Deep learning-assisted preclinical MR fingerprinting for sub-millimeter T1 and T2 mapping of entire macaque brain.

PURPOSE: Preclinical MR fingerprinting (MRF) suffers from long acquisition time for organ-level coverage due to demanding image resolution and limited undersampling capacity. This study aims to develop a deep learning-assisted fast MRF framework for sub-millimeter T1 and T2 mapping of entire macaque brain on a preclinical 9.4 T MR system. METHODS: Three dimensional MRF images were reconstructed by singular value decomposition (SVD) compressed reconstruction. T1 and T2 mapping for each axial slice exploited a self-attention assisted residual U-Net to suppress aliasing-induced quantification errors, and the transmit-field (B1 + ) measurements for robustness against B1 + inhomogeneity. Supervised network training used MRF images simulated via virtual parametric maps and a desired undersampling scheme. This strategy bypassed the difficulties of acquiring fully sampled preclinical MRF data to guide network training. The proposed fast MRF framework was tested on experimental data acquired from ex vivo and in vivo macaque brains. RESULTS: The trained network showed reasonable adaptability to experimental MRF images, enabling robust delineation of various T1 and T2 distributions in the brain tissues. Further, the proposed MRF framework outperformed several existing fast MRF methods in handling the aliasing artifacts and capturing detailed cerebral structures in the mapping results. Parametric mapping of entire macaque brain at nominal resolution of 0.35 × $$ \times $$ 0.35 × $$ \times $$ 1 mm3 can be realized via a 20-min 3D MRF scan, which was sixfold faster than the baseline protocol. CONCLUSION: Introducing deep learning to MRF framework paves the way for efficient organ-level high-resolution quantitative MRI in preclinical applications.

Elevated magnetic resonance imaging measures of adipose tissue deposition in women with breast cancer treatment-related lymphedema.

PURPOSE: Breast cancer treatment-related lymphedema (BCRL) is a common co-morbidity of breast cancer therapies, yet factors that contribute to BCRL progression remain incompletely characterized. We investigated whether magnetic resonance imaging (MRI) measures of subcutaneous adipose tissue were uniquely elevated in women with BCRL. METHODS: MRI at 3.0 T of upper extremity and torso anatomy, fat and muscle tissue composition, and T2 relaxometry were applied in left and right axillae of healthy control (n = 24) and symptomatic BCRL (n = 22) participants to test the primary hypothesis that fat-to-muscle volume fraction is elevated in symptomatic BCRL relative to healthy participants, and the secondary hypothesis that fat-to-muscle volume fraction is correlated with MR relaxometry of affected tissues and BCRL stage (significance criterion: two-sided p < 0.05). RESULTS: Fat-to-muscle volume fraction in healthy participants was symmetric in the right and left sides (p = 0.51); in BCRL participants matched for age, sex, and BMI, fat-to-muscle volume fraction was elevated on the affected side (fraction = 0.732 ± 0.184) versus right and left side in controls (fraction = 0.545 ± 0.221, p < 0.001). Fat-to-muscle volume fraction directly correlated with muscle T2 (p = 0.046) and increased with increasing level of BCRL stage (p = 0.041). CONCLUSION: Adiposity quantified by MRI is elevated in the affected upper extremity of women with BCRL and may provide a surrogate marker of condition onset or severity. CLINICAL TRIAL: NCT02611557.

Rapid MR relaxometry using deep learning: An overview of current techniques and emerging trends.

Quantitative mapping of MR tissue parameters such as the spin-lattice relaxation time (T1 ), the spin-spin relaxation time (T2 ), and the spin-lattice relaxation in the rotating frame (T1ρ ), referred to as MR relaxometry in general, has demonstrated improved assessment in a wide range of clinical applications. Compared with conventional contrast-weighted (eg T1 -, T2 -, or T1ρ -weighted) MRI, MR relaxometry provides increased sensitivity to pathologies and delivers important information that can be more specific to tissue composition and microenvironment. The rise of deep learning in the past several years has been revolutionizing many aspects of MRI research, including image reconstruction, image analysis, and disease diagnosis and prognosis. Although deep learning has also shown great potential for MR relaxometry and quantitative MRI in general, this research direction has been much less explored to date. The goal of this paper is to discuss the applications of deep learning for rapid MR relaxometry and to review emerging deep-learning-based techniques that can be applied to improve MR relaxometry in terms of imaging speed, image quality, and quantification robustness. The paper is comprised of an introduction and four more sections. Section 2 describes a summary of the imaging models of quantitative MR relaxometry. In Section 3, we review existing "classical" methods for accelerating MR relaxometry, including state-of-the-art spatiotemporal acceleration techniques, model-based reconstruction methods, and efficient parameter generation approaches. Section 4 then presents how deep learning can be used to improve MR relaxometry and how it is linked to conventional techniques. The final section concludes the review by discussing the promise and existing challenges of deep learning for rapid MR relaxometry and potential solutions to address these challenges.

Toward quantitative neuroimaging biomarkers for Friedreich's ataxia at 7 Tesla: Susceptibility mapping, diffusion imaging, R2 and R1 relaxometry.

Friedreich's ataxia (FRDA) is a rare genetic disorder leading to degenerative processes. So far, no effective treatment has been found. Therefore, it is important to assist the development of medication with imaging biomarkers reflecting disease status and progress. Ten FRDA patients (mean age 37 ± 14 years; four female) and 10 age- and sex-matched controls were included. Acquisition of magnetic resonance imaging (MRI) data for quantitative susceptibility mapping, R1 , R2 relaxometry and diffusion imaging was performed at 7 Tesla. Results of volume of interest (VOI)-based analyses of the quantitative data were compared with a voxel-based morphometry (VBM) evaluation. Differences between patients and controls were assessed using the analysis of covariance (ANCOVA; p < 0.01) with age and sex as covariates, effect size of group differences, and correlations with disease characteristics with Spearman correlation coefficient. For the VBM analysis, a statistical threshold of 0.001 for uncorrected and 0.05 for corrected p-values was used. Statistically significant differences between FRDA patients and controls were found in five out of twelve investigated structures, and statistically significant correlations with disease characteristics were revealed. Moreover, VBM revealed significant white matter atrophy within regions of the brainstem, and the cerebellum. These regions overlapped partially with brain regions for which significant differences between healthy controls and patients were found in the VOI-based quantitative MRI evaluation. It was shown that two independent analyses provided overlapping results. Moreover, positive results on correlations with disease characteristics were found, indicating that these quantitative MRI parameters could provide more detailed information and assist the search for effective treatments.

Reproducibility of liver iron concentration estimates in MRI through R2* measurement determined by least-squares curve fitting.

Measuring transverse relaxation rate (R2* = 1/T2*) via MRI allows for noninvasive evaluation of multiple clinical parameters, including liver iron concentration (LIC) and fat fraction. Both fat and iron contribute to diffuse liver disease when stored in excess in the liver. This liver damage leads to fibrosis and cirrhosis with an increased risk of developing hepatocellular carcinoma. Liver iron concentration is linearly related to R2* measurements using MRI. A phantom was constructed to assess R2* quantification variability on 1.5 and 3 T MRI systems. Quantification was executed using least-squares curve fitting techniques. The phantom was created using readily available, low-cost materials. It contains four vials with R2* values that cover a clinically relevant range (100 to 420 Hz at 1.5 T). Iron content was achieved using ferric chloride solutions contained in glass vials, each affixed in a three-dimensional (3D)-printed polylactide (PLA) structure, surrounded by distilled water, all housed in a sealed acrylic cylinder. Multiple phantom stands were also 3D-printed using PLA for precise orientation of the phantom with respect to the direction of the static magnetic field. Acquisitions at different phantom angles, across multiple MRI systems, and with different pulse sequence parameters were evaluated. The variability between any two R2* measurements, taken in the same vial under these various acquisition conditions, on a 1.5 T MRI system, was <7% for each of the four vials. For 3 T MRI systems, variability was less than 14% in all cases. Variability was <6% for both 1.5 and 3 T acquisitions when unchanged pulse sequence parameters were used. The phantom can be used to mimic a range of clinically relevant levels of R2* relaxation rates, as measured using MRI. These measurements were found to be reproducible relative to the gold-standard method, liver biopsy, across several different image acquisition conditions.

Measuring non-parametric distributions of intravoxel mean diffusivities using a clinical MRI scanner.

We measure spectra of water mobilities (i.e., mean diffusivities) from intravoxel pools in brain tissues of healthy subjects with a non-parametric approach. Using a single-shot isotropic diffusion encoding (IDE) preparation, we eliminate signal confounds caused by anisotropic diffusion, including microscopic anisotropy, and acquire in vivo diffusion-weighted images (DWIs) over a wide range of diffusion sensitizations. We analyze the measured IDE signal decays using a regularized inverse laplace transform (ILT) to derive a probability distribution of mean diffusivities of tissue water in each voxel. Based on numerical simulations we assess the sensitivity and accuracy of our ILT analysis and optimize an experimental protocol for use with clinical MRI scanners. In vivo spectra of intravoxel mean diffusivities measured in healthy subjects generally show single-peak distributions throughout the brain parenchyma, with small differences in peak location and shape among white matter, cortical and subcortical gray matter, and cerebrospinal fluid. Mean diffusivity distributions (MDDs) with multiple peaks are observed primarily in voxels at tissue interfaces and are likely due to partial volume contributions. To quantify tissue-specific MDDs with improved statistical power, we average voxel-wise normalized MDDs in corresponding regions-of-interest (ROIs). This non-parametric, rotation-invariant assessment of isotropic diffusivities of tissue water may reflect important microstructural information, such as cell packing and cell size, and active physiological processes, such as water transport and exchange, which may enhance biological specificity in the clinical diagnosis and characterization of ischemic stroke, cancer, neuroinflammation, and neurodegenerative disorders and diseases.

Mapping the human brainstem: Brain nuclei and fiber tracts at 3 T and 7 T.

Structural high-resolution imaging of the brainstem can be of high importance in clinical practice. However, ultra-high field magnetic resonance imaging (MRI) is still restricted in use due to limited availability. Therefore, quantitative MRI techniques (quantitative susceptibility mapping [QSM], relaxation measurements [ R2* , R1 ], diffusion tensor imaging [DTI]) and T2 - and proton density (PD)-weighted imaging in the human brainstem at 3 T and 7 T are compared. Five healthy volunteers (mean age: 21.5 ± 1.9 years) were measured at 3 T and 7 T using multi-echo gradient echo sequences for susceptibility mapping and R2* relaxometry, magnetization-prepared 2 rapid acquisition gradient echo sequences for R1 relaxometry, turbo-spin echo sequences for PD- and T2 -weighted imaging and readout-segmented echo planar sequences for DTI. Susceptibility maps were computed using Laplacian-based phase unwrapping, V-SHARP for background field removal and the streaking artifact reduction for QSM algorithm for dipole inversion. Contrast-to-noise ratios (CNRs) were determined at 3 T and 7 T in ten volumes of interest (VOIs). Data acquired at 7 T showed higher CNR. However, in four VOIs, lower CNR was observed for R2* at 7 T. QSM was shown to be the contrast with which the highest number of structures could be identified. The depiction of very fine tracts such as the medial longitudinal fasciculus throughout the brainstem was only possible in susceptibility maps acquired at 7 T. DTI effectively showed the main tracts (crus cerebri, transverse pontine fibers, corticospinal tract, middle and superior cerebellar peduncle, pontocerebellar tract, and pyramid) at both field strengths. Assessing the brainstem with quantitative MRI methods such as QSM, R2* , as well as PD- and T2 -weighted imaging with great detail, is also possible at 3 T, especially when using susceptibility mapping calculated from a gradient echo sequence with a wide range of echo times from 10.5 to 52.5 ms. However, tracing smallest structures strongly benefits from imaging at ultra-high field.

Specific visualization of neuromelanin-iron complex and ferric iron in the human post-mortem substantia nigra using MR relaxometry at 7T.

Neuromelanin (NM) is an endogenous iron chelating molecule of pigmented neurons in the human substantia nigra (SN). Along with the increase in iron deposition, the reduction in NM-containing dopaminergic neurons and the variation of iron load on NM are generally considered to be important factors participating to pathogenesis of Parkinson's disease (PD). The aim of this study was to non-invasively delineate the spatial distributions of paramagnetic magnetic susceptibility perturbers, such as NM-iron complex and ferric iron in SN. Multiple quantitative MR parameters of T1, T2, T2*, susceptibility weighted image (SWI), quantitative susceptibility map (QSM), and T1 weighted image with magnetization transfer (MT) effects were acquired for six post-mortem SN samples without a history of neurological disease. Co-registered quantitative histological validations were performed to identify and correlate NM pigments, iron deposits, and myelin distributions with respect to associated MR parameters. The regions with NM pigments and iron deposits showed positive magnetic susceptibility (paramagnetic) values, while myelinated areas showed negative magnetic susceptibility (diamagnetic) values from the QSM. The region of reduced T2 values in SN mostly coincided with high iron deposits, but not necessarily with the NM pigments. The correlations between T2*/T2 (or T2*/T22) values and NM pigments were higher than those between T2* values and NM pigments, due to the effective size differences between NM-iron complex and ferric iron. Consequently, separate segmentations of ferric iron from the T2 map and NM-iron complex from the T2*/T2 map (or T2*/T22 map) were possible with the boundary of the SN determined from the T1 weighted image.

Improved liver R2* mapping by pixel-wise curve fitting with adaptive neighborhood regularization.

PURPOSE: To improve liver R2* mapping by incorporating adaptive neighborhood regularization into pixel-wise curve fitting. METHODS: Magnetic resonance imaging R2* mapping remains challenging because of the serial images with low signal-to-noise ratio. In this study, we proposed to exploit the neighboring pixels as regularization terms and adaptively determine the regularization parameters according to the interpixel signal similarity. The proposed algorithm, called the pixel-wise curve fitting with adaptive neighborhood regularization (PCANR), was compared with the conventional nonlinear least squares (NLS) and nonlocal means filter-based NLS algorithms on simulated, phantom, and in vivo data. RESULTS: Visually, the PCANR algorithm generates R2* maps with significantly reduced noise and well-preserved tiny structures. Quantitatively, the PCANR algorithm produces R2* maps with lower root mean square errors at varying R2* values and signal-to-noise-ratio levels compared with the NLS and nonlocal means filter-based NLS algorithms. For the high R2* values under low signal-to-noise-ratio levels, the PCANR algorithm outperforms the NLS and nonlocal means filter-based NLS algorithms in the accuracy and precision, in terms of mean and standard deviation of R2* measurements in selected region of interests, respectively. CONCLUSIONS: The PCANR algorithm can reduce the effect of noise on liver R2* mapping, and the improved measurement precision will benefit the assessment of hepatic iron in clinical practice. Magn Reson Med 80:792-801, 2018. © 2018 International Society for Magnetic Resonance in Medicine.

A diffusion model-free framework with echo time dependence for free-water elimination and brain tissue microstructure characterization.

PURPOSE: The compartmental nature of brain tissue microstructure is typically studied by diffusion MRI, MR relaxometry or their correlation. Diffusion MRI relies on signal representations or biophysical models, while MR relaxometry and correlation studies are based on regularized inverse Laplace transforms (ILTs). Here we introduce a general framework for characterizing microstructure that does not depend on diffusion modeling and replaces ill-posed ILTs with blind source separation (BSS). This framework yields proton density, relaxation times, volume fractions, and signal disentanglement, allowing for separation of the free-water component. THEORY AND METHODS: Diffusion experiments repeated for several different echo times, contain entangled diffusion and relaxation compartmental information. These can be disentangled by BSS using a physically constrained nonnegative matrix factorization. RESULTS: Computer simulations, phantom studies, together with repeatability and reproducibility experiments demonstrated that BSS is capable of estimating proton density, compartmental volume fractions and transversal relaxations. In vivo results proved its potential to correct for free-water contamination and to estimate tissue parameters. CONCLUSION: Formulation of the diffusion-relaxation dependence as a BSS problem introduces a new framework for studying microstructure compartmentalization, and a novel tool for free-water elimination.

Quantitative MR Evaluation of Chronic Pancreatitis: Extracellular Volume Fraction and MR Relaxometry.

OBJECTIVE: The purpose of this study was to determine if extracellular volume fraction and T1 mapping can be used to diagnose chronic pancreatitis (CP). MATERIALS AND METHODS: This HIPAA-compliant study analyzed 143 consecutive patients with and without CP who underwent MR imaging between May 2016 and February 2017. Patients were selected for the study according to inclusion and exclusion criteria that considered history and clinical and laboratory findings. Eligible patients (n = 119) were grouped as normal (n = 60) or with mild (n = 22), moderate (n = 27), or severe (n = 10) CP on the basis of MRCP findings using the Cambridge classification as the reference standard. T1 maps were acquired in unenhanced and late contrast-enhanced phases using a 3D dual flip-angle gradient-echo sequence. All patients were imaged on the same 3-T scanner using the same imaging parameters, contrast agent, and dosage. RESULTS: Mean extracellular volume fractions and T1 relaxation times were significantly different within the study groups (one-way ANOVA, p < 0.001). Using the AUC curve analysis, extracellular volume fraction of > 0.27 showed 92% sensitivity (54/59) and 77% specificity (46/60) for the diagnosis of CP (AUC = 0.90). A T1 relaxation time of > 950 ms revealed 64% sensitivity (38/59) and 88% specificity (53/60) (AUC = 0.80). Combining extracellular volume fraction and T1 mapping yielded sensitivity of 85% (50/59) and specificity of 92% (55/60) (AUC = 0.94). CONCLUSION: Extracellular volume fraction and T1 mapping may provide quantitative metrics for determining the presence and severity of acinar cell loss and aid in the diagnosis of CP.

Myocardial T1 mapping at 3.0 tesla using an inversion recovery spoiled gradient echo readout and bloch equation simulation with slice profile correction (BLESSPC) T1 estimation algorithm.

BACKGROUND: To develop an accurate and precise myocardial T1 mapping technique using an inversion recovery spoiled gradient echo readout at 3.0 Tesla (T). THEORY AND METHODS: The modified Look-Locker inversion-recovery (MOLLI) sequence was modified to use fast low angle shot (FLASH) readout, incorporating a BLESSPC (Bloch Equation Simulation with Slice Profile Correction) T1 estimation algorithm, for accurate myocardial T1 mapping. The FLASH-MOLLI with BLESSPC fitting was compared with different approaches and sequences with regards to T1 estimation accuracy, precision and image artifact based on simulation, phantom studies, and in vivo studies of 10 healthy volunteers and three patients at 3.0 Tesla. RESULTS: The FLASH-MOLLI with BLESSPC fitting yields accurate T1 estimation (average error = -5.4 ± 15.1 ms, percentage error = -0.5% ± 1.2%) for T1 from 236-1852 ms and heart rate from 40-100 bpm in phantom studies. The FLASH-MOLLI sequence prevented off-resonance artifacts in all 10 healthy volunteers at 3.0T. In vivo, there was no significant difference between FLASH-MOLLI-derived myocardial T1 values and "ShMOLLI+IE" derived values (1458.9 ± 20.9 ms versus 1464.1 ± 6.8 ms, P = 0.50); However, the average precision by FLASH-MOLLI was significantly better than that generated by "ShMOLLI+IE" (1.84 ± 0.36% variance versus 3.57 ± 0.94%, P < 0.001). CONCLUSION: The FLASH-MOLLI with BLESSPC fitting yields accurate and precise T1 estimation, and eliminates banding artifacts associated with bSSFP at 3.0T.

Rapid look-up table method for noise-corrected curve fitting in the R2* mapping of iron loaded liver.

PURPOSE: Fitting the measured decay signal to the first moment in the presence of noncentral chi noise (M(1) NCM) can correctly address the effect of noise on the effective transverse relaxation rate (R2*) relaxometry of iron loaded liver. However, this method requires intensive computation, which restricts its application to R2* mapping. This work aims to develop a rapid implementation of the M(1) NCM method for R2* mapping. METHODS: The computation of the confluent hypergeometric function in the M(1) NCM model was approximated using cubic spline interpolation with breakpoints and coefficients precalculated and stored in a look-up table (M(1) NCM-LUT). The performance of the proposed M(1) NCM-LUT method was evaluated through simulation and based on in vivo liver R2* relaxometry data. RESULTS: In both simulation and in vivo studies, the maximum absolute difference between R2* maps generated by the M(1) NCM and M(1) NCM-LUT methods was nearly 10(-3) s(-1) or less, and the M(1) NCM-LUT method obtained a R2* map in approximately 1 s and achieved an acceleration of approximately five orders of magnitude. CONCLUSION: The proposed M(1) NCM-LUT method can significantly increase the speed of the liver R2* mapping using the M(1) NCM model. This development is important in promoting application of this R2* mapping technique for tissue iron quantification.

Instantaneous signal loss simulation (InSiL): an improved algorithm for myocardial T1 mapping using the MOLLI sequence.

PURPOSE: To propose a T1 mapping algorithm for the modified Look-Locker inversion-recovery (MOLLI) sequence that can improve T1 estimation accuracy. MATERIALS AND METHODS: The modified T1 mapping algorithm (InSiL) is based on the simulation of MOLLI signal evolution and simulates the longitudinal magnetization signal perturbation by each single-shot image acquisition in MOLLI as an instantaneous signal loss. InSiL was evaluated against original MOLLI using Bloch simulations, phantom studies, and in 15 healthy volunteers at 1.5T. RESULTS: In phantom studies, the maximum absolute error by InSiL is less than 2%, while that by MOLLI is more than 20% for T1 values from 221 msec to 1539 msec. The benefit of InSiL is greatest at heart rate (HR) >80 bpm and T1 >1000 msec, and InSiL reduced MOLLI T1 error from 14.9 ± 4.5% to 0.4 ± 0.3%. Average InSiL-derived native myocardial T1 values at 1.5T in healthy volunteers were significantly higher than MOLLI-derived values by 236.9 ± 11.7 msec (1160.3 ± 25.1 msec vs. 923.4 ± 22.3 msec, P < 0.001) at an average HR of 65.1 ± 14.7 bpm. CONCLUSION: The proposed InSiL approach yields better T1 mapping accuracy than MOLLI, and is less sensitive to HR variation in tissues with longer T1 values.

Quantification of liver iron with MRI: state of the art and remaining challenges.

Liver iron overload is the histological hallmark of hereditary hemochromatosis and transfusional hemosiderosis, and can also occur in chronic hepatopathies. Iron overload can result in liver damage, with the eventual development of cirrhosis, liver failure, and hepatocellular carcinoma. Assessment of liver iron levels is necessary for detection and quantitative staging of iron overload and monitoring of iron-reducing treatments. This article discusses the need for noninvasive assessment of liver iron and reviews qualitative and quantitative methods with a particular emphasis on magnetic resonance imaging (MRI). Specific MRI methods for liver iron quantification include signal intensity ratio as well as R2 and R2* relaxometry techniques. Methods that are in clinical use, as well as their limitations, are described. Remaining challenges, unsolved problems, and emerging techniques to provide improved characterization of liver iron deposition are discussed.

Myocardial T1 mapping using an instantaneous signal loss simulation modeling and a Bayesian estimation method: A robust T1 extraction method free of tuning parameters.

The Instantaneous Signal Loss Simulation (InSiL) model is a promising alternative to the classical mono-exponential fitting of the Modified Look-Locker Inversion-recovery (MOLLI) sequence in cardiac T1 mapping applications, which achieves better accuracy and is less sensitive to heart rate (HR) variations. Classical non-linear least squares (NLLS) estimation methods require some parameters of the model to be fixed a priori in order to give reliable T1 estimations and avoid outliers. This introduces further bias in the estimation, reducing the advantages provided by the InSiL model. In this paper, a novel Bayesian estimation method using a hierarchical model is proposed to fit the parameters of the InSiL model. The hierarchical Bayesian modeling has a shrinkage effect that works as a regularizer for the estimated values, by pulling spurious estimated values toward the group-mean, hence reducing greatly the number of outliers. Simulations, physical phantoms, and in-vivo human cardiac data have been used to show that this approach estimates accurately all the InSiL parameters, and achieve high precision estimation of the T1 compared to the classical MOLLI model and NLLS InSiL estimation.

MR Relaxometry for Discriminating Malignant Ovarian Cystic Tumors: A Prospective Multicenter Cohort Study.

BACKGROUND: Endometriosis-associated ovarian cancer (EAOC) is a well-known type of cancer that arises from ovarian endometrioma (OE). OE contains iron-rich fluid in its cysts due to repeated hemorrhages in the ovaries. However, distinguishing between benign and malignant tumors can be challenging. We conducted a retrospective study on magnetic resonance (MR) relaxometry of cyst fluid to distinguish EAOC from OE and reported that this method showed good accuracy. The purpose of this study is to evaluate the accuracy of a non-invasive method in re-evaluating pre-surgical diagnosis of malignancy by a prospective multicenter cohort study. METHODS: After the standard diagnosis process, the R2 values were obtained using a 3T system. Data on the patients were then collected through the Case Report Form (CRF). Between December 2018 and March 2023, six hospitals enrolled 109 patients. Out of these, 81 patients met the criteria required for the study. RESULTS: The R2 values calculated using MR relaxometry showed good discriminating ability with a cut-off of 15.74 (sensitivity 80.6%, specificity 75.0%, AUC = 0.750, p < 0.001) when considering atypical or borderline tumors as EAOC. When atypical and borderline cases were grouped as OE, EAOC could be distinguished with a cut-off of 16.87 (sensitivity 87.0%, specificity 61.1%). CONCLUSIONS: MR relaxometry has proven to be an effective tool for discriminating EAOC from OE. Regular use of this method is expected to provide significant insights for clinical practice.

Improved MRI R2 * relaxometry of iron-loaded liver with noise correction.

Accurate and reproducible MRI R2 * relaxometry for tissue iron quantification is important in managing transfusion-dependent patients. MRI data are often acquired using array coils and reconstructed by the root-sum-square algorithm, and as such, measured signals follow the noncentral chi distribution. In this study, two noise-corrected models were proposed for the liver R2 * quantification: fitting the signal to the first moment and fitting the squared signal to the second moment in the presence of the noncentral chi noise. These two models were compared with the widely implemented offset and truncation models on both simulation and in vivo data. The results demonstrated that the "slow decay component" of the liver R2 * was mainly caused by the noise. The offset model considerably overestimated R2 * values by incorrectly adding a constant to account for the slow decay component. The truncation model generally produced accurate R2 * measurements by only fitting the initial data well above the noise level to remove the major source of errors, but underestimated very high R2 * values due to the sequence limit of obtaining very short echo time images. Both the first and second-moment noise-corrected models constantly produced accurate and precise R2 * measurements by correctly addressing the noise problem.

Cascade of Denoising and Mapping Neural Networks for MRI R2* Relaxometry of Iron-Loaded Liver.

MRI of effective transverse relaxation rate (R2*) measurement is a reliable method for liver iron concentration quantification. However, R2* mapping can be degraded by noise, especially in the case of iron overload. This study aimed to develop a deep learning method for MRI R2* relaxometry of an iron-loaded liver using a two-stage cascaded neural network. The proposed method, named CadamNet, combines two convolutional neural networks separately designed for image denoising and parameter mapping into a cascade framework, and the physics-based R2* decay model was incorporated in training the mapping network to enforce data consistency further. CadamNet was trained using simulated liver data with Rician noise, which was constructed from clinical liver data. The performance of CadamNet was quantitatively evaluated on simulated data with varying noise levels as well as clinical liver data and compared with the single-stage parameter mapping network (MappingNet) and two conventional model-based R2* mapping methods. CadamNet consistently achieved high-quality R2* maps and outperformed MappingNet at varying noise levels. Compared with conventional R2* mapping methods, CadamNet yielded R2* maps with lower errors, higher quality, and substantially increased efficiency. In conclusion, the proposed CadamNet enables accurate and efficient iron-loaded liver R2* mapping, especially in the presence of severe noise.

A systematic review demonstrating correlation of MRI compositional parameters with clinical outcomes following articular cartilage repair interventions in the knee.

Objective: Compositional-MRI parameters enable the assessment of cartilage ultrastructure. Correlation of these parameters with clinical outcomes is unclear. This systematic review investigated the correlation of various compositional- MRI parameters with clinical outcome measures following cartilage repair or regeneration interventions in the knee. Design: This study was registered with PROSPERO and reported in accordance with PRISMA. PubMed, Institute of Science Index, Scopus, Cochrane Central Register of Controlled Trials, and Embase databases were searched. All studies, regardless of type, that presented correlation of compositional- MRI parameters with clinical outcome measures were included. Two researchers independently performed data extraction and QUADAS-2 analysis. Compositional-MRI parameter change following intervention and correlation with clinical outcome measures were evaluated. Results: 19 studies were included. Risk of bias was generally low. 5 different compositional parameters were observed from the included studies. However, due to the significant variability in the reporting of compositional-MRI parameters across studies, meta-analyses were possible only for T2 values and T2 index values (T2 value of repair cartilage relative to normal cartilage). Correlation of T2 values of repair cartilage with clinical outcome score was r â€‹= â€‹0.33 [0.15, 0.52]. Correlation of T2 index with clinical outcome score was r â€‹= â€‹0.52 [0.32, 0.77]. Conclusions: Correlation between T2 values and clinical outcome scores following knee cartilage repair were found. The heterogeneity of the correlations extracted from the included studies limited the scope for the meta-analysis. Thus, standardised, high-quality studies are required for better assessment of correlation between compositional MRI parameters and clinical outcome measures after cartilage repair. Registration number: PROSPERO CRD42021287364.Study protocol available on PROSPERO website.