Search Results (273)

To address the challenges of poor adaptability to spatial heterogeneity, easy breakage of amplitude–phase coupling relationships, and insufficient physical consistency in complex optical wavefield reconstruction, this paper proposes the DdONN-PINNs hybrid framework. Focused on preserving the intrinsic symmetries of wave physics, the framework achieves deep integration of optical neural networks and physics-informed information. Centered on an architecture of “SIREN shared encoding–domain-specific output”, it utilizes the periodic activation property of SIREN encoders to maintain the spatial symmetry of wavefield distribution, incorporates learnable Fourier diffraction layers to model physical propagation processes, and adopts native complex-domain modeling to avoid splitting the real and imaginary parts of complex amplitudes—effectively adapting to spatial heterogeneity while fully preserving amplitude-phase coupling in wavefields. Validated on rogue wavefields governed by the Nonlinear Schrödinger Equation (NLSE), experimental results demonstrate that DdONN-PINNs achieve an amplitude Mean Squared Error (MSE) of $2.94\times {10}^{-3}$ and a phase MSE of $5.86\times {10}^{-4}$, outperforming non-domain-decomposed models and ReLU-activated variants significantly. Robustness analysis shows stable reconstruction performance even at a noise level of $\sigma =0.1$. This framework provides a balanced solution for wavefield reconstruction that integrates precision, physical interpretability, and robustness, with potential applications in fiber-optic communication and ocean optics. Full article

The Allen-Cahn equation is a fundamental model in materials science for describing phase separation phenomena. This paper introduces an Energy-Stabilized Scaled Deep Neural Network (ES-ScaDNN) framework to solve the Allen-Cahn equation by energy minimization. Unlike traditional numerical methods, our approach directly approximates the solution of steady-state solution the Allen-Cahn equation by minimizing the associated energy functional using a deep neural network. ES-ScaDNN incorporates two key innovations. The first is a scaling layer designed to map the network output to the physical range of the Allen-Cahn phase variable. The second is a variance-based regularization term designed to promote clear phase separation. We demonstrate the accuracy and efficiency of ES-ScaDNN through comprehensive numerical experiments in both one and two dimensions. Our results show that ReLU activation functions are particularly well-suited for one-dimensional cases, while tanh functions are more suitable for two-dimensional problems due to their superior ability to maintain solution smoothness. Furthermore, we investigate how training epochs and the interface parameter ε influence the behavior of the solution. ES-ScaDNN provides a novel, accurate, and efficient deep learning framework for solving the Allen-Cahn equation, paving the way for tackling more complex phase-field problems. Full article

Human immunodeficiency virus (HIV-1) and SARS-CoV-2 continue to co-burden global health, motivating discovery of broad-spectrum small molecules. Conessine, a steroidal alkaloid, has reported membrane-active and antimicrobial properties but remains underexplored as a dual antiviral chemotype. To interrogate conessine’s multi-target antiviral potential against key enzymatic and entry determinants of HIV-1 and SARS-CoV-2 and to benchmark performance versus approved comparators. Eight targets were modeled: HIV-1 reverse transcriptase (RT, 3V81), protease (PR, 1HVR), integrase (IN, 3LPT), gp120–gp41 trimer (4NCO); and SARS-CoV-2 main protease (M^pro, 6LU7), papain-like protease (PL^pro, 6W9C), RNA-dependent RNA polymerase (RdRp, 7BV2), spike RBD (6M0J). Ligands (conessine; positive controls: dolutegravir for HIV-1, nirmatrelvir for SARS-CoV-2) were prepared with standard protonation, minimized, and docked using AutoDock Vina v 1.2.0exhaustiveness 4; 20 poses). Binding modes were profiled in 2D/3D. Protocol robustness was verified by re-docking co-crystallized ligands (RMSD ≤ 2.0 Å). Atomistic MD (explicit TIP3P, OPLS4, 300 K/1 atm, NPT; 50–100 ns) assessed pose stability (RMSD/RMSF), pocket compaction (Rg, volume), and interaction persistence; MM/GBSA provided qualitative energy decomposition. ADMET was predicted in silico. Conessine showed coherent, hydrophobically anchored binding across both viral panels. Best docking scores (kcal·mol⁻¹) were: HIV-1—PR −6.910, RT −6.672, IN −5.733; SARS-CoV-2—spike RBD −7.025, M^pro −5.745, RdRp −5.737, PL^pro −5.024. Interaction maps were dominated by alkyl/π-alkyl packing to catalytic corridors (e.g., PR Ile50/Val82, RT Tyr181/Val106; M^pro His41/Met49; RBD L455/F486/Y489) with occasional carbon-/water-mediated H-bonds guiding orientation. MD sustained low ligand RMSD (typically ≤1.6–2.2 Å) and damped RMSF at catalytic loops, indicating pocket rigidification; MM/GBSA trends (≈ −30 to −40 kcal·mol⁻¹, dispersion-driven) supported persistent nonpolar stabilization. Benchmarks behaved as expected: dolutegravir bound strongly to IN (−6.070) and PR (−7.319) with stable MD; nirmatrelvir was specific for M^pro and displayed weaker, discontinuous engagement at PL^pro/RdRp/RBD under identical settings. ADMET suggested conessine has excellent permeability/BBB access (high logP), but liabilities include poor aqueous solubility, predicted hERG risk, and CYP2D6 substrate dependence.Conessine operates as a hydrophobic, multi-target wedge with the most favorable computed engagement at HIV-1 PR/RT and the SARS-CoV-2 spike RBD, while maintaining stable poses at M^pro and RdRp. The scaffold merits medicinal-chemistry optimization to improve solubility and de-risk cardiotoxicity/CYP interactions, followed by biochemical and cell-based validation against prioritized targets. Full article

In recent years, academic performance prediction has evolved as a research field thanks to its development and exploration in the educational context. Early student performance prediction is crucial for enhancing educational outcomes and implementing timely interventions. Conventional approaches frequently struggle on behalf of the complexity of student profiles as a consequence of single activation functions, which prevent them from effectively learning intricate patterns. In addition, these models could experience obstacles such as the vanishing gradient problem and computational complexity. Therefore, this research study designed an Activation Ensemble Deep Neural Network (AcEnDNN) model to gain control of the previously mentioned challenges. The main contribution is the creation of a credible student performance prediction model that comprises extensive data preprocessing, feature extraction, and an Activation Ensemble DNN. By utilizing various methods of activation functions, such as ReLU, tanh, sigmoid, and swish, the ensembled activation functions are able to learn the complex structure of student data, which leads to more accurate performance prediction. The AcEn-DNN model is trained and evaluated based on the publicly available Student-mat.csv dataset, Student-por.csv dataset, and a real-time dataset. The experimental results revealed that the AcEn-DNN model achieved lower error rates, with an MAE of 1.28, MAPE of 2.36, MSE of 4.55, and RMSE of 2.13 based on a training percentage of 90%, confirming its robustness in modeling nonlinear relationships within student data. The proposed model also gained the minimum error values MAE of 1.28, MAPE of 2.97, MSE of 4.77, and RMSE of 2.18, based on a K-fold value of 10, utilizing the Student-mat.csv dataset. These findings highlight the model’s potential in early identification of at-risk students, enabling educators to develop targeted learning strategies. This research contributes to educational data mining by advancing predictive modeling techniques that evaluate student performance. Full article

Medical imaging, particularly chest X-rays, plays a vital role in radiological diagnosis. However, interpreting these images and generating detailed diagnostic reports is a time-consuming task for clinicians. To address this challenge, this study proposes an automated image captioning framework for chest X-ray images, aiming to reduce clinical workload and enhance diagnostic efficiency. The proposed approach employs convolutional neural networks (CNNs) for visual feature extraction and a modified Transformer architecture—referred to as the Medical Transformer—for structured report generation. Three CNN models, namely InceptionV3, ResNet152V2, and Inception–ResNetV2, were evaluated as feature extractors. The attention mechanisms, Bahdanau, Luong, and scaled dot product, were activated by ReLU or Tanh functions to identify the optimal configuration, i.e., the maximum attention is used. Experiments were conducted using the Indiana University Chest X-ray dataset, which contains 7466 images paired with corresponding diagnostic reports. The proposed approach employs image augmentation to accommodate input variability, utilizes Inception–ResNetV2 for feature extraction, and integrates the Medical Transformer with maximum attention mechanisms to achieve optimal performance in medical report generation. Evaluation metrics include BLEU (BLEU-1 to BLEU-4 scores of 0.720, 0.669, 0.648, and 0.600, respectively), METEOR (0.741), and BERTScore (F_BERT = 0.787), demonstrating superior performance compared to baseline models and the state of the art. These results validate the effectiveness of the proposed Medical Transformer framework in generating accurate and clinically relevant medical image captions. Full article

To address the challenges of high model complexity, substantial computational resource consumption, and insufficient classification accuracy in existing soybean seed identification research, we first perform soybean seed segmentation based on polygon features, constructing a dataset comprising five categories: whole seeds, broken seeds, seeds with epidermal damage, immature seeds, and spotted seeds. The MobileViT module is then optimized by employing Depthwise Separable Convolution (DSC) in place of standard convolutions, applying Transformer Half-Dimension (THD) for dimensional reconstruction, and integrating Dynamic Channel Recalibration (DCR) to reduce model parameters and enhance inter-channel interactions. Furthermore, by incorporating the CBAM attention mechanism into the MV2 module and replacing the ReLU6 activation function with the Mish activation function, the model’s feature extraction capability and generalization performance are further improved. These enhancements culminate in a novel soybean seed detection model, MobileViT-SD (MobileViT for Soybean Detection). Experimental results demonstrate that the proposed MobileViT-SD model contains only 2.09 million parameters while achieving a classification accuracy of 98.39% and an F1 score of 98.38%, representing improvements of 2.86% and 2.88%, respectively, over the original MobileViT model. Comparative experiments further show that MobileViT-SD not only outperforms several representative lightweight models in both detection accuracy and efficiency but also surpasses a number of mainstream heavyweight models. Its highly optimized, lightweight architecture combines efficient inference performance with low resource consumption, making it well-suited for deployment in computing-constrained environments, such as edge devices. Full article

Skid resistance is a key factor in road safety, directly affecting vehicle stability and braking efficiency. To enhance predictive accuracy, this study develops a multilayer perceptron (MLP) model for forecasting the Sideway Force Coefficient (SFC) of asphalt pavements and systematically examines the role of activation functions and optimizers. Seven activation functions (Sigmoid, Tanh, ReLU, Leaky ReLU, ELU, Mish, Swish) and three optimizers (SGD, RMSprop, Adam) are evaluated using regression metrics (MSE, RMSE, MAE, R²) and loss-curve analysis. Results show that ReLU and Mish provide notable improvements over Sigmoid, with ReLU increasing goodness of fit and accuracy by 13–15%, and Mish further enhancing nonlinear modeling by 12–14%. For optimizers, Adam achieves approximately 18% better performance than SGD, offering faster convergence, higher accuracy, and stronger stability, while RMSprop shows moderate performance. The findings suggest that combining ReLU or Mish with Adam yields highly precise and robust predictions under multi-source heterogeneous inputs. This study offers a reliable methodological reference for intelligent pavement condition monitoring and supports safety management in highway transportation systems. Full article

Collisions between ships and sea ice pose a significant threat to maritime safety, making it essential to detect sea ice and perform safety-oriented path planning for polar navigation. This paper utilizes an optimized You Only Look Once version 5 (YOLOv5) model, designated as YOLOv5-ICE, for the detection of sea ice in satellite imagery, with the resultant detection data being employed to input obstacle coordinates into a ship path planning system. The enhancements include the Squeeze-and-Excitation (SE) attention mechanism, improved spatial pyramid pooling, and the Flexible ReLU (FReLU) activation function. The improved YOLOv5-ICE shows enhanced performance, with its mAP increasing by 3.5% compared to the baseline YOLOv5 and also by 1.3% compared to YOLOv8. YOLOv5-ICE demonstrates robust performance in detecting small sea ice targets within large-scale satellite images and excels in high ice concentration regions. For path planning, the Any-Angle Path Planning on Grids algorithm is applied to simulate routes based on detected sea ice floes. The objective function incorporates the path length, number of ship turns, and sea ice risk value, enabling path planning under varying ice concentrations. By integrating detection and path planning, this work proposes a novel method to enhance navigational safety in polar regions. Full article

With the advancement of intelligent technology and the rise in labor costs, manual identification and cutting of 3D cotton cup indentations can no longer meet modern demands. The increasing variety and shape of 3D cotton cups due to personalized requirements make the use of fixed molds for cutting inefficient, leading to a large number of molds and high costs. Therefore, this paper proposes a UNet-based indentation segmentation algorithm to automatically extract 3D cotton cup indentation data. By incorporating the VGG16 network and Leaky-ReLU activation function into the UNet model, the method improves the model’s generalization capability, convergence speed, detection speed, and reduces the risk of overfitting. Additionally, attention mechanisms and an Atrous Spatial Pyramid Pooling (ASPP) module are introduced to enhance feature extraction, improving the network’s spatial feature extraction ability. Experiments conducted on a self-made 3D cotton cup dataset demonstrate a precision of 99.53%, a recall of 99.69%, a mIoU of 99.18%, and an mPA of 99.73%, meeting practical application requirements. The extracted 3D cotton cup indentation contour data is automatically input into an intelligent CNC cutting machine to cut 3D cotton cup. The cutting results of 400 data points show an 0.20 mm ± 0.42 mm error, meeting the cutting accuracy requirements for flexible material 3D cotton cups. This study may serve as a reference for machine vision, image segmentation, improvements to deep learning architectures, and automated cutting machinery for flexible materials such as fabrics. Full article

We present a deep learning model for the recognition of speech commands in the English language. The dataset is based on the Google Speech Commands Dataset by Warden P., version 0.01, and it consists of ten distinct commands (“left”, “right”, “go”, “stop”, “up”, “down”, “on”, “off”, “yes”, and “no”) along with additional “silence” and “unknown” classes. The dataset is split in a speaker-independent manner, with 70% of speakers assigned to the training set and 15% to the test set and validation set. All audio clips are sampled at 16 kHz, with a total of 46 146 clips. Audio files are converted into Mel spectrogram representations, which are then used as input to a deep learning model composed of a four-layer convolutional neural network followed by two fully connected layers. The model employs Rectified Linear Unit (ReLU) activation, the Adam optimizer, and dropout regularization to improve generalization. The achieved testing accuracy is 96.05%. Micro- and macro-averaged precision, recall, and F1-score of 95% are reported to reflect class-wise performance, and a confusion matrix is also provided. The proposed model has been deployed on a Raspberry Pi 5 as a Fog computing device for real-time speech recognition applications. Full article

Background: Titanium dental implants are widely used, but their long-term mechanical reliability under fatigue loading remains a key concern. Traditional finite element analysis is accurate but computationally intensive. This study explores the integration of finite element analysis data with neural networks to predict fatigue-related responses efficiently. Methods: A dataset of 200 finite element analysis simulations was generated, varying load intensity, load angle, and implant size. Each simulation provided three outputs: maximum von Mises stress, maximum displacement, and fatigue safety factor. A feedforward neural network with two hidden layers (64 neurons each, ReLU activation) was trained using 160 simulations, with 40 reserved for testing. Results: The neural network achieved high accuracy across all outputs, with R² values of 0.97 for stress, 0.95 for deformation, and 0.92 for the fatigue safety factor. Mean errors across the test set were below 5%, indicating strong predictive performance under diverse conditions. Conclusions: The findings demonstrate that neural networks can reliably replicate finite element analysis outcomes with significantly reduced computational time. This approach offers a promising tool for accelerating implant assessment and supports the growing role of AI in biomechanical design and analysis. Full article

The increasing difficulty of handling industrial and agricultural wastes has generated interest in reusing materials such as Cement Kiln Dust (CKD) and Rice Husk Ash (RHA) for sustainable soil stabilization. This study examined the enhancement of lateritic soil with the incorporation of CKD (0–12%) and RHA (0–25%) by weight. An integrated experimental and Artificial Neural Network (ANN) methodology was utilized to evaluate and forecast geotechnical features. Laboratory assessments were conducted to measure Atterberg limits, Maximum Dry Density (MDD), Optimum Moisture Content (OMC), and Unconfined Compressive Strength (UCS) at 0, 7, and 28 days of curing. The results indicated significant enhancements in soil characteristics with CKD-RHA combinations. Artificial Neural Network models, including GELU, LOGSIG-3, and Leaky ReLU activation functions, accurately predicted the UCS, MDD, and OMC, achieving R² values as high as 0.980. This work underscores the efficacy of CKD-RHA mixtures in improving soil stability and the promise of ANN models as excellent prediction instruments, fostering sustainable and economical construction methodologies. Full article

To improve the performance of image super-resolution reconstruction, this paper optimizes the classical SRGAN model architecture. The original SRResNet is replaced with the EDSR network as the generator, which effectively enhances the ability to restore image details. To address the issue of insufficient multi-scale feature extraction in SRGAN during image reconstruction, an LSK attention mechanism is introduced into the generator. By fusing features from different receptive fields through parallel multi-scale convolution kernels, the model improves its ability to capture key details. To mitigate the instability and overfitting problems in the discriminator training, the Mish activation function is used instead of LeakyReLU to improve gradient flow, and a Dropout layer is introduced to enhance the discriminator’s generalization ability, preventing overfitting to the generator. Additionally, a staged training strategy is employed during adversarial training. Experimental results show that the improved model effectively enhances image reconstruction quality while maintaining low complexity. The generated results exhibit clearer details and more natural visual effects. On the public datasets Set5, Set14, and BSD100, compared to the original SRGAN, the PSNR and SSIM metrics improved by 13.4% and 5.9%, 9.9% and 6.0%, and 6.8% and 5.8%, respectively, significantly enhancing the reconstruction of super-resolution images, achieving more refined and realistic image quality improvement. The model also demonstrates stronger generalization ability on complex cross-domain data, such as remote sensing images and medical images. The improved model achieves higher-quality image reconstruction and more natural visual effects while maintaining moderate computational overhead, validating the effectiveness of the proposed improvements. Full article

This study proposes a hybrid deep learning framework for predicting vehicle crash frequency (F_i) using nationwide traffic accident and traffic volume data from the United States (2019–2022). Crash frequency is defined as the product of exposure frequency ($N_a$) and crash risk rate ($\lambda$), a structure widely adopted for its ability to separate physical exposure from the crash likelihood. $N_a$ was computed using an extended Safety Performance Function (SPF) that incorporates roadway traffic volume, segment length, number of lanes, and traffic density, while $\lambda$ was estimated using a multilayer perceptron-based deep neural network (DNN) with inputs such as impact speed, road surface condition, and vehicle characteristics. The DNN integrates rectified linear unit (ReLU) activation, batch normalization, dropout layers, and the Huber loss function to capture nonlinearity and over-dispersion beyond the capability of traditional statistical models. Model performance, evaluated through five-fold cross-validation, achieved R² = 0.7482, MAE = 0.1242, and MSE = 0.0485, demonstrating a strong capability to identify high-risk areas. Compared to traditional regression approaches such as Poisson and negative binomial models, which are often constrained by equidispersion assumptions and limited flexibility in capturing nonlinear effects, the proposed framework demonstrated substantially improved predictive accuracy and robustness. Unlike prior studies that loosely combined SPF terms with machine learning, this study explicitly decomposes F_i into $N_a$ and $\lambda$, ensuring interpretability while leveraging DNN flexibility for crash risk estimation. This dual-layer integration provides a unique methodological contribution by jointly achieving interpretability and predictive robustness, validated with a nationwide dataset, and highlights its potential for evidence-based traffic safety assessments and policy development. Full article

In modern industry, the spring full-lift safety valve is a key device for safe pressure relief of pressure-bearing systems. Its valve seat sealing surface is easily damaged after long-term use, causing internal leakage, resulting in safety hazards and economic losses. Therefore, it is of great significance to quickly and accurately diagnose its internal leakage state. Among the current methods for identifying fluid machinery faults, model-based methods have difficulties in parameter determination. Although the data-driven convolutional neural network (CNN) has great potential in the field of fault diagnosis, it has problems such as hyperparameter selection relying on experience, insufficient capture of time series and multi-scale features, and lack of research on valve internal leakage type identification. To this end, this study proposes a safety valve internal leakage identification method based on high-frequency FPGA data acquisition and improved CNN. The acoustic emission signals of different internal leakage states are obtained through the high-frequency FPGA acquisition system, and the two-dimensional time–frequency diagram is obtained by short-time Fourier transform and input into the improved model. The model uses the leaky rectified linear unit (LReLU) activation function to enhance nonlinear expression, introduces random pooling to prevent overfitting, optimizes hyperparameters with the help of horned lizard optimization algorithm (HLOA), and integrates the bidirectional gated recurrent unit (BiGRU) and selective kernel attention module (SKAM) to enhance temporal feature extraction and multi-scale feature capture. Experiments show that the average recognition accuracy of the model for the internal leakage state of the safety valve is 99.7%, which is better than the comparison model such as ResNet-18. This method provides an effective solution for the diagnosis of internal leakage of safety valves, and the signal conversion method can be extended to the fault diagnosis of other mechanical equipment. In the future, we will explore the fusion of lightweight networks and multi-source data to improve real-time and robustness. Full article