Search

IntroductionSpatio-temporal MRI methods enable whole-brain multi-parametric mapping at ultra-fast acquisition times through efficient k-space encoding, but can have very long reconstruction times, which limit their integration into clinical practice. Deep learning (DL) is a promising approach to accelerate reconstruction, but can be computationally intensive to train and deploy due to the large dimensionality of spatio-temporal MRI. DL methods also need large training data sets and can produce results that don’t match the acquired data if data consistency is not enforced. The aim of this project is to reduce reconstruction time using DL whilst simultaneously limiting the risk of deep learning induced hallucinations, all with modest hardware requirements.MethodsDeep Learning Initialized Compressed Sensing (Deli-CS) is proposed to reduce the reconstruction time of iterative reconstructions by “kick-starting” the iterative reconstruction with a DL generated starting point. The proposed framework is applied to volumetric multi-axis spiral projection MRF that achieves whole-brain T1 and T2 mapping at 1-mm isotropic resolution for a 2-minute acquisition. First, the traditional reconstruction is optimized from over two hours to less than 40 minutes while using more than 90% less RAM and only 4.7 GB GPU memory, by using a memory-efficient GPU implementation. The Deli-CS framework is then implemented and evaluated against the above reconstruction.ResultsDeli-CS achieves comparable reconstruction quality with 50% fewer iterations bringing the full reconstruction time to 20 minutes.ConclusionDeli-CS reduces the reconstruction time of subspace reconstruction of volumetric spatio-temporal acquisitions by providing a warm start to the iterative reconstruction algorithm.

This work proposed a systematic fabrication process towards high-performance SERS-active substrates as the chemical sensors, based on a combination of the hydrothermal-synthesized ZnO nanorods (ZnO NRs) and the decorated Au nano-surface. The fabrication process was first focused on the preparation of the Au/ZnO NRs hybrids. The ZnO NR templates were grown on silicon (1 0 0) wafer substrates by the hydrothermal method. Then, the Au films were deposited on the ZnO NR templates by the DC pulse magnetron sputtering technique. The obtained SERS substrates were utilized towards trace detection and classifications of pentaerythritol tetranitrate (PETN), which is commonly used in industrial explosives and improvised explosive devices. The results from field-emission electron microscopy (FE-SEM) and transmission electron microscopy (TEM) indicated the changes in physical morphologies and crystal structure, respectively, based on the variation of the fabrication parameters. The FE-SEM results showed that well-vertical ZnO NRs demonstrated the length of the nanorods before and after the Au decorations. In addition, the hexagonal structures, before and after the Au decorations, were observed with the average diameter of 50 and 181 nm, the nanorod distances of 178 and 81 nm, respectively. The TEM results also confirmed the average size and d-spacing of the ZnO NRs and Au/ZnO NR hybrids, with validated elements/compounds from the EDX mapping technique. Furthermore, the sample surfaces were observed with Raman mapping spectroscopy towards uniformly distributed examination. Finally, the fabricated samples were investigated with trace concentration of PETN towards the limit of detection. The results that our SERS substrates could detect PETN of less than 1 µg/ml, which indicated excellent performance of the trace detection of the explosive substance. The SERS substrates were further used to successfully classify different sources of the PETN substances, with 10 mg/ml, based on the machine learning approach with the principal component analysis.

Mild cognitive impairment (MCI) is an early stage of age-inappropriate cognitive decline, which could develop into dementia – an untreatable neurodegenerative disorder. An early detection of MCI is a crucial step for timely prevention and intervention. To tackle this problem, recent studies have developed deep learning models to detect MCI and various types of dementia using data obtained from the classic clock-drawing test (CDT), a popular neuropsychological screening tool that can be easily and rapidly implemented for assessing cognitive impairments in an aging population. While these models succeed at distinguishing severe forms of dementia, it is still difficult to predict the early stage of the disease using the CDT data alone. Also, the state-of-the-art deep learning techniques still face the black-box challenges, making it questionable to implement them in the clinical setting. Here, we propose a novel deep learning modeling framework that incorporates data from multiple drawing tasks including the CDT, cube-copying, and trail-making tasks obtained from a digital platform. Using self-attention and soft-label methods, our model achieves much higher classification performance at detecting MCI compared to those of a well-established convolutional neural network model. Moreover, our model can highlight features of the MCI data that considerably deviate from those of the healthy aging population, offering accurate predictions for detecting MCI along with visual explanation that aids the interpretation of the deep learning model.

Colorimetric loop-mediated DNA isothermal amplification-based assays have gained momentum in the diagnosis of COVID-19 owing to their unmatched feasibility in low-resource settings. However, the vast majority of them are restricted to proprietary pH-sensitive dyes that limit downstream assay optimization or hinder efficient result interpretation. To address this problem, we developed a novel dual colorimetric RT-LAMP assay using in-house pH-dependent indicators to maximize the visual detection and assay simplicity, and further integrated it with the artificial intelligence (AI) operated tool (RT-LAMP-DETR) to enable a more precise and rapid result analysis in large scale testing. The dual assay leverages xylenol orange (XO) and a newly formulated lavender green (LG) dye for distinctive colorimetric readouts, which enhance the test accuracy when performed and analyzed simultaneously. Our RT-LAMP assay has a detection limit of 50 viral copies/reaction with the cycle threshold (Ct) value ≤ 39.7 ± 0.4 determined by the WHO-approved RT-qPCR assay. RT-LAMP-DETR exhibited a complete concordance with the results from naked-eye observation and RT-qPCR, achieving 100% sensitivity, specificity, and accuracy that altogether render it suitable for ultrasensitive point-of-care COVID-19 screening efforts. From the perspective of pandemic preparedness, our method offers a simpler, faster, and cheaper (∼$8/test) approach for COVID-19 testing and other emerging pathogens with respect to RT-qPCR.

Early and precise diagnosis of cancer is critical for a better prognosis. Here, we describe a simple and cost-effective plasmonic paper as a surface enhanced Raman scattering (SERS) substrate in combination with magnetic separation for cancer screening. The plasmonic paper was fabricated by immersing plain filter paper into gold nanorod solution and the SERS property of the paper was evaluated using 4-mercaptobenzoic acid (4-MBA) and rhodamine 6 G (R6 G), which showed an enhancement factor (EF) in the range of 106–108. HT-29, a colorectal cancer cell line that highly expresses epithelial cell adhesion molecule (EpCAM), served as the target cells; non-EpCAM-expressing cells, namely fibroblasts and red blood cells (RBCs), were used as negative controls. Intrinsic SERS spectra of the target and control cells showed distinctive patterns on the plasmonic paper due to differences in their structure and components. A combination of principal component analysis (PCA) and k-nearest-neighbor algorithm (k-NN) was employed to analyze and distinguish the acquired HT-29 and fibroblast SERS spectra, demonstrating a diagnostic sensitivity and specificity of 84.8% and 82.6%, respectively, whereas the differentiation between HT-29 and RBCs SERS spectra showed a sensitivity and specificity of 96.4% and 100%, respectively. The magnetic separation was applied to capture the target cells from cell mixtures followed by PCA and k-NN analysis. The identification of the captured cells as cancerous cells from the HT-29 and fibroblast mixture indicated an accuracy of 83.7%, while that from a mixture of HT-29 and RBCs was 98.2%. Thus, the simple paper-based SERS substrate with the assistance of magnetic enrichment and multivariate analysis offers a potent new platform for cancer cell detection and screening.

We propose Nonlinear Dipole Inversion (NDI) for high-quality Quantitative Susceptibility Mapping (QSM) without regularization tuning, while matching the image quality of state-of-the-art reconstruction techniques. In addition to avoiding over-smoothing that these techniques often suffer from, we also obviate the need for parameter selection. NDI is flexible enough to allow for reconstruction from an arbitrary number of head orientations and outperforms COSMOS even when using as few as 1-direction data. This is made possible by a nonlinear forward-model that uses the magnitude as an effective prior, for which we derived a simple gradient descent update rule. We synergistically combine this physics-model with a Variational Network (VN) to leverage the power of deep learning in the VaNDI algorithm. This technique adopts the simple gradient descent rule from NDI and learns the network parameters during training, hence requires no additional parameter tuning. Further, we evaluate NDI at 7 T using highly accelerated Wave-CAIPI acquisitions at 0.5 mm isotropic resolution and demonstrate high-quality QSM from as few as 2-direction data., National Institutes of Health (U.S.) (Grants R01EB020613, R01EB019437, P41EB015896, U01EB025162, S10RR023401, S10RR019307, S10RR019254, S10RR023043))

Purpose To describe a model-based reconstruction strategy for single-shot echo planar imaging (EPI) that intrinsically accounts for k-space nonuniformity, Nyquist ghosting, and geometric distortions during rather than before or after image reconstruction. Methods Ramp sampling and inhomogeneous B0 field-induced distortion cause the EPI samples to lie on a non-Cartesian grid, thus requiring the nonuniform fast Fourier transform. Additionally, a 2D Nyquist ghost phase correction without the need for extra navigator acquisition is included in the proposed reconstruction. Coil compression is also incorporated to reduce the computational load. The proposed method is applied to phantom and human brain MRI data. Results The results demonstrate that Nyquist ghosting and geometric distortions are reduced by the proposed reconstruction. The proposed 2D phase correction is superior to a conventional 1D correction. The reductions of both artifacts lead to improved temporal signal-to-noise ratio (tSNR). The virtual coil results suggest that the processing time can be reduced by up to 75%, with a mean tSNR loss of only 3.2% when using 8-virtual instead of 32-physical coils for twofold undersampled data. Conclusion The proposed reconstruction improves the quality (ghosting, geometry, and tSNR) of EPI without requiring calibration data for Nyquist ghost correction. Magn Reson Med 78:2250–2264, 2017. © 2017 International Society for Magnetic Resonance in Medicine.

Purpose To develop and implement an efficient reconstruction technique to improve accelerated multi-channel multi-contrast MRI. Theory and Methods The vectorial total generalized variation (TGV) operator is used as a regularizer for the sensitivity encoding (SENSE) technique to improve image quality of multi-channel multi-contrast MRI. The alternating direction method of multipliers (ADMM) is used to efficiently reconstruct the data. The performance of the proposed method (MC-TGV-SENSE) is assessed on two healthy volunteers at several acceleration factors. Results As demonstrated on the in vivo results, MC-TGV-SENSE had the lowest root-mean-square error (RMSE), highest structural similarity index, and best visual quality at all acceleration factors, compared to other methods under consideration. MC-TGV-SENSE yielded up to 17.3% relative RMSE reduction compared to the widely used total variation regularized SENSE. Furthermore, we observed that the reconstruction time of MC-TGV-SENSE is reduced by approximately a factor of two with comparable RMSEs by using the proposed ADMM-based algorithm as opposed to the more commonly used Chambolle–Pock primal-dual algorithm for the TGV-based reconstruction. Conclusion MC-TGV-SENSE is a better alternative than the existing reconstruction methods for accelerated multi-channel multi-contrast MRI. The proposed method exploits shared information among the images (MC), mitigates staircasing artifacts (TGV), and uses the encoding power of multiple receiver coils (SENSE).

PURPOSE To introduce a combined machine learning (ML)- and physics-based image reconstruction framework that enables navigator-free, highly accelerated multishot echo planar imaging (msEPI) and demonstrate its application in high-resolution structural and diffusion imaging. METHODS Single-shot EPI is an efficient encoding technique, but does not lend itself well to high-resolution imaging because of severe distortion artifacts and blurring. Although msEPI can mitigate these artifacts, high-quality msEPI has been elusive because of phase mismatch arising from shot-to-shot variations which preclude the combination of the multiple-shot data into a single image. We utilize deep learning to obtain an interim image with minimal artifacts, which permits estimation of image phase variations attributed to shot-to-shot changes. These variations are then included in a joint virtual coil sensitivity encoding (JVC-SENSE) reconstruction to utilize data from all shots and improve upon the ML solution. RESULTS Our combined ML + physics approach enabled Rinplane × multiband (MB) = 8- × 2-fold acceleration using 2 EPI shots for multiecho imaging, so that whole-brain T2 and T2 * parameter maps could be derived from an 8.3-second acquisition at 1 × 1 × 3-mm3 resolution. This has also allowed high-resolution diffusion imaging with high geometrical fidelity using 5 shots at Rinplane × MB = 9- × 2-fold acceleration. To make these possible, we extended the state-of-the-art MUSSELS reconstruction technique to simultaneous multislice encoding and used it as an input to our ML network. CONCLUSION Combination of ML and JVC-SENSE enabled navigator-free msEPI at higher accelerations than previously possible while using fewer shots, with reduced vulnerability to poor generalizability and poor acceptance of end-to-end ML approaches.

Deep neural networks have demonstrated promising potential for the field of medical image reconstruction. In this work, an MRI reconstruction algorithm, which is referred to as quantitative susceptibility mapping (QSM), has been developed using a deep neural network in order to perform dipole deconvolution, which restores magnetic susceptibility source from an MRI field map. Previous approaches of QSM require multiple orientation data (e.g. Calculation of Susceptibility through Multiple Orientation Sampling or COSMOS) or regularization terms (e.g. Truncated K-space Division or TKD; Morphology Enabled Dipole Inversion or MEDI) to solve the ill-conditioned deconvolution problem. Unfortunately, they either require long multiple orientation scans or suffer from artifacts. To overcome these shortcomings, a deep neural network, QSMnet, is constructed to generate a high quality susceptibility map from single orientation data. The network has a modified U-net structure and is trained using gold-standard COSMOS QSM maps. 25 datasets from 5 subjects (5 orientation each) were applied for patch-wise training after doubling the data using augmentation. Two additional datasets of 5 orientation data were used for validation and test (one dataset each). The QSMnet maps of the test dataset were compared with those from TKD and MEDI for image quality and consistency in multiple head orientations. Quantitative and qualitative image quality comparisons demonstrate that the QSMnet results have superior image quality to those of TKD or MEDI and have comparable image quality to those of COSMOS. Additionally, QSMnet maps reveal substantially better consistency across the multiple orientations than those from TKD or MEDI. As a preliminary application, the network was tested for two patients. The QSMnet maps showed similar lesion contrasts with those from MEDI, demonstrating potential for future applications., Comment: This work is accepted in neuroimage on 8 June, 2018 and soon will be published. The pubmed link is https://www.ncbi.nlm.nih.gov/pubmed/29894829

Purpose To develop a reconstruction pipeline that intrinsically accounts for both simultaneous multislice echo planar imaging (SMS-EPI) reconstruction and dynamic slice-specific Nyquist ghosting correction in time-series data. Methods After 1D slice-group average phase correction, the separate polarity (i.e., even and odd echoes) SMS-EPI data were unaliased by slice GeneRalized Autocalibrating Partial Parallel Acquisition. Both the slice-unaliased even and odd echoes were jointly reconstructed using a model-based framework, extended for SMS-EPI reconstruction that estimates a 2D self-phase map, corrects dynamic slice-specific phase errors, and combines data from all coils and echoes to obtain the final images. Results The percentage ghost-to-signal ratios (%GSRs) and its temporal variations for MB3Ry 2 with a field of view/4 shift in a human brain obtained by the proposed dynamic 2D and standard 1D phase corrections were 1.37 ± 0.11 and 2.66 ± 0.16, respectively. Even with a large regularization parameter λ applied in the proposed reconstruction, the smoothing effect in fMRI activation maps was comparable to a very small Gaussian kernel size 1 × 1 × 1 mm3 . Conclusion The proposed reconstruction pipeline reduced slice-specific phase errors in SMS-EPI, resulting in reduction of GSR. It is applicable for functional MRI studies because the smoothing effect caused by the regularization parameter selection can be minimal in a blood-oxygen-level-dependent activation map.

Purpose To develop and evaluate the performance of an acquisition and reconstruction method for accelerated MR spectroscopic imaging (MRSI) through undersampling of spiral trajectories. Theory and Methods A randomly undersampled spiral acquisition and sensitivity encoding (SENSE) with total variation (TV) regularization, random SENSE+TV, is developed and evaluated on single-slice numerical phantom, in vivo single-slice MRSI, and in vivo three-dimensional (3D)-MRSI at 3 Tesla. Random SENSE+TV was compared with five alternative methods for accelerated MRSI. Results For the in vivo single-slice MRSI, random SENSE+TV yields up to 2.7 and 2 times reduction in root-mean-square error (RMSE) of reconstructed N-acetyl aspartate (NAA), creatine, and choline maps, compared with the denoised fully sampled and uniformly undersampled SENSE+TV methods with the same acquisition time, respectively. For the in vivo 3D-MRSI, random SENSE+TV yields up to 1.6 times reduction in RMSE, compared with uniform SENSE+TV. Furthermore, by using random SENSE+TV, we have demonstrated on the in vivo single-slice and 3D-MRSI that acceleration factors of 4.5 and 4 are achievable with the same quality as the fully sampled data, as measured by RMSE of reconstructed NAA map, respectively. Conclusion With the same scan time, random SENSE+TV yields lower RMSEs of metabolite maps than other methods evaluated. Random SENSE+TV achieves up to 4.5-fold acceleration with comparable data quality as the fully sampled acquisition. Magn Reson Med 74:13–24, 2015. © 2014 Wiley Periodicals, Inc.

Purpose: We introduce L2-regularized reconstruction algorithms with closed-form solutions that achieve dramatic computational speed-up relative to state of the art L1- and L2-based iterative algorithms while maintaining similar image quality for various applications in MRI reconstruction. Materials and Methods: We compare fast L2-based methods to state of the art algorithms employing iterative L1- and L2-regularization in numerical phantom and in vivo data in three applications; (i) Fast Quantitative Susceptibility Mapping (QSM), (ii) Lipid artifact suppression in Magnetic Resonance Spectroscopic Imaging (MRSI), and (iii) Diffusion Spectrum Imaging (DSI). In all cases, proposed L2-based methods are compared with the state of the art algorithms, and two to three orders of magnitude speed up is demonstrated with similar reconstruction quality. Results: The closed-form solution developed for regularized QSM allows processing of a three-dimensional volume under 5 s, the proposed lipid suppression algorithm takes under 1 s to reconstruct single-slice MRSI data, while the PCA based DSI algorithm estimates diffusion propagators from undersampled q-space for a single slice under 30 s, all running in Matlab using a standard workstation. Conclusion: For the applications considered herein, closed-form L2-regularization can be a faster alternative to its iterative counterpart or L1-based iterative algorithms, without compromising image quality.

Copyright © 2016 John Wiley & Sons, Ltd. Quantitative susceptibility mapping (QSM) estimates the underlying tissue magnetic susceptibility from the gradient echo (GRE) phase signal through background phase removal and dipole inversion steps. Each of these steps typically requires the solution of an ill-posed inverse problem and thus necessitates additional regularization. Recently developed single-step QSM algorithms directly relate the unprocessed GRE phase to the unknown susceptibility distribution, thereby requiring the solution of a single inverse problem. In this work, we show that such a holistic approach provides susceptibility estimation with artifact mitigation and develop efficient algorithms that involve simple analytical solutions for all of the optimization steps. Our methods employ total variation (TV) and total generalized variation (TGV) to jointly perform the background removal and dipole inversion in a single step. Using multiple spherical mean value (SMV) kernels of varying radii permits high-fidelity background removal whilst retaining the phase information in the cortex. Using numerical simulations, we demonstrate that the proposed single-step methods reduce the reconstruction error by up to 66% relative to the multi-step methods that involve SMV background filtering with the same number of SMV kernels, followed by TV- or TGV-regularized dipole inversion. In vivo single-step experiments demonstrate a dramatic reduction in dipole streaking artifacts and improved homogeneity of image contrast. These acquisitions employ the rapid three-dimensional echo planar imaging (3D EPI) and Wave-CAIPI (controlled aliasing in parallel imaging) trajectories for signal-to-noise ratio-efficient whole-brain imaging. Herein, we also demonstrate the multi-echo capability of the Wave-CAIPI sequence for the first time, and introduce an automated, phase-sensitive coil sensitivity estimation scheme based on a 4-s calibration acquisition. Copyright © 2016 John Wiley & Sons, Ltd.

Quantitative Susceptibility Mapping (QSM) aims to estimate the tissue susceptibility distribution that gives rise to subtle changes in the main magnetic field, which are captured by the image phase in a gradient echo (GRE) experiment. The underlying susceptibility distribution is related to the acquired tissue phase through an ill-posed linear system. To facilitate its inversion, spatial regularization that imposes sparsity or smoothness assumptions can be employed. This paper focuses on efficient algorithms for regularized QSM reconstruction. Fast solvers that enforce sparsity under Total Variation (TV) and Total Generalized Variation (TGV) constraints are developed using Alternating Direction Method of Multipliers (ADMM). Through variable splitting that permits closed-form iterations, the computation efficiency of these solvers are dramatically improved. An alternative approach to improve the conditioning of the ill-posed inversion is to acquire multiple GRE volumes at different head orientations relative to the main magnetic field. The phase information from such multi-orientation acquisition can be combined to yield exquisite susceptibility maps and obviate the need for regularized reconstruction, albeit at the cost of increased data acquisition time.

Clinical magnetic resonance imaging (MRI) protocols typically include multiple acquisitions of the same region of interest under different contrast settings. This paper presents an efficient algorithm to jointly reconstruct a set of undersampled images with different contrasts. The proposed method has faster reconstruction time and better quality as measured by the normalized root-mean-square error (RMSE) compared to the existing methods consisting of multi-contrast Fast Composite Splitting Algorithm (FCSA-MT), Multiple measurement vectors FOCal Underdetermined System Solver (M-FOCUSS), and total variation regularized compressed sensing (SparseMRI). To efficiently solve the £2, 1-regularized optimization problem, our proposed algorithm adopts the Split Bregman (SB) technique to divide the problem into sub-problems. We efficiently compute a closed-form solution to each of the sub-problems by implementing a 3D spatial gradient operator as element-wise multiplication in k-space. As demonstrated by the in vivo results, the proposed algorithm (SB-L21) offers 2x, 32x, and 66x faster reconstruction with lower RMSE averaged across all contrasts and slices compared to FCSA-MT, M-FOCUSS, and SparseMRI, respectively.

Therapies based on biologics involving delivery of proteins, DNA, and RNA are currently among the most promising approaches. However, although large combinatorial libraries of biologics and delivery vehicles can be readily synthesized, there are currently no means to rapidly characterize them in vivo using animal models. Here, we demonstrate high-throughput in vivo screening of biologics and delivery vehicles by automated delivery into target tissues of small vertebrates with developed organs. Individual zebrafish larvae are automatically oriented and immobilized within hydrogel droplets in an array format using a microfluidic system, and delivery vehicles are automatically microinjected to target organs with high repeatability and precision. We screened a library of lipid-like delivery vehicles for their ability to facilitate the expression of protein-encoding RNAs in the central nervous system. We discovered delivery vehicles that are effective in both larval zebrafish and rats. Our results showed that the in vivo zebrafish model can be significantly more predictive of both false positives and false negatives in mammals than in vitro mammalian cell culture assays. Our screening results also suggest certain structure–activity relationships, which can potentially be applied to design novel delivery vehicles., National Institutes of Health (U.S.) (Transformative Research Award R01 NS073127), National Institutes of Health (U.S.) (Director's Innovator Award DP2 OD002989), David & Lucile Packard Foundation (Award in Science and Engineering), Sanofi Aventis (Firm), Foxconn International Holdings Ltd., Hertz Foundation (Fellowship), University Grants Committee (Hong Kong, China) (Early Career Award 125012), National Natural Science Foundation (China) (81201164), ITC (ITS/376/13), Chinese University of Hong Kong (Grant 9610215), Chinese University of Hong Kong (Grant 7200269)

To develop and evaluate the performance of an acquisition and reconstruction method for accelerated MR spectroscopic imaging (MRSI) through undersampling of spiral trajectories.A randomly undersampled spiral acquisition and sensitivity encoding (SENSE) with total variation (TV) regularization, random SENSE+TV, is developed and evaluated on single-slice numerical phantom, in vivo single-slice MRSI, and in vivo three-dimensional (3D)-MRSI at 3 Tesla. Random SENSE+TV was compared with five alternative methods for accelerated MRSI.For the in vivo single-slice MRSI, random SENSE+TV yields up to 2.7 and 2 times reduction in root-mean-square error (RMSE) of reconstructed N-acetyl aspartate (NAA), creatine, and choline maps, compared with the denoised fully sampled and uniformly undersampled SENSE+TV methods with the same acquisition time, respectively. For the in vivo 3D-MRSI, random SENSE+TV yields up to 1.6 times reduction in RMSE, compared with uniform SENSE+TV. Furthermore, by using random SENSE+TV, we have demonstrated on the in vivo single-slice and 3D-MRSI that acceleration factors of 4.5 and 4 are achievable with the same quality as the fully sampled data, as measured by RMSE of reconstructed NAA map, respectively.With the same scan time, random SENSE+TV yields lower RMSEs of metabolite maps than other methods evaluated. Random SENSE+TV achieves up to 4.5-fold acceleration with comparable data quality as the fully sampled acquisition. Magn Reson Med 74:13-24, 2015. © 2014 Wiley Periodicals, Inc.

Diffusion Spectrum Imaging (DSI) reveals detailed local diffusion properties at the expense of substantially long imaging times. It is possible to accelerate acquisition by undersampling in q-space, followed by image reconstruction that exploits prior knowledge on the diffusion probability density functions (pdfs). Previously proposed methods impose this prior in the form of sparsity under wavelet and total variation (TV) transforms, or under adaptive dictionaries that are trained on example datasets to maximize the sparsity of the representation. These compressed sensing (CS) methods require full-brain processing times on the order of hours using Matlab running on a workstation. This work presents two dictionary-based reconstruction techniques that use analytical solutions, and are two orders of magnitude faster than the previously proposed dictionary-based CS approach. The first method generates a dictionary from the training data using Principal Component Analysis (PCA), and performs the reconstruction in the PCA space. The second proposed method applies reconstruction using pseudoinverse with Tikhonov regularization with respect to a dictionary. This dictionary can either be obtained using the K-SVD algorithm, or it can simply be the training dataset of pdfs without any training. All of the proposed methods achieve reconstruction times on the order of seconds per imaging slice, and have reconstruction quality comparable to that of dictionary-based CS algorithm.

Patient-specific abdominal aortic aneurysms (AAAs) are characterized by local curvature changes, which we assess using a feature-based approach on topologies representative of the AAA outer wall surface. The application of image segmentation methods yields 3D reconstructed surface polygons that contain low-quality elements, unrealistic sharp corners, and surface irregularities. To optimize the quality of the surface topology, an iterative algorithm was developed to perform interpolation of the AAA geometry, topology refinement, and smoothing. Triangular surface topologies are generated based on a Delaunay triangulation algorithm, which is adapted for AAA segmented masks. The boundary of the AAA wall is represented using a signed distance function prior to triangulation. The irregularities on the surface are minimized by an interpolation scheme and the initial coarse triangulation is refined by forcing nodes into equilibrium positions. A surface smoothing algorithm based on a low-pass filter is applied to remove sharp corners. The optimal number of iterations needed for polygon refinement and smoothing is determined by imposing a minimum average element quality index with no significant AAA sac volume change. This framework automatically generates high-quality triangular surface topologies that can be used to characterize local curvature changes of the AAA wall.