Magnetic Resonance Materials in Physics, Biology and Medicine最新文献

Objective: Acquiring fully sampled training data is challenging for many MRI applications. We present a self-supervised image reconstruction method, termed ReSiDe, capable of recovering images solely from undersampled data.

Materials and methods: ReSiDe is inspired by plug-and-play (PnP) methods, but unlike traditional PnP approaches that utilize pre-trained denoisers, ReSiDe iteratively trains the denoiser on the image or images that are being reconstructed. We introduce two variations of our method: ReSiDe-S and ReSiDe-M. ReSiDe-S is scan-specific and works with a single set of undersampled measurements, while ReSiDe-M operates on multiple sets of undersampled measurements and provides faster inference. Studies I, II, and III compare ReSiDe-S and ReSiDe-M against other self-supervised or unsupervised methods using data from T1- and T2-weighted brain MRI, MRXCAT digital perfusion phantom, and first-pass cardiac perfusion, respectively.

Results: ReSiDe-S and ReSiDe-M outperform other methods in terms of peak signal-to-noise ratio and structural similarity index measure for Studies I and II, and in terms of expert scoring for Study III.

Discussion: We present a self-supervised image reconstruction method and validate it in both static and dynamic MRI applications. These developments can benefit MRI applications where the availability of fully sampled training data is limited.

Objective: Tissue heating near the implanted deep brain stimulation (DBS) during magnetic resonance imaging (MRI) poses a significant safety constraint. This study aimed to evaluate the performance of parallel transmit (pTx) head transmit radiofrequency (RF) coils in DBS patients, with a focus on excitation fidelity under specific absorption rate (SAR) control for brain imaging at 3T MRI.

Materials and methods: We employed electromagnetic simulations to assess different coil configurations, including multi-row pTx coils of 16-24 channels arranged in 1, 2, and 3 rows, and compared these to a circularly polarised and pTx birdcage coil using a realistic human model without and with DBS leads and electrodes.

Results: Two- and three-row pTx coils with overlapping loop elements exhibited similar performance, which was superior in excitation homogeneity and local SAR compared to the single-row coil and the birdcage coil both without and with DBS.

Discussion: These findings suggest that multi-row coils can enhance the safety and efficacy of MRI in patients with DBS devices, so potentially improving imaging performance in this expanding patient population. There was a minimal difference in performance between the 2 and 3-row coils, favouring the simpler, lower channel count design for practical implementation.

Object: Spatio-temporal MRI methods offer rapid whole-brain multi-parametric mapping, yet they are often hindered by prolonged reconstruction times or prohibitively burdensome hardware requirements. The aim of this project is to reduce reconstruction time using deep learning.

Materials and methods: This study focuses on accelerating the reconstruction of volumetric multi-axis spiral projection MRF, aiming for whole-brain T1 and T2 mapping, while ensuring a streamlined approach compatible with clinical requirements. To optimize reconstruction time, the traditional method is first revamped with a memory-efficient GPU implementation. Deep Learning Initialized Compressed Sensing (Deli-CS) is then introduced, which initiates iterative reconstruction with a DL-generated seed point, reducing the number of iterations needed for convergence.

Results: The full reconstruction process for volumetric multi-axis spiral projection MRF is completed in just 20 min compared to over 2 h for the previously published implementation. Comparative analysis demonstrates Deli-CS's efficiency in expediting iterative reconstruction while maintaining high-quality results.

Discussion: By offering a rapid warm start to the iterative reconstruction algorithm, this method substantially reduces processing time while preserving reconstruction quality. Its successful implementation paves the way for advanced spatio-temporal MRI techniques, addressing the challenge of extensive reconstruction times and ensuring efficient, high-quality imaging in a streamlined manner.

Objectives: Caffeine, a known neurostimulant and adenosine antagonist, affects brain physiology by decreasing cerebral blood flow. It interacts with adenosine receptors to induce vasoconstriction, potentially disrupting brain homeostasis. However, the impact of caffeine on blood-brain barrier (BBB) permeability to water remains underexplored. This study investigated the water exchange via the BBB in a perturbed physiological condition caused by caffeine ingestion, using the multiple echo time (multi-TE) arterial spin labeling (ASL) technique.

Material and methods: Ten healthy, regular coffee drinkers (age = 31 ± 9 years, 3 females) were scanned to acquire five measurements before and six measurements after caffeine ingestion. Data were analyzed with a multi-TE two-compartment model to estimate exchange time (Tex), serving as a proxy for BBB permeability to water. Additionally, cerebral blood flow (CBF), arterial transit time (ATT), and intravoxel transit time (ITT) were investigated.

Results: Following caffeine intake, mean gray matter CBF showed a significant time-dependent decrease (P < 0.01). In contrast, Tex, ATT, and ITT did not exhibit significant time-dependent change. However, a non-significant decreasing trend was observed for Tex and ITT, respectively, while ATT showed an increasing trend over time.

Discussion: The observed decreasing trend in Tex after caffeine ingestion suggests a potential increase in water flux across the BBB, which may represent a compensatory mechanism to maintain brain homeostasis in response to the caffeine-induced reduction in CBF. Further studies with larger sample sizes are needed to validate and expand upon these findings.

Objective: Invasive multimodal fMRI in rodents is often compromised by susceptibility artifacts from adhesives used to secure cranial implants. We hypothesized that adhesive type, shape, and field strength significantly affect susceptibility artifacts, and systematically evaluated various adhesives.

Materials and methods: Thirty-one adhesives were applied in constrained/unconstrained geometries and imaged with T2*-weighted EPI at 7.0 and 9.4 T to assess artifact depths. Spherical and flat patch shapes, both unconstrained geometries, were compared for artifact depth in vitro and in vivo. Adhesion strength was assessed on post-mortem mouse crania. Finally, an integrative scoring system rated adhesive properties, including artifact depth, handling, and adhesion strength.

Results: Susceptibility artifacts were two times larger at 9.4 than at 7.0 T (p < 0.001), strongest at the patch edges, and deeper with spherical than flat patches (p < 0.05). Artifact size depended more on shape and volume after curing than adhesive type. Our integrative scoring system showed resins, bonding agents, and acrylics offered the best overall properties, while silicones and cements were less favorable.

Discussion: Adhesive selection requires balancing handling, curing time, strength, and artifact depth. To minimize artifacts, adhesives should be applied in a spread-out, flat and thin layer. Our integrative scoring system supports classification of future classes of adhesives.

Objective: Segmentation of individual thigh muscles in MRI images is essential for monitoring neuromuscular diseases and quantifying relevant biomarkers such as fat fraction (FF). Deep learning approaches such as U-Net have demonstrated effectiveness in this field. However, the impact of reducing neural network complexity remains unexplored in the FF quantification in individual muscles.

Material and methods: U-Net architectures with different complexities have been compared for the quantification of the fat fraction in each muscle group selected in the central part of the thigh region. The corresponding performance has been assessed in terms of Dice score (DSC) and FF quantification error. The database contained 1450 thigh images from 59 patients and 14 healthy subjects (age: 47 ± 17 years, sex: 36F, 37M). Ten individual muscles were segmented in each image. The performance of each model was compared to nnU-Net, a complex architecture with 4.35 $\times$ 10⁷ parameters, 12.8 Gigabytes of peak memory usage and 167 h of training time.

Results: As expected, nnU-Net achieved the highest DSC (94.77 ± 0.13%). A simpler U-Net (5.81 $\times$ 10⁵ parameters, 2.37 Gigabytes, 14 h of training time) achieved a lower DSC but still above 90%. Surprisingly, both models achieved a comparable FF estimation.

Discussion: The poor correlation between observed DSC and FF indicates that less complex architectures, reducing GPU memory utilization and training time, can still accurately quantify FF.

Objective: To establish an arterial spin labeling (ASL) protocol for rat livers that improves data reliability and reproducibility for perfusion quantification.

Methods: This study used respiratory-gated, single-slice, FAIR-based ASL imaging with multiple inversion times (TI) in rat livers. Quality assurance measures included: (1) introduction of mechanical ventilation to ensure consistent respiratory cycles by controlling the respiratory rate (45 bpm), tidal volume (10 ml/kg), and inspiration: expiration ratio (I:E ratio, 1:2), (2) optimization of the trigger window for consistent trigger points, and (3) use of fit residual map and coefficient of variance as metrics to assess data quality. We compared image quality, perfusion maps, and fit residual maps between mechanically ventilated and non-ventilated animals, as well as repeated ASL measurements (session = 4 per animal) in two mechanically ventilated animals.

Results: Perfusion measurements over multiple sessions in mechanically ventilated rats exhibited low perfusion data variability and high reproducibility both within and between liver lobes. Image quality and perfusion maps were significantly improved in mechanically ventilated animals compared to non-ventilated animals.

Discussion: The implementation of mechanical ventilation and optimized quality assurance protocols enhanced the reliability and reproducibility of FAIR-based multi-TI-ASL imaging in rat livers. Our findings demonstrate these measures as a robust approach for achieving consistent liver perfusion quantification in preclinical settings.

Objectives: Phosphorus-31 magnetic resonance spectroscopic imaging (³¹P-MRSI) is a non-invasive tool for assessing cellular high-energy metabolism in-vivo. However, its acquisition suffers from a low sensitivity, which necessitates large voxel sizes or multiple averages to achieve an acceptable signal-to-noise ratio (SNR), resulting in long scan times.

Materials and methods: To overcome these limitations, we propose an acquisition and reconstruction scheme for FID-MRSI sequences. Specifically, we employed Compressed Sensing (CS) and Low-Rank (LR) with Total Generalized Variation (TGV) regularization in a combined CS-LR framework. Additionally, we used a novel approach to k-space undersampling that utilizes distinct pseudo-random patterns for each average. To evaluate the proposed method's performance, we performed a retrospective analysis on healthy volunteers' brains and ex-vivo perfused kidneys.

Results: The presented method effectively improves the SNR two-to-threefold while preserving spectral and spatial quality even with threefold acceleration. We were able to recover signal attenuation of anatomical information, and the SNR improvement was obtained while maintaining the metabolites peaks linewidth.

Conclusions: We presented a novel combined CS-LR acceleration and reconstruction method for FID-MRSI sequences, utilizing a unique approach to k-space undersampling. Our proposed method has demonstrated promising results in enhancing the SNR making it applicable for reducing acquisition time.

Purposes: This research highlights the need for affordable phantoms for MRI education. Current options are either expensive or limited. A phantom, easy to manufacture and distribute, is proposed to demonstrate various pedagogical concepts, aiding students in understanding MRI image quality concepts.

Methods: We designed a cylindrical MRI phantom that comprises sections that can be filled with chosen liquids and gels. The dimensions were chosen to fit most consumer-grade 3D printers, facilitating widespread dissemination. It includes five modular sections for evaluating spatial resolution, geometrical accuracy, slice thickness accuracy, homogeneity, and contrast.

Results: The modular cylindrical MRI phantom was successfully fabricated. Each section of the phantom was tested to ensure it met the specified pedagogical needs. The spatial resolution section provided clear images for evaluating fine details. The geometrical accuracy section allowed for precise measurement of distortions. The slice thickness accuracy section confirmed the consistency of slice thickness across different MRI sequences. The homogeneity section demonstrated uniform signal distribution, and the contrast section effectively displayed varying contrast levels.

Conclusions: This modular MRI phantom offers a cost-effective tool for educational purposes in MRI. Its design enables educators to demonstrate multiple pedagogical scenarios with a single object. The phantom's compatibility with consumer-grade 3D printers and its modularity makes it accessible and adaptable to various educational settings. Future work could explore further customization and enhancement of the phantom to cover additional educational needs. This tool represents a significant step toward improving MRI education and training by providing a practical, hands-on learning experience.

Objective: To propose a deep learning-based low-field mobile MRI strategy for fast, high-quality, unshielded imaging using minimal hardware resources.

Methods: Firstly, we analyze the correlation of EMI signals between the sensing coil and the MRI coil to preliminarily verify the feasibility of active EMI shielding using a single sensing coil. Then, a powerful deep learning EMI elimination model is proposed, which can accurately predict the EMI components in the MRI coil signals using EMI signals from at least one sensing coil. Further, deep learning models with different task objectives (super-resolution and denoising) are strategically stacked for multi-level post-processing to enable fast and high-quality low-field MRI. Finally, extensive phantom and brain experiments were conducted on a home-built 0.2 T mobile brain scanner for the evaluation of the proposed strategy.

Results: 20 healthy volunteers were recruited to participate in the experiment. The results show that the proposed strategy enables the 0.2 T scanner to generate images with sufficient anatomical information and diagnostic value under unshielded conditions using a single sensing coil. In particular, the EMI elimination outperforms the state-of-the-art deep learning methods and numerical computation methods. In addition, 2 × super-resolution (DDSRNet) and denoising (SwinIR) techniques enable further improvements in imaging speed and quality.

Discussion: The proposed strategy enables low-field mobile MRI scanners to achieve fast, high-quality imaging under unshielded conditions using minimal hardware resources, which has great significance for the widespread deployment of low-field mobile MRI scanners.