Magnetic Resonance in Medicine最新文献

Purpose: To confirm ihMT's specificity to myelin, an ihMT presaturation module was combined with a Poon-Henkelman multi-echo spin-echo readout to separate the ihMT signal in myelin water from intra-/extra-cellular water. This study explored the relationship between two quantitative myelin imaging techniques and measured the ihMT signal of myelin water.

Methods: Six rats were injured; three were sacrificed three weeks post-injury, and three were sacrificed eight weeks post-injury, and three healthy control rats were also sacrificed. The nine formalin-fixed rat spinal cords were imaged using a Poon-Henkelman multi-echo spin-echo readout with an ihMT prepulse at different strengths of $T_{1D}$ filtering at 7T.

Results: The proposed model was able to characterize the ihMT decay signal in myelin water and intra-/extra-cellular water pool. From this proposed four-pool model with dipolar order reservoirs, we see a drop in the $T_{1D}\left(\text{myelin}\right)$ fit parameter in the fasciculus gracilis white matter region of the three-week post-injury cord. $T_{1D}\left(\text{myelin}\right)$ and $T_{1D}$ (non-myelin) were estimated to be approx. 8 and 1.5 ms, respectively.

Conclusion: The drop in $T_{1D}\left(\text{myelin}\right)$ in the three-week post-injury cords suggests that $T_{1D}\left(\text{myelin}\right)$ could potentially distinguish between functional myelin and myelin debris; however, more studies are needed to confirm this.

Purpose: To quantitatively compare signal-to-noise ratios (SNRs), linewidth values, and Cramér-Rao lower bounds (CRLBs) for amide resonances in the human brain measured at 3T and 7T using downfield MR spectroscopic imaging (DF-MRSI).

Methods: Seven normal volunteers (25-52 years, 5 female) were imaged using Philips 3T Elition and 7T Achieva scanners. Both systems have 32-channel receive head coils and 16-channel local shim arrays (MRShim GmbH) in addition to high-order spherical harmonic shims. Three-dimensional DF-MRSI data were collected using a previously developed pulse sequence with spectral-spatial excitation and frequency-selective refocusing pulses. Matched imaging protocols on both field strengths were applied to achieve a nominal voxel size of 7 × 7 × 15 mm in a scan time of 10.6 min. Spectral analysis was performed using the "LCModel" software package. SNR and CRLB values (%) were compared between 3T and 7T data using univariate general linear models.

Results: Significantly increased amide SNR and decreased CRLB values (p < 0.05) were found at 7T. Averaged over all brain regions, SNR was 2.9 ± 1.1 at 3T and 5.4 ± 1.5 at 7T, and CRLBs were 11.4 ± 3.9 and 4.9 ± 1.5 respectively. 7T MRI and amide images did show some regional signal dropoff due to transmit B₁ inhomogeneity, however.

Conclusion: Three-dimensional DF-MRSI at 7T showed 86% increased SNR and 57% decreased CRLB values compared with 3T, confirming the expected improvements at higher field. Improvements are probably due to multiple factors, including higher magnetization at 7T, the shorter minimum echo time available, among others.

Purpose: Focused ultrasound-induced blood-brain barrier (BBB) opening is a promising method for neurotherapeutic delivery. The standard for quantifying induced BBB permeability is the $K^{\text{trans}}$ parameter, which reflects both permeability and plasma flow. The influence of plasma flow can be eliminated by estimating the PS parameter. However, this parameter has been largely unexplored in this application. This study aims to compare permeability estimates based on $K^{\text{trans}}$ and PS in focused ultrasound-induced BBB opening experiments.

Methods: We used the extended Tofts model (ETM) and the two-compartment exchange model (2CXM) to estimate $K^{\text{trans}}$ and PS parameters, respectively. Permeability estimates were compared using simulated concentration curves, simulated DCE-MRI data, and real datasets. We explored the influence of spatially-regularized model fitting on the results.

Results: For opened BBB, $K^{\text{trans}}$ was minimally influenced by plasma flow under the tested conditions. However, fitting the ETM often introduced outliers in $K^{\text{trans}}$ estimates in regions with closed BBB. The 2CXM outperformed the ETM at high signal-to-noise ratios, but its higher complexity led to lower precision at low signal-to-noise ratios. Both these issues were successfully compensated by spatially-regularized model fitting.

Conclusion: Both $K^{\text{trans}}$ and PS seem to be eligible options for the quantification of BBB opening, and the correct choice depends on the specifics of the acquired DCE-MRI data. Additionally, spatial regularization has demonstrated its importance in enhancing the accuracy and reproducibility of results for both models.

Purpose: Field monitoring using field probes allows for accurate measurement of magnetic field perturbations, such as from eddy currents, during MRI scanning. However, errors may result when the spatial variation of the fields is not well-described by the conventionally used spherical harmonics model that has the maximum order constrained by the number of probes. The objective of this work was to develop and validate a field monitoring approach that compresses higher order spherical harmonics into a smaller set of new basis functions that can be characterized using fewer probes.

Methods: Field monitoring of acquisitions was repeated with probes in different locations. High-order field dynamics were computed from this "calibration" data assembled from provided scans, from which compression matrices could be devised using principal component analysis. Compression matrices were then used to fit field dynamics using "compressed" basis functions with data from 16 probes, which were then used in image reconstruction. Performance was evaluated by assessing the accuracy of computed field dynamics as well as in vivo image quality. Technique generalizability was also assessed by using various acquisition and diffusion encoding strategies in the calibration.

Results: Qualitative and quantitative improvements in accuracy were observed when using the proposed fitting method compared to the conventional approach. However, compression effectiveness was influenced by the probe quantity and arrangement, and the specific acquisition data included in the calibration.

Conclusion: The ability to tailor basis functions to more compactly describe the spatial variation of field perturbations enables improved characterization of fields with rapid spatial variations.

Purpose: Although gated first-pass contrast-enhanced sequences are the clinical standard for cardiovascular MR perfusion, some patient conditions necessitate using ungated steady-state sequences. However, through-plane cardiac motion and blood flow into the left ventricle can disrupt the magnetization steady state of the tissue, and perfusion quantification based on a steady-state assumption will contain errors. The tissue magnetization steady-state disruption can be eliminated with a proposed sequence modification that simultaneously excites transition bands with no change in the sequence resolution or timing parameters.

Theory and methods: The proposed sequence modification simultaneously excites two transition bands adjacent to the imaged region. The transition bands experience the same consistent excitation history as the imaged slices. Thus, any tissue that moves into an imaged slice location from a transition band location will not disrupt the magnetization steady state. Gradient dephasing and radiofrequency spoiling are used to null the transition band signal so that it does not contribute to the reconstructed images. Transition bands were added to a two-dimensional ungated steady-state radial FLASH (fast low-angle shot) sequence with simultaneous multiband imaging on a PRISMA 3T MRI scanner. Phantom, normal canine, and human subject data are presented.

Results: Transition bands reduce the amount of tissue magnetization disruption to the steady state without adding artifacts to the imaged slices. Myocardial blood flow maps from a selected normal canine study and a normal human subject show good uniformity and consistency to literature values for all slices. Perfusion estimates with the proposed method also demonstrate good consistency with saturation-recovery methods commonly used for myocardial perfusion imaging.

Conclusion: We have demonstrated that the proposed transition bands can reduce quantification errors resulting from blood flow into the left ventricle and through-plane cardiac and respiratory motion. There is no loss of image-acquisition efficiency, and temporal resolution is unchanged with this technique.

Purpose: Commonly used MR image quality (IQ) metrics have poor concordance with radiologist-perceived diagnostic IQ. Here, we develop and explore deep feature distances (DFDs)-distances computed in a lower-dimensional feature space encoded by a convolutional neural network (CNN)-as improved perceptual IQ metrics for MR image reconstruction. We further explore the impact of distribution shifts between images in the DFD CNN encoder training data and the IQ metric evaluation.

Methods: We compare commonly used IQ metrics (PSNR and SSIM) to two "out-of-domain" DFDs with encoders trained on natural images, an "in-domain" DFD trained on MR images alone, and two domain-adjacent DFDs trained on large medical imaging datasets. We additionally compare these with several state-of-the-art but less commonly reported IQ metrics, visual information fidelity (VIF), noise quality metric (NQM), and the high-frequency error norm (HFEN). IQ metric performance is assessed via correlations with five expert radiologist reader scores of perceived diagnostic IQ of various accelerated MR image reconstructions. We characterize the behavior of these IQ metrics under common distortions expected during image acquisition, including their sensitivity to acquisition noise.

Results: All DFDs and HFEN correlate more strongly with radiologist-perceived diagnostic IQ than SSIM, PSNR, and other state-of-the-art metrics, with correlations being comparable to radiologist inter-reader variability. Surprisingly, out-of-domain DFDs perform comparably to in-domain and domain-adjacent DFDs.

Conclusion: A suite of IQ metrics, including DFDs and HFEN, should be used alongside commonly-reported IQ metrics for a more holistic evaluation of MR image reconstruction perceptual quality. We also observe that general vision encoders are capable of assessing visual IQ even for MR images.

Purpose: Accurate analysis of metabolite levels from ¹H MRS data is a significant challenge, typically requiring the estimation of approximately 100 parameters from a single spectrum. Signal overlap, spectral noise, and common artifacts further complicate the analysis, leading to instability and reports of poor agreement between different analysis approaches. One inconsistently used method to improve analysis stability is known as regularization, where poorly determined parameters are partially constrained to take a predefined value. In this study, we examine how regularization of frequency and linewidth parameters influences analysis accuracy.

Methods: The accuracy of three MRS analysis methods was compared: (1) ABfit, (2) ABfit-reg, and (3) LCModel, where ABfit-reg is a modified version of ABfit incorporating regularization. Accuracy was assessed on synthetic MRS data generated with random variability in the frequency shift and linewidth parameters applied to each basis signal. Spectra ( $N=1000$ ) were generated across a range of SNR values (10, 30, 60, 100) to evaluate the impact of variable data quality.

Results: Comparison between ABfit and ABfit-reg demonstrates a statistically significant (p <  0.0005) improvement in accuracy associated with regularization for each SNR regime. An approximately 10% reduction in the mean squared metabolite errors was found for ABfit-reg compared to LCModel for SNR >10 (p <  0.0005). Furthermore, Bland-Altman analysis shows that incorporating regularization into ABfit enhances its agreement with LCModel.

Conclusion: Regularization is beneficial for MRS fitting and accurate characterization of the frequency and linewidth variability in vivo may yield further improvements.

Purpose: To propose and implement a new method to study water exchange between brain tissue and fluids.

Methods: An MLEV T₂-preparation combined with a cerebrospinal fluid (CSF) nulling inversion recovery was implemented in combination with an ultralong-echo time (TE) three-dimensional fast spin-echo readout. To handle systematic imperfections and isolate the exchange signal, T₂-prepared images were subtracted from one of two control images. The first control turned off the T₂ preparation and adjusted inversion timing to correct for relaxation. The second control used the same T₂ preparation but shifted in time. Preparations were implemented on a 3T scanner and tested in 14 healthy volunteers. We evaluated the exchange signal magnitude and distribution, as well as robustness against B₁ imperfection and intrasession reproducibility. We also compared the signal to that measured with ultralong-TE arterial spin labeling, another suggested marker of water exchange.

Results: Initial experiments using the T₂-preparation off control demonstrated a detectable exchange signal especially in the choroid plexus, but with substantial residual signal in CSF spaces, suggesting imperfect subtraction of non-exchanging spins. When using the time shifted control, we greatly reduced subtraction errors. Signal was consistently measured in the choroid plexus and at the boundaries between cortex and CSF with much higher signal-to-noise ratio and spatial resolution than ultralong-TE arterial spin labeling.

Conclusions: The measured water exchange distribution appears consistent with the localization of aquaporin channels at the CSF boundary. Because aquaporin activity may reflect CSF production and glymphatic clearance, our method may provide a noninvasive marker of these functions.

Purpose: Conventional MRI coils offer suboptimal parallel imaging performance for young children. Our goal was to enhance imaging acceleration by dedicated, flexible high-density coil design for pediatric patients at 3T.

Methods: We design, construct, and evaluate a highly flexible small loop array. Key design notes include full-wave simulation and analysis of the dual-turn loop, miniature feedboard allocation at the loop center, and cable management. Phantom experiments and adult and pediatric volunteer case studies were conducted to evaluate the small loop array imaging performance compared to commercial reference coils.

Results: Dual-turn loop configuration forms higher preamp decoupling impedance than the same size single-turn, supporting a flexible form factor that requires a wide range of critical overlap. Both phantom and in-vivo studies demonstrate superior parallel imaging performance or high spatial resolution imaging using the small loop, compared to commercial reference coils.

Conclusion: A dedicated high-density coil array with a minimum inter-component interference layout design allows a flexible form factor and higher imaging accelerations. Phantom and in-vivo volunteer case studies demonstrate promising results in improving efficiency for pediatric patients in routine clinical imaging procedures.