3D Kernel Research Articles

Analysis of temporal and spatial dynamics and driving factors in the aquaculture industry of Fuding City, China

BackgroundThe sustainable development of aquaculture is a crucial determinant of food security, the well-being of aquaculture practitioners, and economic growth within coastal regions. Considering the existing gap in research regarding spatial and temporal development of aquaculture, this study investigates the progression of aquaculture practices over time and across various locations in Fuding City, China. MethodsWe retrospectively collected temporal and spatial data on aquaculture, as well as demographic, social, and economic data for Fuding City from 2010 to 2020. By employing 3D kernel density analysis, we illustrated the temporal and spatial changes in aquaculture. Furthermore, we utilized Ordinary Least Squares regression to investigate the driving factors behind the spatial changes in the aquaculture industry. ResultsOver the past decades, we observed that in Fuding City, both the number of fishing rafts and aquaculture households initially decreased and then increased. The spatial distribution of aquaculture experienced a shift from the west (inner bay area) to the east (coastal area). Additionally, the type of fishing rafts also varied by region, with traditional rafts dominating the western inner bay and plastic rafts prevalent in the eastern offshore areas. Analysis of driving factors revealed that at least six factors have a significant positive correlation with the eastward shift of the aquaculture industry's center, including the proportion of migrant population, proportion of aquaculture to total fishery output, average temperature, investment in aquaculture technology, total fish sales, and GDP of Fuding City. ConclusionThis study examines the spatial and temporal dynamics of aquaculture in Fuding City from 2010 to 2020, proposing an innovative approach to spatial optimization that integrates both horizontal and vertical strategies. These insights aim to guide the development of coastal aquaculture policies and support sustainable regional development, fostering a balanced coexistence between human activities and marine environments.

3D geometric kernel computation in polygon mesh structures

This paper introduces a novel approach to compute the geometric kernel of a polygon mesh embedded in 3D. The geometric kernel defines the set of points inside or on the shape’s boundary, ensuring visibility of the entire shape. The proposed method utilizes scattered rays to identify a sufficient number of sample points on the kernel surface and subsequently leverages these points to locate as many surface vertices as possible. By computing the convex hull of these identified points, we derive an approximation of the kernel. Notably, the output of our method consists exclusively of interior or boundary points of the actual kernel. Comparative evaluations against established CGAL and Polyhedron Kernel algorithms highlight our method’s superior computational speed and high approximation accuracy. The parametric structure of our solution allows for different levels of accuracy to be obtained, enabling the user to tailor the approximation to their specific needs. This property sets our algorithm apart from others and provides greater flexibility in its use. Additionally, adjusting the algorithmic settings also enables the computation of the kernel itself with a trade-off in computational speed. Furthermore, our algorithm swiftly and accurately identifies an empty kernel for non-star-shaped configurations.

Characterizing the voxel-based approaches in radioembolization dosimetry with reDoseMC.

Yttrium-90 ( ) represents the primary radioisotope used in radioembolization procedures, while holmium-166 ( ) is hypothesized to serve as a viable substitute for due to its comparable therapeutic potential and improved quantitative imaging. Voxel-based dosimetry for these radioisotopes relies on activity images obtained through PET or SPECT and dosimetry methods, including the voxel S-value (VSV) and the local deposition method (LDM). However, the evaluation of the accuracy of absorbed dose calculations has been limited by the use of non-ideal reference standards and investigations restricted to the liver. The objective of this study was to expand upon these dosimetry characterizations by investigating the impact of image resolutions, voxel sizes, target volumes, and tissue materials on the accuracy of and dosimetry techniques. A specialized radiopharmaceutical dosimetry software called reDoseMC was developed using the Geant4 Monte Carlo toolkit and validated by benchmarking the generated kernels with published data. The decay spectra of both and were also compared. Multiple VSV kernels were generated for the liver, lungs, soft tissue, and bone for isotropic voxel sizes of 1 mm, 2 mm, and 4 mm. Three theoretical phantom setups were created with 20 or 40mm activity and mass density inserts for the same three voxel sizes. To replicate the limited spatial resolutions present in PET and SPECT images, image resolutions were modeled using a 3D Gaussian kernel with a Full Width at Half Maximum (FWHM) ranging from 0 to 16mm and with no added noise. The VSV and LDM dosimetry methods were evaluated by characterizing their respective kernels and analyzing their absorbed dose estimates calculated on theoretical phantoms. The ground truth for these estimations was calculated usingreDoseMC. The decay spectra obtained through reDoseMC showed less than a 1% difference when compared to previously published experimental data for energies below 1.9MeV in the case of and less than 1% for energies below 1.5MeV for . Additionally, the validation kernels for VSV exhibited results similar to those found in published Monte Carlo codes, with source dose depositions having less than a 3% error margin. Resolution thresholds ( ), defined as resolutions that resulted in similar dose estimates between the LDM and VSV methods, were observed for . They were 1.5mm for bone, 2.5 mm for soft tissue and liver, and 8.5 mm for lungs. For , the accuracy of absorbed dose deposition was found to be dependent on the contributions of absorbed dose from photons. Volume errors due to variations in voxel size impacted the final dose estimates. Larger target volumes yielded more accurate mean doses than smaller volumes. For both radioisotopes, the radial dose profiles for the VSV and LDM approximated but never matched the referencestandard. reDoseMC was developed and validated for radiopharmaceutical dosimetry. The accuracy of voxel-based dosimetry was found to vary widely with changes in image resolutions, voxel sizes, chosen target volumes, and tissue material; hence, the standardization of dosimetry protocols was found to be of great importance for comparable dosimetryanalysis.

Efficient multiscale spatial attention 3D abdominal multiorgan segmentation model

AbstractAccurate and efficient 3D medical image segmentation models are critical for clinical applications. However, many models which performed well in recent years have consumed significant computational cost, especially those based on self‐attention. For more efficient and accurate segmentation of medical 3D images, we develop an efficient multiscale spatial attention model for 3D abdominal multiorgan segmentation. We use 1D and 2D kernels in the encoder and decoder modules to simulate 3D kernels to reduce the number of parameters and computational cost. Furthermore, we use the lightweight spatial attention and mixed pooling module to capture cross‐dimensional information, multiscale features, and long‐range contextual information. We achieved a Dice Similarity Coefficient (DSC) of 89.86 and a Normalized Surface Dice (NSD) of 78.2 on the FLARE dataset with only 1183 M GPU memory usage, 9 M parameters, and 193G floating point operations(FLOPs).

Feature selection based on unsupervised clustering evaluation for predicting neoadjuvant chemoradiation response for patients with locally advanced rectal cancer

Accurate response prediction allows for personalized cancer treatment of locally advanced rectal cancer (LARC) with neoadjuvant chemoradiation. In this work, we designed a convolutional neural network (CNN) feature extractor with switchable 3D and 2D convolutional kernels to extract deep learning features for response prediction. Compared with radiomics features, convolutional kernels may adaptively extract local or global image features from multi-modal MR sequences without the need of feature predefinition. We then developed an unsupervised clustering based evaluation method to improve the feature selection operation in the feature space formed by the combination of CNN features and radiomics features. While normal process of feature selection generally includes the operations of classifier training and classification execution, the process needs to be repeated many times after new feature combinations were found to evaluate the model performance, which incurs a significant time cost. To address this issue, we proposed a cost effective process to use a constructed unsupervised clustering analysis indicator to replace the classifier training process by indirectly evaluating the quality of new found feature combinations in feature selection process. We evaluated the proposed method using 43 LARC patients underwent neoadjuvant chemoradiation. Our prediction model achieved accuracy, area-under-curve (AUC), sensitivity and specificity of 0.852, 0.871, 0.868, and 0.735 respectively. Compared with traditional radiomics methods, the prediction models (AUC = 0.846) based on deep learning-based feature sets are significantly better than traditional radiomics methods (AUC = 0.714). The experiments also showed following findings: (1) the features with higher predictive power are mainly from high-order abstract features extracted by CNN on ADC images and T2 images; (2) both ADC_Radiomics and ADC_CNN features are more advantageous for predicting treatment responses than the radiomics and CNN features extracted from T2 images; (3) 3D CNN features are more effective than 2D CNN features in the treatment response prediction. The proposed unsupervised clustering indicator is feasible with low computational cost, which facilitates the discovery of valuable solutions by highlighting the correlation and complementarity between different types of features.

3D-KCPNet: Efficient 3DCNNs based on tensor mapping theory

As the deep neural networks (DNNs) with satisfied expression ability usually require large scale to gain adequate performance, and deploying large-scale DNNs on resource-limited environments is still a challenge, neural network compression becomes a hot topic nowadays. Among the multiple compression methods, tensor decomposition reveals many specific advantages, such as regular data structure comes from linear algebra, convenient approach of training from scratch, ideal compression ratio, etc. Nevertheless, for some compact neural modules such as two-dimensional/three-dimensional (2D/3D) convolutional kernels, traditional tensor decomposition in the way of approximation has encountered intractable obstacles. Fortunately, some works utilize the tensor mapping approach, which just regards the data structure of tensor decomposition as neural layers, to reconstruct the convolutional kernel into a new lightweight module with several thinner convolutional kernels. The only two flies in the ointment are there lack necessary theories of tensor mapping, and there is still no tensor mapping way to compress high-order three-dimensional convolutional neural networks (3DCNNs). In this paper, we first deeply analyze the tensor mapping theory including convergence and precision, which separately establishes the rationality of tensor mapping and its superiority over the traditional tensor approximation, according to the Lottery Ticket Hypothesis. Then we propose an efficient method termed as 3D-KCPNet, to compress 3DCNNs based on the novel Kronecker canonical polyadic (KCP) tensor decomposition. The proposed method can not only compress the 3D convolutional kernels, but also convert a 3D convolution to efficient 1 × 1 × 1 and 2D depthwise convolutions. The experiments on the video recognition datasets VIVA Challenge, UCF11, UCF50, and UCF101 show that the accuracy of 3D-KCPNet can surpass its original baseline model and the corresponding tensor approximation model.

Accelerating CEST imaging using a model-based deep neural network with synthetic training data.

To develop a model-based deep neural network for high-quality image reconstruction of undersampled multi-coil CEST data. Inspired by the variational network (VN), the CEST image reconstruction equation is unrolled into a deep neural network (CEST-VN) with a k-space data-sharing block that takes advantage of the inherent redundancy in adjacent CEST frames and 3D spatial-frequential convolution kernels that exploit correlations in the x-ω domain. Additionally, a new pipeline based on multiple-pool Bloch-McConnell simulations is devised to synthesize multi-coil CEST data from publicly available anatomical MRI data. The proposed network is trained on simulated data with a CEST-specific loss function that jointly measures the structural and CEST contrast. The performance of CEST-VN was evaluated on four healthy volunteers and five brain tumor patients using retrospectively or prospectively undersampled data with various acceleration factors, and then compared with other conventional and state-of-the-art reconstruction methods. The proposed CEST-VN method generated high-quality CEST source images and amide proton transfer-weighted maps in healthy and brain tumor subjects, consistently outperforming GRAPPA, blind compressed sensing, and the original VN. With the acceleration factors increasing from 3 to 6, CEST-VN with the same hyperparameters yielded similar and accurate reconstruction without apparent loss of details or increase of artifacts. The ablation studies confirmed the effectiveness of the CEST-specific loss function and data-sharing block used. The proposed CEST-VN method can offer high-quality CEST source images and amide proton transfer-weighted maps from highly undersampled multi-coil data by integrating the deep learning prior and multi-coil sensitivity encoding model.

Gate-Shift-Fuse for Video Action Recognition.

Convolutional Neural Networks are the de facto models for image recognition. However 3D CNNs, the straight forward extension of 2D CNNs for video recognition, have not achieved the same success on standard action recognition benchmarks. One of the main reasons for this reduced performance of 3D CNNs is the increased computational complexity requiring large scale annotated datasets to train them in scale. 3D kernel factorization approaches have been proposed to reduce the complexity of 3D CNNs. Existing kernel factorization approaches follow hand-designed and hard-wired techniques. In this paper we propose Gate-Shift-Fuse (GSF), a novel spatio-temporal feature extraction module which controls interactions in spatio-temporal decomposition and learns to adaptively route features through time and combine them in a data dependent manner. GSF leverages grouped spatial gating to decompose input tensor and channel weighting to fuse the decomposed tensors. GSF can be inserted into existing 2D CNNs to convert them into an efficient and high performing spatio-temporal feature extractor, with negligible parameter and compute overhead. We perform an extensive analysis of GSF using two popular 2D CNN families and achieve state-of-the-art or competitive performance on five standard action recognition benchmarks.

3D Sensitivity Kernels With Full Attenuation Computed by a Combination of the Strong Stability Preserving Runge‐Kutta Method and the Scattering‐Integral Method

AbstractTo date developments of seismic attenuation models of the Earth have lagged those of velocity models. This is partly due to difficulties in isolating waveform perturbations caused by attenuation and velocity heterogeneities and partly due to different theories (e.g., ray theory, finite frequency theory) used in most previous and current studies to approximate sensitivity kernels and invert for the attenuation structure. We present in this paper a new method for computing the 3D waveform Fréchet kernels that account for full physical‐dispersion and dissipation attenuation. For solving the 3D isotropic anelastic wave equation described by the generalized Maxwell body model, we extend our previously proposed full waveform modeling method in Cartesian coordinates to that in spherical coordinates, which can provide stable numerical solutions even in the presence of strong attenuation. Then, we apply the scattering‐integral method for calculating 3D travel time and amplitude sensitivity kernels with respect to velocity and attenuation structures. We demonstrate the accuracy of our forward method and the effectiveness of the implementation of absorbing and free surface boundary conditions through numerical tests. Moreover, by choosing the Northwestern United States region as a realistic example, we verify the accuracy of the computed 3D sensitivity kernels through comparing the waveform measurements with predictions from the kernels. Finally, we discuss the importance of calculating full anelastic sensitivity kernels including both effects of physical dispersion and dissipation, where we specially explore the effect of scattering due to random velocity and attenuation heterogeneities on waveform measurements.

Analysis of the Recursive Locally-Adaptive Filtration of 3D Tensor Images

This work is focused on the computational complexity (CC) reduction of the locally-adaptive processing of 3D tensor images, based on recursive approaches. As a basis, a local averaging operation is used, implemented as a sliding mean 3D filter (SM3DF) with a central symmetric 3D kernel. Symmetry plays a very important role in constructing the working window. The presented theoretical approach could be adopted in various algorithms for locally-adaptive processing, such as additive noise reduction, sharpness enhancement, texture segmentation, etc. The basic characteristics of the recursive SM3DF are analyzed, together with the main features of the adaptive algorithms for filtration of Gaussian noises and unsharp masking where the filter is aimed at. In the paper, the ability of SM3DF implementation through recursive sliding mean 1D filters, sequentially bonded together, is also introduced. The computational complexity of the algorithms is evaluated for the recursive and non-recursive mode. The recursive SM3DF also suits the 3D convolutional neural networks which comprise sliding locally-adaptive 3D filtration in their layers. As a result of the lower CC, a promising opportunity is opened for higher efficiency of the 3D image processing through tensor neural networks.

Analytical expressions for singular integrals arising from the 3D Laplace and Stokes kernels when using constant or linear triangular and quadrilateral boundary elements

In this paper we present analytical expressions for the computation of singular integrals obtained in the discretisation of boundary integral equations for the Laplace and creeping flow (Stokes) problems with triangular or quadrilateral boundary elements with linear interpolation of the potential and constant interpolation of the flux. We compare the singular integrals computed with the presented analytical expressions with the same integrals computed with numerical quadrature and find that a considerably larger computational effort has to be made for numerical quadrature to achieve high accuracy than with the analytical expression. Furthermore, we show that the accuracy in solving a Laplace test case and a creeping flow test case using analytical expressions for singular integrals is better than the accuracy achieved with numerical quadrature. The analytical expressions are listed in the appendix of the paper and their implementation in computer code is available online.

Human activity recognition from UAV videos using a novel DMLC-CNN model

Human Activity Recognition (HAR) remains a challenging issue that requires to be resolved. Utilizing images, smart phones, or sensors, HAR could be done. Recent studies have explored HAR utilizing Unmanned Aerial Vehicle (UAV) videos. But, owing to numerous restrictions correlated to the platform, HAR from videos captured by drones remains a challenge. To resolve this issue, a novel mechanism for HAR from UAV videos utilizing the Diminutive Multi-Dimensional Locality Coding based Convolutional Neural Network (DMLC-CNN) model is proposed. Primarily, the input video is converted into frames; also, Bounding Boxes (BBs) are produced for humans. For attaining the number of clusters grounded on human presence, the BB is given to the Robust Lucrative Sensitive Amalgam K-Nearest Neighbor (RoLSA-KNN) clustering. After that, utilizing the LR-fit-centric Superior Photogrammetric Silhouette (LR-SPS) segmentation, the humans' silhouettes are segmented. The attained output is divided into patches; then, it is converted into stacked layer frames. Afterward, these frames are converted into a 3D kernel. The Diminutive Multi-Dimensional Hyper Graph (DMHG) extracted the features; then, the Locality-constrained Orthonormal Feature Space Coding (LOFSC) Algorithm encoded the extracted features. Lastly, for recognizing Human Activity (HA), DMLC-CNN is wielded. The proposed framework is contrasted with the conventional techniques. After experimental evaluation, the proposed mechanism was found to be more efficient in HAR.

Development of a collapsed cone convolution/superposition dose calculation algorithm with a mass density-specific water kernel for magnetic resonance-guided radiotherapy.

This study aimed to develop and validate a collapsed cone convolution for magnetic resonance-guided radiotherapy (CCCMR). The 3D energy deposition kernels (EDKs) were generated in water in a 1.5-T transverse magnetic field. The CCCMR corrects the inhomogeneity in simulation geometry by referring to the EDKs according to the mass density between the interaction and energy deposition points in addition to density scaling. Dose distributions in a water phantom and in slab phantoms with inserted inhomogeneities were calculated using the Monte Carlo (MC) and CCCMR. The percentage depth dose (PDD) and off-axis ratio (OAR) were compared, and the gamma passing rate (3%/2mm) was evaluated. The CCCMR simulated asymmetric dose distributions in the simulation phantoms, especially the water phantom, and all PDD and OAR profiles were in good agreement with the findings of the MC. The gamma passing rates were >99% for each field size and for the entire region. In the inhomogeneity phantoms, although the CCCMR underestimated dose in the low mass density regions, it could reconstruct dose changes at mass density boundaries. The gamma passing rate for the entire region was >95% for the field size of 2 × 2cm2, but it was 68.9-86.7% for the field sizes of ≥5 × 5cm2. Conclusively, in water, the CCCMR can obtain dose distributions comparable to those with the MC. Although the dose differences between them were mainly in inhomogeneity regions, the possibility of the effective use of the CCCMR in small field sizes was demonstrated.

Area-Efficient Mapping of Convolutional Neural Networks to Memristor Crossbars Using Sub-Image Partitioning.

Memristor crossbars can be very useful for realizing edge-intelligence hardware, because the neural networks implemented by memristor crossbars can save significantly more computing energy and layout area than the conventional CMOS (complementary metal-oxide-semiconductor) digital circuits. One of the important operations used in neural networks is convolution. For performing the convolution by memristor crossbars, the full image should be partitioned into several sub-images. By doing so, each sub-image convolution can be mapped to small-size unit crossbars, of which the size should be defined as 128 × 128 or 256 × 256 to avoid the line resistance problem caused from large-size crossbars. In this paper, various convolution schemes with 3D, 2D, and 1D kernels are analyzed and compared in terms of neural network's performance and overlapping overhead. The neural network's simulation indicates that the 2D + 1D kernels can perform the sub-image convolution using a much smaller number of unit crossbars with less rate loss than the 3D kernels. When the CIFAR-10 dataset is tested, the mapping of sub-image convolution of 2D + 1D kernels to crossbars shows that the number of unit crossbars can be reduced almost by 90% and 95%, respectively, for 128 × 128 and 256 × 256 crossbars, compared with the 3D kernels. On the contrary, the rate loss of 2D + 1D kernels can be less than 2%. To improve the neural network's performance more, the 2D + 1D kernels can be combined with 3D kernels in one neural network. When the normalized ratio of 2D + 1D layers is around 0.5, the neural network's performance indicates very little rate loss compared to when the normalized ratio of 2D + 1D layers is zero. However, the number of unit crossbars for the normalized ratio = 0.5 can be reduced by half compared with that for the normalized ratio = 0.

Prediction of Peak Pressure by Blast Wave Propagation Between Buildings Using a Conditional 3D Convolutional Neural Network

To predict the damage resulting from an explosion in the middle of a city, where buildings are concentrated, the peak pressure reaching the walls of the buildings or in between buildings should be accurately and rapidly calculated. However, predicting peak pressure between buildings is known to be very difficult because of the diffraction and reflection of blast waves, which have generally been analyzed by numerical analysis methods. However, numerical analysis is not suitable in a military operation environment which requires rapid analysis, because it takes considerable time and resources. This study proposes a deep neural network that quickly and accurately predicts the peak pressure caused by the propagation of blast waves, for the effective analysis of weapon effectiveness and damage in urban environments. The proposed deep learning model is based on a 3-dimensional convolutional neural network (3D CNN) model that processes the spatial information of explosion and measurement in the 3D spaces using 3D kernels. To predict the peak pressure between buildings separated by an arbitrary distance using a single model, we also propose using conditional convolution, which modulates the prediction output according to the building distance. The proposed models were trained with a dataset constructed through finite element analysis with various building distances, explosion locations, and explosive weights. The experiment with a fixed building distance showed that the relative error of the proposed 3D CNN is less than 7%, which is 2.5 times more accurate than a simple multi-layer perceptron (MLP) model. For unseen building layouts, the conditional 3D convolution showed 3.6 times lower error than the MLP model, demonstrating the effectiveness of the conditional convolution for prediction in arbitrary building layouts. Most importantly, the proposed deep learning models took less than one minute per prediction, which is significantly faster than finite element analysis, which takes 6 to 8 hours to analyze a single simulation case.

3-D Structure-Oriented Adaptive Gaussian Pyramid for Seismic Multiscale Fracture Detection

Gaussian pyramid (GP) has the property of decomposing the 2D data into multiple scales. It has been successfully applied in image processing, but rarely used in seismic data processing and interpretation. Specifically, seismic data is not just a 2D image in one time slice or horizon. Moreover, the pattern of geological body varies rapidly in 3D spaces. Hence, the classical GP method exists a limitation in processing 3D seismic data. Especially in fracture detection, the 2D isotropic Gaussian kernel used in GP tends to blur the fracture details. In this letter, to match the high dimensional seismic data, we propose a structure-oriented adaptive Gaussian pyramid (SOA-GP) algorithm, which expand the 2D GP method to 3D situation. In this case, to eliminate the influence of transverse change of wave impedance and consider stratigraphic characteristics, the 2D Gaussian filter is also substituted by the 3D structure adaptive anisotropic Gaussian kernel, which is constructed by instantaneous phase-based gradient structure tensor (GST) method. Meanwhile, seismic attributes are applied to the multi-resolution seismic data decomposed by SOA-GP method to realize multi-scale fracture detection. Finally, we apply this new 3D SOA-GP method to a field data. It demonstrates that the multi-scale decomposition of the new method could produce more details of fracture and less disturbs from noise.

Large time behavior of solutions to the 3D anisotropic Navier–Stokes equation

We consider the large time behavior of the solution to the 3D Navier–Stokes equation with horizontal viscosity Δhu=∂12u+∂22u and show that the Lp decay rate of the horizontal components of the velocity field coincides to that of the 2D heat kernel, while the vertical component decays like the 3D heat kernel. Moreover, we consider the asymptotic expansion of the solution and find that a portion of the nonlinear term affect the leading term of the horizontal components of the velocity field, whereas the leading term of the vertical component is given by only the linear solution.

GRASPNET: Fast spatiotemporal deep learning reconstruction of golden-angle radial data for free-breathing dynamic contrast-enhanced magnetic resonance imaging.

The purpose of the current study was to develop a deep learning technique called Golden-angle RAdial Sparse Parallel Network (GRASPnet) for fast reconstruction of dynamic contrast-enhanced 4D MRI acquired with golden-angle radial k-space trajectories. GRASPnet operates in the image-time space and does not use explicit data consistency to minimize the reconstruction time. Three different network architectures were developed: (1) GRASPnet-2D: 2D convolutional kernels (x,y) and coil and contrast dimensions collapsed into a single combined dimension; (2) GRASPnet-3D: 3D kernels (x,y,t); and (3) GRASPnet-2D + time: two 3D kernels to first exploit spatial correlations (x,y,1) followed by temporal correlations (1,1,t). The networks were trained using iterative GRASP reconstruction as the reference. Free-breathing 3D abdominal imaging with contrast injection was performed on 33 patients with liver lesions using a T1-weighted golden-angle stack-of-stars pulse sequence. Ten datasets were used for testing. The three GRASPnet architectures were compared with iterative GRASP results using quantitative and qualitative analysis, including impressions from two body radiologists. The three GRASPnet techniques reduced the reconstruction time to about 13 s with similar results with respect to iterative GRASP. Among the GRASPnet techniques, GRASPnet-2D + time compared favorably in the quantitative analysis. Spatiotemporal deep learning enables reconstruction of dynamic 4D contrast-enhanced images in a few seconds, which would facilitate translation to clinical practice of compressed sensing methods that are currently limited by long reconstruction times.

A treecode algorithm based on tricubic interpolation

Treecode algorithms efficiently approximate N-body interactions in O(N) or O(NlogN). In order to treat general 3D kernels, recent developments employ polynomial interpolation to approximate the kernels. The polynomials are a tensor product of 1-dimensional polynomials. Here, we develop an O(NlogN) tricubic interpolation based treecode method for 3D kernels. The tricubic interpolation is inherently three-dimensional and as such does not employ a tensor product. The form allows for easy evaluation of the derivatives of the kernel, required in dynamical simulations, which is not the case for the tensor product approach. We develop both a particle-cluster and cluster-particle variants and present results for the Coulomb, screened Coulomb and the real space Ewald kernels. We also present results of an MD simulation of a Lennard-Jones liquid using the tricubic treecode.

A novel strategy for classifying spectral-spatial shallow and deep hyperspectral image features using 1D-EWT and 3D-CNN

Hyperspectral images (HIs) are used in diverse disciplines, such as resource handling, land cover analysis, food science, anomaly detection, and precision agriculture. Researchers have been working on a number of visual processing and machine intelligence algorithms to handle this type of data as efficiently as feasible. Deep learning approaches have advanced significantly in the field of machine vision, which is also having a big impact on the analysis of hyperspectral data. To increase its discriminative potential for HI classification, this work suggests a powerful 3D-CNNs (Convolutional Neural Networks) architecture, where the shallow features extracted using 1D-EWT (Empirical Wavelet Transform) are served as input, and the ultimate output of the CNN are projected class-related outcomes. The framework is known as PEC, where P stands for PCA, E for 1D-EWT, and C for 3D-CNN model. Prior to features extraction, the HI undergoes spectral dimension reduction via Principal Component Anaalysis (PCA). To forecast segmentation for a volumetric area of a 3D HI sample, 3D CNNs use 3D convolutional kernels. The usage of more parameters by these CNNs means that the ability to employ interslice context that can boost performance. The CNN model’s parameters are optimised using a limited training set. The newly proposed PEC framework achieves a considerable overall accuracy of 99.58% and 99.94% (percentage increase of 0.08% and 0.54%) and Kappa score of 99.51% and 99.92 (percentage increase of 1.5% and 0.8%) compared to similar approaches for two real-world data sets IP and PU respectively with 30% training samples.