Search Results (28,453)

Wireless sensor network (WSN) coverage optimization is a challenging high-dimensional and nonlinear problem that directly affects network performance, including sensing quality, energy efficiency, and system reliability. Although metaheuristic algorithms have been widely applied to this problem, many existing methods still suffer from premature convergence, insufficient population diversity, and an imbalance between exploration and exploitation. To address these issues, this paper proposes a multi-strategy enhanced enterprise development optimization algorithm (MEEDOA) inspired by business management mechanisms. The proposed method integrates a hybrid population initialization strategy, an adaptive activity switching mechanism based on performance feedback, a multi-elite collaborative learning strategy, and a Lévy flight-based stagnation escape mechanism. These strategies are tightly coupled within a unified adaptive framework to improve global search capability, convergence speed, and robustness. Furthermore, a mathematical model for WSN deployment is constructed based on a binary sensing model and discrete coverage evaluation. From the perspective of symmetry, the sensing regions of sensor nodes exhibit significant geometric symmetry in both two-dimensional and three-dimensional deployment spaces. In the two-dimensional case, the sensing and communication regions are modeled as concentric circular structures, while in the three-dimensional case, the sensing regions are represented by isotropic spheres with symmetric spatial distributions. Such symmetry properties provide an effective basis for describing coverage behavior, reducing redundant overlap, and improving the uniformity of node deployment. Meanwhile, the proposed MEEDOA preserves population diversity and enhances search balance, enabling the algorithm to better capture symmetric coverage patterns and more effectively explore complex spatial deployment configurations. Extensive experiments on CEC2014, CEC2017, CEC2020, and CEC2022 benchmark functions demonstrate that MEEDOA achieves superior convergence accuracy, faster convergence speed, and stronger robustness compared with several state-of-the-art algorithms. Additional simulation results in WSN deployment applications verify its effectiveness in improving coverage performance under both symmetric and irregular spatial deployment scenarios. The results indicate that the proposed MEEDOA provides a reliable and efficient solution for complex global optimization problems and practical engineering applications. Full article

Event cameras capture sparse, high-temporal-resolution visual information, making them attractive for challenging scenarios with fast motion and severe illumination changes. However, event-based depth models trained on one real-world benchmark often degrade substantially when transferred to another, revealing a practical cross-dataset domain shift between real sensor datasets. In this work, we study parameter-efficient adaptation from MVSEC to DSEC using a frozen VFM-based recurrent depth backbone. We systematically compare several parameter-efficient fine-tuning (PEFT) strategies, including Bias-only, Adapter, Decoder Weight Tuning, ConvLSTM-only, and FiLM-based modulation, under labeled few-shot adaptation. Across three random seeds, Bias-only achieves the best few-shot accuracy, reaching 0.189 AbsRel with 150 calibration samples. Decoder-side FiLM provides the best accuracy–efficiency trade-off, maintaining stable performance while updating only 2048 parameters, and reaches 0.176 AbsRel when trained with the full DSEC training set under our protocol. Our study shows that tuning native pretrained parameters is a strong baseline in this specific MVSEC → DSEC event-depth adaptation setting, whereas higher-capacity auxiliary modules are less effective under limited target-domain supervision. These results establish a controlled MVSEC → DSEC benchmark and provide practical guidance for adapting event-based monocular depth models under cross-dataset transfer. Full article

To address the limitations of conventional mobile robot path planning results in terms of geometric continuity, kinematic executability, and adaptability to dynamic environments, this study proposes a reference trajectory generation method oriented toward trajectory tracking. First, the A* algorithm is employed to search for an initial collision-free path, and key-point sparsification is applied to remove redundant nodes. Then, a geometrically continuous reference path is constructed using cubic B-splines. On this basis, by considering the kinematic constraints of the differential-drive mobile robot together with the local curvature characteristics of the path, a local trackability index is introduced, and the reference velocity is adaptively corrected under the maximum angular velocity constraint to improve trajectory executability and tracking smoothness. To address local path invalidation caused by dynamic obstacles, a collision-risk-triggered local replanning and trajectory stitching mechanism is further developed to achieve smooth transition between the original and updated trajectories. Simulation and real-world experimental results demonstrate that the proposed method can effectively reduce path redundancy, improve trajectory smoothness and executability, and achieve rapid local path updating and stable trajectory stitching in dynamic environments. Full article

The exhaust gas temperature (EGT) of the gas generator is a critical indicator for the health management system of a marine gas turbine engine. Therefore, EGT prediction can not only support predictive maintenance decision-making but also serves as a reliable virtual sensor for EGT measurement. However, the engine EGT exhibits strongly nonlinear coupling relationships with other gas path variables, which causes challenges for data-driven prediction. Graph neural networks (GNNs) are particularly effective in capturing the coupling relationships among gas path sensor variables. However, conventional static graph structures fail to characterize the varying coupling strengths under different operating conditions. In this study, a thermodynamic knowledge-driven graph attention network (TKD-GAT) method is proposed for accurate and robust EGT prediction. First, a physics-guided graph topology is constructed based on the gas turbine thermodynamic equations. Subsequently, a multi-head attention mechanism is introduced to generate edge weights that capture the varying thermodynamic coupling strengths under different operation conditions. The proposed model is evaluated on a real-world LM2500 gas turbine, which is widely used in modern propulsion systems of commercial and military ships. The ablation study confirms that the thermodynamic knowledge-driven graph topology and the attention mechanism-based edge weights are both necessary to enhance the EGT prediction performance. The TKD-GAT model shows the best performance with an RMSE of 0.446% and an R² of 0.971 compared with state-of-the-art models. The paired t-test and effect size measurement (Cohen’s d) statistically confirm the significance of performance improvements. The statistical results from multiple independent experiments prove the stability of the TKD-GAT model. Additionally, the model achieves a competitive computational cost despite the integration of a physics-guided graph topology and attention mechanisms. Crucially, an interpretability analysis confirms that the learned attention weights adhere to thermodynamic principles under different operation conditions. The proposed TKD-GAT model provides an effective solution for EGT prediction in health management systems. Full article

This paper introduced the Enhanced Puzzle Optimization Algorithm (EPOA), a hybrid metaheuristic that augmented the original Puzzle Optimization Algorithm (POA) with uniform crossover, random-resetting mutation, and explicit elitism. The contribution does not lie in inventing these operators individually, since they are classical search components, but in integrating them into POA’s two-phase search dynamics to address premature convergence, diversity loss, and best-solution preservation in a targeted manner. This paper formalized EPOA’s update rules, provided pseudocode and flow diagrams, and enforced bound handling for box-constrained problems. Comprehensive tests on the CEC2022 single-objective benchmark suite (F₁–F₁₂) showed that EPOA attained rank 1 on 11 of 12 functions and rank 3 on the remaining case, with large error reductions relative to baseline POA (e.g., on F₁, the mean error dropped from 62.836 to 0.004; on F₆, the mean error dropped from 2370.962 to 7.239). The method was further evaluated on six classical constrained engineering design problems (welded beam, tension/compression spring, speed reducer, pressure vessel, three-bar truss, and cantilever beam). Statistical indicators such as the mean and standard deviation were used to assess robustness. The results showed that EPOA delivered a strong exploration–exploitation balance and robust solution quality across rugged landscapes and real-world constraints. Full article

Early detection of suicidal ideation and depressive risk remains a critical challenge, particularly when individuals express distress implicitly through metaphorical or obfuscated language. Existing approaches primarily rely on explicit linguistic signals, limiting their effectiveness in real-world settings. This paper proposes a unified multi-aspect transformer-based framework that integrates multi-source learning, multi-task optimization, affective feature fusion, and adversarial training to detect implicit psychological risk indicators in textual data. The model jointly learns suicidal ideation detection, depression severity classification, and perceived threat detection, while incorporating emotional representations derived from valence, arousal, and polarity signals. To improve robustness, an adversarial training strategy is employed to simulate obfuscated expressions, enhancing robustness and generalization under linguistic perturbations. Interpretability is ensured through a hybrid explainable AI approach combining attention mechanisms and SHAP-based feature attribution. Extensive experiments conducted on four benchmark datasets demonstrate that the proposed approach achieves state-of-the-art performance (F1-score = 0.91), with statistically significant improvements over strong baselines. Additional analyses, including ablation studies, adversarial evaluation, and calibration assessment, confirm the effectiveness, robustness, and reliability of the proposed framework. These results highlight the potential of the model for deployment in high-stakes applications such as clinical triage and online risk monitoring, where early and interpretable detection of concealed psychological distress is essential. Full article

Data-driven exoskeletons offer the potential for adaptive augmentation of human mobility. Yet their widespread adoption is hindered by labor-intensive biomechanical data collection and manual tuning. Herein, this study presents a highly efficient synthetic data approach to facilitate data-driven pipelines. We leveraged an Adversarial Motion Priors (AMP) agent to learn stylized walking within a massively parallel, physics-based simulation. The resulting high-fidelity data were collected and validated against OpenSim inverse dynamics pipelines. Further, we trained an end-to-end torque prediction algorithm using the collected data. A novel CNN-Transformer architecture was developed to map contralateral swing-phase data to variable-length push-off torque profiles. This enabled real-time, adaptive torque assistance of exoskeletons for variable-speed walking. A custom ankle exoskeleton was used to demonstrate robust sim-to-real transferability. Our system achieved an average root mean square error of approximately 0.081 ± 0.015 newton-meters per kilogram and an average R² of 0.836 ± 0.050 across speeds ranging from 0.6 to 1.75 m·s⁻¹. The controller significantly reduced user-positive ankle mechanical work by up to 14 ± 6.30%. Finally, our multi-sensor configuration exhibited inherent fault tolerance, ensuring safe operation even under partial sensor failure. By taking a scalable, data-driven approach, this work offers a practical pathway toward deploying autonomous exoskeletons in versatile, real-world environments. Full article

Classification of vehicle engines using the chemical composition of the exhaust from these engines can be used to identify the engine’s design and verify compliance with environmental regulations through the vehicle’s emissions. This paper describes a method to identify the type of vehicles using machine learning (ML), where low-cost MQ series sensors measure the gases and particle emissions from a vehicle exhaust system, while simultaneously collecting and measuring the vehicle’s temperature and humidity levels. A custom-designed multi-sensor exhaust sensing module is employed to capture real-time exhaust emissions prior to entering the atmosphere. Exhaust samples are collected from vehicles representing three major engine categories: petrol, diesel, and compressed natural gas (CNG). In addition, fresh air samples are collected as a baseline environmental reference for comparison. All exhaust measurements are collected under controlled and consistent engine operating conditions to ensure comparable emission profiling across vehicle classes. To ensure consistent combustion-based emission profiling, this study focuses on conventional fuel-powered vehicles. MQ-series gas sensors are sensitive to combustion by-products emitted during engine operation, such as carbon monoxide (CO), hydrocarbons (HC), while also exhibiting cross-sensitivity to other gaseous components present in exhaust mixtures. Nevertheless, the proposed system performs pattern-based classification using relative sensor response signatures. Standardization of data is achieved through z-score normalization. The best models developed (based on three separate experimental designs) are trained and validated using six supervised machine learning algorithms such as Logistic Regression, Support Vector Machine (RBF), k-Nearest Neighbors, Random Forest, Gradient Boosting Decision Tree, and XGBoost and are compared against one another. Evaluation of the tested algorithms using various evaluation metrics demonstrated that ensemble models outperformed all other algorithms, achieving the highest accuracy of 99.5%. Furthermore, noise analysis confirms that the proposed solution maintains high classification accuracy (more than 89%) even under substantial sensor perturbations mimicking the real-world deployment. The solution proposed below illustrates how using gas sensors and advanced algorithms can provide accurate exhaust identification and identify engines in real-time. Full article

This paper presents a comprehensive review of bridge health identification (BHI) within the emerging paradigm of intelligent infrastructure, with a particular focus on modal analysis and artificial intelligence (AI)-driven methodologies. Aging bridge networks, increasing traffic demands, and environmental stressors have significantly accelerated structural deterioration, necessitating advanced monitoring and diagnostic frameworks. Modal parameters, including natural frequencies, mode shapes, and damping ratios, are widely recognized as reliable indicators of structural condition and form the foundation of vibration-based BHI. This study systematically reviews operational modal analysis (OMA) techniques, including frequency-domain, time-domain, and hybrid approaches, highlighting their capabilities and limitations under real-world conditions. Furthermore, the integration of AI and machine learning (ML) methods, ranging from supervised and unsupervised learning to deep learning (DL) and reinforcement learning (RL), is critically examined in the context of data-driven damage detection, feature extraction, and predictive maintenance. Special attention is given to Automated Operational Modal Analysis (AOMA), where recent advances in FDD- and SSI-based frameworks have enabled scalable and user-independent modal identification. Despite significant progress, key challenges remain, including environmental variability, data scarcity, lack of interpretability, and deployment constraints. Finally, the paper identifies major research gaps and outlines future directions toward physics-informed AI, multi-modal data fusion, uncertainty-aware decision-making, and digital twin integration. The study provides a unified perspective bridging structural dynamics and intelligent data-driven approaches, contributing to the development of next-generation smart bridge monitoring systems. Full article

Smart metering of high-time-resolution energy data enables efficient power grid management. However, it also raises significant privacy concerns by revealing users’ consumption patterns. In this paper, a novel privacy-preserving idea is introduced by utilizing a rechargeable battery (RB) to reshape the smart meter readings to statistically align with random target readings, which are preset independently of the private user energy consumption data. For the long-term privacy-preserving and cost-efficient objectives, we formulate a sequential energy management unit (EMU) policy design as a constrained Markov decision process (CMDP), where the cost-efficient objective is optimized subject to the constraint on privacy preservation. We then develop a novel variational-constraint adversarial proximal policy optimization (VCA-PPO) algorithm to solve the CMDP without requiring prior knowledge of probabilistic models. Experimental results on a standard real-world dataset demonstrate the effectiveness of the proposed method and its superiority to the load-flatness benchmark method. Full article

Background/Objectives: The survival of patients experiencing transformation of follicular lymphoma (tFL) is generally inferior to that of their non-transformed FL counterparts. While rituximab (R) has been shown to reduce transformation rates, data on Obinutuzumab (O), a third generation anti-CD20 monoclonal antibody, are limited. Methods: This retrospective study analyzed risk factors for tFL and evaluated outcomes following transformation in 1145 consecutive patients with FL (2010–2023). Results: Over a median follow-up of 70 months, 9% (n = 103) of FL patients developed tFL, with a median time of 36 months from FL diagnosis to transformation. In multivariate analysis, O-based, compared to R-based regimens and maintenance therapy (compared to no maintenance), were independently associated with reduced risk of histologic transformation (HR 0.40, 95% CI 0.19–0.90, p = 0.026 and HR 0.42, 95% CI 0.23–0.77, p = 0.005, respectively). Transformed FL was associated with shorter overall survival (OS) from FL diagnosis compared to non-transformed FL (median not reached, p < 0.001), while prior exposure to anti-FL chemoimmunotherapy predicted shorter OS following transformation (median 34.6 months vs. not reached, p = 0.001). Multivariate analysis confirmed prior exposure to anti-FL therapy (HR 5.27, p < 0.001), male sex (HR 2.36, p = 0.014), and age over 65 (HR 2.54, p = 0.019) to be associated with shorter OS among patients with tFL. Conclusions: These findings, although requiring further validation, suggest that O-based regimens may reduce the risk of transformation and support prior studies demonstrating adverse survival outcomes in previously treated tFL patients. Full article

The exponential growth of genomic and proteomic data has made computational protein–protein interaction (PPI) prediction indispensable, driving the need for a comprehensive and method-aware evaluation of supervised learning approaches. PPIs are fundamental to understanding cellular processes and disease mechanisms, yet experimental identification remains slow, costly, and difficult to scale. This survey systematically investigates ten supervised learning models—Extreme Learning Machine (ELM), Convolutional Neural Networks (CNNs), Graph Neural Networks (GNNs), Deep Neural Networks (DNNs), Naïve Bayes, Probabilistic Decision Tree, Support Vector Machine (SVM), Least Squares SVM (LS-SVM), K-Nearest Neighbor (KNN), and Weighted K-Nearest Neighbor (WKNN)—through a tri-layered framework that integrates Comparative Quantitative Analysis, Comparative Observational Analysis, and Experimental Evaluations. Beyond conventional accuracy summaries, this work provides critical commentary tied to real-world use, analyzing where techniques succeed or fail in practice—for instance, when instance-based methods bottleneck during inference, when kernel choices influence SVM variance, or when deep architectures trade accuracy for computational cost. The survey also offers concrete deployment guidance, such as calibration insights for WKNN versus KNN under varying feature noise or dataset curation quality, delivering operational perspectives that typical surveys omit. Comparative Quantitative Analysis consolidates metrics such as accuracy, F1-score, and computational time from the existing literature, while Comparative Observational Analysis evaluates interpretability, scalability, dataset suitability, and efficiency. Complementing these, Experimental Evaluations conducted by the authors empirically validate model performance on benchmark datasets. Together, these layers provide a unified and evidence-backed perspective on algorithmic strengths, weaknesses, and practical applicability. Findings show that GNNs and DNNs achieve the highest predictive accuracy due to their ability to capture structural and topological relationships, whereas ELM and Naïve Bayes offer superior efficiency. SVM and LS-SVM maintain robust stability under noisy conditions, and CNNs are well-suited for sequence-based prediction tasks. By combining empirical validation, critical insights, and deployment-focused recommendations, this survey delivers decision-grade guidance that bridges theoretical understanding with real-world implementation, thus clarifying the trade-offs among accuracy, efficiency, and scalability in PPI detection research. Full article

Background. Artificial intelligence (AI) is increasingly recognized as a transformative technology in healthcare, with growing interest in its applications within pediatric dentistry. Given the unique clinical, developmental, and behavioral characteristics of pediatric patients, AI-based systems may offer valuable support in improving diagnosis, prevention, and treatment planning. Methods. A narrative review was conducted to synthesize current evidence on AI applications in pediatric dentistry. A comprehensive search strategy, including predefined keywords and free terms, was applied across multiple databases (Embase, Scopus, PubMed, and Web of Science) up to 1 January 2026. Reviews addressing AI-based technologies in pediatric dental care were selected and analyzed. Results. The available literature indicates that AI is being progressively applied across multiple domains of pediatric dentistry, although with varying levels of evidence. More extensively investigated areas include diagnostic imaging, caries detection, orthodontic assessment, and growth evaluation, where AI systems—particularly those based on machine learning and deep learning—have demonstrated high accuracy and reproducibility. Other emerging fields, such as remote monitoring, behavioral management, preventive strategies, and patient education, show promising potential but remain less explored. Overall, AI-based tools appear to enhance diagnostic support, enable early detection of oral conditions, and contribute to more personalized and efficient clinical workflows. Conclusions. AI represents a rapidly evolving adjunct in pediatric dentistry with the potential to improve clinical decision-making, preventive care, and patient management. Despite encouraging results, further validation in real-world settings, along with careful consideration of ethical, legal, and data-related challenges, is required to support its responsible integration into routine clinical practice. Full article

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder. Early and accurate diagnosis is critical to delaying disease progression, alleviating clinical symptoms, and improving the long-term quality of life for the affected patients. The deep integration of artificial intelligence (AI) and medical imaging enables efficient early AD screening, overcoming traditional limitations. This study presents a systematic review of AI-driven applications in the early diagnosis of AD with a dual focus on single-modal and multimodal analytical frameworks, comprehensively analyzing core technical components across existing research including data preprocessing pipelines, mainstream deep learning and machine learning diagnostic models, standard performance evaluation metrics, and widely adopted public research datasets, while further qualitatively comparing the diagnostic efficacy and applicability of diverse methodologies across distinct imaging and non-imaging modalities. In addition, this review systematically delineates and compares the application merits, technical bottlenecks, and clinical suitability of AI-enabled diagnostic methods across diverse modalities, providing robust methodological guidance and clear directional references for future research on the early diagnosis of AD and facilitating the advancement of the field toward higher diagnostic precision, broader population applicability, and tighter integration with real-world clinical practice. Full article

This prospective observational study aimed to estimate the annual service-based incidence of individuals with First Episode Psychosis (FEP) and high-risk states for psychosis presenting to a public Community Mental Health Center within a defined urban catchment area in Northwestern Greece. It offers novel real-world insights into early intervention in psychosis within a resource-constrained, post-crisis health care setting. All individuals aged ≥16 years who presented to the Community Mental Health Center of the University of Ioannina between January 2023 and December 2024 were assessed. Those diagnosed with FEP or identified as being at a high risk for psychosis using the Comprehensive Assessment of At-Risk Mental States were included, while duration of untreated psychosis (DUP) was estimated with the Symptom Onset in Schizophrenia inventory. Among 1115 service users, 51 (4.6%) met criteria for FEP (N = 33) or high-risk states (N = 18), rising to 7.5% among those aged 16–36 years. The annual service-based incidence of FEP was 10.26 per 100,000 in the general population, increasing to 51.62 in individuals aged 16–36 and 63.17 in those aged 16–26. Including high-risk cases, service-based incidence reached 109.71 per 100,000 in the 16–26 age group. Mean DUP was 39.4 weeks but was 7.0 weeks among 80% with DUP < 1 year. Most FEP patients (63.6%) required brief hospitalization, and over half reported family history of mental illness. These findings highlight substantial community caseloads and the need to strengthen early intervention services. Full article