, p=0.030) on DMRD than clinical examination, while there was no significant difference regarding uterine prolapse (p=0.789). Grading of cystocele and rectal prolapse as well as diagnosis of enterocele/peritoneocele, rectocele, and intussusception could be done only with DMRD. DMRD had a strong and significant correlation with clinical examination regarding cystocele (r=0.943, p=0.003), uterine prolapse (r=0.972, p=0.001), and rectal prolapse (r=0.951, p=0.001).
CONCLUSIONS: DMRD demonstrated significantly better performance in the diagnosis of multiple compartment involvement, cystocele, and rectal prolapse. DMRD and clinical examination were significantly correlated regarding the diagnosis of cystocele, uterine prolapse, and rectal prolapse. Thus, DMRD provides information, in addition to the clinical examination, and should be used in symptomatic patients.

Dynamic contrast-enhanced (DCE) cardiac magnetic resonance imaging (CMRI) is a widely used modality for diagnosing myocardial blood flow (perfusion) abnormalities. During a typical free-breathing DCE-CMRI scan, close to 300 time-resolved images of myocardial perfusion are acquired at various contrast \"wash in/out\" phases. Manual segmentation of myocardial contours in each time-frame of a DCE image series can be tedious and time-consuming, particularly when non-rigid motion correction has failed or is unavailable. While deep neural networks (DNNs) have shown promise for analyzing DCE-CMRI datasets, a \"dynamic quality control\" (dQC) technique for reliably detecting failed segmentations is lacking. Here we propose a new space-time uncertainty metric as a dQC tool for DNN-based segmentation of free-breathing DCE-CMRI datasets by validating the proposed metric on an external dataset and establishing a human-in-the-loop framework to improve the segmentation results. In the proposed approach, we referred the top 10% most uncertain segmentations as detected by our dQC tool to the human expert for refinement. This approach resulted in a significant increase in the Dice score (p < 0.001) and a notable decrease in the number of images with failed segmentation (16.2% to 11.3%) whereas the alternative approach of randomly selecting the same number of segmentations for human referral did not achieve any significant improvement. Our results suggest that the proposed dQC framework has the potential to accurately identify poor-quality segmentations and may enable efficient DNN-based analysis of DCE-CMRI in a human-in-the-loop pipeline for clinical interpretation and reporting of dynamic CMRI datasets.

The cervical spinal canal has a wide range of motion and specific biomechanics involved with different pathologies that may cause dynamic cord compressions. This study has introduced new protocol for acquiring an extension view of cervical MRI to assess dynamic cervical spinal canal compromise. We posit that dynamic MRI comprising extension view in prone position could be a practical option when deciding the best approach in treating challenging patients.

Diagnosing cervical foraminal stenosis with intermittent arm radiculopathy is challenging due to discrepancies between MRI findings and symptoms. This can be attributed to the fact that MRI images are often obtained in a relaxed supine position. This study aims to evaluate the feasibility of the Dynamic MRI Compression System (DMRICS) and to assess possible changes in cervical foramina, with both quantitative measurements and qualitative grading systems, with MRI during a simulated Spurling test. Ten patients (five women and five men, ages 29-45) with previously confirmed cervical foraminal stenosis underwent MRI scans using DMRICS. MRI images were acquired in both relaxed and provoked states. A radiologist assessed 30 foramina (C4-C7) on the symptomatic side in both patient positions. Quantitative and qualitative measures were performed, including the numeric rating scale (NRS) and the Park and Kim grading systems. The provoked state induced concordant neck and arm pain in 9 of 10 patients. Significant shifts in Park and Kim foraminal gradings were noted: 13 of 27 Park gradings and 9 of 27 Kim gradings escalated post provocation. No quantitative changes were observed. This pilot study indicates that the DMRICS device has the potential to improve diagnostic accuracy for cervical radiculopathy, demonstrating induced cervical foraminal changes during a simulated Spurling test while performing MRI.

MRI is the gold standard modality for speech imaging. However, it remains relatively slow, which complicates imaging of fast movements. Thus, an MRI of the vocal tract is often performed in 2D. While 3D MRI provides more information, the quality of such images is often insufficient. The goal of this study was to test the applicability of super-resolution algorithms for dynamic vocal tract MRI. In total, 25 sagittal slices of 8 mm with an in-plane resolution of 1.6 × 1.6 mm2 were acquired consecutively using a highly-undersampled radial 2D FLASH sequence. The volunteers were reading a text in French with two different protocols. The slices were aligned using the simultaneously recorded sound. The super-resolution strategy was used to reconstruct 1.6 × 1.6 × 1.6 mm3 isotropic volumes. The resulting images were less sharp than the native 2D images but demonstrated a higher signal-to-noise ratio. It was also shown that the super-resolution allows for eliminating inconsistencies leading to regular transitions between the slices. Additionally, it was demonstrated that using visual stimuli and shorter text fragments improves the inter-slice consistency and the super-resolved image sharpness. Therefore, with a correct speech task choice, the proposed method allows for the reconstruction of high-quality dynamic 3D volumes of the vocal tract during natural speech.

OBJECTIVE: To develop a novel deep learning approach for 4D-MRI reconstruction, named Movienet, which exploits space-time-coil correlations and motion preservation instead of k-space data consistency, to accelerate the acquisition of golden-angle radial data and enable subsecond reconstruction times in dynamic MRI.
METHODS: Movienet uses a U-net architecture with modified residual learning blocks that operate entirely in the image domain to remove aliasing artifacts and reconstruct an unaliased motion-resolved 4D image. Motion preservation is enforced by sorting the input image and reference for training in a linear motion order from expiration to inspiration. The input image was collected with a lower scan time than the reference XD-GRASP image used for training. Movienet is demonstrated for motion-resolved 4D MRI and motion-resistant 3D MRI of abdominal tumors on a therapeutic 1.5T MR-Linac (1.5-fold acquisition acceleration) and diagnostic 3T MRI scanners (2-fold and 2.25-fold acquisition acceleration for 4D and 3D, respectively). Image quality was evaluated quantitatively and qualitatively by expert clinical readers.
RESULTS: The reconstruction time of Movienet was 0.69 s (4 motion states) and 0.75 s (10 motion states), which is substantially lower than iterative XD-GRASP and unrolled reconstruction networks. Movienet enables faster acquisition than XD-GRASP with similar overall image quality and improved suppression of streaking artifacts.
CONCLUSIONS: Movienet accelerates data acquisition with respect to compressed sensing and reconstructs 4D images in less than 1 s, which would enable an efficient implementation of 4D MRI in a clinical setting for fast motion-resistant 3D anatomical imaging or motion-resolved 4D imaging.