Search Results (120)

The classification of very-low-frequency and low-frequency (VLF/LF) lightning-radiation electric-field waveforms is of paramount importance for lightning-disaster prevention and mitigation. However, traditional waveform classification methods suffer from the complex characteristics of lightning waveforms, such as non-stationarity, strong noise interference, and feature coupling, limiting classification accuracy and generalization. To address this problem, a novel framework is proposed for VLF/LF lightning-radiated electric-field waveform classification. Firstly, an improved Kalman filter (IKF) is meticulously designed to eliminate possible high-frequency interferences (such as atmospheric noise, electromagnetic radiation from power systems, and electronic noise from measurement equipment) embedded within the waveforms based on the maximum entropy criterion. Subsequently, an attention-based multi-fusion convolutional neural network (AMCNN) is developed for waveform classification. In the AMCNN architecture, waveform information is comprehensively extracted and enhanced through an optimized feature fusion structure, which allows for a more thorough consideration of feature diversity, thereby significantly improving the classification accuracy. An actual dataset from Anhui province in China is used to validate the proposed classification framework. Experimental results demonstrate that our framework achieves a classification accuracy of 98.9% within a processing time of no more than 5.3 ms, proving its superior classification performance for lightning-radiation electric-field waveforms. Full article

In image-guided radiation therapy (IGRT), deformable image registration between computed tomography (CT) and cone beam computed tomography (CBCT) images remain challenging due to the computational cost of iterative algorithms and the data dependence of supervised deep learning methods. Implicit neural representation (INR) provides a promising alternative, but conventional multilayer perceptron (MLP) might struggle to efficiently represent complex, nonlinear deformations. This study introduces a novel INR-based registration framework that models the deformation as a continuous, time-varying velocity field, parameterized by a Kolmogorov–Arnold Network (KAN) constructed using Jacobi polynomials. To our knowledge, this is the first integration of KAN into medical image registration, establishing a new paradigm beyond standard MLP-based INR. For improved efficiency, the KAN estimates low-dimensional principal components of the velocity field, which are reconstructed via inverse principal component analysis and temporally integrated to derive the final deformation. This approach achieves a ~70% improvement in computational efficiency relative to direct velocity field modeling while ensuring smooth and topology-preserving transformations through velocity regularization. Evaluation on a publicly available pelvic CT–CBCT dataset demonstrates up to 6% improvement in registration accuracy over traditional iterative methods and ~3% over MLP-based INR baselines, indicating the potential of the proposed method as an efficient and generalizable alternative for deformable registration. Full article

Nanotechnology has become a transformative field in modern science and engineering, offering innovative approaches to enhance conventional thermal and fluid systems. Heat and mass transfer phenomena, particularly fluid motion across various geometries, play a crucial role in industrial and engineering processes. The inclusion of nanoparticles in base fluids significantly improves thermal conductivity and enables advanced phase-change technologies. The current work examines Powell–Eyring nanofluid’s heat transmission properties on a stretched Riga plate, considering the effects of magnetic fields, porosity, Darcy–Forchheimer flow, thermal radiation, and activation energy. Using the proper similarity transformations, the pertinent governing boundary-layer equations are converted into a set of ordinary differential equations (ODEs), which are then solved using the boundary value problem fourth-order collocation (BVP4C) technique in the MATLAB program. Tables and graphs are used to display the outcomes. Due to their significance in the industrial domain, the Nusselt number and skin friction are also evaluated. The velocity of the nanofluid is shown to decline with a boost in the Hartmann number, porosity, and Darcy–Forchheimer parameter values. Moreover, its energy curves are increased by boosting the values of thermal radiation and the Biot number. A stronger Hartmann number M decelerates the flow (thickening the momentum boundary layer), whereas increasing the Riga forcing parameter Q can locally enhance the near-wall velocity due to wall-parallel Lorentz forcing. Visual comparisons and numerical simulations are used to validate the results, confirming the durability and reliability of the suggested approach. By using a systematic design technique that includes training, testing, and validation, the fluid dynamics problem is solved. The model’s performance and generalization across many circumstances are assessed. In this work, an artificial neural network (ANN) architecture comprising two hidden layers is employed. The model is trained with the Levenberg–Marquardt scheme on reliable numerical datasets, enabling enhanced prediction capability and computational efficiency. The ANN demonstrates exceptional accuracy, with regression coefficients $R\approx 1.0$ and the best validation mean squared errors of $8.52\times {10}^{-10}$, $7.91\times {10}^{-9}$, and $1.59\times {10}^{-8}$ for the Powell–Eyring, heat radiation, and thermophoresis models, respectively. The ANN-predicted velocity, temperature, and concentration profiles show good agreement with numerical findings, with only minor differences in insignificant areas, establishing the ANN as a credible surrogate for quick parametric assessment and refinement in magnetohydrodynamic (MHD) nanofluid heat transfer systems. Full article

Three-dimensional (3D) reconstruction technology is not only a core and key technology in computer vision and graphics, but also a key force driving the flourishing development of many cutting-edge applications such as virtual reality (VR), augmented reality (AR), autonomous driving, and digital earth. With the rise in novel view synthesis technologies such as Neural Radiation Field (NeRF) and 3D Gaussian Splatting (3DGS), 3D reconstruction is facing unprecedented development opportunities. This article introduces the basic principles of traditional 3D reconstruction methods, including Structure from Motion (SfM) and Multi View Stereo (MVS) techniques, and analyzes the limitations of these methods in dealing with complex scenes and dynamic environments. Focusing on implicit 3D scene reconstruction techniques related to NeRF, this paper explores the advantages and challenges of using deep neural networks to learn and generate high-quality 3D scene rendering from limited perspectives. Based on the principles and characteristics of 3DGS-related technologies that have emerged in recent years, the latest progress and innovations in rendering quality, rendering efficiency, sparse view input support, and dynamic 3D reconstruction are analyzed. Finally, the main challenges and opportunities faced by current 3D reconstruction technology and novel view synthesis technology were discussed in depth, and possible technological breakthroughs and development directions in the future were discussed. This article aims to provide a comprehensive perspective for researchers in 3D reconstruction technology in fields such as digital twins and smart cities, while opening up new ideas and paths for future technological innovation and widespread application. Full article

This systematic review examines the methodologies and applications of integrated risk indicators in positron emission tomography (PET) radiopharmaceutical production, focusing on occupational, technological, and environmental risks. Conducted in accordance with PRISMA 2020 guidelines and utilizing the Ryyan software 2023 for article screening, the review synthesizes findings from 70 studies published between 2020 and 2025 in English and Spanish, including articles, conference papers, and reviews. The review was registered on PROSPERO (CRD420251078221). Key disciplines contributing to risk assessment frameworks include environmental science, occupational health and safety, civil engineering, mining engineering, maritime safety, financial/economic risk, and systems engineering. Predominant risk assessment methods identified are probabilistic modeling (e.g., Monte Carlo simulations), machine learning (e.g., neural networks), multi-criteria decision-making (e.g., AHP and TOPSIS), and failure mode and effects analysis (FMEA), each offering strengths, such as uncertainty quantification and systematic hazard identification, alongside limitations like data dependency and subjectivity. The review explores how frameworks from other industries can be adapted to address PET-specific risks, such as radiation exposure to workers, equipment failure, and waste management, and how studies integrate these factors into unified risk indicators using weighted scoring, probabilistic methods, and fuzzy logic. Gaps in the literature include limited stakeholder engagement, lack of standardized frameworks, insufficient real-time monitoring, and under-represented environmental risks. Future research directions propose developing PET-specific tools, integrating AI and IoT for real-time data, establishing standardized frameworks, and expanding environmental assessments to enhance risk management in PET radiopharmaceutical production. This review highlights the interdisciplinary nature of risk assessment and the critical need for comprehensive, tailored approaches to ensure safety and sustainability in this field. Full article

To model the response of neural networks to electromagnetic radiation in real-world environments, this study proposes a memristive dual-wing fractional-order Hopfield neural network (MDW-FOMHNN) model, utilizing a fractional-order memristor to simulate neuronal responses to electromagnetic radiation, thereby achieving complex chaotic dynamics. Analysis reveals that within specific ranges of the coupling strength, the MDW-FOMHNN lacks equilibrium points and exhibits hidden chaotic attractors. Numerical solutions are obtained using the Adomian Decomposition Method (ADM), and the system’s chaotic behavior is confirmed through Lyapunov exponent spectra, bifurcation diagrams, phase portraits, and time series. The study further demonstrates that the coupling strength and fractional order significantly modulate attractor morphologies, revealing diverse attractor structures and their coexistence. The complexity of the MDW-FOMHNN output sequence is quantified using spectral entropy, highlighting the system’s potential for applications in cryptography and related fields. Based on the polynomial form derived from ADM, a field programmable gate array (FPGA) implementation scheme is developed, and the expected chaotic attractors are successfully generated on an oscilloscope, thereby validating the consistency between theoretical analysis and numerical simulations. Finally, to link theory with practice, a simple and efficient MDW-FOMHNN-based encryption/decryption scheme is presented. Full article

The main goal of this study is to create a computational solver that analyzes the effects of magnetohydrodynamics (MHD) on heat radiation in Cu–water-based Prandtl nanofluid flow using artificial neural networks. Copper nanoparticles are utilized to boost the water-based fluid’s thermal effect. This study primarily focuses on heat transfer over a horizontal sheet, exploring different scenarios by varying key factors such as the magnetic field and thermal radiation properties. The mathematical model is formulated using partial differential equations (PDEs), which are then transformed into a corresponding set of ordinary differential equations (ODEs) through appropriate similarity transformations. The bvp4c solver is then used to simulate the numerical behavior. The effects of relevant parameters on the temperature, velocity, skin friction, and local Nusselt number profiles are examined. It is discovered that the parameters of the Prandtl fluid have a considerable impact. The local skin friction and the local Nusselt number are improved by increasing these parameters. The dataset is split into 70% training, 15% validation, and 15% testing. The ANN model successfully predicts skin friction and Nusselt number profiles, showing good agreement with numerical simulations. This hybrid framework offers a robust predictive approach for heat management systems in industrial applications. This study provides important insights for researchers and engineers aiming to comprehend flow characteristics and their behavior and to develop accurate predictive models. Full article

High-quality green gardens can markedly enhance the quality of life and mental well-being of their users. However, health and lifestyle constraints make it difficult for people to enjoy urban gardens, and traditional methods struggle to offer the high-fidelity experiences they need. This study introduces a 3D scene reconstruction and rendering strategy based on implicit neural representation through the efficient and removable neural radiation fields model (NeRF-RE). Leveraging neural radiance fields (NeRF), the model incorporates a multi-resolution hash grid and proposal network to improve training efficiency and modeling accuracy, while integrating a segment-anything model to safeguard public privacy. Take the crabapple tree, extensively utilized in urban garden design across temperate regions of the Northern Hemisphere. A dataset comprising 660 images of crabapple trees exhibiting three distinct geometric forms is collected to assess the NeRF-RE model’s performance. The results demonstrated that the ‘harvest gold’ crabapple scene had the highest reconstruction accuracy, with PSNR, LPIPS and SSIM of 24.80 dB, 0.34 and 0.74, respectively. Compared to the Mip-NeRF 360 model, the NeRF-RE model not only showed an up to 21-fold increase in training efficiency for three types of crabapple trees, but also exhibited a less pronounced impact of dataset size on reconstruction accuracy. This study reconstructs real scenes with high fidelity using virtual reality technology. It not only facilitates people’s personal enjoyment of the beauty of natural gardens at home, but also makes certain contributions to the publicity and promotion of urban landscapes. Full article

Breast cancer detection through non-invasive and accurate techniques remains a critical challenge in medical diagnostics. This study introduces a deep learning-based framework that leverages a microwave radar system equipped with an arc-shaped array of six antennas to estimate key tumor parameters, including position, size, and depth. This research begins with the evolutionary design of an ultra-wideband octagram ring patch antenna optimized for enhanced tumor detection sensitivity in directional near-field coupling scenarios. The antenna is fabricated and experimentally evaluated, with its performance validated through S-parameter measurements, far-field radiation characterization, and efficiency analysis to ensure effective signal propagation and interaction with breast tissue. Specific Absorption Rate (SAR) distributions within breast tissues are comprehensively assessed, and power adjustment strategies are implemented to comply with electromagnetic exposure safety limits. The dataset for the deep learning model comprises simulated self and mutual S-parameters capturing tumor-induced variations over a broad frequency spectrum. A core innovation of this work is the development of the Attention-Based Feature Separation (ABFS) model, which dynamically identifies optimal frequency sub-bands and disentangles discriminative features tailored to each tumor parameter. A multi-branch neural network processes these features to achieve precise tumor localization and size estimation. Compared to conventional attention mechanisms, the proposed ABFS architecture demonstrates superior prediction accuracy and interpretability. The proposed approach achieves high estimation accuracy and computational efficiency in simulation studies, underscoring the promise of integrating deep learning with conformal microwave imaging for safe, effective, and non-invasive breast cancer detection. Full article

Image segmentation is an important method in the field of image processing, while infrared (IR) image segmentation is one of the challenges in this field due to the unique characteristics of IR data. Infrared imaging utilizes the infrared radiation emitted by objects to produce images, which can supplement the performance of visible-light images under adverse lighting conditions to some extent. However, the low spatial resolution and limited texture details in IR images hinder the achievement of high-precision segmentation. To address these issues, an attention mechanism based on symmetrical cross-channel interaction—motivated by symmetry principles in computer vision—was integrated into a Mask Region-Based Convolutional Neural Network (Mask R-CNN) framework. A Bottleneck-enhanced Squeeze-and-Attention (BNSA) module was incorporated into the backbone network, and novel loss functions were designed for both the bounding box (Bbox) regression and mask prediction branches to enhance segmentation performance. Furthermore, a dedicated infrared image dataset was constructed to validate the proposed method. The experimental results demonstrate that the optimized model achieves higher segmentation accuracy and better segmentation performance compared to the original network and other mainstream segmentation models on our dataset, demonstrating how symmetrical design principles can effectively improve complex vision tasks. Full article

The article presents a method of shaping unconventional building envelopes characterized by their effective solar performance in a temperate climate. An analysis related to the impact of geometric shape on the solar direct radiation falling on building envelopes was presented in terms of polyhedral forms. It was based on interdisciplinary issues located in the fields of solar radiation, unconventional forms of buildings, numerical simulations, and artificial neural networks. The elaborated method’s algorithm was employed to describe the relationships between the envelope systems and the amount of the radiation falling on these systems, identified during the performed simulations. Two novel parametric models were defined to execute the simulations. The first was an initial geometric model defined by a number of arbitrary independent variables. The second was defined by one dependent variable representing the quantity of the solar radiation falling on each envelope. The analysis carried out showed that the invented trapezoidal forms of envelopes allowed for better control of the incident solar radiation in relation to the forms generated with other methods. The invented trapezoidal forms of envelopes can increase the amount of their direct solar irradiation up to 63%, compared to prior pyramidal forms. Although the climatic loads used were related to Strzyżów characterized by the geographical coordinates 49.9 N and 21.9 E and located in Central Europe, it was possible to adapt the method to other meteorological boundary conditions by changing the values of the defined parameters. The resultant parametric solar model can be employed to search for many diversified discrete solar envelopes of buildings and rational arrangements of their external planes so that the direct solar radiation falling on these envelopes can be increased during cold periods and restricted during hot summer periods of the year. Full article

Solar flares, caused by magnetic field reconnection in the sun’s atmosphere, are intense bursts of electromagnetic radiation that can disrupt the Earth’s space environment, affecting communication systems, GPSs, and satellites. Traditional physics-based methods for solar flare forecasting have utilized the statistical relationships between solar activity indicators, such as sunspots and magnetic field properties, employing techniques like Poisson distributions and discriminant analysis to estimate probabilities and identify critical parameters. While these methods provide valuable insights, limitations in predictive accuracy have driven the integration of deep learning approaches. With the accumulation of solar observation data and the development of data-driven algorithms, deep learning methods have been widely used to build solar flare prediction models. Most research has focused on designing or selecting the right deep network for the task. However, the influence of the magnetic field height on deep-learning-based prediction models has not been studied. This paper investigates how different magnetic field heights affect solar flare prediction performance. Active regions were observed using HMI magnetograms from 2010 to 2019. The magnetic field heights were stratified to create a database of active regions, and deep neural networks like AlexNet, ResNet-18, and SqueezeNet were used to evaluate prediction performance. The results show that predictions at around 7200 km above the photosphere outperform other heights, aligning with physical method analysis. At this altitude, the average AUC of the predictions from the three models reaches 0.788. Full article

Background: Terahertz (THz) waves, lying between millimeter waves and infrared light, may interact with biomolecules due to their unique energy characteristics. However, whether THz waves are neurally regulated remains controversial, and the underlying mechanism is elusive. Methods: Mouse brain slices were exposed to 1.94 THz waves for 1 h. Synaptic plasticity was evaluated via transmission electron microscopy (TEM), long-term potentiation (LTP), and neuronal class III β-tubulin (Tuj1) and synaptophysin (SYN) expression. Immunofluorescence (IF) and electrophysiology were used to identify neurons sensitive to THz waves. The calcium activity of excitatory neurons, glutamate receptor currents, and glutamate neuron marker expression was also assessed using calcium imaging, a patch clamp, and Western blotting (WB). Optogenetics and chemogenetics were used to determine the role of excitatory neurons in synaptic plasticity impairment after THz wave exposure. NMDA receptor 2B (GluN2B) was overexpressed in the ventral hippocampal CA1 (vCA1) by a lentivirus to clarify the role of GluN2B in THz wave-induced synaptic plasticity impairment. Results: Exposure to 1.94 THz waves increased postsynaptic density (PSD) thickness and reduced the field excitatory postsynaptic potential (fEPSP) slope and Tuj1 and SYN expression. THz waves diminished vCA1 glutamatergic neuron activity and excitability, neural electrical activity, and glutamate transporter function. THz waves reduced N-methyl-D-aspartate receptor (NMDAR) current amplitudes and NMDAR subunit expression. Activating vCA1 glutamatergic neurons through optogenetics and chemogenetics mitigated THz wave-induced synaptic plasticity impairment. GluN2B subunit overexpression improved synaptic plasticity marker expression, synaptic ultrastructure, and the fEPSP slope. Conclusions: Exposure to 1.94 THz waves decreased synaptic plasticity, glutamatergic neuron excitability, and glutamatergic synaptic transmission in the vCA1. Glutamatergic neuron activation and GluN2B overexpression alleviated THz wave-induced synaptic plasticity impairment; thus, neuromodulation could be a promising therapeutic strategy to mitigate the adverse effects of THz radiation. Full article

Retrieving LST from infrared spectral observations is challenging because it needs separation from emissivity in surface radiation emission, which is feasible only when the state of the surface–atmosphere system is known. Thanks to its high spectral resolution, the Infrared Atmospheric Sounding Interferometer (IASI) instrument onboard Metop polar-orbiting satellites is the only sensor that can simultaneously retrieve LST, the emissivity spectrum, and atmospheric composition. Still, it cannot penetrate thick cloud layers, making observations blind to surface emissions under cloudy conditions, with surface and atmospheric parameters being flagged as voids. The present paper aims to discuss a downscaling–fusion methodology to retrieve LST missing values on a spatial field retrieved from spatially scattered IASI observations to yield level 3, regularly gridded data, using as proxy data LST from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) flying on Meteosat Second Generation (MSG) platform, a geostationary instrument, and from the Advanced Very High-Resolution Radiometer (AVHRR) onboard Metop polar-orbiting satellites. We address this problem by using machine learning techniques, i.e., Gradient Boosting, Random Forest, Gaussian Process Regression, Neural Network, and Stacked Regression. We applied the methodology over the Po Valley region, a very heterogeneous area that allows addressing the trained models’ robustness. Overall, the methods significantly enhanced spatial sampling, keeping errors in terms of Root Mean Square Error (RMSE) and bias (Mean Absolute Error, MAE) very low. Although we demonstrate and assess the results primarily using IASI data, the paper is also intended for applications to the IASI follow-on, that is, IASI Next Generation (IASI-NG), and much more to the Infrared Sounder (IRS), which is planned to fly this year, 2025, on the Meteosat Third Generation platform (MTG). Full article

Within the context of “peak carbon and carbon neutrality”, reducing carbon emissions from coal-fired power plants and increasing the proportion of renewable energy in electricity generation have become critical issues in the transition to renewable energy. Based on the principles of cascaded energy utilization, this paper improves the coupling methodology of an integrated solar thermal and coal-fired power generation system based on existing research. A parabolic trough collector field and a three-tank molten salt thermal energy storage system are connected in series and then in parallel with the outlet of the reheater. ASPEN PLUS V14 and MATLAB R2018b software were used to simulate a steady-state model and numerical model, respectively, so as to study the feasibility of the improved complementary framework in enhancing the peak load capacity of coal-fired units and reducing their carbon emissions. Actual solar radiation data from a specific location in Inner Mongolia were gathered to train a neural network predictive model. Then, the peak-shaving performance of the complementary system in matching load demands under varying hours of thermal energy storage was simulated. The findings demonstrate that, under constant boiler load conditions, optimizing the complementary system with a thermal energy storage duration of 5 h and 50 min results in an energy utilization efficiency of 88.82%, accompanied by a daily reduction in coal consumption by 36.49 tonnes. This indicates that when operated under the improved coupling framework with optimal parameters, the peak regulation capabilities of coal-fired power units can be improved and carbon emission can be reduced. Full article