Search

Purpose To develop an end-to-end deep learning solution for quickly reconstructing radial simultaneous multi-slice (SMS) myocardial perfusion datasets with comparable quality to the pixel tracking spatiotemporal constrained reconstruction (PT-STCR) method. Methods Dynamic contrast enhanced (DCE) radial SMS myocardial perfusion data were obtained from 20 subjects who were scanned at rest and/or stress with or without ECG gating using a saturation recovery radial CAIPI turboFLASH sequence. Input to the networks consisted of complex coil combined images reconstructed using the inverse Fourier transform of undersampled radial SMS k-space data. Ground truth images were reconstructed using the PT-STCR pipeline. The performance of the residual booster 3D U-Net was tested by comparing it to state-of-the-art network architectures including MoDL, CRNN-MRI, and other U-Net variants. Results Results demonstrate significant improvements in speed requiring approximately 8 seconds to reconstruct one radial SMS dataset which is approximately 200 times faster than the PT-STCR method. Images reconstructed with the residual booster 3D U-Net retain quality of ground truth PT-STCR images (0.963 SSIM/40.238 PSNR/0.147 NRMSE). The residual booster 3D U-Net has superior performance compared to existing network architectures in terms of image quality, temporal dynamics, and reconstruction time. Conclusion Residual and booster learning combined with the 3D U-Net architecture was shown to be an effective network for reconstructing high-quality images from undersampled radial SMS datasets while bypassing the reconstruction time of the PT-STCR method.

Background The brain is the most cholesterol‐rich organ and myelin contains 70% of total brain cholesterol. Statins are potent cholesterol‐lowing medications used by millions of adults for prevention of vascular disease, yet the effect of statins on cholesterol‐rich brain white matter (WM) is largely unknown. Methods We used longitudinal neuroimaging data acquired from 73 healthy, cognitively unimpaired, statin‐naïve, middle‐aged adults during an 18‐month randomized controlled trial of simvastatin 40 mg daily (n = 35) or matching placebo (n = 38). ANCOVA models (covariates: age, sex, APOE‐ɛ4) tested the effect of treatment group on percent change in WM, gray matter (GM), and WM hyperintensity (WMH) neuroimaging measures at each study visit. Mediation analysis tested the indirect effects of simvastatin on WM microstructure through change in serum total cholesterol levels. Results At 18 months, the simvastatin group showed a significant preservation in global WM fractional anisotropy (β = 0.88%, 95% CI 0.27 to 1.50, P = 0.005), radial diffusivity (β = −1.10%, 95% CI −2.13 to −0.06, P = 0.039), and WM volume (β = 0.72%, 95% CI 0.13 to 1.32, P = 0.018) relative to the placebo group. There was no significant effect of simvastatin on GM or WMH volume. Change in serum total cholesterol mediated approximately 30% of the effect of simvastatin on WM microstructure. Conclusions Simvastatin treatment in healthy, middle‐aged adults resulted in preserved WM microstructure and volume at 18 months. The partial mediation by serum cholesterol reduction suggests both peripheral and central mechanisms. Future studies are needed to determine whether these effects persist and translate to cognitive outcomes. Trial Registration NCT00939822 (ClinicalTrials.gov).

Introduction: The utilization of HAMs is crucial in emergency and intensive care departments, as they can cause a significant amount of damage to the patient and health care members if we could not follow the standard treatment guidelines. Drug utilization evaluation/review involves a comprehensive review of the patient’s prescription and medication data before, during, and after dispensing to ensure appropriate medication decision making and positive patient outcomes. Objective: This study was taken up given finding the utilization patterns and rectifying the issues with the usage of high alert medications (HAMs) and improving their utilization. Methodology: A cross-sectional study was conducted for 6 months at a south Indian tertiary care hospital. Treatment guidelines were prepared to compare the actual drug use. Data were collected both retrospectively and prospectively by patients and care taker’s interview, medication chart review, and discussion with prescribers and applied WHO DUE indicators to evaluate utilization patterns. Results: Of 362 cases, 57.73 % were males/ and the majority geriatrics. Among all HAMs Insulin is frequently prescribed (34.5 %) and the costly drug is Enoxaparin. Generic names were used in writing prescriptions and parenteral formulations were mostly used. Around 9 ADRs were identified and managed, and a total of 133 moderate to severe Drug-Drug Interactions were found, of them, only 2 were actual. Conclusion: With this study, we conclude that the use of HAMs was found to be appropriate as per the guidelines as we observed very few DRPs with the study drugs. Keywords: HAM, DRP’s, DUE, ICU, DDD

Throughout adulthood, the brain undergoes an array of structural and functional changes during the typical aging process. These changes involve decreased brain volume, reduced synaptic density, and alterations in white matter (WM). Although there have been some previous neuroimaging studies that have measured the ability of adult language production and its correlations to brain function, structural gray matter volume, and functional differences between young and old adults, the structural role of WM in adult language production in individuals across the life span remains to be thoroughly elucidated. This study selected 38 young adults and 35 old adults for diffusion tensor imaging (DTI) and performed the Controlled Oral Word Association Test to assess verbal fluency (VF). Tract-Based Spatial Statistics were employed to evaluate the voxel-based group differences of diffusion metrics for the values of fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), radial diffusivity (RD), and local diffusion homogeneity (LDH) in 12 WM regions of interest associated with language production. To investigate group differences on each DTI metric, an analysis of covariance (ANCOVA) controlling for sex and education level was performed, and the statistical threshold was considered at p < 0.00083 (0.05/60 labels) after Bonferroni correction for multiple comparisons. Significant differences in DTI metrics identified in the ANCOVA were used to perform correlation analyses with VF scores. Compared to the old adults, the young adults had significantly (1) increased FA values on the bilateral anterior corona radiata (ACR); (2) decreased MD values on the right ACR, but increased MD on the left uncinate fasciculus (UF); and (3) decreased RD on the bilateral ACR. There were no significant differences between the groups for AD or LDH. Moreover, the old adults had only a significant correlation between the VF score and the MD on the left UF. There were no significant correlations between VF score and DTI metrics in the young adults. This study adds to the growing body of research that WM areas involved in language production are sensitive to aging.

A comprehensive mapping of the structural and functional circuitry of the brain is a major unresolved problem in contemporary neuroimaging research. Diffusion-weighted and functional MRI have provided investigators with the capability to assess structural and functional connectivity in-vivo, driven primarily by methods of white matter tractography and resting-state fMRI, respectively. These techniques have paved the way for the construction of the functional and structural connectomes, which are quantitative representations of brain architecture as neural networks, comprised of nodes and edges. The connectomes, typically depicted as matrices or graphs, possess topological properties that inherently characterize the strength, efficiency, and organization of the connections between distinct brain regions. Graph theory, a general mathematical framework for analyzing networks, can be implemented to derive metrics from the connectomes that are sensitive to changes in brain connectivity associated with age, sex, cognitive function, and disease. These quantities can be assessed at either the global (whole brain) or local levels, allowing for the identification of distinct regional connectivity hubs and associated localized brain networks, which together serve crucial roles in establishing the structural and functional architecture of the brain. As a result, structural and functional connectomes have each been employed to study the brain circuitry underlying early brain development, neuroplasticity, developmental disorders, psychopathology, epilepsy, aging, neurodegenerative disorders, and traumatic brain injury. While these studies have yielded important insights into brain structure, function, and pathology, a precise description of the innate relationship between functional and structural networks across the brain remains unachieved. To date, connectome research has merely scratched the surface of potential clinical applications and related characterizations of brain-wide connectivity. Continued advances in diffusion and functional MRI acquisition, the delineation of functional and structural networks, and the quantification of neural network properties in specific brain regions, will be invaluable to future progress in neuroimaging science.

Introduction Neurite orientation dispersion and density imaging (NODDI), a multi-compartment diffusion-weighted imaging (DWI) model, may be useful for detecting early cortical microstructural alterations in Alzheimer's disease prior to cognitive impairment. Methods Using neuroimaging (NODDI and T1-weighted magnetic resonance imaging [MRI]) and cerebrospinal fluid (CSF) biomarker data (measured using Elecsys® CSF immunoassays) from 219 cognitively unimpaired participants, we tested the main and interactive effects of CSF amyloid beta (Aβ)42 /Aβ40 and phosphorylated tau (p-tau) on cortical NODDI metrics and cortical thickness, controlling for age, sex, and apolipoprotein E e4. Results We observed a significant CSF Aβ42 /Aβ40 × p-tau interaction on cortical neurite density index (NDI), but not orientation dispersion index or cortical thickness. The directionality of these interactive effects indicated: (1) among individuals with lower CSF p-tau, greater amyloid burden was associated with higher cortical NDI; and (2) individuals with greater amyloid and p-tau burden had lower cortical NDI, consistent with cortical neurodegenerative changes. Discussion NDI is a particularly sensitive marker for early cortical changes that occur prior to gross atrophy or development of cognitive impairment.

Objective: Myocardial first-pass perfusion imaging with MRI is well-established clinically. However, it is potentially weakened by limited myocardial coverage compared to nuclear medicine. Clinical evaluations of whole-heart MRI perfusion by 3D methods, while promising, have to date had the limit of breathhold requirements at stress. This work aims to develop a new free-breathing 3D myocardial perfusion method, and to test its performance in a small patient population. Methods: This work required tolerance to respiratory motion for stress investigations, and therefore employed a “stack-of-stars” hybrid Cartesian-radial MRI acquisition method. The MRI sequence was highly optimised for rapid acquisition and combined with a compressed sensing reconstruction. Stress and rest datasets were acquired in four healthy volunteers, and in six patients with coronary artery disease (CAD), which were compared against clinical reference information.Results: This free-breathing method produced datasets that appeared consistent with clinical reference data in detecting moderate-to-strong induced perfusion abnormalities. However, the majority of the mild defects identified clinically were not detected by the method, potentially due to the presence of transient myocardial artefacts present in the images. Discussion: The feasibility of detecting CAD using this 3D first-pass perfusion sequence during free-breathing is demonstrated. Good agreement on typical moderate-to-strong CAD cases is promising, however, questions still remain on the sensitivity of the technique to milder cases.

Improved understanding of neuroimaging signal changes and their relation to patient outcomes after ischemic stroke is needed to improve ability to predict motor improvement and make therapy recommendations. The posterior limb of the internal capsule (PLIC) is a hub of afferent and efferent motor signaling and this work proposes new, image-based methods for prognosis based on interhemispheric differences in the PLIC. In this work, nine acute supratentorial ischemic stroke patients with motor impairment received a baseline, 203-direction diffusion brain MRI and a clinical assessment 3–12 days post-stroke and were compared to nine age-matched healthy controls. Asymmetries based on the mean and Kullback-Leibler divergence in the ipsilesional and contralesional PLIC were calculated for diffusion tensor imaging (DTI) and diffusion spectrum imaging (DSI) measures from the baseline MRI. Predictions of upper extremity Fugl-Meyer (FM) scores at 5-weeks follow-up from baseline measures of PLIC asymmetry in diffusion tensor imaging (DTI) and diffusion spectrum imaging (DSI) models were evaluated. For the stroke participants, the baseline asymmetry measures in the PLIC for the orientation dispersion index of the neurite orientation dispersion and density imaging (NODDI) model were highly correlated with upper extremity FM outcomes (r2 = 0.83). Use of DSI and the NODDI orientation dispersion index parameter shows promise of being more predictive of stroke recovery and to help better understand white matter changes in stroke, beyond DTI measures. The new finding that baseline interhemispheric differences in the PLIC calculated from the orientation dispersion index of the NODDI model are highly correlated with upper extremity functional outcomes may lead to improved image-based motor-outcome prediction after middle cerebral artery ischemic stroke.

Computational tools in the form of software packages are burgeoning in the field of medical imaging and biomedical research. These tools enable biomedical researchers to analyze a variety of data using modern machine learning and statistical analysis techniques. While these publicly available software packages are a great step towards a multiplicative increase in the biomedical research productivity, there are still many open issues related to validation and reproducibility of the results. A key gap is that while scientists can validate domain insights that are implicit in the analysis, the analysis itself is coded in a programming language and that domain scientist may not be a programmer. Thus, there is no/limited direct validation of the program that carries out the desired analysis. We propose a novel solution, building upon recent successes in natural language understanding, to address this problem. Our platform allows researchers to perform, share, reproduce and interpret the analysis pipelines and results via natural language. While this approach still requires users to have a conceptual understanding of the techniques, it removes the burden of programming syntax and thus lowers the barriers to advanced and reproducible neuroimaging and biomedical research.

Performing large scale hypothesis testing on brain imaging data to identify group-wise differences (e.g., between healthy and diseased subjects) typically leads to a large number of tests (one per voxel). Multiple testing adjustment (or correction) is necessary to control false positives, which may lead to lower detection power in detecting true positives. Motivated by the use of so-called “independent filtering” techniques in statistics (for genomics applications), this paper investigates the use of independent filtering for manifold-valued data (e.g., Diffusion Tensor Imaging, Cauchy Deformation Tensors) which are broadly used in neuroimaging studies. Inspired by the concept of variance of a Riemannian Gaussian distribution, a type of non-specific data-dependent Riemannian variance filter is proposed. In practice, the filter will select a subset of the full set of voxels for performing the statistical test, leading to a more appropriate multiple testing correction. Our experiments on synthetic/simulated manifold-valued data show that the detection power is improved when the statistical tests are performed on the voxel locations that “pass” the filter. Given the broadening scope of applications where manifold-valued data are utilized, the scheme can serve as a general feature selection scheme.

Recent advances in deep learning have improved the segmentation accuracy of subcortical brain structures, which would be useful in neuroimaging studies of many neurological disorders. However, most existing deep learning based approaches in neuroimaging do not investigate the specific difficulties that exist in segmenting extremely small but important brain regions such as the subnuclei of the amygdala. To tackle this challenging task, we developed a dual-branch dilated residual 3D fully convolutional network with parallel convolutions to extract more global context and alleviate the class imbalance issue by maintaining a small receptive field that is just the size of the regions of interest (ROIs). We also conduct multi-scale feature fusion in both parallel and series to compensate the potential information loss during convolutions, which has been shown to be important for small objects. The serial feature fusion enabled by residual connections is further enhanced by a proposed top-down attention-guided refinement unit, where the high-resolution low-level spatial details are selectively integrated to complement the high-level but coarse semantic information, enriching the final feature representations. As a result, the segmentations resulting from our method are more accurate both volumetrically and morphologically, compared with other deep learning based approaches. To the best of our knowledge, this work is the first deep learning-based approach that targets the subregions of the amygdala. We also demonstrated the feasibility of using a cycle-consistent generative adversarial network (CycleGAN) to harmonize multi-site MRI data, and show that our method generalizes well to challenging traumatic brain injury (TBI) datasets collected from multiple centers. This appears to be a promising strategy for image segmentation for multiple site studies and increased morphological variability from significant brain pathology.

A sizable body of work on relative attributes provides evidence that relating pairs of images along a continuum of strength pertaining to a visual attribute yields improvements in a variety of vision tasks. In this paper, we show how emerging ideas in graph neural networks can yield a solution to various problems that broadly fall under relative attribute learning. Our main idea is the observation that relative attribute learning naturally benefits from exploiting the graph of dependencies among the different relative attributes of images, especially when only partial ordering is provided at training time. We use message passing to perform end to end learning of the image representations, their relationships as well as the interplay between different attributes. Our experiments show that this simple framework is effective in achieving competitive accuracy with specialized methods for both relative attribute learning and binary attribute prediction, while relaxing the requirements on the training data and/or the number of parameters, or both.

Historically, the aerospace industry has been a principal driver of the development of the study of durability. This chapter will identify the foundations of the general subject and specific developments that motivate the science foundations as well as the engineering practice of durability of material systems using diverse approaches such as discrete damage modeling for certification, critical path methods for damage accumulation, and machine learning methods for materials-based system control.

Dynamic contrast-enhanced (DCE)-MRI has evolved into a mature imaging modality that is used to study the physiological processes and to reveal a variety of diseases, tumor stages, and organ functionalities. Quantitative DCE-MRI requires high spatiotemporal resolution to trace the contrast agent changes over time to obtain perfusion maps in the area of interest, where specific imaging challenges emerge. This chapter gives an overview of the basic principles and techniques involved in quantitative DCE-MRI. A review of the contrast agent mechanism is provided. An imaging protocol with advanced acquisition and reconstruction methods, including data undersampling, non-Cartesian sampling, parallel imaging, and constrained reconstruction, are also discussed. These techniques can achieve a higher spatiotemporal resolution for quantitative analysis. Available options for postprocessing of DCE-MRI data are also discussed, involving motion compensation, area- and pixel-wise methods, and tracer kinetic models.

We present a unified heat kernel smoothing framework for modeling 3D anatomical surface data extracted from medical images. Due to image acquisition and preprocessing noises, it is expected the medical imaging data is noisy. The surface data of the anatomical structures is regressed using the weighted linear combination of Laplace-Beltrami (LB) eigenfunctions to smooth out noisy data and perform statistical analysis. The method is applied in characterizing the 3D growth pattern of human hyoid bone between ages 0 and 20 obtained from CT images. We detected a significant age effect on localized parts of the hyoid bone.

The thalamus is a key sensorimotor relay area that is implicated in autism spectrum disorder (ASD). However, it is unknown how the thalamus and white-matter structures that contain thalamo-cortical fiber connections (e.g., the internal capsule) develop from childhood into adulthood and whether this microstructure relates to basic motor challenges in ASD. We used diffusion weighted imaging in a cohort-sequential design to assess longitudinal development of the thalamus, and posterior- and anterior-limbs of the internal capsule (PLIC and ALIC, respectively) in 89 males with ASD and 56 males with typical development (3-41 years; all verbal). Our results showed that the group with ASD exhibited different developmental trajectories of microstructure in all regions, demonstrating childhood group differences that appeared to approach and, in some cases, surpass the typically developing group in adolescence and adulthood. The PLIC (but not ALIC nor thalamus) mediated the relation between age and finger-tapping speed in both groups. Yet, the gap in finger-tapping speed appeared to widen at the same time that the between-group gap in the PLIC appeared to narrow. Overall, these results suggest that childhood group differences in microstructure of the thalamus and PLIC become less robust in adolescence and adulthood. Further, finger-tapping speed appears to be mediated by the PLIC in both groups, but group differences in motor speed that widen during adolescence and adulthood suggest that factors beyond the microstructure of the thalamus and internal capsule may contribute to atypical motor profiles in ASD. Autism Res 2018, 11: 450-462. © 2017 International Society for Autism Research, Wiley Periodicals, Inc. Lay summary Microstructure of the thalamus, a key sensory and motor brain area, appears to develop differently in individuals with autism spectrum disorder (ASD). Microstructure is important because it informs us of the density and organization of different brain tissues. During childhood, thalamic microstructure was distinct in the ASD group compared to the typically developing group. However, these group differences appeared to narrow with age, suggesting that the thalamus continues to dynamically change in ASD into adulthood.

Autism spectrum disorder (ASD) is a neurodevelopmental disorder with unknown brain etiology. Our knowledge to date about structural brain development across the lifespan in ASD comes mainly from cross-sectional studies, thereby limiting our understanding of true age effects within individuals with the disorder that can only be gained through longitudinal research. The present study describes FreeSurfer-derived volumetric findings from a longitudinal dataset consisting of 607 T1-weighted magnetic resonance imaging (MRI) scans collected from 105 male individuals with ASD (349 MRIs) and 125 typically developing male controls (258 MRIs). Participants were six to forty-five years of age at their first scan, and were scanned up to 5 times over a period of 16 years (average inter-scan interval of 3.7 years). Atypical age-related volumetric trajectories in ASD included enlarged gray matter volume in early childhood that approached levels of the control group by late childhood, an age-related increase in ventricle volume resulting in enlarged ventricles by early adulthood and reduced corpus callosum age-related volumetric increase resulting in smaller corpus callosum volume in adulthood. Larger corpus callosum volume was related to a lower (better) ADOS score at the most recent study visit for the participants with ASD. These longitudinal findings expand our knowledge of volumetric brain-based abnormalities in males with ASD, and highlight the need to continue to examine brain structure across the lifespan and well into adulthood.

Background: Quantifying myocardial perfusion is complicated by the complexity of pharmacokinetic model being used and the reliability of perfusion parameter estimates. More complex modeling provides more information about the underlying physiology, but too many parameters in complex models introduce a new problem of reliable estimation. To overcome the problem of multiple parameters, we have developed a technique that combines knowledge from two different cardiac magnetic resonance (MR) imaging techniques: dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) and T1 mapping. Using extracellular volume (ECV) estimates from T1 mapping may allow more robust model parameter estimates. Methods: Simulations and human scans were performed. The myocardial perfusion scans used an ungated saturation recovery prepared TurboFLASH pulse sequence. Four short-axis (SA) slices were acquired after a single saturation pulse with a saturation recovery time of ~25 ms before the first slice. Gadoteridol was injected and ~240 frames were acquired over a minute with shallow breathing and no electrocardiograph (ECG) gating. This was followed 20±5 minutes later by an injection of regadenoson to induce hyperemia. The data were acquired using an under-sampled golden angle radial acquisition. Modified look-locker inversion recovery (MOLLI) T1 mapping was performed in 3 slices pre- and post-contrast. The pre- and post-contrast T1 maps were used for ECV estimation. Quantification of perfusion was done using a 4-parameter model with additional information about ECV supplied during model fitting. Phase contrast scans of the coronary sinus (CS) were acquired at rest and immediately after the stress perfusion acquisition to estimate global flow. Results: Without ECV information, the 5-parameter model fails to converge to a unique solution and often gives incorrect estimates for the perfusion parameters. The myocardial blood flow (MBF) estimates during rest and stress were 0.9±0.1 and 2.3±0.6 mL/min/g, respectively. The extraction fraction estimates were 0.49±0.04 and 0.34±0.05 during rest and stress, respectively. Conclusions: These results show that it is possible to successfully fit a dynamic perfusion model with an extraction fraction parameter by using information from T1 mapping scans. This hybrid approach is especially important when the 5-parameter model alone fails to converge on a unique solution. This work is a good example of exploiting information overlaps between various cardiac MR imaging techniques.

Discrete Damage Modeling is applied to complex sub-elements loaded in static and fatigue. Layered 8-node brick elements called LC3D8 elements were implemented for the first time in DDM models to reduce the computational requirements for modeling large and complex parts. Three sub-elements were tested and modeled: Fastener Pull Through (FPT), Hat Pull Off (HPO), and Three Point Bend with Flange (3PB-F). All three sub-elements were tested and simulated in quasi-static loading, while the 3PB-F was simulated in constant amplitude fatigue as well. Results demonstrate that DDM is capable of capturing peak load response accurately and damage initiation locations fairly accurately even with relatively coarse meshes.

Purpose The purpose of this study was to further develop and combine several innovative sequence designs to achieve quantitative 3D myocardial perfusion. These developments include an optimized 3D stack-of-stars readout (150 ms per beat), efficient acquisition of a 2D arterial input function, tailored saturation pulse design, and potential whole heart coverage during quantitative stress perfusion. Theory and methods All studies were performed free-breathing on a Prisma 3T MRI scanner. Phantom validation was used to verify sequence accuracy. A total of 21 subjects (3 patients with known disease) were scanned, 12 with a rest only protocol and 9 with both stress (regadenoson) and rest protocols. First pass quantitative perfusion was performed with gadoteridol (0.075 mmol/kg). Results Implementation and quantitative perfusion results are shown for healthy subjects and subjects with known coronary disease. Average rest perfusion for the 15 included healthy subjects was 0.79 ± 0.19 mL/g/min, the average stress perfusion for 6 healthy subject studies was 2.44 ± 0.61 mL/g/min, and the average global myocardial perfusion reserve ratio for 6 healthy subjects was 3.10 ± 0.24. Perfusion deficits for 3 patients with ischemia are shown. Average resting heart rate was 59 ± 7 bpm and the average stress heart rate was 81 ± 10 bpm. Conclusion This work demonstrates that a quantitative 3D myocardial perfusion sequence with the acquisition of a 2D arterial input function is feasible at high stress heart rates such as during stress. T1 values and gadolinium concentrations of the sequence match the reference standard well in a phantom, and myocardial rest and stress perfusion and myocardial perfusion reserve values are consistent with those published in literature.

Purpose Dynamic contrast enhanced MRI of the heart typically acquires 2-4 short-axis (SA) slices to detect and characterize coronary artery disease. This acquisition scheme is limited by incomplete coverage of the left ventricle. We studied the feasibility of using radial simultaneous multi-slice (SMS) technique to achieve SA, 2-chamber and/or 4-chamber long-axis (2CH LA and/or 4CH LA) coverage with and without electrocardiography (ECG) gating using a motion-robust reconstruction framework. Methods 12 subjects were scanned at rest and/or stress, free breathing, with or without ECG gating. Multiple sets of radial SMS k-space were acquired within each cardiac cycle, and each SMS set sampled 3 parallel slices that were either SA, 2CH LA, or 4CH LA slices. The radial data was interpolated onto Cartesian space using an SMS GRAPPA operator gridding method. Self-gating and respiratory states binning of the data were done. The binning information as well as a pixel tracking spatiotemporal constrained reconstruction method were applied to obtain motion-robust image reconstructions. Reconstructions with and without the pixel tracking method were compared for signal-to-noise ratio and contrast-to-noise ratio. Results Full coverage of the heart (at least 3 SA and 3 LA slices) during the first pass of contrast at every heartbeat was achieved by using the radial SMS acquisition. The proposed pixel tracking reconstruction improves the average SNR and CNR by 21% and 30% respectively, and reduces temporal blurring for both gated and ungated acquisitions. Conclusion Acquiring simultaneous multi-slice SA, 2CH LA and/or 4CH LA myocardial perfusion images in every heartbeat is feasible in both gated and ungated acquisitions. This can add confidence when detecting and characterizing coronary artery disease by revealing ischemia in different views, and by providing apical coverage that is improved relative to SA slices alone. The proposed pixel tracking framework improves the reconstruction while adding little computational cost.

Current late gadolinium enhancement (LGE) imaging of left atrial (LA) scar or fibrosis is relatively slow and requires 5-15min to acquire an undersampled (R=1.7) 3D navigated dataset. The GeneRalized Autocalibrating Partially Parallel Acquisitions (GRAPPA) based parallel imaging method is the current clinical standard for accelerating 3D LGE imaging of the LA and permits an acceleration factor ~R=1.7. Two compressed sensing (CS) methods have been developed to achieve higher acceleration factors: a patch based collaborative filtering technique tested with acceleration factor R~3, and a technique that uses a 3D radial stack-of-stars acquisition pattern (R~1.8) with a 3D total variation constraint. The long reconstruction time of these CS methods makes them unwieldy to use, especially the patch based collaborative filtering technique. In addition, the effect of CS techniques on the quantification of percentage of scar/fibrosis is not known. We sought to develop a practical compressed sensing method for imaging the LA at high acceleration factors. In order to develop a clinically viable method with short reconstruction time, a Split Bregman (SB) reconstruction method with 3D total variation (TV) constraints was developed and implemented. The method was tested on 8 atrial fibrillation patients (4 pre-ablation and 4 post-ablation datasets). Blur metric, normalized mean squared error and peak signal to noise ratio were used as metrics to analyze the quality of the reconstructed images, Quantification of the extent of LGE was performed on the undersampled images and compared with the fully sampled images. Quantification of scar from post-ablation datasets and quantification of fibrosis from pre-ablation datasets showed that acceleration factors up to R~3.5 gave good 3D LGE images of the LA wall, using a 3D TV constraint and constrained SB methods. This corresponds to reducing the scan time by half, compared to currently used GRAPPA methods. Reconstruction of 3D LGE images using the SB method was over 20 times faster than standard gradient descent methods.

Purpose To evaluate the interstudy repeatability of multislice quantitative cardiovascular magnetic resonance myocardial blood flow (MBF), myocardial perfusion reserve (MPR), and extracellular volume (ECV). A unique saturation recovery self-gated acquisition was used for the perfusion scans. Materials and Methods An ungated golden angle radial turboFLASH pulse sequence was used to scan 10 subjects on two separate days on a 3T scanner. A single saturation pulse was followed by a set of four slices. Rest and hyperemia scans were acquired during free breathing. The images were reconstructed using an iterative algorithm with spatiotemporal constraints. The ungated images were retrospectively binned (self-gated) into near-systole and near-diastole. Deformable registration was performed to adjust for respiratory and residual cardiac motion, and the data were fit with a Fermi model to estimate the interstudy repeatability of quantitative self-gated MBF and MPR. Results The coefficient of variation (CoV) of the territorial MPR using the self-gated near-systole data was 18.6%. The self-gated near-diastole data gave less good CoV of MPR, equal to 46.2%. For MBFs, and using smaller (segmental) regions, the CoVs were 20.1% and 22.7% for the estimation of myocardial blood flow at stress and rest, respectively, using the self-gated near-systole data. The self-gated near-diastole data gave CoV = 48.6% and 44.9% for stress and rest. Conclusion The self-gated free-breathing technique for quantification of myocardial blood flow showed good repeatability for near-systole, with results comparable to published studies on interstudy repeatability of quantitative myocardial perfusion MRI using ECG-gating and breath-holds. Self-gated near-diastole data results were less repeatable. J. Magn. Reson. Imaging 2015.

Clamped Tapered Beam (CTB) sub-element, proposed by Advanced Composite Project Team at the NASA Langley Research Center was studied under monotonic fatigue loading. The specimen was designed to study matrix crack initiation from pristine condition, delamination initiation and propagation including migration from one interface to other. Regularized eXtended Finite Element Method (Rx-FEM), which allows modeling displacement discontinuities associated with matrix cracks independent of the mesh orientation was used for analysis. Previously developed fatigue Cohesive Zone Method (CZM) was modified to include the S-N information in the Newton-Rapson equilibrium loop. The analysis was conducted in two phases, before and after the experimental data was revealed. The blind predictions were conducted at R=0.1 load ratio for 80% and 70% values of the fatigue load amplitude with respect to static strength and overestimated the rate of delamination growth. The corrected predictions took into account the R=0.2 load ratio, thermal residual stresses and were based on the modified algorithm. The failure event sequence and the respective cycle counts were consistent with experimental data.

When a brain network is constructed by an existing parcellation method, the topological structure of the network changes depending on the scale of the parcellation. To avoid the scale dependency, we propose to construct a nested hierarchical structural brain network by subdividing the existing parcellation hierarchically. The method is applied in diffusion tensor imaging study of 111 twins in characterizing the topology of the brain network. The genetic contribution of the whole brain structural connectivity is determined and shown to be robustly present over different network scales.

Post-mortem studies of Parkinson’s disease (PD) suggest that Lewy body pathology accumulates in a predictable topographical sequence, beginning in the olfactory bulb, followed by caudal brainstem, substantia nigra, limbic cortex, and neocortex. Diffusion weighted imaging (DWI) is sensitive, if not specific, to early disease-related white matter (WM) change in a variety of traumatic and degenerative brain diseases. Although numerous cross-sectional studies have reported DWI differences in cerebral WM in PD, only a few longitudinal studies have investigated whether DWI change exceeds that of normal aging or coincides with regional Lewy body accumulation. The present study mapped regional differences in the rate of DWI-based microstructural change between 29 PD patients and 43 age-matched controls over 18 months. Iterative within- and between-subject tensor-based registration was completed on motion- and eddy current-corrected DWI images, then baseline versus follow-up difference maps of fractional anisotropy, mean, radial, and axial diffusivity were analyzed in the Biological Parametric Mapping toolbox for MATLAB. This analysis showed that PD patients had a greater decline in WM integrity in the rostral brainstem, caudal subcortical WM, and cerebellar peduncles, compared with controls. In addition, patients with unilateral clinical signs at baseline experienced a greater rate of WM change over the 18-month study than patients with bilateral signs. These findings suggest that rate of WM microstructural change in PD exceeds that of normal aging and is maximal during early-stage disease. In addition, the neuroanatomic locations (rostral brainstem and subcortical WM) of accelerated WM change fit with current theories of topographic disease progression.

Purpose To develop a robust multidimensional deep-learning based method to simultaneously generate accurate neurite orientation dispersion and density imaging (NODDI) and generalized fractional anisotropy (GFA) parameter maps from undersampled q-space datasets for use in stroke imaging. Methods Traditional diffusion spectrum imaging (DSI) capable of producing accurate NODDI and GFA parameter maps requires hundreds of q-space samples which renders the scan time clinically untenable. A convolutional neural network (CNN) was trained to generated NODDI and GFA parameter maps simultaneously from 10× undersampled q-space data. A total of 48 DSI scans from 15 stroke patients and 14 normal subjects were acquired for training, validating, and testing this method. The proposed network was compared to previously proposed voxel-wise machine learning based approaches for q-space imaging. Network-generated images were used to predict stroke functional outcome measures. Results The proposed network achieves significant performance advantages compared to previously proposed machine learning approaches, showing significant improvements across image quality metrics. Generating these parameter maps using CNNs also comes with the computational benefits of only needing to generate and train a single network instead of multiple networks for each parameter type. Post-stroke outcome prediction metrics do not appreciably change when using images generated from this proposed technique. Over three test participants, the predicted stroke functional outcome scores were within 1-6% of the clinical evaluations. Conclusions Estimates of NODDI and GFA parameters estimated simultaneously with a deep learning network from highly undersampled q-space data were improved compared to other state-of-the-art methods providing a 10-fold reduction scan time compared to conventional methods.

In cardiac perfusion imaging, choice of flip angle is an important factor for steady state acquisition. This work focuses on presenting an analytical framework for understanding how non-ideal slice excitation profiles affect contrast in ungated 2D steady state cardiac perfusion studies, and to study a technique for estimating flip angle that maximizes enhanced/unenhanced myocardial contrast-to-noise ratio (CNR) in single slice and multi-slice acquisitions. A numerical simulation of ungated 2D golden ratio radial spoiled gradient echo (SPGR) was created that takes into consideration the actual (Bloch simulated) slice excitation profile. The effect of slice excitation profile on myocardial CNR as a function of flip angle was assessed in phantoms and in vivo. For fast RF pulses, the flip angle that yields maximum CNR (considering the actual slice excitation profile) was considerably higher than expected, assuming an ideal excitation. The simulation framework presented accurately predicts the flip angle yielding maximum CNR when the actual slice excitation profile is taken into consideration. The prescribed flip angle for optimal contrast in ungated 2D steady-state SPGR cardiac perfusion studies can vary significantly from that calculated when an ideal slice excitation profile is assumed. Consideration of the actual slice excitation can yield a more optimal flip angle estimate in both the single slice and multi-slice cases.

There is significant interest, both from basic and applied research perspectives, in understanding how structural/functional connectivity changes can explain behavioral symptoms and predict decline in neurodegenerative diseases such as Alzheimer's disease (AD). The first step in most such analyses is to encode the connectivity information as a graph; then, one may perform statistical inference on various ‘global’ graph theoretic summary measures (e.g., modularity, graph diameter) and/or at the level of individual edges (or connections). For AD in particular, clear differences in connectivity at the dementia stage of the disease (relative to healthy controls) have been identified. Despite such findings, AD-related connectivity changes in preclinical disease remain poorly characterized. Such preclinical datasets are typically smaller and group differences are weaker. In this paper, we propose a new multi-resolution method for performing statistical analysis of connectivity networks/graphs derived from neuroimaging data. At the high level, the method occupies the middle ground between the two contrasts — that is, to analyze global graph summary measures (global) or connectivity strengths or correlations for individual edges similar to voxel based analysis (local). Instead, our strategy derives a Wavelet representation at each primitive (connection edge) which captures the graph context at multiple resolutions. We provide extensive empirical evidence of how this framework offers improved statistical power by analyzing two distinct AD datasets. Here, connectivity is derived from diffusion tensor magnetic resonance images by running a tractography routine. We first present results showing significant connectivity differences between AD patients and controls that were not evident using standard approaches. Later, we show results on populations that are not diagnosed with AD but have a positive family history risk of AD where our algorithm helps in identifying potentially subtle differences between patient groups. We also give an easy to deploy open source implementation of the algorithm for use within studies of connectivity in AD and other neurodegenerative disorders.

Purpose: To improve rank constrained reconstructions for undersampled multi-image MRI acquisitions. Methods: Motivated by the recent developments in low-rank matrix completion theory and its applicability to rapid dynamic MRI, a new reordering-based rank constrained reconstruction of undersampled multi-image data that uses prior image information is proposed. Instead of directly minimizing the nuclear norm of a matrix of estimated images, the nuclear norm of reordered matrix values is minimized. The reordering is based on the prior image estimates. The method is tested on brain diffusion imaging data and dynamic contrast enhanced myocardial perfusion data. Results: Good quality images from data undersampled by a factor of three for diffusion imaging and by a factor of 3.5 for dynamic cardiac perfusion imaging with respiratory motion were obtained. Reordering gave visually improved image quality over standard nuclear norm minimization reconstructions. Root mean squared errors with respect to ground truth images were improved by ∼18% and ∼16% with reordering for diffusion and perfusion applications, respectively. Conclusions: The reordered low-rank constraint is a way to inject prior image information that offers improvements over a standard low-rank constraint for undersampled multi-image MRI reconstructions.

Laminated composite materials are used in stiffness critical applications and contain various stress concentration features such as holes, tapers and stiffener attachments. It is these areas which often determine the load carrying capacity of the structural elements and thickness scaling of critical parts. A representative subelement is studied in this work with the aim of evaluating the ability to predict the mechanical response of such critical regions by using discrete damage modeling. Clamped Tapered Beam Specimen (CTBS) designed by Advanced Composite Project Team at NASA LaRC is considered. This specimen allows studying the matrix crack initiation from pristine condition as well as delamination initiation and evolution including migration from one interface to another. Regularized eXtended Finite Element Method (Rx-FEM) is employed for analysis. The Rx-FEM allows modeling the displacement discontinuity associated with individual matrix cracks in individual plies of a composite without regard to mesh orientation by inserting additional degrees of freedom in the process of the simulation. The propagation of the mesh independent crack is then performed by using Cohesive Zone Method (CZM). A simultaneous experimental program was conducted at NASA LaRC and used for the analysis validation. The predicted results showed good agreement with experimental data for ply level transverse strength parameter obtained by using three point bend test method, whereas the results obtained by using a lower value resulting from tensile testing of 90o coupons resulted in under-prediction of the peak load. The distance between the migration location and the load application point is consistent with experimental data.

We present a new structural brain network parcellation scheme that can subdivide existing parcellations into smaller subregions in a hierarchically nested fashion. The hierarchical parcellation was used to build multilayer convolutional structural brain networks that preserve topology across different network scales. As an application, we applied the method to diffusion weighted imaging study of 111 twin pairs. The genetic contribution of the whole brain structural connectivity was determined. We showed that the overall heritability is consistent across different network scales.

Analyzing data representing multifarious trajectories is central to the many fields in Science and Engineering; for example, trajectories representing a tennis serve, a gymnast’s parallel bar routine, progression/remission of disease and so on. We present a novel geometric algorithm for performing statistical analysis of trajectories with distinct number of samples representing longitudinal (or temporal) data. A key feature of our proposal is that unlike existing schemes, our model is deployable in regimes where each participant provides a different number of acquisitions (trajectories have different number of sample points or temporal span). To achieve this, we develop a novel method involving the parallel transport of the tangent vectors along each given trajectory to the starting point of the respective trajectories and then use the span of the matrix whose columns consist of these vectors, to construct a linear subspace in R(m). We then map these linear subspaces (possibly of distinct dimensions) of R(m) on to a single high dimensional hypersphere. This enables computing group statistics over trajectories by instead performing statistics on the hypersphere (equipped with a simpler geometry). Given a point on the hypersphere representing a trajectory, we also provide a “reverse mapping” algorithm to uniquely (under certain assumptions) reconstruct the subspace that corresponds to this point. Finally, by using existing algorithms for recursive Fréchet mean and exact principal geodesic analysis on the hypersphere, we present several experiments on synthetic and real (vision and medical) data sets showing how group testing on such diversely sampled longitudinal data is possible by analyzing the reconstructed data in the subspace spanned by the first few principal components.

Objective We present a registration-based semiautomatic mandible segmentation (SAMS) pipeline designed to process a large number of computed tomography studies to segment 3-dimensional mandibles. Method The pipeline consists of a manual preprocessing step, an automatic segmentation step, and a final manual postprocessing step. The automatic portion uses a nonlinear diffeomorphic method to register each preprocessed input computed tomography test scan on 54 reference templates, ranging in age from birth to 19 years. This creates 54 segmentations, which are then combined into a single composite mandible. Results This pipeline was assessed using 20 mandibles from computed tomography studies with ages 1 to 19 years, segmented using both SAMS-processing and manual segmentation. Comparisons between the SAMS-processed and manually-segmented mandibles revealed 97% similarity agreement with comparable volumes. The resulting 3-dimensional mandibles were further enhanced with manual postprocessing in specific regions. Conclusions Findings are indicative of a robust pipeline that reduces manual segmentation time by 75% and increases the feasibility of large-scale mandibular growth studies.

Purpose To validate an optimal 12-fold accelerated real-time cine MRI pulse sequence with radial k-space sampling and compressed sensing (CS) in patients at 1.5T and 3T. Methods We used two strategies to reduce image artifacts arising from gradient delays and eddy currents in radial k-space sampling with balanced steady-state free precession readout. We validated this pulse sequence against a standard breath-hold cine sequence in two patient cohorts: a myocardial infarction (n = 16) group at 1.5T and chronic kidney disease group (n = 18) at 3T. Two readers independently performed visual analysis of 68 cine sets in four categories (myocardial definition, temporal fidelity, artifact, noise) on a 5-point Likert scale (1 = nondiagnostic, 2 = poor, 3 = adequate or moderate, 4 = good, 5 = excellent). Another reader calculated left ventricular (LV) functional parameters, including ejection fraction. Results Compared with standard cine, real-time cine produced nonsignificantly different visually assessed scores, except for the following categories: 1) temporal fidelity scores were significantly lower (P = 0.013) for real-time cine at both field strengths, 2) artifacts scores were significantly higher (P = 0.013) for real-time cine at both field strengths, and 3) noise scores were significantly (P = 0.013) higher for real-time cine at 1.5T. Standard and real-time cine pulse sequences produced LV functional parameters that were in good agreement (e.g., absolute mean difference in ejection fraction

Statistical machine learning models that operate on manifold-valued data are being extensively studied in vision, motivated by applications in activity recognition, feature tracking and medical imaging. While non-parametric methods have been relatively well studied in the literature, efficient formulations for parametric models (which may offer benefits in small sample size regimes) have only emerged recently. So far, manifold-valued regression models (such as geodesic regression) are restricted to the analysis of cross-sectional data, i.e., the so-called fixed effects in statistics. But in most longitudinal analysis (e.g., when a participant provides multiple measurements, over time) the application of fixed effects models is problematic. In an effort to answer this need, this paper generalizes non-linear mixed effects model to the regime where the response variable is manifold-valued, i.e., f: Rd → M. We derive the underlying model and estimation schemes and demonstrate the immediate benefits such a model can provide — both for group level and individual level analysis — on longitudinal brain imaging data. The direct consequence of our results is that longitudinal analysis of manifold-valued measurements (especially, the symmetric positive definite manifold) can be conducted in a computationally tractable manner.

Cardiovascular magnetic resonance (CMR) perfusion has been established as a useful imaging modality for the detection of coronary artery disease (CAD). However, there are several limitations when applying standard, ECG-gated stress/rest perfusion CMR to patients with atrial fibrillation (AF). In this study we investigate an approach with no ECG gating and a rapid rest/stress perfusion protocol to determine its accuracy for detection of CAD in patients with AF. 26 patients with AF underwent a rapid rest/regadenoson stress CMR perfusion imaging protocol, and all patients had X-ray coronary angiography. An ungated radial myocardial perfusion sequence was used. Imaging protocol included: rest perfusion image acquisition, followed nearly immediately by administration of regadenoson to induce hyperemia, 60 s wait, and stress image acquisition. CMR perfusion images were interpreted by three blinded readers as normal or abnormal. Diagnostic accuracy was evaluated by comparison to X-ray angiography. 21 of the CMR rest/stress perfusion scans were negative, and 5 were positive by angiography criteria. Majority results of the ungated datasets from all of the readers showed a sensitivity, specificity and accuracy of 80, 100 and 96%, respectively, for detection of CAD. An ungated, rapid rest/stress regadenoson perfusion CMR protocol appears to be useful for the diagnosis of obstructive CAD in patients with AF.