Search NASA⌕ Search

SEARCH · Search NASA

Results for “sparse”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 271 records · Page 15

SU-E-I-45: Reconstruction of CT Images From Sparsely-Sampled Data Using the Logarithmic Barrier Method

Purpose: To develop and investigate whether the logarithmic barrier (LB) method can result in high-quality reconstructed CT images using sparsely-sampled noisy projection data Methods: The objective function is typically formulated as the sum of the total variation (TV) and a data fidelity (DF) term with a parameter λ that governs the relative weight between them. Finding the optimized value of λ is a critical step for this approach to give satisfactory results. The proposed LB method avoid using λ by constructing the objective function as the sum of the TV and a log function whose augment is the DF term. Newton's method was used to solve the optimization problem. The algorithm was coded in MatLab2013b. Both Shepp-Logan phantom and a patient lung CT image were used for demonstration of the algorithm. Measured data were simulated by calculating the projection data using radon transform. A Poisson noise model was used to account for the simulated detector noise. The iteration stopped when the difference of the current TV and the previous one was less than 1%. Results: Shepp-Logan phantom reconstruction study shows that filtered back-projection (FBP) gives high streak artifacts for 30 and 40 projections. Although visually the streak artifacts are less pronounced for 64 and 90 projections in FBP, the 1D pixel profiles indicate that FBP gives noisier reconstructed pixel values than LB does. A lung image reconstruction is presented. It shows that use of 64 projections gives satisfactory reconstructed image quality with regard to noise suppression and sharp edge preservation. Conclusion: This study demonstrates that the logarithmic barrier method can be used to reconstruct CT images from sparsely-amped data. The number of projections around 64 gives a balance between the over-smoothing of the sharp demarcation and noise suppression. Future study may extend to CBCT reconstruction and improvement on computation speed.

60 APPLIED LIFE SCIENCES↗

SU-E-I-87: Automated Liver Segmentation Method for CBCT Dataset by Combining Sparse Shape Composition and Probabilistic Atlas Construction

Purpose: The aiming of this study was to extract liver structures for daily Cone beam CT (CBCT) images automatically. Methods: Datasets were collected from 50 intravenous contrast planning CT images, which were regarded as training dataset for probabilistic atlas and shape prior model construction. Firstly, probabilistic atlas and shape prior model based on sparse shape composition (SSC) were constructed by iterative deformable registration. Secondly, the artifacts and noise were removed from the daily CBCT image by an edge-preserving filtering using total variation with L1 norm (TV-L1). Furthermore, the initial liver region was obtained by registering the incoming CBCT image with the atlas utilizing edge-preserving deformable registration with multi-scale strategy, and then the initial liver region was converted to surface meshing which was registered with the shape model where the major variation of specific patient was modeled by sparse vectors. At the last stage, the shape and intensity information were incorporated into joint probabilistic model, and finally the liver structure was extracted by maximum a posteriori segmentation.Regarding the construction process, firstly the manually segmented contours were converted into meshes, and then arbitrary patient data was chosen as reference image to register with the rest of training datasets by deformable registration algorithm for constructing probabilistic atlas and prior shape model. To improve the efficiency of proposed method, the initial probabilistic atlas was used as reference image to register with other patient data for iterative construction for removing bias caused by arbitrary selection. Results: The experiment validated the accuracy of the segmentation results quantitatively by comparing with the manually ones. The volumetric overlap percentage between the automatically generated liver contours and the ground truth were on an average 88%–95% for CBCT images. Conclusion: The experiment demonstrated that liver structures of CBCT with artifacts can be extracted accurately for following adaptive radiation therapy. This work is supported by National Natural Science Foundation of China (No. 61201441), Research Fund for Excellent Young and Middle-aged Scientists of Shandong Province (No. BS2012DX038), Project of Shandong Province Higher Educational Science and Technology Program (No. J12LN23), Jinan youth science and technology star (No.20120109)

60 APPLIED LIFE SCIENCES↗

SU-E-J-212: Identifying Bones From MRI: A Dictionary Learnign and Sparse Regression Approach

Purpose: To develop an efficient and robust scheme to identify bony anatomy based on MRI-only simulation images. Methods: MRI offers important soft tissue contrast and functional information, yet its lack of correlation to electron-density has placed it as an auxiliary modality to CT in radiotherapy simulation and adaptation. An effective scheme to identify bony anatomy is an important first step towards MR-only simulation/treatment paradigm and would satisfy most practical purposes. We utilize a UTE acquisition sequence to achieve visibility of the bone. By contrast to manual + bulk or registration-to identify bones, we propose a novel learning-based approach for improved robustness to MR artefacts and environmental changes. Specifically, local information is encoded with MR image patch, and the corresponding label is extracted (during training) from simulation CT aligned to the UTE. Within each class (bone vs. nonbone), an overcomplete dictionary is learned so that typical patches within the proper class can be represented as a sparse combination of the dictionary entries. For testing, an acquired UTE-MRI is divided to patches using a sliding scheme, where each patch is sparsely regressed against both bone and nonbone dictionaries, and subsequently claimed to be associated with the class with the smaller residual. Results: The proposed method has been applied to the pilot site of brain imaging and it has showed general good performance, with dice similarity coefficient of greater than 0.9 in a crossvalidation study using 4 datasets. Importantly, it is robust towards consistent foreign objects (e.g., headset) and the artefacts relates to Gibbs and field heterogeneity. Conclusion: A learning perspective has been developed for inferring bone structures based on UTE MRI. The imaging setting is subject to minimal motion effects and the post-processing is efficient. The improved efficiency and robustness enables a first translation to MR-only routine. The scheme generalizes to multiple tissue classes.

60 APPLIED LIFE SCIENCES↗

SU-F-BRF-02: Automated Lung Segmentation Method Using Atlas-Based Sparse Shape Composition with a Shape Constrained Deformable Model

Purpose: To develop an automated lung segmentation method, which combines the atlas-based sparse shape composition with a shape constrained deformable model in thoracic CT for patients with compromised lung volumes. Methods: Ten thoracic computed tomography scans for patients with large lung tumors were collected and reference lung ROIs in each scan was manually segmented to assess the performance of the method. We propose an automated and robust framework for lung tissue segmentation by using single statistical atlas registration to initialize a robust deformable model in order to perform fine segmentation that includes compromised lung tissue. First, a statistical image atlas with sparse shape composition is constructed and employed to obtain an approximate estimation of lung volume. Next, a robust deformable model with shape prior is initialized from this estimation. Energy terms from ROI edge potential and interior ROI region based potential as well as the initial ROI are combined in this model for accurate and robust segmentation. Results: The proposed segmentation method is applied to segment right lung on three CT scans. The quantitative results of our segmentation method achieved mean dice score of (0.92–0.95), mean accuracy of (0.97,0.98), and mean relative error of (0.10,0.16) with 95% CI. The quantitative results of previously published RASM segmentation method achieved mean dice score of (0.74,0.96), mean accuracy of (0.66,0.98), and mean relative error of (0.04, 0.38) with 95% CI. The qualitative and quantitative comparisons show that our proposed method can achieve better segmentation accuracy with less variance compared with a robust active shape model method. Conclusion: The atlas-based segmentation approach achieved relatively high accuracy with less variance compared to RASM in the sample dataset and the proposed method will be useful in image analysis applications for lung nodule or lung cancer diagnosis and radiotherapy assessment in thoracic computed tomography.

60 APPLIED LIFE SCIENCES↗

Sparse-view proton computed tomography using modulated proton beams

Purpose: Proton imaging that uses a modulated proton beam and an intensity detector allows a relatively fast image acquisition compared to the imaging approach based on a trajectory tracking detector. In addition, it requires a relatively simple implementation in a conventional proton therapy equipment. The model of geometric straight ray assumed in conventional computed tomography (CT) image reconstruction is however challenged by multiple-Coulomb scattering and energy straggling in the proton imaging. Radiation dose to the patient is another important issue that has to be taken care of for practical applications. In this work, the authors have investigated iterative image reconstructions after a deconvolution of the sparsely view-sampled data to address these issues in proton CT. Methods: Proton projection images were acquired using the modulated proton beams and the EBT2 film as an intensity detector. Four electron-density cylinders representing normal soft tissues and bone were used as imaged object and scanned at 40 views that are equally separated over 360°. Digitized film images were converted to water-equivalent thickness by use of an empirically derived conversion curve. For improving the image quality, a deconvolution-based image deblurring with an empirically acquired point spread function was employed. They have implemented iterative image reconstruction algorithms such as adaptive steepest descent-projection onto convex sets (ASD-POCS), superiorization method–projection onto convex sets (SM-POCS), superiorization method–expectation maximization (SM-EM), and expectation maximization-total variation minimization (EM-TV). Performance of the four image reconstruction algorithms was analyzed and compared quantitatively via contrast-to-noise ratio (CNR) and root-mean-square-error (RMSE). Results: Objects of higher electron density have been reconstructed more accurately than those of lower density objects. The bone, for example, has been reconstructed within 1% error. EM-based algorithms produced an increased image noise and RMSE as the iteration reaches about 20, while the POCS-based algorithms showed a monotonic convergence with iterations. The ASD-POCS algorithm outperformed the others in terms of CNR, RMSE, and the accuracy of the reconstructed relative stopping power in the region of lung and soft tissues. Conclusions: The four iterative algorithms, i.e., ASD-POCS, SM-POCS, SM-EM, and EM-TV, have been developed and applied for proton CT image reconstruction. Although it still seems that the images need to be improved for practical applications to the treatment planning, proton CT imaging by use of the modulated beams in sparse-view sampling has demonstrated its feasibility.

60 APPLIED LIFE SCIENCES↗

SU-F-BRB-12: A Novel Haar Wavelet Based Approach to Deliver Non-Coplanar Intensity Modulated Radiotherapy Using Sparse Orthogonal Collimators

Purpose: Existing efforts to replace complex multileaf collimator (MLC) by simple jaws for intensity modulated radiation therapy (IMRT) resulted in unacceptable compromise in plan quality and delivery efficiency. We introduce a novel fluence map segmentation method based on compressed sensing for plan delivery using a simplified sparse orthogonal collimator (SOC) on the 4π non-coplanar radiotherapy platform. Methods: 4π plans with varying prescription doses were first created by automatically selecting and optimizing 20 non-coplanar beams for 2 GBM, 2 head & neck, and 2 lung patients. To create deliverable 4π plans using SOC, which are two pairs of orthogonal collimators with 1 to 4 leaves in each collimator bank, a Haar Fluence Optimization (HFO) method was used to regulate the number of Haar wavelet coefficients while maximizing the dose fidelity to the ideal prescription. The plans were directly stratified utilizing the optimized Haar wavelet rectangular basis. A matching number of deliverable segments were stratified for the MLC-based plans. Results: Compared to the MLC-based 4π plans, the SOC-based 4π plans increased the average PTV dose homogeneity from 0.811 to 0.913. PTV D98 and D99 were improved by 3.53% and 5.60% of the corresponding prescription doses. The average mean and maximal OAR doses slightly increased by 0.57% and 2.57% of the prescription doses. The average number of segments ranged between 5 and 30 per beam. The collimator travel time to create the segments decreased with increasing leaf numbers in the SOC. The two and four leaf designs were 1.71 and 1.93 times more efficient, on average, than the single leaf design. Conclusion: The innovative dose domain optimization based on compressed sensing enables uncompromised 4π non-coplanar IMRT dose delivery using simple rectangular segments that are deliverable using a sparse orthogonal collimator, which only requires 8 to 16 leaves yet is unlimited in modulation resolution. This work is supported in part by Varian Medical Systems, Inc. and NIH R43 CA18339.

60 APPLIED LIFE SCIENCES↗

Characterizing spatiotemporal information loss in sparse-sampling-based dynamic MRI for monitoring respiration-induced tumor motion in radiotherapy

Purpose: Sparse-sampling and reconstruction techniques represent an attractive strategy to achieve faster image acquisition speeds, while maintaining adequate spatial resolution and signal-to-noise ratio in rapid magnetic resonance imaging (MRI). The authors investigate the use of one such sequence, broad-use linear acquisition speed-up technique (k-t BLAST) in monitoring tumor motion for thoracic and abdominal radiotherapy and examine the potential trade-off between increased sparsification (to increase imaging speed) and the potential loss of “true” information due to greater reliance on a priori information. Methods: Lung tumor motion trajectories in the superior–inferior direction, previously recorded from ten lung cancer patients, were replayed using a motion phantom module driven by an MRI-compatible motion platform. Eppendorf test tubes filled with water which serve as fiducial markers were placed in the phantom. The modeled rigid and deformable motions were collected in a coronal image slice using balanced fast field echo in conjunction with k-t BLAST. Root mean square (RMS) error was used as a metric of spatial accuracy as measured trajectories were compared to input data. The loss of spatial information was characterized for progressively increasing acceleration factor from 1 to 16; the resultant sampling frequency was increased approximately from 2.5 to 19 Hz when the principal direction of the motion was set along frequency encoding direction. In addition to the phantom study, respiration-induced tumor motions were captured from two patients (kidney tumor and lung tumor) at 13 Hz over 49 s to demonstrate the impact of high speed motion monitoring over multiple breathing cycles. For each subject, the authors compared the tumor centroid trajectory as well as the deformable motion during free breathing. Results: In the rigid and deformable phantom studies, the RMS error of target tracking at the acquisition speed of 19 Hz was approximately 0.3–0.4 mm, which was smaller than the reconstructed pixel resolution of 0.67 mm. In the patient study, the dynamic 2D MRI enabled the monitoring of cycle-to-cycle respiratory variability present in the tumor position. It was seen that the range of centroid motion as well as the area covered due to target motion during each individual respiratory cycle was underestimated compared to the entire motion range observed over multiple breathing cycles. Conclusions: The authors’ initial results demonstrate that sparse-sampling- and reconstruction-based dynamic MRI can be used to achieve adequate image acquisition speeds without significant information loss for the task of radiotherapy guidance. Such monitoring can yield spatial and temporal information superior to conventional offline and online motion capture methods used in thoracic and abdominal radiotherapy.

60 APPLIED LIFE SCIENCES↗

SU-D-207A-05: Investigating Sparse-Sampled MRI for Motion Management in Thoracic Radiotherapy

Purpose: Sparse sampling and reconstruction-based MRI techniques represent an attractive strategy to achieve sufficiently high image acquisition speed while maintaining image quality for the task of radiotherapy guidance. In this study, we examine rapid dynamic MRI using a sparse sampling sequence k-t BLAST in capturing motion-induced, cycle-to-cycle variations in tumor position. We investigate the utility of long-term MRI-based motion monitoring as a means of better characterizing respiration-induced tumor motion compared to a single-cycle 4DCT. Methods: An MRI-compatible, programmable, deformable lung motion phantom with eleven 1.5 ml water marker tubes was placed inside a 3.0 T whole-body MR scanner (Philips Ingenia). The phantom was programmed with 10 lung tumor motion traces previously recorded using the Synchrony system. 2D+t image sequences of a coronal slice were acquired using a balanced-SSFP sequence combined with k-t BLAST (accn=3, resolution=0.66×0.66×5 mm3; acquisition time = 110 ms/slice). kV fluoroscopic (ground truth) and 4DCT imaging was performed with the same phantom setup and motion trajectories. Marker positions in all three modalities were segmented and tracked using an opensource deformable image registration package, NiftyReg. Results: Marker trajectories obtained from rapid MRI exhibited <1 mm error compared to kv Fluoro trajectories in the presence of complex motion including baseline shifts and changes in respiratory amplitude, indicating the ability of MRI to monitor motion with adequate geometric fidelity for the purpose of radiotherapy guidance. In contrast, the trajectory derived from 4DCT exhibited significant errors up to 6 mm due to cycle-to-cycle variations and baseline shifts. Consequently, 4DCT was found to underestimate the range of marker motion by as much as 50%. Conclusion: Dynamic MRI is a promising tool for radiotherapy motion management as it permits for longterm, dose-free, soft-tissue-based monitoring of motion, yielding richer and more accurate information about tumor position and motion range compared to the current state-of-the-art, 4DCT. This work was partially supported through research funding from National Institutes of Health (R01CA169102).

60 APPLIED LIFE SCIENCES↗

MO-DE-207A-11: Sparse-View CT Reconstruction Via a Novel Non-Local Means Method

Purpose: Sparse-view computed tomography (CT) reconstruction is an effective strategy to reduce the radiation dose delivered to patients. Due to its insufficiency of measurements, traditional non-local means (NLM) based reconstruction methods often lead to over-smoothness in image edges. To address this problem, an adaptive NLM reconstruction method based on rotational invariance (RIANLM) is proposed. Methods: The method consists of four steps: 1) Initializing parameters; 2) Algebraic reconstruction technique (ART) reconstruction using raw projection data; 3) Positivity constraint of the image reconstructed by ART; 4) Update reconstructed image by using RIANLM filtering. In RIANLM, a novel similarity metric that is rotational invariance is proposed and used to calculate the distance between two patches. In this way, any patch with similar structure but different orientation to the reference patch would win a relatively large weight to avoid over-smoothed image. Moreover, the parameter h in RIANLM which controls the decay of the weights is adaptive to avoid over-smoothness, while it in NLM is not adaptive during the whole reconstruction process. The proposed method is named as ART-RIANLM and validated on Shepp-Logan phantom and clinical projection data. Results: In our experiments, the searching neighborhood size is set to 15 by 15 and the similarity window is set to 3 by 3. For the simulated case with a resolution of 256 by 256 Shepp-Logan phantom, the ART-RIANLM produces higher SNR (35.38dB<24.00dB) and lower MAE (0.0006<0.0023) reconstructed image than ART-NLM. The visual inspection demonstrated that the proposed method could suppress artifacts or noises more effectively and preserve image edges better. Similar results were found for clinical data case. Conclusion: A novel ART-RIANLM method for sparse-view CT reconstruction is presented with superior image. Compared to the conventional ART-NLM method, the SNR and MAE from ART-RIANLM increases 47% and decreases 74%, respectively.

60 APPLIED LIFE SCIENCES↗

A Generalized Sampling and Preconditioning Scheme for Sparse Approximation of Polynomial Chaos Expansions

We propose an algorithm for recovering sparse orthogonal polynomial expansions via collocation. A standard sampling approach for recovering sparse polynomials uses Monte Carlo sampling, from the density of orthogonality, which results in poor function recovery when the polynomial degree is high. Our proposed approach aims to mitigate this limitation by sampling with respect to the weighted equilibrium measure of the parametric domain and subsequently solves a preconditioned $\ell^1$-minimization problem, where the weights of the diagonal preconditioning matrix are given by evaluations of the Christoffel function. Our algorithm can be applied to a wide class of orthogonal polynomial families on bounded and unbounded domains, including all classical families. We present theoretical analysis to motivate the algorithm and numerical results that show our method is superior to standard Monte Carlo methods in many situations of interest. In conclusion, numerical examples are also provided to demonstrate that our proposed algorithm leads to comparable or improved accuracy even when compared with Legendre- and Hermite-specific algorithms.

97 MATHEMATICS AND COMPUTING↗

Explicit Quantum Circuits for Block Encodings of Certain Sparse Matrices

Many standard linear algebra problems can be solved on a quantum computer by using recently developed quantum linear algebra algorithms that make use of block encodings and quantum eigenvalue/singular value transformations. A block encoding embeds a properly scaled matrix of interest A in a larger unitary transformation U that can be decomposed into a product of simpler unitaries and implemented efficiently on a quantum computer. Although quantum algorithms can potentially achieve exponential speedup in solving linear algebra problems compared to the best classical algorithm, such a gain in efficiency ultimately hinges on our ability to construct an efficient quantum circuit for the block encoding of A, which is difficult in general, and not trivial even for well structured sparse matrices. Here, in this paper, we give a few examples on how efficient quantum circuits can be explicitly constructed for some well structured sparse matrices and discuss a few strategies used in these constructions. We also provide implementations of these quantum circuits in MATLAB.

97 MATHEMATICS AND COMPUTING↗

Sparse Hardware Embedding of Spiking Neuron Systems for Community Detection

We study the applicability of spiking neural networks and neuromorphic hardware for solving general optimization problems without the use of adaptive training or learning algorithms. We leverage the dynamics of Hopfield networks and spin-glass systems to construct a fully connected spiking neural system to generate synchronous spike responses indicative of the underlying community structure in an undirected, unweighted graph. Mapping this fully connected system to current generation neuromorphic hardware is done by embedding sparse tree graphs to generate only the leading-order spiking dynamics. Here, we demonstrate that for a chosen set of benchmark graphs, the spike responses generated on a current generation neuromorphic processor can improve the stability of graph partitions and non-overlapping communities can be identified even with the loss of higher-order spiking behavior if the graphs are sufficiently dense. For sparse graphs, the loss of higher-order spiking behavior improves the stability of certain graph partitions but does not retrieve the known community memberships.

97 MATHEMATICS AND COMPUTING↗

A Sparse Tensor Benchmark Suite for CPUs and GPUs

Tensor computations present significant performance chal- lenges that impact a wide spectrum of applications ranging from machine learning, healthcare analytics, social network analysis, data mining to quantum chemistry and signal processing. Efforts to improve the perfor- mance of tensor computations include exploring data layout, execution scheduling, and parallelism in common tensor kernels. This work presents a benchmark suite for arbitrary-order sparse tensor kernels using state- of-the-art tensor formats: coordinate (COO) and hierarchical coordinate (HiCOO) on CPUs and GPUs. It presents a set of reference tensor kernel implementations that are compatible with real-world tensors and power law tensors extended from synthetic graph generation techniques. We also propose Roofline performance models for these kernels to provide insights of computer platforms from sparse tensor view. This benchmark suite along with the synthetic tensor generator is publicly available.

Li, Jiajia↗

Fast and Scalable Sparse Triangular Solver for Multi-GPU Based HPC Architectures

Designing efficient and scalable sparse linear algebra kernels on modern multi-GPU based HPC systems is a daunting task due to significant irregular memory references and workload imbalance across the GPUs. This is particularly the case for \textit{Sparse Triangular Solver (SpTRSV)} which introduces additional two-dimensional computation dependencies among subsequent computation steps. Dependency information is exchanged and shared among GPUs, thus warrant for efficient memory allocation, data partitioning, and workload distribution as well as fine-grained communication and synchronization support. In this work, we demonstrate that directly adopting unified memory can adversely affect the performance of SpTRSV on multi-GPU architectures, despite linking via fast interconnect like NVLinks and NVSwitches. Alternatively, we employ the latest NVSHMEM technology based on Partitioned Global Address Space programming model to enable efficient fine-grained communication and drastic synchronization overhead reduction. Furthermore, to handle workload imbalance, we propose a malleable task-pool execution model which can further enhance the utilization of GPUs. By applying these techniques, our experiments on the NVIDIA multi-GPU supernode V100-DGX-1 and DGX-2 systems demonstrate that our design can achieve on average 3.53x (up to 9.86x) speedup on a DGX-1 system and 3.66x (up to 9.64x) speedup on a DGX-2 system with 4-GPUs over the Unified-Memory design. The comprehensive sensitivity and scalability studies also show that the proposed zero-copy SpTRSV is able to fully utilize the computing and communication resources of the multi-GPU system.

Xie, Chenhao↗

A Length Adaptive Algorithm-Hardware Co-design of Transformer on FPGA Through Sparse Attention and Dynamic Pipelining

Transformers are considered one of the most important deep learning models since 2018, in part because it establishes state-of-the-art (SOTA) records and could potentially replace existing Deep Neural Networks (DNNs). Despite the remarkable triumphs, the prolonged turnaround time of Transformer models is a widely recognized roadblock. The variety of sequence lengths imposes additional computing overhead where inputs need to be zero-padded to the maximum sentence length in the batch to accommodate the parallel computing platforms. This paper targets the field-programmable gate array (FPGA) and proposes a coherent sequence length adaptive algorithm–hardware co-design for Transformer acceleration. Particularly, we develop a hardware-friendly sparse attention operator and a length-aware hardware resource scheduling algorithm. The proposed sparse attention operator brings the complexity of attention-based models down to linear complexity and alleviates the off-chip memory traffic. The proposed length-aware resource hardware scheduling algorithm dynamically allocates the hardware resources to fill up the pipeline slots and eliminates bubbles for NLP tasks. Experiments show that our design has very small accuracy loss and has 80.2 × and 2.6 × speedup compared to CPU and GPU implementation, and 4 × higher energy efficiency than state-of-the-art GPU accelerator optimized via CUBLAS GEMM.

Peng, Hongwu↗

Brief Announcement: Communication Optimal Sparse LU Factorization for Planar Matrices

We introduce a new parallel algorithm for solving sparse LU factorization of planar matrices, which commonly arise in the finite element method for 2D PDEs. Existing scalable methods, such as the multifrontal approach with subtree-to-subcube mapping by Gupta et al. [1] and right-looking with 3D mapping by Sao et al. [2] fail to achieve optimal communication costs for these matrices. Our new algorithm combines 3D mapping and subtree-to-subcube mapping to minimize communication costs while allowing trade-offs between extra memory and reduced communication. We demonstrate that our proposed algorithm attains the communication lower bound up to a factor of O(log log n) in the memory-optimal case and up to a factor of O(log P) in the memory-independent case for an n-dimensional planar sparse matrix on P processors.

Sao, Piyush↗

A Novel Fault Diagnosis Approach for Rolling Bearing Based on CWT and Adaptive Sparse Representation

Extraction and enhancement of weak impulse signature is the key of rolling bearing fault prognostics in which case the features are often weak and covered by noise. Tunable Q-factor wavelet transform (TQWT), as an emerging wavelet construction theory developed in a frequency domain explicitly, has the advantages of matching with the specific oscillation behavior of signal components. In this article, an adaptive sparse representation (ASR) method is proposed, which integrates the sparse code shrinkage (SCS) and parameter optimization into TQWT. However, direct application of ASR is difficult to extract fault signatures at the early stage or low-speed operation due to weak fault symptoms and background noise. A novel fault diagnosis strategy based on continuous wavelet transform (CWT) and ASR is investigated. CWT owns significant advantages on multiscale subdivision and weak signal detection. The results of simulated and experimental vibration signal analyses verify the effectiveness of the proposed method in accurately extracting weak impulse features from the noise environment.

Yuan, Xing↗

Amesos2 Templated Direct Sparse Solver Package

Amesos2 is a templated direct sparse solver package. Amesos2 provides interfaces to direct sparse solvers, rather than providing native solver capabilities. Amesos2 is a derivative work of the Trilinos package Amesos.

Bavier, Eric↗