Fast Compressed Sensing of 3D Radial T1 Mapping with Different Sparse and Low-Rank Models.

Knowledge of the relative performance of the well-known sparse and low-rank compressed sensing models with 3D radial quantitative magnetic resonance imaging acquisitions is limited. We use 3D radial T1 relaxation time mapping data to compare the total variation, low-rank, and Huber penalty function approaches to regularization to provide insights into the relative performance of these image reconstruction models. Simulation and ex vivo specimen data were used to determine the best compressed sensing model as measured by normalized root mean squared error and structural similarity index. The large-scale compressed sensing models were solved by combining a GPU implementation of a preconditioned primal-dual proximal splitting algorithm to provide high-quality T1 maps within a feasible computation time. The model combining spatial total variation and locally low-rank regularization yielded the best performance, followed closely by the model combining spatial and contrast dimension total variation. Computation times ranged from 2 to 113 min, with the low-rank approaches taking the most time. The differences between the compressed sensing models are not necessarily large, but the overall performance is heavily dependent on the imaged object.

3D T1 relaxation time measurements in an equine model of subtle post-traumatic osteoarthritis using MB-SWIFT.

The aim of this study is to assess whether articular cartilage changes in an equine model of post-traumatic osteoarthritis (PTOA), induced by surgical creation of standard (blunt) grooves, and very subtle sharp grooves, could be detected with ex vivo T1 relaxation time mapping utilizing three-dimensional (3D) readout sequence with zero echo time. Grooves were made on the articular surfaces of the middle carpal and radiocarpal joints of nine mature Shetland ponies and osteochondral samples were harvested at 39 weeks after being euthanized under respective ethical permissions. T1 relaxation times of the samples (n = 8 + 8 for experimental and n = 12 for contralateral controls) were measured with a variable flip angle 3D multiband-sweep imaging with Fourier transform sequence. Equilibrium and instantaneous Young's moduli and proteoglycan (PG) content from OD of Safranin-O-stained histological sections were measured and utilized as reference parameters for the T1 relaxation times. T1 relaxation time was significantly (p < 0.05) increased in both groove areas, particularly in the blunt grooves, compared with control samples, with the largest changes observed in the superficial half of the cartilage. T1 relaxation times correlated weakly (Rs ≈ 0.33) with equilibrium modulus and PG content (Rs ≈ 0.21). T1 relaxation time in the superficial articular cartilage is sensitive to changes induced by the blunt grooves but not to the much subtler sharp grooves, at the 39-week timepoint post-injury. These findings support that T1 relaxation time has potential in detection of mild PTOA, albeit the most subtle changes could not be detected.

Machine Learning Prediction of Collagen Fiber Orientation and Proteoglycan Content From Multiparametric Quantitative MRI in Articular Cartilage.

BACKGROUND: Machine learning models trained with multiparametric quantitative MRIs (qMRIs) have the potential to provide valuable information about the structural composition of articular cartilage. PURPOSE: To study the performance and feasibility of machine learning models combined with qMRIs for noninvasive assessment of collagen fiber orientation and proteoglycan content. STUDY TYPE: Retrospective, animal model. ANIMAL MODEL: An open-source single slice MRI dataset obtained from 20 samples of 10 Shetland ponies (seven with surgically induced cartilage lesions followed by treatment and three healthy controls) yielded to 1600 data points, including 10% for test and 90% for train validation. FIELD STRENGTH/SEQUENCE: A 9.4 T MRI scanner/qMRI sequences: T1 , T2 , adiabatic T1ρ and T2ρ , continuous-wave T1ρ and relaxation along a fictitious field (TRAFF ) maps. ASSESSMENT: Five machine learning regression models were developed: random forest (RF), support vector regression (SVR), gradient boosting (GB), multilayer perceptron (MLP), and Gaussian process regression (GPR). A nested cross-validation was used for performance evaluation. For reference, proteoglycan content and collagen fiber orientation were determined by quantitative histology from digital densitometry (DD) and polarized light microscopy (PLM), respectively. STATISTICAL TESTS: Normality was tested using Shapiro-Wilk test, and association between predicted and measured values was evaluated using Spearman's Rho test. A P-value of 0.05 was considered as the limit of statistical significance. RESULTS: Four out of the five models (RF, GB, MLP, and GPR) yielded high accuracy (R2  = 0.68-0.75 for PLM and 0.62-0.66 for DD), and strong significant correlations between the reference measurements and predicted cartilage matrix properties (Spearman's Rho = 0.72-0.88 for PLM and 0.61-0.83 for DD). GPR algorithm had the highest accuracy (R2  = 0.75 and 0.66) and lowest prediction-error (root mean squared [RMSE] = 1.34 and 2.55) for PLM and DD, respectively. DATA CONCLUSION: Multiparametric qMRIs in combination with regression models can determine cartilage compositional and structural features, with higher accuracy for collagen fiber orientation than proteoglycan content. EVIDENCE LEVEL: 2 TECHNICAL EFFICACY: Stage 2.

New methods for robust continuous wave T1ρ relaxation preparation.

Measurement of the longitudinal relaxation time in the rotating frame of reference (T1ρ ) is sensitive to the fidelity of the main imaging magnetic field (B0 ) and that of the RF pulse (B1 ). The purpose of this study was to introduce methods for producing continuous wave (CW) T1ρ contrast with improved robustness against field inhomogeneities and to compare the sensitivities of several existing and the novel T1ρ contrast generation methods with the B0 and B1 field inhomogeneities. Four hard-pulse and four adiabatic CW-T1ρ magnetization preparations were investigated. Bloch simulations and experimental measurements at different spin-lock amplitudes under ideal and non-ideal conditions, as well as theoretical analysis of the hard-pulse preparations, were conducted to assess the sensitivity of the methods to field inhomogeneities, at low (ω1 << ΔB0 ) and high (ω1 >> ΔB0 ) spin-locking field strengths. In simulations, previously reported single-refocus and new triple-refocus hard-pulse and double-refocus adiabatic preparation schemes were found to be the most robust. The mean normalized absolute deviation between the experimentally measured relaxation times under ideal and non-ideal conditions was found to be smallest for the refocused preparation schemes and broadly in agreement with the sensitivities observed in simulations. Experimentally, all refocused preparations performed better than those that were non-refocused. The findings promote the use of the previously reported hard-pulse single-refocus ΔB0 and B1 insensitive T1ρ as a robust method with minimal RF energy deposition. The double-refocus adiabatic B1 insensitive rotation-4 CW-T1ρ preparation offers further improved insensitivity to field variations, but because of the extra RF deposition, may be preferred for ex vivo applications.

Deep-Learning-Based Contrast Synthesis From MRF Parameter Maps in the Knee Joint.

BACKGROUND: Magnetic resonance fingerprinting (MRF) is a method to speed up acquisition of quantitative MRI data. However, MRF does not usually produce contrast-weighted images that are required by radiologists, limiting reachable total scan time improvement. Contrast synthesis from MRF could significantly decrease the imaging time. PURPOSE: To improve clinical utility of MRF by synthesizing contrast-weighted MR images from the quantitative data provided by MRF, using U-nets that were trained for the synthesis task utilizing L1- and perceptual loss functions, and their combinations. STUDY TYPE: Retrospective. POPULATION: Knee joint MRI data from 184 subjects from Northern Finland 1986 Birth Cohort (ages 33-35, gender distribution not available). FIELD STRENGTH AND SEQUENCE: A 3 T, multislice-MRF, proton density (PD)-weighted 3D-SPACE (sampling perfection with application optimized contrasts using different flip angle evolution), fat-saturated T2-weighted 3D-space, water-excited double echo steady state (DESS). ASSESSMENT: Data were divided into training, validation, test, and radiologist's assessment sets in the following way: 136 subjects to training, 3 for validation, 3 for testing, and 42 for radiologist's assessment. The synthetic and target images were evaluated using 5-point Likert scale by two musculoskeletal radiologists blinded and with quantitative error metrics. STATISTICAL TESTS: Friedman's test accompanied with post hoc Wilcoxon signed-rank test and intraclass correlation coefficient. The statistical cutoff P <0.05 adjusted by Bonferroni correction as necessary was utilized. RESULTS: The networks trained in the study could synthesize conventional images with high image quality (Likert scores 3-4 on a 5-point scale). Qualitatively, the best synthetic images were produced with combination of L1- and perceptual loss functions and perceptual loss alone, while L1-loss alone led to significantly poorer image quality (Likert scores below 3). The interreader and intrareader agreement were high (0.80 and 0.92, respectively) and significant. However, quantitative image quality metrics indicated best performance for the pure L1-loss. DATA CONCLUSION: Synthesizing high-quality contrast-weighted images from MRF data using deep learning is feasible. However, more studies are needed to validate the diagnostic accuracy of these synthetic images. EVIDENCE LEVEL: 4. TECHNICAL EFFICACY: Stage 1.

Dual-contrast computed tomography enables detection of equine posttraumatic osteoarthritis in vitro.

To prevent the progression of posttraumatic osteoarthritis, assessment of cartilage composition is critical for effective treatment planning. Posttraumatic changes include proteoglycan (PG) loss and elevated water content. Quantitative dual-energy computed tomography (QDECT) provides a means to diagnose these changes. Here, we determine the potential of QDECT to evaluate tissue quality surrounding cartilage lesions in an equine model, hypothesizing that QDECT allows detection of posttraumatic degeneration by providing quantitative information on PG and water contents based on the partitions of cationic and nonionic agents in a contrast mixture. Posttraumatic osteoarthritic samples were obtained from a cartilage repair study in which full-thickness chondral defects were created surgically in both stifles of seven Shetland ponies. Control samples were collected from three nonoperated ponies. The experimental (n = 14) and control samples (n = 6) were immersed in the contrast agent mixture and the distributions of the agents were determined at various diffusion time points. As a reference, equilibrium moduli, dynamic moduli, and PG content were measured. Significant differences (p < 0.05) in partitions between the experimental and control samples were demonstrated with cationic contrast agent at 30 min, 60 min, and 20 h, and with non-ionic agent at 60 and 120 min. Significant Spearman's rank correlations were obtained at 20 and 24 h (ρ = 0.482-0.693) between the partition of cationic contrast agent, cartilage biomechanical properties, and PG content. QDECT enables evaluation of posttraumatic changes surrounding a lesion and quantification of PG content, thus advancing the diagnostics of the extent and severity of cartilage injuries.

Epiphyseal cartilage canal architecture and extracellular matrix remodeling in mucopolysaccharidosis VII dogs at the onset of postnatal growth.

Purpose: Mucopolysaccharidosis (MPS) VII is a genetic, lysosomal storage disease characterized by abnormal accumulation of glycosaminoglycans in cells and tissues. MPS VII patients exhibit multiple failures of endochondral ossification during postnatal growth, including markedly delayed cartilage-to-bone conversion in the vertebrae and long bones. Cartilage canals provide the template for vascularization at the onset of secondary ossification. The objective of this study was to investigate whether abnormal cartilage canal architecture and enzyme-mediated extracellular matrix (ECM) remodeling contribute to delayed cartilage-to-bone conversion in MPS VII.Materials and Methods: The epiphyseal cartilage canal networks of 9-day-old healthy control and MPS VII-affected dog vertebrae were characterized using high-resolution, contrast-free quantitative susceptibility mapping magnetic resonance imaging. Relative expression levels of matrix metalloproteinases (MMPs) 9, 13 and 14 were examined using immunohistochemistry, while tartrate-resistant acid phosphatase (TRAP) and alkaline phosphatase (ALP) were examined using in situ enzyme staining.Results: Interestingly, the density, number, connectivity and thickness of cartilage canals was not significantly different between MPS VII and control vertebrae. Immunohistochemistry revealed diminished MMP-9, but normal MMP-13 and 14 expression by epiphyseal cartilage chondrocytes, while ALP and TRAP enzyme expression by chondrocytes and chondroclasts, respectively, were both diminished in MPS VII.Conclusions: Our findings suggest that while the epiphyseal cartilage canal network in MPS VII is normal at the onset of secondary ossification, expression of enzymes required for cartilage resorption and replacement with mineralized ECM, and initiation of angiogenesis, is impaired.

Quantitative susceptibility mapping of the rat brain after traumatic brain injury.

The primary lesion arising from the initial insult after traumatic brain injury (TBI) triggers a cascade of secondary tissue damage, which may also progress to connected brain areas in the chronic phase. The aim of this study was, therefore, to investigate variations in the susceptibility distribution related to these secondary tissue changes in a rat model after severe lateral fluid percussion injury. We compared quantitative susceptibility mapping (QSM) and R2 * measurements with histological analyses in white and grey matter areas outside the primary lesion but connected to the lesion site. We demonstrate that susceptibility variations in white and grey matter areas could be attributed to reduction in myelin, accumulation of iron and calcium, and gliosis. QSM showed quantitative changes attributed to secondary damage in areas located rostral to the lesion site that appeared normal in R2 * maps. However, combination of QSM and R2 * was informative in disentangling the underlying tissue changes such as iron accumulation, demyelination, or calcifications. Therefore, combining QSM with R2 * measurement can provide a more detailed assessment of tissue changes and may pave the way for improved diagnosis of TBI, and several other complex neurodegenerative diseases.

Orientation anisotropy of quantitative MRI parameters in degenerated human articular cartilage.

Quantitative magnetic resonance (MR) relaxation parameters demonstrate varying sensitivity to the orientation of the ordered tissues in the magnetic field. In this study, the orientation dependence of multiple relaxation parameters was assessed in cadaveric human cartilage with varying degree of natural degeneration, and compared with biomechanical testing, histological scoring, and quantitative histology. Twelve patellar cartilage samples were imaged at 9.4 T MRI with multiple relaxation parameters, including T1 , T2 , CW - T1ρ , and adiabatic T1ρ , at three different orientations with respect to the main magnetic field. Anisotropy of the relaxation parameters was quantified, and the results were compared with the reference measurements and between samples of different histological Osteoarthritis Research Society International (OARSI) grades. T2 and CW - T1ρ at 400 Hz spin-lock demonstrated the clearest anisotropy patterns. Radial zone anisotropy for T2 was significantly higher for samples with OARSI grade 2 than for grade 4. The proteoglycan content (measured as optical density) correlated with the radial zone MRI orientation anisotropy for T2 (r = 0.818) and CW - T1ρ with 400 Hz spin-lock (r = 0.650). Orientation anisotropy of MRI parameters altered with progressing cartilage degeneration. This is associated with differences in the integrity of the collagen fiber network, but it also seems to be related to the proteoglycan content of the cartilage. Samples with advanced OA had great variation in all biomechanical and histological properties and exhibited more variation in MRI orientation anisotropy than the less degenerated samples. Understanding the background of relaxation anisotropy on a molecular level would help to develop new MRI contrasts and improve the application of previously established quantitative relaxation contrasts.

Evaluation of articular cartilage with quantitative MRI in an equine model of post-traumatic osteoarthritis.

Chondral lesions lead to degenerative changes in the surrounding cartilage tissue, increasing the risk of developing post-traumatic osteoarthritis (PTOA). This study aimed to investigate the feasibility of quantitative magnetic resonance imaging (qMRI) for evaluation of articular cartilage in PTOA. Articular explants containing surgically induced and repaired chondral lesions were obtained from the stifle joints of seven Shetland ponies (14 samples). Three age-matched nonoperated ponies served as controls (six samples). The samples were imaged at 9.4 T. The measured qMRI parameters included T1 , T2 , continuous-wave T1ρ (CWT1ρ ), adiabatic T1ρ (AdT1ρ ), and T2ρ (AdT2ρ ) and relaxation along a fictitious field (TRAFF ). For reference, cartilage equilibrium and dynamic moduli, proteoglycan content and collagen fiber orientation were determined. Mean values and profiles from full-thickness cartilage regions of interest, at increasing distances from the lesions, were used to compare experimental against control and to correlate qMRI with the references. Significant alterations were detected by qMRI parameters, including prolonged T1 , CWT1ρ , and AdT1ρ in the regions adjacent to the lesions. The changes were confirmed by the reference methods. CWT1ρ was more strongly associated with the reference measurements and prolonged in the affected regions at lower spin-locking amplitudes. Moderate to strong correlations were found between all qMRI parameters and the reference parameters (ρ = -0.531 to -0.757). T1 , low spin-lock amplitude CWT1ρ , and AdT1ρ were most responsive to changes in visually intact cartilage adjacent to the lesions. In the context of PTOA, these findings highlight the potential of T1 , CWT1ρ , and AdT1ρ in evaluation of compositional and structural changes in cartilage.

Bright ultrashort echo time SWIFT MRI signal at the osteochondral junction is not located in the calcified cartilage.

In this study, we aimed to precisely localize the hyperintense signal that is generated at the osteochondral junction when using ultrashort echo time magnetic resonance imaging (MRI) and to investigate the osteochondral junction using sweep imaging with Fourier transformation (SWIFT) MRI. Furthermore, we seek to evaluate what compositional properties of the osteochondral junction are the sources of this signal. In the study, we obtained eight samples from a tibial plateau dissected from a 68-year-old male donor, and one additional osteochondral sample of bovine origin. The samples were imaged using high-resolution ultrashort echo time SWIFT MRI and microcomputed tomography (µCT) scans. Localization of the bright signal in the osteochondral junction was performed using coregistered data sets. Potential sources of the signal feature were examined by imaging the bovine specimen with variable receiver bandwidths and by performing variable flip angle T1 relaxation time mapping. The results of the study showed that the hyperintense signal was found to be located entirely in the deep noncalcified articular cartilage. The intensity of this signal at the interface varied between the specimens. Further tests with bovine specimens indicated that the imaging bandwidth and T1 relaxation affect the properties of the signal. Based on the present results, the calcified cartilage has low signal intensity even in SWIFT imaging. Concomitantly, it appears that the bright signal seen in ultrashort echo time imaging resides within the noncalcified cartilage. Furthermore, the most likely sources of this signal are the rapid T1 relaxation of the deep cartilage and the susceptibility-induced effects arising from the calcified tissues.

Arthroscopic Determination of Cartilage Proteoglycan Content and Collagen Network Structure with Near-Infrared Spectroscopy.

Conventional arthroscopic evaluation of articular cartilage is subjective and insufficient for assessing early compositional and structural changes during the progression of post-traumatic osteoarthritis. Therefore, in this study, arthroscopic near-infrared (NIR) spectroscopy is introduced, for the first time, for in vivo evaluation of articular cartilage thickness, proteoglycan (PG) content, and collagen orientation angle. NIR spectra were acquired in vivo and in vitro from equine cartilage adjacent to experimental cartilage repair sites. As reference, digital densitometry and polarized light microscopy were used to evaluate superficial and full-thickness PG content and collagen orientation angle. To relate NIR spectra and cartilage properties, ensemble neural networks, each with two different architectures, were trained and evaluated by using Spearman's correlation analysis (ρ). The ensemble networks enabled accurate predictions for full-thickness reference properties (PG content: ρin vitro, Val= 0.691, ρin vivo= 0.676; collagen orientation angle: ρin vitro, Val= 0.626, ρin vivo= 0.574) from NIR spectral data. In addition, the networks enabled reliable prediction of PG content in superficial (25%) cartilage (ρin vitro, Val= 0.650, ρin vivo= 0.613) and cartilage thickness (ρin vitro, Val= 0.797, ρin vivo= 0.596). To conclude, NIR spectroscopy could enhance the detection of initial cartilage degeneration and thus enable demarcation of the boundary between healthy and compromised cartilage tissue during arthroscopic surgery.

Evaluation of the Suitability of Miniature Pigs as an Animal Model of Juvenile Osteochondritis Dissecans.

Juvenile osteochondritis dissecans (JOCD) is a developmental disease characterized by formation of intra-articular (osteo)chondral flaps or fragments. Evidence-based treatment guidelines for JOCD are currently lacking. An animal model would facilitate study of JOCD and evaluation of diagnostic and treatment approaches. The purpose of this study was to assess the suitability of miniature pigs as a model of JOCD at the distal femur. First, stifle (knee) joints harvested from three juvenile miniature pigs underwent magnetic resonance imaging (MRI) to establish the vascular architecture of the distal femoral epiphyseal cartilage. Second, vessels supplying the axial or abaxial aspects of the medial femoral condyle were surgically interrupted in four additional juvenile miniature pigs, and the developing epiphyseal cartilage lesions were monitored using three consecutive MRI examinations over nine weeks. The miniature pigs were then euthanized, and their distal femora were harvested for histological evaluation. Vascular architecture of the distal femoral epiphyseal cartilage in the miniature pigs was found to be nearly identical to that of juvenile human subjects, characterized by separate vascular beds supplying the axial and abaxial aspects of the condyles. Surgical interruption of the vascular supply to the abaxial aspect of the medial femoral condyle resulted in ischemic cartilage necrosis (a precursor lesion of JOCD) in 75% (3/4) of the miniature pigs. Cartilage lesions were identified during the first MRI performed 3 weeks post-operatively. No clinically apparent JOCD-like lesions developed. In conclusion, miniature pigs are suitable for modeling JOCD precursor lesions. Further investigation of the model is warranted to assess induction of clinically apparent JOCD lesions. © 2019 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 37:2130-2137, 2019.

Quantitative susceptibility mapping of articular cartilage: Ex vivo findings at multiple orientations and following different degradation treatments.

PURPOSE: We investigated the feasibility of quantitative susceptibility mapping (QSM) for assessing degradation of articular cartilage by measuring ex vivo bovine cartilage samples subjected to different degradative treatments. Specimens were scanned at several orientations to study if degradation affects the susceptibility anisotropy. T2*-mapping, histological stainings, and polarized light microscopy were used as reference methods. Additionally, simulations of susceptibility in layered geometry were performed. METHODS: Samples (n = 9) were harvested from the patellae of skeletally mature bovines. Three specimens served as controls, and the rest were artificially degraded. MRI was performed at 9.4T using a 3D gradient echo sequence. QSM and T2* images and depth profiles through the centers of the samples were compared with each other and the histological findings. A planar isotropic model with depth-wise susceptibility variation was used in the simulations. RESULTS: A strong diamagnetic contrast was seen in the deep and calcified layers of cartilage, while T2* maps reflected the typical trilaminar structure of the collagen network. Anisotropy of susceptibility in cartilage was observed and was found to differ from the T2* anisotropy. Slight changes were observed in QSM and T2* following the degradative treatments. In simulations, anisotropy was observed. CONCLUSIONS: The results suggest that QSM is not sensitive to cartilage proteoglycan content, but shows sensitivity to the amount of calcification and to the integrity of the collagen network, providing potential for assessing osteoarthritis. The simulations suggested that the anisotropy of susceptibility might be partially explained by the layered geometry of susceptibility in cartilage.

Quantitative photoacoustic tomography augmented with surface light measurements.

Quantitative photoacoustic tomography is an imaging modality in which distributions of optical parameters inside tissue are estimated from photoacoustic images. This optical parameter estimation is an ill-posed problem and it needs to be approached in the framework of inverse problems. In this work, utilising surface light measurements in quantitative photoacoustic tomography is studied. Estimation of absorption and scattering is formulated as a minimisation problem utilising both internal quantitative photoacoustic data and surface light data. The image reconstruction problem is studied with two-dimensional numerical simulations in various imaging situations using the diffusion approximation as the model for light propagation. The results show that quantitative photoacoustic tomography augmented with surface light data can improve both absorption and scattering estimates when compared to the conventional quantitative photoacoustic tomography.