Structural Features Research Articles

EGST: Enhanced Geometric Structure Transformer for Point Cloud Registration.

We explore the effect of geometric structure descriptors on extracting reliable correspondences and obtaining accurate registration for point cloud registration. The point cloud registration task involves the estimation of rigid transformation motion in unorganized point cloud, hence it is crucial to capture the contextual features of the geometric structure in point cloud. Recent coordinates-only methods ignore numerous geometric information in the point cloud which weaken ability to express the global context. We propose Enhanced Geometric Structure Transformer to learn enhanced contextual features of the geometric structure in point cloud and model the structure consistency between point clouds for extracting reliable correspondences, which encodes three explicit enhanced geometric structures and provides significant cues for point cloud registration. More importantly, we report empirical results that Enhanced Geometric Structure Transformer can learn meaningful geometric structure features using none of the following: (i) explicit positional embeddings, (ii) additional feature exchange module such as cross-attention, which can simplify network structure compared with plain Transformer. Extensive experiments on the synthetic dataset and real-world datasets illustrate that our method can achieve competitive results.

Numerical simulation of the influence of wave parameters on the horizontal cylinder in the nonlinear coupling of the wave–current

Considering that marine structures are frequently subjected to the combined effects of waves and currents, the wave–current coupling environment largely determines the structural load and the surrounding water–air mixed flow, which is a typical feature of offshore structures, such as ships and offshore platforms. This study focuses on the interaction between a horizontal cylinder and a free surface in a wave–current nonlinear coupled environment using numerical simulations. The numerical method is based on the incompressible Navier–Stokes equation, with the volume-of-fluid method used to capture the free surface. Based on the second-order Stokes wave theory, we studied the impact of wave height and steepness on the cylinder force, vorticity field, free-surface deformation, and spatial–temporal distribution characteristics of entrainment bubbles. The results showed that the wave height and wave steepness have opposite effects on the root mean square (RMS) value of the force and influence the amplitude and period of the force curve. The stretching of the negative vortex led to varying degrees of double-frequency oscillation modes in the force curves. The main sources of bubbles in the wake are the breaking of the free surface and the entrainment caused by the cylinder vortex, and the bubbles caused by the former account for the majority. Compared with the wave height, an increase in wave steepness can cause a more severe interface breaking, resulting in more entrainment bubbles.

Towards Inductive and Efficient Explanations for Graph Neural Networks.

Despite recent progress in Graph Neural Networks (GNNs), explaining predictions made by GNNs remains a challenging and nascent problem. The leading method mainly considers the local explanations, i.e., important subgraph structure and node features, to interpret why a GNN model makes the prediction for a single instance, e.g. a node or a graph. As a result, the explanation generated is painstakingly customized at the instance level. The unique explanation interpreting each instance independently is not sufficient to provide a global understanding of the learned GNN model, leading to the lack of generalizability and hindering it from being used in the inductive setting. Besides, training the explanation model explaining for each instance is time-consuming for large-scale real-life datasets. In this study, we address these key challenges and propose PGExplainer, a parameterized explainer for GNNs. PGExplainer adopts a deep neural network to parameterize the generation process of explanations, which renders PGExplainer a natural approach to multi-instance explanations. Compared to the existing work, PGExplainer has better generalization ability and can be utilized in an inductive setting without training the model for new instances. Thus, PGExplainer is much more efficient than the leading method with significant speed-up. In addition, the explanation networks can also be utilized as a regularizer to improve the generalization power of existing GNNs when jointly trained with downstream tasks. Experiments on both synthetic and real-life datasets show highly competitive performance with up to 24.7% relative improvement in AUC on explaining graph classification over the leading baseline.

Private housing compliance with public seismic safety measures after 2015 Gorkha earthquake in Nepal

After the Gorkha earthquake in 2015, the Nepalese government introduced a series of seismic-resistant building designs to reconstruct and repair residential dwellings. The objective of this research was to explore the extent of compliance with the new prescribed or recommended seismic-resistant features in rebuilt residential structures in two Nepal districts that were severely damaged during the earthquake. The primary source of data was a field survey conducted with a structural questionnaire among 332 affected households during October and November 2022. The survey also utilized informal interviews at three study sites in August 2023, with results showing that 95%–97 % of total respondents reconstructed their houses according to earthquake-resistant standards. The same percentage of respondents withdrew the final payment from the National Reconstruction Authority (NRA). Most reconstructed houses were smaller than before the earthquake; many respondents rebuilt one-room, one-story structures. In addition, housing typologies at the study sites changed from brick/stone and mud-based masonry to cement-based construction. Key Words: Gorkha earthquake, seismic-resistant guidelines, compliance with government safety measures, reconstructed houses, NRA.

Minimal mass optimization of tensegrity torsional structures

The advantage of high strength-to-mass ratio renders tensegrity structures suitable for application as lightweight structures. Therefore, the minimal mass optimization problem is one of the focal points in the research of tensegrity structures. Various minimal mass optimization configurations have been proposed for different types of loads. In this paper, we will continue this topic to deal with the minimal mass optimization problem of the tensegrity structure under torsional loads. The inspiration for this study stems from the constant radial angle feature of Michell structures, which is also present in the logarithmic spiral. By employing inverse stereographic projection with conformality, the logarithmic spiral can be mapped onto the sphere , the angle between the curve and the lines of latitude remains constant. On this basis, a type of tensegrity sphere structure combining the spherical logarithmic spiral and DHT (Double Helix Tensegrity) topology is proposed. Compared to the other two spheres with identical topology, the proposed tensegrity sphere structure significantly reduces the minimal mass under torsional loads. Furthermore, we have observed that when logarithmic spiral curves with opposite chirality are orthogonal, the optimal solution with minimal mass occurs. These findings demonstrate the potential of the combination of logarithmic spirals and conformal transformations, and positive effect of the orthogonal arrangement of members for mass optimization of tensegrity structures.

Multistate nonpolar resistive switching in nickel embedded polyoxovanadate for high density data storage

The evolution of the electronic industry constantly relies on downscaling of electronic devices and integrating novel materials in active regions to accomplish ever-higher speeds and new features in device structures. Employing materials that display multistate switching for resistive-random-access-memory or simply resistive memory could be a simple and effective way to realize high density data storage. In this context, we report multistate “nonpolar” resistive switching in a nickel embedded polyoxovanadate cluster, (K2H5[NiV14O40]) – a molecule that belongs to the larger polyoxometalate family. We observed unique and distinctive nonpolar resistive switching behaviour for the first time in a multi-redox polyoxometalate cluster. The switching characteristics were repeatable for more than 200 cycles. Our two terminal Al/K2H5[NiV14O40])/ITO memory cells exhibited considerably high resistance window (105) and also long retention time (2000 s). This work holds promise for a novel strategy in order to achieve multilevel storage by exploiting different varieties of polyoxometalate molecules as active switching element that can possibly connect memory with neuromorphic computing.

Scaled boundary finite element based two-level learning approach for structural flaw identification

A comprehensive framework that combines classification and regression techniques through machine learning (ML) algorithms to address the inverse problem of flaw identification and quantification is proposed. The framework uses the scaled boundary finite element method (SBFEM) and quadtree (octree mesh in 3D) to compute displacements for various flaw shapes and specified boundary conditions. The flaw data, such as their numbers, location, and size, along with the displacement data, serve as the input for ML models. A random forest model is used to predict the number of flaws, while an artificial neural network (ANN) employing a fully connected multi-layer perceptron is utilized to predict flaw locations and sizes. The framework underwent rigorous testing to identify the number and location of multiple flaws (circular and elliptical) in 2D structures. The accuracy of the classification and regression models for flaw detection, ranging from one to four circular flaws, is observed to be 98%. As an extension of its capabilities, the proposed model successfully identifies multiple flaws and their features in 3D structures as well with 98% accuracy, demonstrating the robustness of the proposed approach.

Long carbon–carbon bonds and beyond

Advances in synthetic organic chemistry have facilitated the preparation and exploration of compounds exhibiting unusual bonding states. This review delves into carbon–carbon single bonds that are exceeding typical length of bonds, elucidating recent advances in understanding their bonding nature, properties, and chemical reactivity. Additionally, we examine factors contributing to the occurrence of such elongated bonds and their effects on other bonding parameters. Furthermore, we shift our focus toward the π-dimers of radical species, surpassing the limit of two-center two-electron (2c/2e) bonds, discussing their formation mechanisms, stability, and inherent properties. A key feature in the electronic structure of π-dimers is the bonding interaction of two unpaired electrons spanning multiple atoms, that is, multicenter two-electron (mc/2e) bonding. This review sheds light on the significant role played by extended carbon–carbon bonds (2c/2e bonds) and radical π-dimers (mc/2e bonding) in organic chemistry, providing valuable insight for future research on new functional materials.

The research of vibration localization under strong disturbances using modal analysis methods

Abstract Vibration localization is a typical feature of aperiodic structures. This work investigates the vibration localization of framed cylindrical shells with a high disturbance degree. First, the spring oscillator system is considered an example to reveal the general law governing how high disturbance degree influences the vibration transmission in a structure. Then the strong localization coefficient is calculated using the modal analysis method. Finally, the conclusions of the one-dimensional systems are applied to the two-dimensional structures based on the passband approximation equivalent method to derive the strong localization coefficient of the aperiodic ring-ribbed cylindrical shell. This study realizes quantitative calculation of the vibration localization coefficient under high disturbance degree. The results of this work provide theoretical support for reducing the vibration transmission performance of a structure by increasing the degree of disturbance.

Brain Image Segmentation for Ultrascale Neuron Reconstruction via an Adaptive Dual-Task Learning Network.

Accurate morphological reconstruction of neurons in whole brain images is critical for brain science research. However, due to the wide range of whole brain imaging, uneven staining, and optical system fluctuations, there are significant differences in image properties between different regions of the ultrascale brain image, such as dramatically varying voxel intensities and inhomogeneous distribution of background noise, posing an enormous challenge to neuron reconstruction from whole brain images. In this paper, we propose an adaptive dual-task learning network (ADTL-Net) to quickly and accurately extract neuronal structures from ultrascale brain images. Specifically, this framework includes an External Features Classifier (EFC) and a Parameter Adaptive Segmentation Decoder (PASD), which share the same Multi-Scale Feature Encoder (MSFE). MSFE introduces an attention module named Channel Space Fusion Module (CSFM) to extract structure and intensity distribution features of neurons at different scales for addressing the problem of anisotropy in 3D space. Then, EFC is designed to classify these feature maps based on external features, such as foreground intensity distributions and image smoothness, and select specific PASD parameters to decode them of different classes to obtain accurate segmentation results. PASD contains multiple sets of parameters trained by different representative complex signal-to-noise distribution image blocks to handle various images more robustly. Experimental results prove that compared with other advanced segmentation methods for neuron reconstruction, the proposed method achieves state-of-the-art results in the task of neuron reconstruction from ultrascale brain images, with an improvement of about 49% in speed and 12% in F1 score.

TGIN: Translation-Based Graph Inference Network for Few-Shot Relational Triplet Extraction.

Extracting relational triplets aims at detecting entity pairs and their semantic relations. Compared with pipeline models, joint models can reduce error propagation and achieve better performance. However, all of these models require large amounts of training data, therefore performing poorly on many long-tail relations in reality with insufficient data. In this article, we propose a novel end-to-end model, called TGIN, for few-shot triplet extraction. The core of TGIN is a multilayer heterogeneous graph with two types of nodes (entity node and relation node) and three types of edges (relation-entity edge, entity-entity edge, and relation-relation edge). On the one hand, this heterogeneous graph with entities and relations as nodes can intuitively extract relational triplets jointly, thereby reducing error propagation. On the other hand, it enables the triplet information of limited labeled data to interact better, thus maximizing the advantage of this information for few-shot triplet extraction. Moreover, we devise a graph aggregation and update method that utilizes translation algebraic operations to mine semantic features while retaining structure features between entities and relations, thereby improving the robustness of the TGIN in a few-shot setting. After updating the node and edge features through layers, TGIN propagates the label information from a few labeled examples to unlabeled examples, thus inferring triplets from these unlabeled examples. Extensive experiments on three reconstructed datasets demonstrate that TGIN can significantly improve the accuracy of triplet extraction by 2.34%~10.74% compared with the state-of-the-art baselines. To the best of our knowledge, we are the first to introduce a heterogeneous graph for few-shot relational triplet extraction.

A machine learning assisted multifidelity modelling methodology to predict 3D stresses in the vicinity of design features in composite structures

Multifidelity global–local finite element (FE) analyses are typically used to predict damage initiation hotspots around repetitive design features in large composite structures, such as composite airframes. We propose the use of machine learning (ML) methods to accelerate these analyses. We demonstrate this ML assisted framework for the stress analysis of a hole in plate feature in an aerospace C-spar structure. To enable this framework, we develop the following original features: a computationally efficient sampling scheme; a work-equivalent boundary condition homogenisation scheme; a volume averaged ply-by-ply stress approach; and a sequential long-short term memory neural network reformulated from a time basis to a stacking sequence basis with further bi-directionality customisation. Overall, we show that the developed method results in high-accuracy prediction of 3D stresses, with over two orders of magnitude reduction in modelling and simulation time compared to FE analyses.

Raman scattering owing to magneto-polaron states in monolayer transition metal dichalcogenides

Magneto-optical measurements are fundamental research tools that allow for studying the hitherto unexplored optical transitions and the related applications of topological two-dimensional (2D) transition metal dichalcogenides (TMDs). A theoretical model is developed for the first-order magneto-resonant Raman scattering in a monolayer of TMD. A significant number of avoided crossing points involving optical phonons in the magneto-polaron (MP) spectrum, a superposition of the electron and hole states in the excitation branches, and their manifestations in optical transitions at various light scattering configurations are unique features for these 2D structures. The Raman intensity reveals three resonant splittings of double avoided-crossing levels. The three excitation branches are present in the MP spectrum provoked by the coupling of the Landau levels in the conduction and valence bands via an out-of-plane A1\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$A_1$$\\end{document}-optical phonon mode. The energy gaps at the anticrossing points in the MP scattering spectrum are revealed as a function of the electron and hole optical deformation potential constants. The resonant MP Raman scattering efficiency profile allows for quantifying the relative contribution of the conduction and valence bands in the formation of MPs. The results obtained are a guideline for controlling MP effects on the magneto-optical properties of TMD semiconductors, which open pathways to novel optoelectronic devices based on 2D TMDs.

An AI and Statistical Shape Analysis combined framework to extract features in complex cardiovascular structures

An AI and Statistical Shape Analysis combined framework to extract features in complex cardiovascular structures

An Exploration of Organoid Technology: Present Advancements, Applications, and Obstacles.

Organoids are in vitro models that exhibit a three-dimensional structure and effectively replicate the structural and physiological features of human organs. The capacity to research complex biological processes and disorders in a controlled setting is laid out by these miniature organ-like structures. This work examines the potential applications of organoid technology, as well as the challenges and future directions associated with its implementation. It aims to emphasize the pivotal role of organoids in disease modeling, drug discovery, developmental biology, precision medicine, and fundamental research. The manuscript was put together by conducting a comprehensive literature review, which involved an in-depth evaluation of globally renowned scientific research databases. The field of organoids has generated significant attention due to its potential applications in tissue development and disease modelling, as well as its implications for personalised medicine, drug screening, and cell-based therapies. The utilisation of organoids has proven to be effective in the examination of various conditions, encompassing genetic disorders, cancer, neurodevelopmental disorders, and infectious diseases. The exploration of the wider uses of organoids is still in its early phases. Research shall be conducted to integrate 3D organoid systems as alternatives for current models, potentially improving both fundamental and clinical studies in the future.

Molecular Mechanism of Tau Misfolding and Aggregation: Insights from Molecular Dynamics Simulation.

Tau dysfunction has a close association with many neurodegenerative diseases, which are collectively referred to as tauopathies. Neurofibrillary tangles (NFTs) formed by misfolding and aggregation of tau are the main pathological process of tauopathy. Therefore, uncovering the misfolding and aggregation mechanism of tau protein will help to reveal the pathogenic mechanism of tauopathies. Molecular dynamics (MD) simulation is well suited for studying the dynamic process of protein structure changes. It provides detailed information on protein structure changes over time at the atomic resolution. At the same time, MD simulation can also simulate various conditions conveniently. Based on these advantages, MD simulations are widely used to study conformational transition problems such as protein misfolding and aggregation. Here, we summarized the structural features of tau, the factors affecting its misfolding and aggregation, and the applications of MD simulations in the study of tau misfolding and aggregation.

Evidence and Prospects of Lycopene as Powerful Red Superfood: Modern Approach to Food Science

Abstract: Lycopene is a plant-derived important nutrient belonging to a group of pigments identified as carotenoids. Lycopene has distinctive chemical and structural features contributing to definite biological properties. Lycopene is a powerful antioxidant and prevents cells and tissues from damage caused by singlet oxygen or free radicals. Research studies have shown the healthpromoting potential of lycopene or its metabolites in the prevention of chronic diseases like diabetes, cancer, asthma, metabolic syndrome, arthritis, and cardiovascular diseases. Lycopene is stable for processing like other natural compounds when present in the plant tissue matrix. Over the last few years, the global lycopene market has increased in different sectors like pharmaceuticals, cosmetics, food and dairy products, and other dietary supplements. The most important aspect of lycopene as a superfood is its synergistic nature with other nutrients. The main aim of this review article is to study chemistry, biosynthetic pathways, biotransformation, sources, and acceptable daily intake. Its characterization as a superfood, therapeutic potential, and wide range of applications in the food industry are also studied.

Metal and solvent dependent formation of imidazolidine or hemiaminal ether complexes

The coordination chemistry of the imidazolidinylpyridine ligand with Ni(II), Cu(II), Cd(II), and Zn(II) ions was investigated in both protic and aprotic solvents. Interestingly, the coordination environment around the metal center varied significantly depending upon the nature of metal ion and solvent used for synthesis. In acetonitrile the imidazolidinylpyridine bound to Cu(II) and Ni(II) remains intact, while in methanol it formed ring-opened hemiaminal ether. The Cd(II) and Zn(II) ion stabilized the imidazolidinylpyridine as a face-capping tridentate ligand regardless of the nature of the solvent. The imidazolidinylpyridine and the complexes formed were characterized by NMR spectroscopy, high resolution electrospray ionization mass spectrometry and single-crystal X-ray crystallography, appropriately. The structural features in the solid-state structures were particularly intriguing, shedding light on the complex interplay between the ligand, metal ions, and solvent environments in coordination chemistry.

Spectral recomposition feature for optimizing seismic velocity model prediction with a neural network

An approach incorporating spectral recomposition results for training a neural network is proposed. The initial procedure is based on an inversion scheme for reconstructing the seismic spectrum of wavelets associated with a reflection event. This allows one to estimate each wavelet's temporal position within a seismogram. Since each calculated wavelet's temporal position is identified, this information is integrated into the seismogram as an additional feature for a trained neural network. A Fully Convolutional Network is trained with models that have features commonly found in offshore hydrocarbon reservoirs. The training step is based on composing a set of an acoustic velocity model set with its seismograms and their position matrices at the time of each wavelet for each seismogram. Next, the network training with a specific number of sets is performed. This approach was able to predict acoustic velocity models more accurately, especially for predicting the shape of salt bodies and the interface of inclined layers, which makes the proposed approach an interesting mean for estimating specific features in offshore hydrocarbon reservoir structures.

Computed tomography technologies to measure key structural features of polymeric biomedical implants from bench to bedside.

Implanted polymeric devices, designed to encourage tissue regeneration, require porosity. However, characterizing porosity, which affects many functional device properties, is non-trivial. Computed tomography (CT) is a quick, versatile, and non-destructive way to gain 3D structural information, yet various CT technologies, such as benchtop, preclinical and clinical systems, all have different capabilities. As system capabilities determine the structural information that can be obtained, seamless monitoring of key device features through all stages of clinical translation must be engineered intentionally. Therefore, in this study we tested feasibility of obtaining structural information in pre-clinical systems and high-resolution micro-CT (μCT) under physiological conditions. To overcome the low CT contrast of polymers in hydrated environments, radiopaque nanoparticle contrast agent was incorporated into porous devices. The size of resolved features in porous structures is highly dependent on the resolution (voxel size) of the scan. As the voxel size of the CT scan increased (lower resolution) from 5 to 50 μm, the measured pore size was overestimated, and percentage porosity was underestimated by nearly 50%. With the homogeneous introduction of nanoparticles, changes to device structure could be quantified in the hydrated state, including at high-resolution. Biopolymers had significant structural changes post-hydration, including a mean increase of 130% in pore wall thickness that could potentially impact biological response. By incorporating imaging capabilities into polymeric devices, CT can be a facile way to monitor devices from initial design stages through to clinical translation.