Search Results (4,022)

Latine immigrants in the U.S. face discrimination, a lack of access to resources, fear of deportation, family separation, and a lack of understanding of the complex factors that have led them to seek safety and survival in a new country, factors that critically impact their mental health. Academic and community partners implemented the Immigrant Well-Being Project (IWP), a mental health intervention that partners community-based organizations (CBOs) and universities to bring Latine immigrants and university students together to engage in mutual learning and social change efforts. We assessed the feasibility of the IWP model and preliminary outcomes from a pilot implementation with three cohorts of Latine immigrant adults (N = 54) between 2018 and 2021. Employing a within-group longitudinal, mixed-methods design with qualitative interviews and quantitative surveys over 12 months (pre-intervention, post-intervention, and 6-month follow-up), we examined engagement and the impact of the 6-month intervention, as well as explored feasibility/acceptability and participants’ experiences. Multilevel growth modeling revealed significant decreases in depression and anxiety symptoms and difficulty accessing resources, with dosage and cohort significantly moderating intervention effects on these variables. Some cohorts also experienced significant decreases in PTSD symptoms and increases in satisfaction with resources. Qualitative results revealed that having student and CBO support navigating services positively impacted mental health, which highlights the importance of connecting immigrants to existing community resources. Findings are discussed in the context of COVID-19 and hostile immigration policies, emphasizing the importance of addressing social–structural determinants of mental health for immigrants. Full article

Persistent concerns about the mathematics achievement of culturally and linguistically diverse students with mathematics difficulties have intensified in recent years, underscoring the need for effective instructional strategies to support this growing student population. Research suggests that culturally and linguistically responsive (CLR) practices can address these disparities; however, many teachers enter the profession without the knowledge or skills to implement CLR practices effectively. Preparing pre-service teachers to support diverse learners before entering the classroom is critical. Accordingly, this study examined the impact of a four-lesson online culturally and linguistically responsive mathematics instructional module (CLR-MI) on pre-service teachers’ knowledge, beliefs, and perceived ability to implement CLR mathematics instruction for students experiencing mathematics difficulties. The module incorporated evidence-based mathematics practices, culturally and linguistically responsive instructional strategies, classroom examples and non-examples, and interactive reflection activities. A randomized controlled trial with a waitlist control design was conducted with 64 undergraduate preservice teachers enrolled in university-level mathematics methods courses. Findings showed participants in both groups demonstrated significant increases in self-efficacy and outcome expectancy beliefs related to CLR mathematics instruction, with improvements over time. However, no significant gains were observed on a measure of Mathematics Knowledge for Teaching (MKT). This suggests that brief, targeted online instructional modules, such as CLR-MI, may enhance preservice teachers’ confidence and belief in their ability to implement CLR practices effectively. In contrast, improvements in mathematics knowledge for teaching may require more sustained learning experiences. Together, the findings highlight the importance of integrating CLR-focused instruction into teacher preparation programs to support the diverse needs of learners in mathematics classrooms. Full article

The increased penetration of renewable energy sources and Inverter-Based Resources (IBRs) has resulted in power systems operating with significantly reduced inertia, thus increasing their susceptibility to frequency instability and posing challenges to the stability and security of modern power grids. Consequently, extensive research efforts have focused on enhancing frequency stability through improved Load Frequency Control (LFC), also referred to as Automatic Generation Control (AGC). Existing studies have proposed a wide range of control strategies that differ in controller structure, optimization techniques, and performance evaluation criteria. Conventional integral-order controllers remain among the most widely investigated approaches. Fractional-order controllers have also attracted considerable attention owing to their enhanced flexibility and dynamic performance. In addition, two-degrees-of-freedom and multi-degrees-of-freedom controllers have been explored to improve frequency regulation through the utilization of multiple control inputs. Fuzzy logic controllers have also been extensively investigated, with approaches varying from sole integration in systems to hybridizing with other control approaches. An emerging body of work has investigated other control approaches, such as sliding mode control, active disturbance rejection, model predictive control, and reinforcement learning. As more Energy Storage Systems (ESSs) are being adopted for ancillary services, recent studies have increasingly focused on integrating ESSs into frequency regulation schemes through both LFC-based control and independent control architectures. This review provides a comprehensive classification and critical analysis of existing LFC topologies, highlighting their characteristics, limitations, and emerging research directions for low-inertia power systems. Full article

Digital technologies are reshaping companion-animal healthcare by expanding access to veterinary services and enabling more connected, data-driven models of care. This narrative review synthesizes current evidence on digital veterinary care, with a focus on the integration of telemedicine, artificial intelligence (AI), and remote patient monitoring (RPM) in companion-animal practice. A targeted narrative literature search of PubMed, Scopus, CAB Abstracts, Google Scholar, and relevant professional guidelines published between 2010 and 2026 was conducted to identify evidence on digital veterinary healthcare; the synthesis was narrative, and no formal risk-of-bias or study-quality assessment was undertaken. Current evidence suggests that telemedicine has shown the greatest value for teletriage, follow-up consultations, chronic disease management, specialist referral, and postoperative care, while RPM extends longitudinal monitoring beyond clinic visits and AI has been evaluated for applications in diagnostic imaging, predictive analytics, clinical decision support, and workflow efficiency, largely in retrospective, pilot-scale, or single-center veterinary studies. However, widespread implementation remains constrained by limited veterinary-specific validation of AI algorithms and commercial biosensors, evolving veterinarian–client–patient relationship (VCPR) regulations, cybersecurity and data privacy concerns, interoperability challenges, and the need for robust clinical evidence. Emerging concepts—including AI-integrated RPM, digital phenotyping, digital twins, federated learning, and One Health interoperability—remain largely prospective or extrapolated from human healthcare but have the potential to advance precision companion-animal medicine, strengthen disease surveillance, and improve preventive healthcare. Overall, digital veterinary care should complement rather than replace conventional veterinary practice. Future progress will depend on rigorous clinical validation, interoperable digital infrastructure, transparent AI governance, harmonized regulatory frameworks, and evidence-based implementation to improve animal welfare, veterinary service delivery, and One Health outcomes. Full article

Lithium-ion batteries are the primary power source for new energy equipment, such as electrochemical energy storage systems and electric vehicles. Accurate state of health (SOH) estimation and remaining useful life (RUL) prediction are essential for ensuring safe operation and reducing lifecycle operation and maintenance costs. Since existing studies are largely based on ideal laboratory data, the proposed algorithms often exhibit limited generalizability and engineering implementation difficulties in real-world scenarios. To address this gap, this paper provides a comprehensive review of the latest research progress with field data as the core focus. First, it defines the scope, acquisition characteristics, and preprocessing strategies of in-service battery data and quantitatively evaluates the applicability of representative domestic and international open-source and industrial proprietary datasets. Second, it clarifies the quantitative standards, intrinsic relationships, and application-specific requirements for SOH and RUL. The performance, applicability, and limitations of model-based, data-driven, and hybrid approaches are comprehensively analyzed based on real-world datasets, with emphasis on their underlying principles, recent technical advances, and engineering potential. Finally, it summarizes challenges, including multi-factor coupled degradation, battery pack consistency assessment, and small-sample transfer learning, and it outlines future directions, including multimodal data fusion, digital twin modeling, and edge intelligence collaboration. Unlike previous reviews that primarily focus on laboratory data, this review establishes an engineering-oriented analytical framework centered on real-world operational data. The findings indicate that hybrid approaches present a highly promising qualitative trend for balancing prediction accuracy and model interpretability, offering valuable guidance for bridging the gap between laboratory research and practical industrial applications. Full article

The use of artificial intelligence (AI) and machine learning (ML) models has increased significantly in recent years, with rapid advances across domains such as natural language processing, computer vision, code generation, and scientific computing. Alongside this growth, a diverse ecosystem of models and services has emerged, raising important questions about efficient and fair mechanisms for accessing and paying for inference. In particular, there is an ongoing debate between subscription-based access and per-request pricing models, the latter becoming especially relevant in the context of autonomous AI agents that dynamically consume external services. One of the emerging open protocols for enabling per-request payments is x402, which uses Hypertext Transfer Protocol (HTTP)-native payment flows. The core payment protocol does not let clients verify that inference was performed using the claimed model. In this paper, we propose a framework for verifiable paid inference that integrates x402-based micropayments with verifiable ML techniques. The framework combines a payment authorization bound to model and input commitments with a common escrow design for synchronous and asynchronous verification. Payment is released after a valid proof or after an optimistic challenge process accepts the result under its stated assumptions. The prototype demonstrates the functional feasibility of this approach, together with measured gas costs and proving times. The measurements indicate that asynchronous verification is more suitable for larger workloads and higher request volumes, whereas full-model proving is practical primarily for smaller models. Full article

The Internet of Medical Things (IoMT) can extend diagnostic observation beyond episodic encounters by connecting wearable, point-of-care, imaging, and bedside devices to artificial-intelligence services. Yet continuous measurement is not itself diagnosis: a clinically useful system must establish signal validity, map an output to a defined diagnostic role, communicate uncertainty, and demonstrate agreement with an appropriate reference standard. This survey analyzes the complete IoMT diagnostic pathway from acquisition and quality control through edge–fog–cloud computation, machine and deep learning, federated training, and clinician-governed action. It distinguishes monitoring, screening, early detection, differential diagnosis, prognosis, and decision support because these tasks require different labels, thresholds, metrics, and evidence. Evidence is synthesized across electrocardiography and cardiovascular risk, glucose and metabolic assessment, connected imaging and oncology, respiratory and infectious-disease detection, electroencephalography and seizure recognition, and decentralized point-of-care testing. The analysis shows how sensor placement, missingness, compression, latency, privacy mechanisms, and site heterogeneity can change diagnostic performance even when the model is unchanged. A diagnostic evidence framework is therefore proposed that links reference standards and patient-level data separation to sensitivity, specificity, calibration, external validation, target-device feasibility, uncertainty-based referral, and prospective workflow utility. Security, explainability, and lifecycle monitoring are treated as conditions of diagnostic reliability rather than independent features. The resulting synthesis provides a clinically centered basis for deciding which IoMT outputs may support screening or diagnosis, which remain monitoring signals, and what evidence is still required before deployment. Full article

Salt marshes are crucial coastal ecosystems that support biodiversity and provide a range of ecosystem services, including nutrient retention, shoreline erosion protection, and carbon storage, through the processes they maintain. However, their dynamic nature, combined with human pressures, has led to their rapid decline, making ongoing monitoring essential. This study presents a pixel-based classification framework for mapping and generating spatially explicit time series of salt marsh extent from satellite imagery, leveraging reflectance values and multi-spectral indices to improve accuracy and temporal consistency. The approach accounts for seasonal and hydrological variability to ensure robust detection. Four machine learning models were evaluated, with the Multi-Layer Perceptron achieving the best performance. The final model was combined with a majority filter and reached a 90% true positive rate for salt marshes, 96% for land and 97% for water, demonstrating its effectiveness for long-term ecological monitoring and conservation. Using the model, seasonal patterns and long-term trends in the salt marsh area were identified, showing strong correlations with tidal cycles and sea level changes. Full article

Wind energy is becoming increasingly important in meeting energy security, grid reliability, and the growing electricity demand. The rapid increase in the number and prevalence of wind turbines has triggered the need for fast, cost-effective, reliable, and easy-to-implement inspection systems. Automated detection of coating defects is important for the maintenance and corrosion prevention of the turbines. In this context, this study aimed to compare and evaluate the accuracy and efficiency of deep learning models for detection of coating defects from images. In this context, the study evaluates five object detection models, RT-DETR-L, YOLOv8-M, YOLOv9-C, YOLO11-L, and YOLO26-L, on an open coating-defect dataset, with inclusion, pinhole, and scratch classes. Each model was trained with 3 random seeds for images of 640 × 640 and 1088 × 1088-pixel resolutions. At 1088 × 1088, YOLO26-L achieved the highest mean mAP50–95 (0.1142), while RT-DETR-L achieved the highest recall (0.5591) and F1 score (0.2217) at the operating confidence threshold of 0.25; RT-DETR-L was also the slowest model (59.8 ms/image). YOLOv8-M achieved a mean mAP50–95 of 0.0976 together with the lowest latency (34.2 ms/image) and lowest true batch-size-1 peak GPU-memory allocation (278 MB). Reducing the resolution to 640 × 640 lowered mAP50–95 by 25.4% for YOLOv8-M and 33.6% for RT-DETR-L while reducing latency and memory use. The results revealed a pronounced accuracy-efficiency trade-off rather than universal superiority of one model for coating defect detection for wind turbine towers. The dataset was acquired during the wind-tower painting process under controlled industrial imaging conditions, so the results characterise in-process coating inspection rather than the inspection of weathered, in-service turbines. Because only three random seeds were used, paired comparisons between individual models were not statistically conclusive (8 of 100 pairwise tests reached p < 0.05) and the reported rankings are descriptive. Applying operating thresholds—determined on the validation set for each model—to the independent test set altered the architectural ranking. At 1088 × 1088 resolution, YOLO26-L surpassed RT-DETR-L to achieve the highest mean F1 score, whereas at 640 × 640 resolution, RT-DETR-L’s ranking dropped significantly. The fact that the selected thresholds range from 0.076 to 0.456 indicates that the differences observed at the common threshold of 0.25 are sensitive to the choice of threshold. While the mAP50–95 decreased by approximately half in the 640 × 640 evaluation—where data leakage was eliminated—the 1088 × 1088 results were considered optimistic, as they were obtained based on the previous segmentation. Full article

Artificial intelligence is included in several parts of education, and it serves as an important tool for it. In this study, we present the initial design and pre-deployment formative evaluation of an AI-powered educational chatbot, called TouriBot, from the perspective of different stakeholders. TouriBot is part of a gamified learning environment created specifically for students in primary education (ages 9–12). The application is intended to support potential engagement with critical literacy-related tasks in the setting of misinformation. The evaluation involved three stakeholder groups, namely, in-service primary school Information and Communication Technology (ICT) teachers (N = 112), ICT educational advisors (N = 13) serving as institutional pedagogical experts and user experience (UX) experts (N = 8). No student data were collected in this evaluation phase. Structured questionnaires assessed perceived usability and pedagogical value as well as behavioral intention to use the tool. The results show that stakeholders evaluated the chatbot positively regarding usability, pedagogical alignment and stated intention to use. ICT teachers indicated high perceived usefulness and ease of use and strong intention to use the system in their teaching environments. The user interface perception-based assessments from the UX experts were mainly positive, and the ICT educational advisors highlighted the chatbot’s alignment with the intended pedagogical goals. However, TouriBot had limitations and drawbacks identified during this initial phase of the study. These stakeholder evaluations provide perception-based feedback that can inform further system refinement and guide student-centered evaluation in classroom settings. Full article

Accurate estimation of crop evapotranspiration (ETc) is fundamental to irrigation planning and sustainable agricultural water management, particularly under increasing climate variability and water scarcity. Conventional flux measurement systems, including eddy covariance towers and lysimeters, provide high-quality observations but are costly, maintenance-intensive, and spatially constrained, limiting their scalability for precision water management. This study presents AquaVolt-AI, a physics-informed machine learning framework that integrates Sentinel-2 optical imagery, NASA ECOSTRESS thermal observations, and meteorological data with the FAO-56 dual crop-coefficient formulation to generate spatially explicit ETc estimates without requiring dedicated on-site sensing infrastructure for routine operation. The framework couples a dynamic residual neural network with physics-based constraints and an automated state-estimation mechanism designed to maintain inference during satellite data gaps and external data-service interruptions. AquaVolt-AI was evaluated over 36 days (28 June–3 August 2026) at the UC Davis Russell Ranch Sustainable Agriculture Facility using ground-based CIMIS observations for validation and ECOSTRESS thermal data as an auxiliary model input. ETc was represented across a 16 × 16 virtual sensing grid comprising 256 spatial sectors at 10 m resolution. The framework achieved a root mean square error of 0.3000 mm day⁻¹ and a mean absolute error of 0.2688 mm day⁻¹. During a consecutive 9-day satellite data gap, the physics-informed state estimator maintained continuous ETc predictions without detectable empirical drift in the evaluated period. These findings demonstrate the feasibility of integrating Earth observation, meteorological information, and physics-informed machine learning within a low-infrastructure computational framework for spatially resolved ETc monitoring. The approach provides a scalable foundation for precision irrigation assessment and data-driven agricultural water management, although broader multi-season and multi-site validation is required to establish transferability across cropping systems and agroclimatic environments. The main novelty of this study is in coupling a bounded residual neural correction to the FAO-56 dual crop coefficient model within a fully serverless architecture, eliminating on-site sensing hardware while preserving physical plausibility during data outages. Full article

Fog–cloud systems allow distributed and latency-sensitive IoT applications to share edge, fog, and cloud resources for efficient computation, storing and delivering services. Nonetheless, the current methods of task scheduling and resource allocation tend to be based on deterministic prediction, heuristic schedules, or response optimization, restricting risk sensitivity and responsiveness to dynamic workloads. To address these issues, we propose a graph network with a reinforcement learning framework to avoid congestion and schedule tasks with a low makespan in fog–IoT systems. The work starts with real-time monitoring of queue states, delay, bandwidth, energy and deadline properties of upcoming IoT tasks. Quantile Regression Long Short-Term Memory (QR-LSTM) predicts a normal task flow and congestion based on the task queue. The congestion tasks are further scheduled using a Delay-Aware Influence Reinforced-Graph Neural Network (DAIR-GNN). Then, Topology-Sensitive Resource Influence Propagation is used to map and analyze the relationship between the task sender and receiver within a network. After that, Dual-Stage Risk-Aware Hierarchical Policy (DRHP) combines Proximal Policy Optimization (PPO) to select the kinds of sources, like fog or cloud, based on the topology. Similarly, Model-Agnostic Meta-Learning (MAML) with Soft Actor–Critic allocates the resources of each tasks within the selected server. Both RL models are trained using Few-Shot Adaptive Policy Transfer to make better predictions. Lastly, Confidence-Uncertainty Regulated Exploration (CURE) is used to compute the prediction score for task allocation improvement. The proposed framework achieves a variance ratio of 80.6%, latency is 0.626 s, and the success rate is 98% in the prediction of congestion, scheduling and allocation of resources. These results demonstrate the improved ability to ensure reliable and effective fog–cloud allocation in response to dynamically changing workloads. Full article

DDoS (Distributed Denial-of-Service) attacks are significant cyber threats that affect network infrastructures and cause massive service disruption. DDoS attacks have become even more threatening, owing to the poor security of IoT devices. This paper proposes two ensemble approaches: (1) Optimized Ensemble Stacking Approach 4 (OESA-4) and (2) Optimized Ensemble Voting Approach 4 (OEVA-4). Both proposed approaches use four supervised machine learning classifiers, such as K-Nearest Neighbors (KNN), Random Forest (RF), Extreme Gradient Boosting (XGB), and Decision Tree (DT). OESA-4 combines the base learners’ outputs via a meta-learner, and OEVA-4 uses weighted soft-voting, assigning different weights to base learners to improve the output’s efficiency. An experimental analysis was conducted on a subset of the CIC-DDoS2019 dataset offered by the Canadian Institute for Cybersecurity to test the performance of state-of-the-art classifiers and proposed approaches. This subset of the dataset contains 325,965 occurrences. The results show that the proposed OESA-4 and OEVA-4 demonstrated reliable performance with accuracies of 99.87% and 99.73%, respectively, on a single split, whereas stratified five-fold cross-validation resulted in mean accuracies of 99.51% ± 0.08% for OESA-4 and 99.28% ± 0.11% for OEVA-4 respectively. The relatively low errors in both cases are indicative of consistent performance in all the five folds tested, demonstrating strong and consistent classification capabilities within the experiments. Thus, the presented paper shows that OESA-4 and OEVA-4 approaches achieve higher performance in identifying various types of DDoS attacks compared to individual classifiers. Full article

Traffic congestion is a persistent challenge in rapidly urbanizing regions, including Penang Island, Malaysia, where growing vehicle demand increasingly exceeds road capacity. Quantitative understanding of traffic dynamics is therefore essential for evidence-based traffic management and infrastructure planning. While previous studies have typically focused on either traffic detection, forecasting, or congestion assessment separately, limited attention has been given to integrating these components into a unified framework for continuous traffic-condition evaluation. This study integrates computer vision-based data extraction with statistical modeling to analyze and predict traffic behavior. Traffic volume is automatically obtained from video streams using a deep learning detection framework and subsequently processed for time-series modeling. Seasonal Autoregressive Integrated Moving Average (SARIMA) is employed to forecast traffic flow, while Ordinary Least Squares (OLS) regression is used to identify factors associated with congestion. Road performance is evaluated using the Volume-to-Capacity ($V/C$) ratio and Level of Service (LOS) indicators. The forecasting model captures daily and weekly traffic patterns with stable predictive performance across observation periods. Regression analysis indicates significant differences in traffic counts across road segments and vehicle types, while temporal traffic analysis identifies recurring peak-period traffic patterns. High $V/C$ ratios consistently correspond to degraded LOS conditions, allowing identification of recurring bottleneck segments within the network. The combined framework demonstrates how automated sensing, statistical prediction, and engineering performance indicators can be jointly used to monitor and interpret urban traffic conditions. The approach provides a reproducible methodology for continuous congestion assessment and supports data-driven planning decisions in medium-sized urban road networks. Full article

In practical power system operation, equipment maintenance and line outages may cause changes in network topology and power flow redistribution. Therefore, fast assessment of the total transfer capability (TTC) of transmission interfaces under N−1 line-outage scenarios is important for operational adjustment and security control. Existing deep-learning-based TTC assessment methods mainly rely on fixed-order vectorized representations, making it difficult to explicitly preserve network connectivity, variations in branch operating states, and regional power transfer characteristics of different target interfaces. To address these issues, this paper proposes a fast multi-interface TTC assessment method based on an edge-feature-enhanced message passing neural network (MPNN). The power system is represented as a directed graph integrating bus operating states, branch electrical states, line in-service states, and target-interface regional information. A line-status-controlled message passing mechanism is combined with branch electrical edge features, enabling node representations to jointly capture effective network connectivity and branch operating states. An interface-aware readout layer is further designed to aggregate system-wide, sending-area, and receiving-area representations and construct interface-specific graph-level features for TTC regression. Case studies on the IEEE 39 bus and IEEE 118 bus systems demonstrate that the proposed model achieves stable TTC prediction across different N−1 topology scenarios and exhibits good adaptability to topology changes. Full article