NMR in Biomedicine最新文献

Ultra-high field (7T) MRI allows scans at sub-millimetre resolution with exquisite signal-to-noise ratio (SNR). As 7T MRI becomes more widely used clinically, the challenge of patient motion must be overcome. Retrospective motion correction is used successfully for some protocols, but for acquisitions such as slice-by-slice scans only prospective motion correction can deliver the full potential of 7T MRI. We report the first implementation of prospective 3D Fat Navigator ("FatNav") motion correction for the Siemens 7T Terra MRI. We implemented a modular Sequence Building Block for FatNav and embedded it into the vendor's gradient-recalled echo (GRE) sequence. We modified the reconstruction pipeline to reconstruct FatNav images online, coregistering them and sending motion updates to the host sequence online. We tested five registration algorithms for performance and accuracy on synthetic FatNav data. We implemented the best three of these in our sequence and tested them online. We acquired T₁ and T₂* weighted brain images of healthy volunteers correcting every other image for motion to visualise the effectiveness of online motion correction. Data were acquired with and without head immobilisation. We also tested performance while correcting every measurement for motion. Our implementation uses a 1.23 s 3D FatNav acquisition module and delivers motion updates in less than 3 s, which is sufficient for motion updates every few k-space lines in typical scans. Corrected images are crisper with fewer visible motion artefacts. This improved sharpness is reflected quantitatively by an increase in the variance of the image Laplacian which is 1.59 x better for corrected vs uncorrected images. Profiles across the cerebral falx are 33% steeper for corrected vs uncorrected images. Prospective FatNav improves GRE image quality in the brain. Our modular Sequence Building Block provides a simple method to integrate motion correction in 7T MRI pulse sequences.

Diffusion tensor imaging (DTI) provides insight into the skeletal muscle microstructure and can be acquired using a stimulated echo acquisition mode (STEAM)-based approach to quantify time-dependent tissue diffusion. This study examined diffusion metrics and signal-to-noise ratio (SNR) in the supraspinatus muscle obtained with a STEAM-DTI sequence with different diffusion encoding times (Δ) and compared them to measures from a spin echo (SE) sequence. Ten healthy subjects (mean age 31.5 ± 4.7 years; five females) underwent 3-Tesla STEAM and SE-DTI of the shoulder in three sessions. STEAM was acquired with Δ of 100/200/400/600 ms. The diffusion encoding time in SE scans was 19 ms (b = 500 s/mm²). Region of interest-based measurement of fractional anisotropy (FA), mean diffusivity (MD), and SNR was performed. Intraclass correlation coefficients (ICCs) were computed to assess test-retest reliability. ANOVA with post-hoc pairwise tests was used to compare measures between different Δ of STEAM as well as STEAM and SE, respectively. FA was significantly higher (FA_STEAM: 0.38-0.46 vs. FA_SE: 0.26) and MD significantly lower (MD_STEAM: 1.20-1.33 vs. MD_SE: 1.62 × 10^-3 mm²/s) in STEAM compared to SE (p < 0.001, respectively). SNR was significantly higher for SE (72.3 ± 8.7) than for STEAM (p < 0.001). ICCs were excellent for FA in STEAM (≥0.911) and SE (0.960). For MD, ICCs were good for STEAM_100ms-600ms (≥0.759) and SE (0.752). STEAM and SE exhibited excellent reliability for FA and good reliability for MD in the supraspinatus muscle. SNR was significantly higher in SE compared to STEAM.

Quality assessment, including inspecting the images for artifacts, is a critical step during magnetic resonance imaging (MRI) data acquisition to ensure data quality and downstream analysis or interpretation success. This study demonstrates a deep learning (DL) model to detect rigid motion in T₁-weighted brain images. We leveraged a 2D convolutional neural network (CNN) trained on motion-synthesized data for three-class classification and tested it on publicly available retrospective and prospective datasets. Grad-CAM heatmaps enabled the identification of failure modes and provided an interpretation of the model's results. The model achieved average precision and recall metrics of 85% and 80% on six motion-simulated retrospective datasets. Additionally, the model's classifications on the prospective dataset showed 93% agreement with the labeling of a radiologist a strong inverse correlation (-0.84) compared to average edge strength, an image quality metric indicative of motion. This model is aimed at inline automatic detection of motion artifacts, accelerating part of the time-consuming quality assessment (QA) process and augmenting expertise on-site, particularly relevant in low-resource settings where local MR knowledge is scarce.

Amide proton transfer (APT) imaging, a technique sensitive to tissue pH, holds promise in the diagnosis of ischemic stroke. Achieving accurate and rapid APT imaging is crucial for this application. However, conventional APT quantification methods either lack accuracy or are time-consuming. Machine learning (ML) has recently been recognized as a potential solution to improve APT quantification. In this paper, we applied an ML model trained on a new type of partially synthetic data, along with an optimization approach utilizing recursive feature elimination, to predict APT imaging in an animal stroke model. This partially synthetic datum is not a simple blend of measured and simulated chemical exchange saturation transfer (CEST) signals. Rather, it integrates the underlying components including all CEST, direct water saturation, and magnetization transfer effects partly derived from measurements and simulations to reconstruct the CEST signals using an inverse summation relationship. Training with partially synthetic data requires less in vivo data compared to training entirely with fully synthetic or in vivo data, making it a more practical approach. Since this type of data closely resembles real tissue, it leads to more accurate predictions than ML models trained on fully synthetic data. Results indicate that an ML model trained on this partially synthetic data can successfully predict the APT effect with enhanced accuracy, providing significant contrast between stroke lesions and normal tissues, thus clearly delineating lesions. In contrast, conventional quantification methods such as the asymmetric analysis method, three-point method, and multiple-pool model Lorentzian fit showed inadequate accuracy in quantifying the APT effect. Moreover, ML methods trained using in vivo data and fully synthetic data exhibited poor predictive performance due to insufficient training data and inaccurate simulation pool settings or parameter ranges, respectively. Following optimization, only 13 frequency offsets were selected from the initial 69, resulting in significantly reduced scan time.

Quantitative MRI and MRS have become important tools for the assessment and management of patients with neuromuscular disorders (NMDs). Despite significant progress, there is a need for new objective measures with improved specificity to the underlying pathophysiological alteration. This would enhance our ability to characterize disease evolution and improve therapeutic development. In this study, qMRI methods that are commonly used in clinical studies involving NMDs, like water T2 (T2_H2O) and T1 and fat-fraction (FF) mapping, were employed to evaluate disease activity and progression in the skeletal muscle of golden retriever muscular dystrophy (GRMD) dogs. Additionally, extracellular volume (ECV) fraction and single-voxel bicomponent water T2 relaxometry were included as potential markers of specific histopathological changes within the tissue. Apart from FF, which was not significantly different between GRMD and control dogs and showed no trend with age, T2_H2O, T1, ECV, and the relative fraction of the long-T2 component, A₂, were significantly elevated in GRMD dogs across all age ranges. Moreover, longitudinal assessment starting at 2 months of age revealed significant decreases in T2_H2O, T1, ECV, A₂, and the T2 of the shorter-T2 component, T2₁, in both control and GRMD dogs during their first year of life. Notably, insights from ECV and bicomponent water T2 indicate that (I) the elevated T2_H2O and T1 values observed in dystrophic muscle are primarily driven by an expansion of the extracellular space, likely driven by the edematous component of inflammatory responses to tissue injury and (II) the significant decrease of T2_H2O and T1 with age in control and GRMD dogs reflects primarily the progressive increase in fiber diameter and protein content during tissue development. Our study underscores the potential of multicomponent water T2 relaxometry and ECV to provide valuable insights into muscle pathology in NMDs.

Quantitative susceptibility mapping (QSM) is a tool for mapping tissue susceptibility. Using QSM for functional brain mapping, it is possible to directly quantify blood-oxygen-level-dependent (BOLD) susceptibility changes. This study presents a submillimeter functional QSM (fQSM) approach compared to BOLD fMRI from data acquired with 3D gradient-echo echo planar imaging (EPI) at ultra-high field. Complex EPI data were acquired in nine healthy subjects with varying temporal and spatial resolutions and used for BOLD fMRI and for fQSM. Right-hand finger tapping experiments were performed as well as one measurement with intentional subject movement. Susceptibility maps were computed using 3D path-based unwrapping, the variable-kernel sophisticated harmonic artifact reduction for phase data, and the streaking artifact reduction for QSM algorithm. Functional data analysis included general linear modeling and computation of z-scores. Submillimeter data were denoised using NOise reduction with DIstribution Corrected (NORDIC), which improved z-scores in the motor cortex for fQSM and fMRI. An expected increase in BOLD fMRI signal and corresponding decrease in magnetic susceptibility was observed in sensorimotor areas during active periods. For all experiments, fQSM showed smaller activation regions compared with fMRI. The percentage of high negative t-values localized in the cortex was higher for fQSM (52%) than for positive or negative t-values for fMRI (45%). For the scans with intentional motion, movement exceeded the size of a voxel, but paradigm dependent signal evolution could be recovered using motion correction. In conclusion, this study demonstrates the feasibility of submillimeter whole-brain fQSM with voxel volume of 0.53 μL. In comparison to traditional BOLD fMRI, fQSM provided improved localization of brain activation within the cortex, especially in submillimeter 3D EPI sequences.

³¹P magnetic resonance spectroscopy (MRS) can spectrally resolve metabolites involved in phospholipid metabolism whose levels are altered in many cancers. Ultra-high field facilitates the detection of phosphomonoesters (PMEs) and phosphodiesters (PDEs) with increased SNR and spectral resolution. Utilizing multi-echo MR spectroscopic imaging (MRSI) further enhances SNR and enables T₂ information estimation per metabolite. To address the specific absorption rate (SAR) challenges associated with high-power demanding adiabatic or composite block pulses in multi-echo phosphorus imaging, we present a dual-band refocusing RF pulse designed for operation at B₁ amplitudes of 14.8 μT which holds potential for integration into multi-echo sequences. Phantom and in vivo experiments conducted in the brain at 7 Tesla validated the effectiveness of this low-power dual-band RF pulse. Furthermore, we implemented the dual-band RF pulse into a multi-echo MRSI sequence where it offered the potential to increase the number of echo pulses within the same acquisition time compared to high-power adiabatic implementation, demonstrating its feasibility and practicality.

Low magnetic field magnetic resonance imaging (MRI) ( $B_0$  < 1 T) is regaining interest in the magnetic resonance (MR) community as a complementary, more flexible, and cost-effective approach to MRI diagnosis. Yet, the impaired signal-to-noise ratio (SNR) per square root of time, or SNR efficiency, leading in turn to prolonged acquisition times, still challenges its relevance at the clinical level. To address this, researchers investigate various hardware and software solutions to improve SNR efficiency at low field, including the leveraging of latest advances in computing hardware. However, there may not be a single recipe for improving SNR at low field, and it is key to embrace the challenges and limitations of each proposed solution. In other words, suitable solutions depend on the final objective or application envisioned for a low-field scanner and, more importantly, on the characteristics of a specific low $B_0$ field. In this review, we aim to provide an overview on software solutions to improve SNR efficiency at low field. First, we cover techniques for efficient k-space sampling and reconstruction. Then, we present post-acquisition techniques that enhance MR images such as denoising and super-resolution. In addition, we summarize recently introduced electromagnetic interference cancellation approaches showing great promises when operating in shielding-free environments. Finally, we discuss the advantages and limitations of these approaches that could provide directions for future applications.

Visualizing neuroimaging data is a key step in evaluating data quality, interpreting results, and communicating findings. This survey focuses on diffusion MRI tractography, which has been widely used in both research and clinical domains within the neuroimaging community. With an increasing number of tractography tools and software, navigating this landscape poses a challenge, especially for newcomers. A systematic exploration of a diverse range of features is proposed across 27 research tools, delving into their main purpose and examining the presence or absence of prevalent visualization and interactive techniques. The findings are structured within a proposed taxonomy, providing a comprehensive overview. Insights derived from this analysis will help (novice) researchers, clinicians, and developers in identifying knowledge gaps and navigating the landscape of tractography visualization tools.