Search Results (22,039)

This research presents the development of a novel object detection algorithm designed to identify humans in natural outdoor environments using minimal computational resources. The proposed system, SARGAS, combines a custom convolutional neural network (CNN) classifier with MOG2 background subtraction and partial affine transformations for camera stabilization. A secondary CNN refines detections and reduces false positives. Unlike conventional supervised models, SARGAS is trained in a partially unsupervised manner, learning to recognize feature patterns without requiring labeled data. The algorithm achieved a recall of 93%, demonstrating strong detection capability even under challenging conditions. However, the overall accuracy reached 65%, due to a higher rate of false positives—an expected trade-off when maximizing recall. This bias is intentional, as missing a human target in search and rescue applications carries a higher cost than producing additional false detections. While supervised models, such as YOLOv5, perform well on data resembling their training sets, they exhibit significant performance degradation on previously unseen footage. In contrast, SARGAS generalizes more effectively, making it a promising candidate for real-world deployment in environments where labeled training data is limited or unavailable. The results establish a solid foundation for further improvements and suggest that unsupervised CNN-based approaches hold strong potential in object detection tasks. Full article

Cold rolling is widely recognized as a key industrial process for enhancing the mechanical properties of materials, particularly hardness, through strain hardening. Despite its importance, accurately predicting the final hardness remains a challenge due to the inherently nonlinear nature of the deformation. While several studies have employed artificial neural networks to predict mechanical properties, architectural parameters still need to be investigated to understand their effects on network behavior and model performance, ultimately supporting the design of more effective architectures. This study investigates hyperparameter tuning in artificial neural networks trained using Resilient Backpropagation by evaluating the impact of varying number of hidden layers and neurons on the prediction accuracy of hardness in 70-30 brass specimens subjected to cold rolling. A dataset of 1000 input–output pairs, containing dimensional and hardness measurements from multiple rolling passes, was used to train and evaluate 819 artificial neural network architectures, each with a different configuration of 1 to 3 hidden layers and 4 to 12 neurons per layer. Each configuration was tested over 50 runs to reduce the influence of randomness and enhance result consistency. Enhancing the network depth from one to two hidden layers improved predictive performance. Architectures with two hidden layers achieved better performance metrics, faster convergence, and lower variation than single-layer networks. Introducing a third hidden layer did not yield meaningful improvements over two-hidden-layer architectures in terms of performance metrics. While the top three-layer model converged in fewer epochs, it required more computational time due to increased model complexity and weight elements. Full article

The synthesis of “green ammonia” from “green hydrogen” represents a critical pathway for renewable energy integration and industrial decarbonization. This study investigates the green ammonia synthesis process using an axial–radial fixed-bed reactor equipped with three catalyst layers. A simplified two-dimensional physical model was developed, and a multiscale simulation approach combining computational fluid dynamics (CFD) with physics-informed neural networks (PINNs) employed. The simulation results demonstrate that the majority of fluid flows axially through the catalyst beds, leading to significantly higher temperatures in the upper bed regions. The reactor exhibits excellent heat exchange performance, ensuring effective preheating of the feed gas. High-pressure zones are concentrated near the top and bottom gas outlets, while the ammonia mole fraction approaches 100% near the bottom outlet, confirming superior conversion efficiency. By integrating PINNs, the prediction accuracy was substantially improved, with flow field errors in the catalyst beds below 4.5% and ammonia concentration prediction accuracy above 97.2%. Key reaction kinetic parameters (pre-exponential factor k₀ and activation energy Ea) were successfully inverted with errors within 7%, while computational efficiency increased by 200 times compared to traditional CFD. The proposed CFD–PINN integrated framework provides a high-fidelity and computationally efficient simulation tool for green ammonia reactor design, particularly suitable for scenarios with fluctuating hydrogen supply. The reactor design reduces energy per unit ammonia and improves conversion efficiency. Its radial flow configuration enhances operational stability by damping feed fluctuations, thereby accelerating green hydrogen adoption. By reducing fossil fuel dependence, it promotes industrial decarbonization. Full article

The surface of the Great Wall harbors a large number of non-vascular plants dominated by cyanobacteria, lichens and mosses as well as microorganisms, and form biocrusts by cementing with the soils and greatly alters the pore structure of the soil and the ecohydrological processes associated with the soil pore space, and thus influences the soil resistance to erosion. However, the microscopic role of the biocrusts in influencing the pore structure of the surface of the Great Wall is not clear. This study chose the Warring States Qin Great Wall in Weiyuan, Gansu Province, China, as research site to quantify thepore structure characteristics of the three-dimensional of bare soil, cyanobacterial-lichen crusts, and moss crusts at the depth of 0–50 mm, by using optical microscopy, scanning electron microscopy, and X-ray computed tomography and image analysis, and the precipitation infiltration process. The results showed that the moss crust layer was dominated by large pores with long extension and good connectivity, which provided preferential seepage channels for precipitation infiltration, while the connectivity between the cyanobacterial-lichen crust voids was poor; The porosity of the cyanobacterial-lichen crust and the moss crust was 500% and 903.27% higher than that of the bare soil, respectively. The porosity of the subsurface layer of cyanobacterial-lichen crust and moss crust was significantly lower than that of the biocrusts layer by 92.54% and 97.96%, respectively, and the porosity of the moss crust was significantly higher than that of the cyanobacterial-lichen crust in the same layer; Cyanobacterial-lichen crusts increased the degree of anisotropy, mean tortuosity, moss crust reduced the degree of anisotropy, mean tortuosity. Biocrusts increased the fractal dimension and Euler number of pores. Compared with bare soil, moss crust and cyanobacterial-lichen crust increased the isolated porosity by 2555% and 4085%, respectively; Biocrusts increased the complexity of the pore network models; The initial infiltration rate, stable infiltration rate, average infiltration rate, and the total amount of infiltration of moss crusted soil was 2.26 and 3.12 times, 1.07 and 1.63 times, respectively, higher than that of the cyanobacterial-lichen crusts and the bare soil, by 1.53 and 2.33 times, and 1.13 and 2.08 times, respectively; CT porosity and clay content are significantly positively correlated with initial soil infiltration rate (|r| ≥ 0.85), while soil type and organic matter content are negatively correlated with initial soil infiltration rate. The soil type and bulk density are directly positively and negatively correlated with CT porosity, respectively (|r| ≥ 0.52). There is a significant negative correlation between soil clay content and porosity Full article

In this paper, we study the classification problem of short-time-window steady-state visual evoked potentials (SSVEPs) and propose a novel deep convolutional network named EEGResNet based on the idea of residual connection to further improve the classification performance. Since the frequency-domain features extracted from short-time-window signals are difficult to distinguish, the EEGResNet starts from the filter bank (FB)-based feature extraction module in the time domain. The FB designed in this paper is composed of four sixth-order Butterworth filters with different bandpass ranges, and the four bandwidths are 19–50 Hz, 14–38 Hz, 9–26 Hz, and 3–14 Hz, respectively. Then, the extracted four feature tensors with the same shape are directly aggregated together. Furthermore, the aggregated features are further learned by a six-layer convolutional neural network with residual connections. Finally, the network output is generated through an adaptive fully connected layer. To prove the effectiveness and superiority of our designed EEGResNet, necessary experiments and comparisons are conducted over two large public datasets. To further verify the application potential of the trained network, a virtual simulation of brain computer interface (BCI) based quadrotor control is presented through V-REP. Full article

The use of informal language on social media and the sheer volume of information make it difficult for a computer system to analyse it automatically. The aim of this work is to design a new group decision-making method that applies two new consensus methods based on sentiment analysis. This method is designed for application in the analysis of texts on social media. To test the method, we will use posts from the so called social network X. The proposed model differs from previous work in this field by defining a new degree of subjectivity and a new degree of reliability associated with user opinions. This work also presents two new consensus measures, one focused on measuring the number of words classified as positive and negative and the other on analysing the percentage of occurrence of those words. Our method allows us to automatically extract preferences from the transcription of the texts used in the debate, avoiding the need for users to explicitly indicate their preferences. The application to a real case of public investment demonstrates the effectiveness of the approach in collaborative contexts that used natural language. Full article

The Surface Water and Ocean Topography (SWOT) mission provides unprecedented high-resolution observations of sea-surface height. However, their direct use in ocean circulation studies is complicated by the presence of small-scale unbalanced motion signals and instrumental noise, which hinder accurate estimation of geostrophic velocities. To address these limitations, we developed an adaptive convolutional neural network (CNN)-based filtering technique that refines SWOT-derived sea-level observations. The network includes multi-head attention layers to exploit information on concurrent wind fields and standard altimetry interpolation errors. We train the model with a custom loss function that accounts for the differences between geostrophic velocities computed from SWOT sea-surface topography and simultaneous in-situ drifter velocities. We compare our method to existing filtering techniques, including a U-Net-based model and a variational noise-reduction filter. Our adaptive-filtering CNN produces accurate velocity estimates while preserving small-scale features and achieving a substantial noise reduction in the spectral domain. By combining satellite and in-situ data with machine learning, this work demonstrates the potential of an adaptive CNN-based filtering approach to enhance the accuracy and reliability of SWOT-derived sea-level and velocity estimates, providing a valuable tool for global oceanographic applications. Full article

The optimization of high-frequency circuits remains a computationally intensive task due to the need for repeated high-fidelity electromagnetic (EM) simulations. To address this challenge, we propose a novel integration of machine learning-generated inverse maps within the space mapping (SM) optimization framework to significantly accelerate circuit optimization while maintaining high accuracy. The proposed approach leverages Bayesian Neural Networks (BNNs) and surrogate modeling techniques to construct an inverse mapping function that directly predicts design parameters from target performance metrics, bypassing iterative forward simulations. The methodology was validated using a low-pass filter optimization scenario, where the inverse surrogate model was trained using electromagnetic simulations from COMSOL Multiphysics 2024 r6.3 and optimized using MATLAB R2024b r24.2 trust region algorithm. Experimental results demonstrate that our approach reduces the number of high-fidelity simulations by over 80% compared to conventional SM techniques while achieving high accuracy with a mean absolute error (MAE) of 0.0262 (0.47%). Additionally, convergence efficiency was significantly improved, with the inverse surrogate model requiring only 31 coarse model simulations, compared to 580 in traditional SM. These findings demonstrate that machine learning-driven inverse surrogate modeling significantly reduces computational overhead, accelerates optimization, and enhances the accuracy of high-frequency circuit design. This approach offers a promising alternative to traditional SM methods, paving the way for more efficient RF and microwave circuit design workflows. Full article

Landslide susceptibility modelling represents an efficient approach to enhance disaster management and mitigation strategies. The focus of this paper lies in the development of a landslide susceptibility evaluation in northwestern Tunisia using the Analytical Hierarchy Process (AHP) and Artificial Neural Network (ANN) approaches. The used database covers 286 landslides, including ten landslide factor maps: rainfall, slope, aspect, topographic roughness index, lithology, land use and land cover, distance from streams, drainage density, lineament density, and distance from roads. The AHP and ANN approaches were applied to classify the factors by analyzing the correlation relationship between landslide distribution and the significance of associated factors. The Landslide Susceptibility Index result reveals five susceptible zones organized from very low to very high risk, where the zones with the highest risks are associated with the combination of extreme amounts of rainfall and steep slope. The performance of the models was confirmed utilizing the area under the Relative Operating Characteristic (ROC) curves. The computed ROC curve (AUC) values (0.720 for ANN and 0.651 for AHP) convey the advantage of the ANN method compared to the AHP method. The overlay of the landslide inventory data locations of historical landslides and susceptibility maps shows the concordance of the results, which is in favor of the established model reliability. Full article

Background/Objectives: The integration of artificial intelligence (AI) in medical imaging is rapidly advancing, yet its application in gynecologic use remains limited. This proof-of-concept study presents the development and validation of a convolutional neural network (CNN) designed to automatically detect and classify endometrial polyps. Methods: A multicenter dataset (n = 3) comprising 65 hysteroscopies was used, yielding 33,239 frames and 37,512 annotated objects. Still frames were extracted from full-length videos and annotated for the presence of histologically confirmed polyps. A YOLOv1-based object detection model was used with a 70–20–10 split for training, validation, and testing. Primary performance metrics included recall, precision, and mean average precision at an intersection over union (IoU) ≥ 0.50 (mAP50). Frame-level classification metrics were also computed to evaluate clinical applicability. Results: The model achieved a recall of 0.96 and precision of 0.95 for polyp detection, with a mAP50 of 0.98. At the frame level, mean recall was 0.75, precision 0.98, and F1 score 0.82, confirming high detection and classification performance. Conclusions: This study presents a CNN trained on multicenter, real-world data that detects and classifies polyps simultaneously with high diagnostic and localization performance, supported by explainable AI features that enhance its clinical integration and technological readiness. Although currently limited to binary classification, this study demonstrates the feasibility and potential of AI to reduce diagnostic subjectivity and inter-observer variability in hysteroscopy. Future work will focus on expanding the model’s capabilities to classify a broader range of endometrial pathologies, enhance generalizability, and validate performance in real-time clinical settings. Full article

Recent advancements have been made in the precise prediction of protein structures within the Protein Folding Problem (PFP), particularly in relation to minimizing the energy function to achieve stable and biologically relevant protein structures. This problem is classified as NP-hard within computational theory, necessitating the development of various techniques and algorithms. Bio-inspired algorithms have proven effective in addressing NP-hard challenges in practical applications. This study introduces a novel hybrid algorithm, termed GRSABio, which integrates the strategies of Jumping Spider Algorithm (JSOA) with the Golden Ratio Simulated Annealing (GRSA) for peptide prediction. Furthermore, the GRSABio algorithm incorporates a Convolutional Neural Network for fragment prediction (FCNN), forms an enhanced methodology called GRSABio-FCNN. This integrated framework achieves improved structure refinement based on energy for protein prediction. The proposed enhanced GRSABio-FCNN approach was applied to a dataset of 60 peptides. The Wilcoxon and Friedman statistics test were employed to compare the GRSABio-FCNN results against recent state-of-the-art-approaches. The results of these tests indicate that the GRSABio-FCNN approach is competitive with state-of-the-art methods for peptides up to 50 amino acids in length and surpasses leading PFP algorithms for peptides with up to 30 amino acids. Full article

The use of a stochastic artificial neural network (SANN) implemented on a Field Programmable Gate Array (FPGA) provides a promising method of performing image recognition on event camera recordings, however, challenges exist due to the fact that event camera data has an inherent unevenness in the timing at which data is sent out of the camera. This paper proposes a sampling system to overcome this challenge, by which all “events” occurring at specific timestamps in an event camera recording are selected (sampled) to be processed and sent to the SANN at regular intervals. This system is implemented on an FPGA in SystemVerilog, and to test it, simulated event camera data is sent to the system from a computer running MATLAB (version 2022+). The sampling system is shown to be functional. Analysis is shown demonstrating its performance regarding data sparsity, time convergence, normalization, repeatability, range, and some characteristics of the hold system. Full article

Accurate forecasting of tropical cyclone (TC) tracks is critical for disaster preparedness and risk mitigation. While traditional numerical weather prediction (NWP) systems have long served as the backbone of operational forecasting, they face limitations in computational cost and sensitivity to initial conditions. In recent years, deep learning (DL) has emerged as a promising alternative, offering data-driven modeling capabilities for capturing nonlinear spatiotemporal patterns. This paper presents a comprehensive review of DL-based approaches for TC track forecasting. We categorize all DL-based TC tracking models according to the architecture, including recurrent neural networks (RNNs), convolutional neural networks (CNNs), Transformers, graph neural networks (GNNs), generative models, and Fourier-based operators. To enable rigorous performance comparison, we introduce a Unified Geodesic Distance Error (UGDE) metric that standardizes evaluation across diverse studies and lead times. Based on this metric, we conduct a critical comparison of state-of-the-art models and identify key insights into their relative strengths, limitations, and suitable application scenarios. Building on this framework, we conduct a critical cross-model analysis that reveals key trends, performance disparities, and architectural tradeoffs. Our analysis also highlights several persistent challenges, such as long-term forecast degradation, limited physical integration, and generalization to extreme events, pointing toward future directions for developing more robust and operationally viable DL models for TC track forecasting. To support reproducibility and facilitate standardized evaluation, we release an open-source UGDE conversion tool on GitHub. Full article

Hormone-dependent breast cancer, particularly in its treatment-resistant forms, remains a significant therapeutic challenge. In this study, we applied a fully computational strategy to design steroid-based compounds capable of simultaneously targeting three key receptors involved in disease progression: progesterone receptor (PR), estrogen receptor alpha (ER-α), and HER2. Using a robust 3D-QSAR model (R² = 0.86; Q²_LOO = 0.86) built from 52 steroidal structures, we identified molecular features associated with high anticancer potential, specifically increased polarizability and reduced electronegativity. From a virtual library of 271 DFT-optimized analogs, 31 compounds were selected based on predicted potency (pIC₅₀ > 7.0) and screened via molecular docking against PR (PDB 2W8Y), HER2 (PDB 7JXH), and ER-α (PDB 6VJD). Seven candidates showed strong binding affinities (ΔG ≤ −9 kcal/mol for at least two targets), with Estero-255 emerging as the most promising. This compound demonstrated excellent conformational stability, a robust hydrogen-bonding network, and consistent multitarget engagement. Molecular dynamics simulations over 100 nanoseconds confirmed the structural integrity of the top ligands, with low RMSD values, compact radii of gyration, and stable binding energy profiles. Key interactions included hydrophobic contacts, π–π stacking, halogen–π interactions, and classical hydrogen bonds with conserved residues across all three targets. These findings highlight Estero-255, alongside Estero-261 and Estero-264, as strong multitarget candidates for further development. By potentially disrupting the PI3K/AKT/mTOR signaling pathway, these compounds offer a promising strategy for overcoming resistance in hormone-driven breast cancer. Experimental validation, including cytotoxicity assays and ADME/Tox profiling, is recommended to confirm their therapeutic potential. Full article

Background: Pancreatic cystic lesions (PCLs), including intraductal papillary mucinous neoplasms (IPMNs) and mucinous cystic neoplasms (MCNs), pose a diagnostic challenge due to their variable malignant potential. Current guidelines, such as Fukuoka and American Gastroenterological Association (AGA), have moderate predictive accuracy and may lead to overtreatment or missed malignancies. Artificial intelligence (AI), incorporating machine learning (ML) and deep learning (DL), offers the potential to improve risk stratification, diagnosis, and management of PCLs by integrating clinical, radiological, and molecular data. This is the first systematic review to evaluate the application, performance, and clinical utility of AI models in the diagnosis, classification, prognosis, and management of pancreatic cysts. Methods: A systematic review was conducted in accordance with PRISMA guidelines and registered on PROSPERO (CRD420251008593). Databases searched included PubMed, EMBASE, Scopus, and Cochrane Library up to March 2025. The inclusion criteria encompassed original studies employing AI, ML, or DL in human subjects with pancreatic cysts, evaluating diagnostic, classification, or prognostic outcomes. Data were extracted on the study design, imaging modality, model type, sample size, performance metrics (accuracy, sensitivity, specificity, and area under the curve (AUC)), and validation methods. Study quality and bias were assessed using the PROBAST and adherence to TRIPOD reporting guidelines. Results: From 847 records, 31 studies met the inclusion criteria. Most were retrospective observational (n = 27, 87%) and focused on preoperative diagnostic applications (n = 30, 97%), with only one addressing prognosis. Imaging modalities included Computed Tomography (CT) (48%), endoscopic ultrasound (EUS) (26%), and Magnetic Resonance Imaging (MRI) (9.7%). Neural networks, particularly convolutional neural networks (CNNs), were the most common AI models (n = 16), followed by logistic regression (n = 4) and support vector machines (n = 3). The median reported AUC across studies was 0.912, with 55% of models achieving AUC ≥ 0.80. The models outperformed clinicians or existing guidelines in 11 studies. IPMN stratification and subtype classification were common focuses, with CNN-based EUS models achieving accuracies of up to 99.6%. Only 10 studies (32%) performed external validation. The risk of bias was high in 93.5% of studies, and TRIPOD adherence averaged 48%. Conclusions: AI demonstrates strong potential in improving the diagnosis and risk stratification of pancreatic cysts, with several models outperforming current clinical guidelines and human readers. However, widespread clinical adoption is hindered by high risk of bias, lack of external validation, and limited interpretability of complex models. Future work should prioritise multicentre prospective studies, standardised model reporting, and development of interpretable, externally validated tools to support clinical integration. Full article