Alternating unbalanced SSFP for 3D R2' mapping of the human brain.

PURPOSE: Measuring the transverse-relaxation rate R2' provides valuable information in quantitative evaluation of tissue microstructure, for example, in terms of oxygenation levels. Here, we propose an alternating unbalanced SSFP pulse sequence for rapid whole-brain 3D R2' mapping. METHODS: Unlike currently practiced, spin echo-based R2' measurement techniques, the proposed method alternates between SSFP-FID and SSFP-ECHO modes for rapid 3D encoding of transverse relaxation rates expressed as R2 + R2' and R2-R2' . Z-shimming gradients embedded into multi-echo trains of each SSFP module are designed to achieve relative immunity to large-scale magnetic-field variations (ΔB0 ). Appropriate models for the temporal evolution of the two groups of SSFP signals were constructed with ΔB0 -induced modulations accounted for, leading to ΔB0 -corrected estimation of R2 , R2' , and R2∗ (= R2 + R2' ). Additionally, relative magnetic susceptibility (Δχ) maps were obtained by quantitative susceptibility mapping of the phase data. Numerical simulations were performed to optimize scan parameters, followed by in vivo studies at 3 T in 7 healthy subjects. Measured parameters were evaluated in six brain regions, and subjected to interparameter correlation analysis. RESULTS: The resultant maps of R2' and additionally derived R2 , R2∗ , and Δχ all demonstrated the expected contrast across brain territories (eg, deep brain structures versus cortex), with the measured values in good agreement with previous reports. Furthermore, regression analyses yielded strong linear relationships for the transverse relaxation parameters ( R2' , R2 , and R2∗ ) against Δχ. CONCLUSION: Results suggest feasibility of the proposed method as a practical and reliable means for measuring R2' , R2 , R2∗ , and Δχ across the entire brain.

Free-Breathing Volumetric Liver R2* and Proton Density Fat Fraction Quantification in Pediatric Patients Using Stack-of-Radial MRI With Self-Gating Motion Compensation.

BACKGROUND: Stack-of-radial multiecho gradient-echo MRI is promising for free-breathing liver R2* quantification and may benefit children. PURPOSE: To validate stack-of-radial MRI with self-gating motion compensation in phantoms, and to evaluate it in children. STUDY TYPE: Prospective. PHANTOMS: Four vials with different R2* driven by a motion stage. SUBJECTS: Sixteen pediatric patients with suspected nonalcoholic fatty liver disease or steatohepatitis (five females, 13 ± 4 years, body mass index 29.2 ± 8.6 kg/m2 ). FIELD STRENGTH/SEQUENCES: Stack-of-radial, and 2D and 3D Cartesian multiecho gradient-echo sequences at 3T. ASSESSMENT: Ungated and gated stack-of-radial proton density fat fraction (PDFF) and R2* maps were reconstructed without and with self-gating motion compensation. Stack-of-radial R2* measurements of phantoms without and with motion were validated against reference 2D Cartesian results of phantoms without motion. In subjects, free-breathing stack-of-radial and reference breath-hold 3D Cartesian were acquired. Subject inclusion for statistical analysis and region of interest placement were determined independently by two observers. STATISTICAL TESTS: Phantom results were fitted with a weighted linear model. Demographic differences between excluded and included subjects were tested by multivariate analysis of variance. PDFF and R2* measurements were compared using Bland-Altman analysis. Interobserver agreement was assessed by the intraclass correlation coefficient (ICC). RESULTS: Ungated stack-of-radial R2* inside moving phantom vials showed a significant positive bias of 64.3 s-1 (P < 0.00001), unlike gated results (P > 0.31). Subject inclusion decisions for statistical analysis from two observers were consistent. No significant differences were found between four excluded and 12 included subjects (P = 0.14). Compared to breath-hold Cartesian, ungated and gated free-breathing stack-of-radial exhibited mean R2* differences of 18.5 s-1 and 3.6 s-1 . Mean PDFF differences were 1.1% and 1.0% for ungated and gated measurements, respectively. Interobserver agreement was excellent (ICC for PDFF = 0.99, ICC for R2* = 0.90; P < 0.0003). DATA CONCLUSION: Stack-of-radial MRI with self-gating motion compensation seems to allow free-breathing liver R2* and PDFF quantification in children. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 2.

Deep learning using a biophysical model for robust and accelerated reconstruction of quantitative, artifact-free and denoised R2* images.

PURPOSE: To introduce a novel deep learning method for Robust and Accelerated Reconstruction (RoAR) of quantitative and B0-inhomogeneity-corrected R2* maps from multi-gradient recalled echo (mGRE) MRI data. METHODS: RoAR trains a convolutional neural network (CNN) to generate quantitative R2∗ maps free from field inhomogeneity artifacts by adopting a self-supervised learning strategy given (a) mGRE magnitude images, (b) the biophysical model describing mGRE signal decay, and (c) preliminary-evaluated F-function accounting for contribution of macroscopic B0 field inhomogeneities. Importantly, no ground-truth R2* images are required and F-function is only needed during RoAR training but not application. RESULTS: We show that RoAR preserves all features of R2* maps while offering significant improvements over existing methods in computation speed (seconds vs. hours) and reduced sensitivity to noise. Even for data with SNR = 5 RoAR produced R2* maps with accuracy of 22% while voxel-wise analysis accuracy was 47%. For SNR = 10 the RoAR accuracy increased to 17% vs. 24% for direct voxel-wise analysis. CONCLUSIONS: RoAR is trained to recognize the macroscopic magnetic field inhomogeneities directly from the input magnitude-only mGRE data and eliminate their effect on R2∗ measurements. RoAR training is based on the biophysical model and does not require ground-truth R2* maps. Since RoAR utilizes signal information not just from individual voxels but also accounts for spatial patterns of the signals in the images, it reduces the sensitivity of R2* maps to the noise in the data. These features plus high computational speed provide significant benefits for the potential usage of RoAR in clinical settings.

Sensitivity of quantitative relaxometry and susceptibility mapping to microscopic iron distribution.

PURPOSE: Determine the impact of the microscopic spatial distribution of iron on relaxometry and susceptibility-based estimates of iron concentration. METHODS: Monte Carlo simulations and in vitro experiments of erythrocytes were used to create different microscopic distributions of iron. Measuring iron with intact erythrocyte cells created a heterogeneous distribution of iron, whereas lysing erythrocytes was used to create a homogeneous distribution of iron. Multi-echo spin echo and spoiled gradient echo acquisitions were then used to estimate relaxation parameters ( R2 and R2* ) and susceptibility. RESULTS: Simulations demonstrate that R2 and R2* measurements depend on the spatial distribution of iron even for the same iron concentration and volume susceptibility. Similarly, in vitro experiments demonstrate that R2 and R2* measurements depend on the microscopic spatial distribution of iron whereas the quantitative susceptibility mapping (QSM) susceptibility estimates reflect iron concentration without sensitivity to spatial distribution. CONCLUSIONS: R2 and R2* for iron quantification depend on the spatial distribution or iron. QSM-based estimation of iron concentration is insensitive to the microscopic spatial distribution of iron, potentially providing a distribution independent measure of iron concentration.

Biophysically motivated efficient estimation of the spatially isotropic R2* component from a single gradient-recalled echo measurement.

PURPOSE: To propose and validate an efficient method, based on a biophysically motivated signal model, for removing the orientation-dependent part of R2* using a single gradient-recalled echo (GRE) measurement. METHODS: The proposed method utilized a temporal second-order approximation of the hollow-cylinder-fiber model, in which the parameter describing the linear signal decay corresponded to the orientation-independent part of R2* . The estimated parameters were compared to the classical, mono-exponential decay model for R2* in a sample of an ex vivo human optic chiasm (OC). The OC was measured at 16 distinct orientations relative to the external magnetic field using GRE at 7T. To show that the proposed signal model can remove the orientation dependence of R2* , it was compared to the established phenomenological method for separating R2* into orientation-dependent and -independent parts. RESULTS: Using the phenomenological method on the classical signal model, the well-known separation of R2* into orientation-dependent and -independent parts was verified. For the proposed model, no significant orientation dependence in the linear signal decay parameter was observed. CONCLUSIONS: Since the proposed second-order model features orientation-dependent and -independent components at distinct temporal orders, it can be used to remove the orientation dependence of R2* using only a single GRE measurement.