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

Objective: To evaluate if co-registering Diffusion-Weighted Imaging (DWI) before generating Apparent Diffusion Coefficient (ADC) maps can improve differentiating benign and malignant breast lesions in MRI based on the A6702 thresholds.

Materials and methods: This IRB-approved study involved an in-house dataset and the publicly available ACRIN-6698 dataset, both including multi b-value DWI. In phase one, 16 ANTs library-based co-registration methods were evaluated on a subset of n = 138 cases from our in-house cohort. The quantitative assessment included mean ADC values of manually segmented lesions (diagnostic metrics using individual and A6702-defined thresholds) and coefficient of Variation. In the second phase, the best-performing methods were tested for generalizability on an unseen set of 40 cases (20 from in-house and 20 from external dataset). Three blinded readers segmented lesions on co-registered and non-co-registered ADC maps. Agreement and consistency were evaluated via Bland-Altman, segmentation distance, and intraclass correlation coefficient.

Results: Rigid co-registration using DWI at b = 750 s/mm² as reference (b750-Rigid) improved accuracy of both optimal/conservative A6702 trial thresholds with sensitivity/specificity increasing from 93%/10% to 97%/30% and 100%/30% respectively. Mean ADC values were not significantly different after co-registration (p > 0.05).

Discussion: Co-registration of DWI images before ADC map generation, particularly using b750-Rigid registration, seems promising for improving lesion classification in breast MRI. Further validation is warranted.

The frequency response and transfer function of a system are closely related, but distinct concepts from a control system theory and signal processing perspective. Unfortunately, these concepts have been used inconsistently in the magnetic resonance imaging (MRI) literature for gradient characterization. This note highlights the differences, with the intention to establish a common naming convention, consistent with other engineering fields. This will facilitate communication between colleagues with a different research background, promoting knowledge transfer and potentially alleviate shortcomings that have resulted from the incorrect usage of the term "transfer function" for gradient characterization in the past.

Background: Deuterium metabolic imaging (DMI) has recently been established as a versatile MR-based technique for in vivo mapping of glucose and other metabolic pathways using safe, non-ionizing ²H-labeled tracers.

Objective: In this review, methodological advances in DMI over the past decade are summarized, spanning hardware, acquisition, reconstruction, and quantification.

Approach and outline: Developments in multinuclear system modifications and dual-tuned head and body coils that enable 3D DMI at clinical and ultra-high field strengths are outlined. Efficient spatial-spectral encoding strategies and balanced steady-state-free-precession-based MRSI, which improve SNR efficiency and spatiotemporal resolution, are reviewed together with temporally interleaved ¹H/²H acquisitions that integrate DMI into standard MRI workflows. Indirect ¹H-observed deuterium detection (QELT) is described as a complementary approach for sites without multinuclear hardware. On the reconstruction side, model-based, low-rank and AI-driven methods are highlighted for de-noising, accelerated sampling, and robust spectral-temporal fitting.

Outlook: Current strategies for concentration estimation, kinetic modeling, and treatment of label loss are discussed, positioning DMI as a promising complement to FDG-PET and ¹³C-MRS for studying metabolism in neurological, oncological and systemic disease.

Motivation: The twisted-pair (TP) coil design is a promising strategy for developing novel, flexible, wearable MRI detectors that can provide SNR gains in various clinical applications of high-field MRI. We hypothesize that the TP coil's receive (Rx) sensitivity can be significantly increased by combining it with two complementary elements, such as dielectric resonators (DRs) and dipole antennas.

Methods: TP coils were combined with DRs made of high-permittivity material (ε_r = 1070) and transceiver (TxRx) dipole antennas. The Tx and Rx performance of six different types of arrays (TP-only, dipole-only, TP with DRs, dipole with DRs, dipole with TPs, and dipole with TPs and DRs) was investigated through numerical simulations involving a cylindrical phantom suitable for lower extremity applications and two human voxel models. MR phantom experiments were conducted using a 7 Tesla whole-body MRI scanner to validate the Tx and Rx performance of all six array types.

Results: The array combining all three types of elements (TP coils, DRs, and dipole antennas) provided the highest overall Rx performance; MR phantom experiments showed that integrating DRs with TP coils increased peripheral SNR by 250% and central SNR by 23% (for a total 38% gain in the center when also using dipole antennas in Rx). Human voxel model simulations confirmed that similar SNR gains can be achieved in vivo. Integrating DRs into TP coils also increased central Tx field efficiency by 4.6% and reduced the peak SAR_10g by 25.8% in the human voxel model Hugo.

Conclusion: DRs and dipole antennas can significantly improve the overall Rx performance of TP coils. This concept can benefit MRI of the human lower extremity at 7 Tesla and encourage exploration of its utility for other clinical applications.

Objective: Liver regeneration after partial hepatectomy (PH) is markedly impaired in liver fibrosis, leading to serious complications. Non-invasive imaging methods for predicting regenerative capacity are crucial for preoperative planning and risk assessment. Gadobenate dimeglumine (Gd-BOPTA)-enhanced MRI at hepatobiliary phase has recently shown promise for assessing liver function. This study investigated the predictive value of Gd-BOPTA-enhanced MRI at hepatobiliary phase for liver regeneration in fibrotic rats.

Materials and methods: Thirty male Sprague-Dawley rats with experimentally induced liver fibrosis underwent Gd-BOPTA-enhanced MRI. The relative enhancement ratios of liver parenchyma (REL) and biliary system (REB) were quantified during the hepatobiliary phase. After 70% PH, the liver regeneration rate (LRR) was calculated on day 3 and 5. Stepwise multivariable linear regression was conducted to identify imaging and biochemical determinants of LRR.

Results: In fibrotic rats, the mean LRRs were 0.80 ± 0.10 (range, 0.64-1.01) and 1.06 ± 0.09 (range, 0.89-1.17) on day 3 and 5 after PH, respectively. Multivariable analysis identified REL (p = 0.002), REB (p = 0.026), and alanine aminotransferase (ALT; p = 0.040) as the strongest determinants of LRR on day 3 (Predicted LRR on day 3 = 0.368 + 1.332 × REL + 0.105 × REB - 0.001 × ALT(IU/l)). On day 5, REL (p < 0.001) and REB (p = 0.023) remained significant determinants of LRR (Predicted LRR on day 5 = 1.107 + 2.601 × REL - 0.173 × REB).

Discussion: Gd-BOPTA-enhanced MRI at hepatobiliary phase can effectively predict LRR on day 3 and 5 after partial hepatectomy in fibrotic rats.

Objective: This work investigated the performance of MRI Faraday cages (FCs) over the lifetime of clinical MRI systems, aiming to better inform an option to repurpose an existing FC when an MRI scanner is replaced.

Materials and methods: FC performance was measured at acceptance testing for 40 MRI systems and for a further 11 MRI systems of various ages. Results were compared with the MRI vendor's FC specification and with measurements made when the FCs were initially built.

Results: The majority of FCs, 63% (n = 25), had at least one measurement below specification at acceptance testing. However, no RF artefacts were observed on MR images. There were significant negative relationships between FC performance and age at the locations of the door and window (p < 0.001).

Discussion: FC performance can degrade between the time of FC manufacture and initial clinical MRI scanning. However, FC attenuation levels may need to be considerably less than specification values before external RF artefacts start appearing on MR images in practice. Further degradation of FC performance may occur over time, but this may be better addressed by maintenance on the MR exam room door rather than a much more costly and time-consuming replacement of the FC.

Objective: Quantitative Deuterium Metabolic Imaging, or DMI, is typically based on the natural abundance deuterium (²H) signal from water that is inherently present in all DMI data. The ²H level in water depends on many geographical and atmospheric factors, whereby the in vivo ²H level can be further modified through the administration and breakdown of deuterated substrates. For water to act as an internal concentration reference, the ²H enrichment needs to be determined on a regional or even per-subject basis.

Materials and methods: An NMR method is presented to quantitatively and robustly determine the ²H enrichment in water using dimethyl sulfoxide (DMSO) as an ¹H/²H internal reference. The method, employing a ¹H/²H ratio of water/DMSO ratios, is independent of the amount of water or reference. The method is readily implemented on any modern NMR spectrometer with signal acquisition based on simple, fully-relaxed pulse-acquire methods and standard NMR tubes.

Results: The double ratio method is validated on samples with known ²H enrichments and variations in ²H water content are demonstrated for bottled spring waters from across the United States, and for human blood plasma during infusions of deuterated glucose and acetate.

Discussion: The presented double ratio method is a robust and practical tool to determine ²H water enrichment on individual subjects and/or specific geographic regions.

Objective: This study compared image quality and lipid suppression efficacy between 5 and 3 T MR systems for lower-extremity non-contrast enhanced magnetic resonance angiography (NCE-MRA) using an optimized 2D time-of-flight (TOF) sequence with spatially separated lipid pre-saturation (SLIP).

Materials and methods: Ten healthy volunteers underwent 2D TOF examination on lower limbs at both field strengths. The SLIP technique was evaluated across field strengths and compared with conventional CHESS to assess lipid suppression efficiency. Subjective scoring was used to assess vessel visualization and image quality, while quantitative analysis of vessel-to-muscle contrast ratios was performed. Statistical significance was determined using paired t-tests. Three patients with Peripheral Arterial Occlusive Disease (PAOD) were evaluated at 5 T and compared to computed tomography angiography (CTA) as the reference standard.

Results: The implementation of SLIP underscores the capability of increased field strength for more effective implementation of chemical shift-based lipid suppression technique. 5 T NCE MRA demonstrated higher Likert scores of radiologists' subjective evaluations of vessel delineation (5 T vs 3 T: 3.73 vs. 3.47, P < 0.001) and image quality (3.58 vs 3.27, P = 0.002) than 3 T. Vessel-to-background ratio (VBR) (14.87 ± 3.80 vs 10.07 ± 2.64, P < 0.001) and vessel contrast-to-background ratio (VCBR) were higher at 5 T (0.84 ± 0.11 vs 0.77 ± 0.12, P < 0 .001), indicating enhanced vessel delineation than 3 T.

Conclusion: 5 T NCE-MRA outperforms 3 T in visualizing lower limb vasculature, with enhanced lipid suppression and reduced in-plane saturation artifacts, offering a non-invasive alternative for peripheral vascular assessment.

Objective: Deuterium metabolic imaging (DMI) is an emerging MR technique providing non-invasive insights into glucose metabolism. Reliable concentration estimation depends on knowledge of tissue specific relaxation times. This study reports T₁ and T₂ relaxation time constants of deuterium-labeled water (HDO) and glucose (Glc) from the human liver and kidney at 7T.

Materials and methods: Twelve healthy volunteers (6f/6 m) were examined using k-space-reordered inversion-recovery and spin-echo DMI with non-Cartesian concentric-ring trajectory (CRT) sampling. Seven volunteers underwent oral ²H-Glc (0.8 g/kg body weight) administration. Data were averaged over organ-specific masks before spectral fitting. One volunteer was measured after oral D₂O (0.5 ml/kg body weight) administration.

Results: Faster longitudinal relaxation but similar transversal relaxation were observed for ²H-labeled Glc in the liver compared to kidney tissue (T₁^liver/kidney = 60 ± 4 ms/85 ± 18 ms, p = 0.016; T₂^liver/kidney = 31 ± 6 ms/35 ± 2 ms, p = 0.283). HDO exhibited significantly shorter liver relaxation times (T₁^liver/kidney = 218 ± 24 ms/324 ± 34 ms, p < 0.001; T₂^liver/kidney = 28 ± 4 ms/39 ± 6 ms, p < 0.001). D₂O loading improved voxelwise SNR enabling renal T₁/T₂ mapping of HDO.

Discussion: Hepatic and renal glucose homeostasis is often impaired in several pathophysiological conditions such as tumors, diabetes and metabolic dysfunction-associated steatotic liver disease. Using organ-specific ²H relaxation times increases the accuracy of concentration estimation and can help to improve the understanding of underlying metabolic processes in future abdominal DMI studies, which can help to push abdominal DMI towards clinical application.

Objective: Structural MRI-based regional volumes are widely used for Alzheimer's disease (AD) classification, but inter-individual variability in intracranial volume (ICV) introduces confounding. Traditional adjustment methods use region-of-interest (ROI)/ICV ratios or residual adjustment during pre-processing, yet no consensus exists on the optimal method. This study tests whether explicitly including ICV as a covariate (ROI + ICV) improves classification compared with ratio, residual adjustment, and the unadjusted baseline.

Materials and methods: T1-weighted MRIs from ADNI1 (n = 1423) and MIRIAD (n = 69) were processed with FreeSurfer to extract eight AD-related ROI volumes and ICV. Four feature configurations (ROI-only, ROI/ICV, Residual ROI, ROI + ICV) were benchmarked across six classifiers for cognitive normal (CN)-AD, CN-mild cognitive impairment (MCI), and MCI-AD. Performance was assessed with AUROC and F1 using Friedman and post hoc tests. In addition, feature attribution was examined with permutation importance and SHAP.

Results: ROI + ICV consistently produced the largest performance gains over ROI-only in CN-AD and CN-MCI, outperforming ratio and residual adjustment across most classifiers. These improvements generalized to the independent MIRIAD dataset. SHAP analyses showed that the directional effect of ICV reversed across strategies: under ratio or residual adjustment, larger ICV decreased AD probability, whereas in ROI + ICV, larger ICV increased it. This highlights ICV's contextual influence on model decisions.

Discussion: Pre-processing-based adjustments do not fully remove ICV effects and can distort ROI-ICV relationships. Explicit covariate inclusion avoids these issues and yields more consistent, generalizable improvements. Thus, ICV should be modeled rather than removed, making ROI + ICV the preferred default ICV-handling strategy for MRI-based AD classification.