Search Results (20,633)

This paper presents a proof-of-concept real-time safety alerting system for safety-critical environments such as construction sites. Key components of the system include Bluetooth Low Energy (BLE) devices for indoor localization, integrated with a customized Android application using the Framework for Internal Navigation and Discovery (FIND). Administrative control and data management are handled by a server-side component, supported by an interactive website for real-time safety monitoring. The architecture supports safety zoning and employs machine learning algorithms, including k-NN, Random Forest, and SVM, for analyzing localization data. Experimental validation in a laboratory setup demonstrates a localization accuracy of 97%, a response time of 1.2 s, and a maximum spatial error of 1.2 m. These results highlight the system’s reliability and potential for enhancing safety compliance in real-world deployment scenarios. Full article

Data stream classification is a critical challenge in data mining, where models must rapidly adapt to evolving data distributions and concept drift in real time, while extreme learning machines offer fast training and strong generalization, most existing methods struggle to jointly address multi-class imbalance, concept drift, and the high cost of label acquisition in streaming settings. In this paper, we present the Adaptive Broad Learning System for Online Imbalanced Classification (ABLS-OIC), which introduces three core innovations: (1) a Class-Adaptive Weight Matrix (CAWM) that dynamically adjusts sample weights according to class distribution, sample density, and difficulty; (2) a Hybrid Memory Retention Mechanism (HMRM) that selectively retains representative samples based on importance and diversity; and (3) a Multi-Objective Adaptive Optimization Framework (MAOF) that balances classification accuracy, class balance, and computational efficiency. Extensive experiments on ten benchmark datasets with varying imbalance ratios and drift patterns show that ABLS-OIC consistently outperforms state-of-the-art methods, with improvements of 5.9% in G-mean, 6.3% in F1-score, and 3.4% in AUC. Furthermore, a real-world credit fraud detection case study demonstrates the practical effectiveness of ABLS-OIC, highlighting its value for early detection of rare but critical events in dynamic, high-stakes applications. Full article

Integration of machine learning (ML) and artificial intelligence (AI) into point-of-care (POC) sensor systems represents a transformative advancement in healthcare. This integration enables sophisticated data analysis and real-time decision-making in emergency and intensive care settings. AI and ML algorithms can process complex biomedical data, improve diagnostic accuracy, and enable early disease detection for better patient outcomes. Predictive analytics in POC devices supports proactive healthcare by analyzing data to forecast health issues and facilitating early intervention and personalized treatment. This review covers the key areas of ML and AI integration in POC devices, including data analysis, pattern recognition, real-time decision support, predictive analytics, personalization, automation, and workflow optimization. Examples of current POC devices that use ML and AI include AI-powered blood glucose monitors, portable imaging devices, wearable cardiac monitors, AI-enhanced infectious disease detection, and smart wound care sensors are also discussed. The review further explores new directions for POC sensors and ML integration, including mental health monitoring, nutritional monitoring, metabolic health tracking, and decentralized clinical trials (DCTs). We also examined the impact of integrating ML and AI into POC devices on healthcare accessibility, efficiency, and patient outcomes. Full article

Training a machine learning (ML) model always needs many computing resources, and cloud-based outsourced training is a good solution to address the issue of a computing resources shortage. However, the cloud may be untrustworthy, and it may pose a privacy threat to the training process. Currently, most work makes use of multi-party computation protocols and lattice-based homomorphic encryption algorithms to solve the privacy problem, but these tools are inefficient in communication or computation. Therefore, in this paper, we focus on the k-means and propose a fast and privacy-preserving method for outsourced clustering of k-means models based on symmetric homomorphic encryption (SHE), which is used to encrypt the clustering dataset and model parameters in our scheme. We design an interactive protocol and use various tools to optimize the protocol time overheads. We perform security analysis and detailed evaluation on the performance of our scheme, and the experimental results show that our scheme has better prediction accuracy, as well as lower computation and total overheads. Full article

Background: Developmental dysplasia of the hip (DDH), if not treated, can lead to osteoarthritis and disability. Ultrasound (US) is a primary screening method for the detection of DDH, but its interpretation remains highly operator-dependent. We propose a supervised machine learning (ML) image segmentation model for the automated recognition of anatomical structures in hip US images. Methods: We conducted a retrospective observational analysis based on a dataset of 10,767 hip US images from 311 patients. All images were annotated for eight key structures according to the Graf method and split into training (75.0%), validation (9.5%), and test (15.5%) sets. Model performance was assessed using the Intersection over Union (IoU) and Dice Similarity Coefficient (DSC). Results: The best-performing model was based on the SegNeXt architecture with an MSCAN_L backbone. The model achieved high segmentation accuracy (IoU; DSC) for chondro-osseous border (0.632; 0.774), femoral head (0.916; 0.956), labrum (0.625; 0.769), cartilaginous (0.672; 0.804), and bony roof (0.725; 0.841). The average Euclidean distance for point-based landmarks (bony rim and lower limb) was 4.8 and 4.5 pixels, respectively, and the baseline deflection angle was 1.7 degrees. Conclusions: This ML-based approach demonstrates promising accuracy and may enhance the reliability and accessibility of US-based DDH screening. Future applications could integrate real-time angle measurement and automated classification to support clinical decision-making. Full article

Purpose: Accurate survival prediction is essential for optimizing the treatment planning in patients with castration-resistant prostate cancer (CRPC). However, the traditional statistical models often underperform due to limited variable inclusion and an inability to account for complex, multidimensional data interactions. Methods: We retrospectively collected 46 clinical, laboratory, and pathological variables from 801 patients with CRPC, covering the disease course from the initial disease diagnosis to CRPC progression. Multiple machine learning (ML) models, including random survival forests (RSFs), XGBoost, LightGBM, and logistic regression, were developed to predict cancer-specific mortality (CSM), overall mortality (OM), and 2- and 3-year survival status. The dataset was split into training and test cohorts (80:20), with 10-fold cross-validation. The performance was assessed using the C-index for regression models and the AUC, accuracy, precision, recall, and F1-score for classification models. Model interpretability was assessed using SHapley Additive exPlanations (SHAP). Results: Over a median follow-up of 24 months, 70.6% of patients experienced CSM. RSFs achieved the highest C-index in the test set for both CSM (0.772) and OM (0.771). For classification tasks, RSFs demonstrated a superior performance in predicting 2-year survival, while XGBoost yielded the highest F1-score for 3-year survival. The SHAP analysis identified time to first-line CRPC treatment and hemoglobin and alkaline phosphatase levels as key predictors of survival outcomes. Conclusion: The RSF and XGBoost ML models demonstrated a superior performance over that of traditional statistical methods in predicting survival in CRPC. These models offer accurate and interpretable prognostic tools that may inform personalized treatment strategies. External validation and the integration of emerging therapies are warranted for broader clinical applicability. Full article

Axial flux permanent magnet (AFPM) machines offer some advantages over conventional radial flux machines for the case of a limited space in the axial direction, such as high torque density, compact structure, and modular design ability. They, therefore, are increasingly used in wind power and electrical vehicles. This paper focuses on developing an effective analytical model and equivalent auto-finite element method (FEM), including rotor linear motion for the design of axial flux permanent magnet generators in vertical axis wind turbines. The initial design of a 1.35 kW-AFPM generator with an outer double rotor and double layer concentrated windings is based on analytical equations, and then it is refined using equivalent time-stepping transient FEM, including rotor linear motion to calculate voltage, electromagnetic force, and torque. The automatic generation of an equivalent transient 2D-FEM model to replace a time-consuming 3D-FEM model is investigated. As a consequence, the influence of slotting the effect on a 1.35 kW-AFPM machine’s performances, such as air gap flux density, internal voltage, and cogging torque, is announced. Full article

Road traffic accidents are the eighth leading cause of human deaths. In order to study two-vehicle accidents, this paper extracted data from 493 two-vehicle accidents from the CIDAS database from 2011 to 2022, used machine learning methods to analyze the accident data, and obtained the significance of two-vehicle accident parameters. Finally, five typical scenarios of two-vehicle accidents were obtained based on this. The results of the significance analysis show that vehicle parameters have a greater impact on occupant injury in the host vehicle; clustering results show that lighting, the number of lanes, the other vehicle’s type, and the speed of the host vehicle have a large impact on occupant injury (for example, the injury rate for the high-speed, nighttime Scenario II was 52.9%, compared to just 10.9% for the lower-speed Scenario IV). Factor analysis results show that precipitation has a large impact on occupant injury, as the frequency of injuries in rainy conditions was 13.4% higher, and the frequency of serious injuries was 7.9% higher, than in accidents without rain. This paper innovatively uses factor analysis to reduce the dimensionality of categorical variables, which provides research ideas for related research. At the same time, the clustering results obtained in this paper also provide references for the establishment of corresponding test scenarios for autonomous driving and the establishment of standards. Full article

With the increasing emergence of intelligent vehicles, novel accident patterns have gradually emerged. In human–machine cooperative driving (HMCD) states, despite driving automation systems being capable of controlling lateral and longitudinal vehicle motions over extended periods, functional limitations persist in specific scenarios due to insufficient expected functionalities. When combined with risk factors, such as driver distraction, these limitations significantly elevate accident risks. This study investigated takeover safety through real vehicle testing in two typical accident scenarios: large-curvature curves and static obstacles. The key findings include the following: (1) in scenarios involving large curvature curves and static obstacles, vehicles are prone to lane departure and missed target detection, which are typical dangerous scenarios; (2) during the human–machine cooperative driving phase, the design of the driving automation system should focus on enhancing driver engagement in driving tasks, while in the autonomous driving phase, the vehicle’s early warning capabilities should be strengthened; (3) the takeover request for longitudinal control requires at least 4.12 s of driver reaction time, while the takeover request for lateral control requires at least 1.87 s. This study provides important theoretical and practical references for safety in designing assisted driving systems and the testing of hazardous scenarios. Full article

Mechanical ventilation is indispensable for patients with severe respiratory conditions, and high-fidelity lung simulators play a pivotal role in ventilator testing, clinical training, and respiratory research. However, most existing simulators are passive, single-lung models with limited and discrete control over respiratory mechanics, which constrains their ability to reproduce realistic breathing dynamics. To overcome these limitations, this study presents a dual-chamber lung simulator that can operate in both active and passive modes. The system integrates a sliding mode controller enhanced by a linear extended state observer, enabling the accurate replication of complex respiratory patterns. In active mode, the simulator allows for the precise tuning of respiratory muscle force profiles, lung compliance, and airway resistance to generate physiologically accurate flow and pressure waveforms. Notably, it can effectively simulate pathological conditions such as acute respiratory distress syndrome (ARDS) and chronic obstructive pulmonary disease (COPD) by adjusting key parameters to mimic the characteristic respiratory mechanics of these disorders. Experimental results show that the absolute flow error remains within ±3L/min, and the response time is under 200ms, ensuring rapid and reliable performance. In passive mode, the simulator emulates ventilator-dependent conditions, providing continuous adjustability of lung compliance from 30 to 100mL/cmH${}_2$O and airway resistance from 2.01 to 14.67$\sqrt{{\mathrm{cmH}}_2\mathrm{O}}/(\mathrm{L}/\mathrm{s})$, with compliance deviations limited to ±5%. This design facilitates fine, continuous modulation of key respiratory parameters, making the system well-suited for evaluating ventilator performance, conducting human–machine interaction studies, and simulating pathological respiratory states. Full article

Sustainable irrigation in water-limited regions requires timely, field-scale estimates of soil water content (SWC). Yet, field-scale SWC studies leveraging near-daily satellite imagery of regenerative systems—particularly cotton under Mediterranean conditions—are lacking, and explainable integrations of Planet SuperDove with agrometeorological inputs remain underexplored. In this study, we evaluated a machine learning framework that integrates near-daily multispectral features from Planet SuperDove with agrometeorological variables to estimate the daily SWC of regenerative cotton under Mediterranean conditions across two seasons (2023–2024). Six regression models were compared; extreme gradient boosting achieved the highest accuracy (R² = 0.73 ± 0.08; RMSE = 4.60 mm ± 0.81; nRMSE = 0.035 ± 0.01), with limited bias and stable performance across the years and moisture conditions. The model interpretability via SHAP indicated that agrometeorological drivers contributed over half of the predictive power, while the NDVI and NIR provided the most informative satellite inputs, followed by the NDRE and PSRI. The results show that combining high-frequency satellite data with meteorological inputs can deliver accurate and interpretable SWC estimates at the homogeneous plot level, supporting irrigation optimization of regenerative systems. This approach is practical, transferable, and suited for operational decision-making where frequent, high-resolution observations are available. Full article

Understanding the distribution of forest aboveground biomass (AGB) is pivotal for carbon monitoring. Field-based inventorying is time-consuming and costly for large-area AGB estimations. The integration of multimodal remote sensing (RS) observations with single-year, field-based forest inventory analysis (FIA) data has the potential to improve the efficiency of large-scale AGB modeling and carbon monitoring initiatives. Our main objective was to systematically compare the AGB prediction accuracies of machine learning algorithms (e.g., random forest (RF) and support vector machine (SVM)) with those of conventional statistical methods (e.g., multiple linear regression (MLR)) using multimodal RS variables as predictors. We implemented a method combining AGB estimates of actual FIA subplot locations with airborne LiDAR, National Agriculture Imagery Program (NAIP) aerial imagery, and Sentinel-2 satellite images for model training, validation, and testing. The hyperparameter-tuned RF model produced a root mean square error (RMSE) of 27.19 Mgha⁻¹ and an R² of 0.41, which outperformed the evaluation metrics of SVM and MLR models. Among the 28 most important explanatory variables used to build the best RF model, 68% of variables were derived from the LiDAR height data. The hyperparameter-tuned linear SVM model exhibited an R² of 0.10 and an RMSE of 32.17 Mgha⁻¹. Additionally, we developed an MLR using eight explanatory variables, which yielded an RMSE of 22.59 Mgha⁻¹ and an R² of 0.22. The linear ensemble model, which was developed using the predictions of all three models, yielded an R² of 0.79. Our results suggested that more field data are required to better generalize the ensemble model. Overall, our findings highlight the importance of variable selection methods, the hyperparameter tuning of ML algorithms, and the integration of multimodal RS data in improving large-area AGB prediction models. Full article

Well integrity is paramount for the safe, environmentally responsible, and economically viable operation of wells throughout their lifecycle, encompassing conventional oil and gas production, unconventional resource extraction (e.g., shale gas and tight oil), and geological storage applications (CO₂, H₂, and natural gas). This review presents a comprehensive synthesis of well integrity challenges, failure mechanisms, monitoring technologies, and management strategies across these operational domains. Key integrity threats—including cement sheath degradation (chemical attack, debonding, cracking, microannuli), casing failures (corrosion, collapse, burst, buckling, fatigue, wear, and connection damage), sustained casing pressure (SCP), and wellhead leaks—are examined in detail. Unique challenges posed by hydraulic fracturing in unconventional wells and emerging risks in CO₂ and hydrogen storage, such as corrosion, carbonation, embrittlement, hydrogen-induced cracking (HIC), and microbial degradation, are also highlighted. The review further explores the evolution of integrity standards (NORSOK, API, ISO), the implementation of Well Integrity Management Systems (WIMS), and the integration of advanced monitoring technologies such as fiber optics, logging tools, and real-time pressure sensing. Particular emphasis is placed on the role of digital technologies—including artificial intelligence, machine learning, and digital twin systems—in enabling predictive maintenance, early failure detection, and lifecycle risk management. The novelty of this review lies in its integrated, cross-domain perspective and its emphasis on digital twin applications for continuous, adaptive well integrity surveillance. It identifies critical knowledge gaps in modeling, materials qualification, and data integration—especially in the context of long-term CO₂ and H₂ storage—and advocates for a proactive, digitally enabled approach to lifecycle well integrity. Full article

Time-Weighted Dynamic Time Warping (TWDTW), adapted from speech recognition, is used in agricultural remote sensing to model crop growth, particularly under limited ground sample conditions. However, most related studies rely on full-season or empirically selected features, overlooking the systematic optimization of features at each observation time to improve TWDTW’s performance. This often introduces a large amount of redundant information that is irrelevant to crop discrimination and increases computational complexity. Therefore, this study focused on maize as the target crop and systematically conducted mapping experiments using Sentinel-1/2 images to evaluate the potential of integrating TWDTW with optimally selected multi-source time series features. The optimal multi-source time series features for distinguishing maize from non-maize were determined using a two-step Jeffries Matusita (JM) distance-based global search strategy (i.e., twelve spectral bands, Normalized Difference Vegetation Index, Enhanced Vegetation Index, and the two microwave backscatter coefficients collected during the maize jointing to tasseling stages). Then, based on the full-season and optimal multi-source time series features, we compared TWDTW with two widely used temporal machine learning models in agricultural remote sensing community. The results showed that TWDTW outperformed traditional supervised temporal machine learning models. In particular, compared with TWDTW driven by the full-season optimal multi-source features, TWDTW using the optimal multi-source time series features improved user accuracy by 0.43% and 2.30%, and producer accuracy by 7.51% and 2.99% for the years 2020 and 2021, respectively. Additionally, it reduced computational costs to only 25% of those driven by the full-season scheme. Finally, maize maps of Yangling District from 2020 to 2023 were produced by optimal multi-source time series features-based TWDTW. Their overall accuracies remained consistently above 90% across the four years, and the average relative error between the maize area extracted from remote sensing images and that reported in the statistical yearbook was only 6.61%. This study provided guidance for improving the performance of TWDTW in large-scale crop mapping tasks, which is particularly important under conditions of limited sample availability. Full article

UAV delivery and ground transfer scheduling in emergency scenarios represent critical technological systems for enhancing disaster response capabilities and safeguarding lives and property. This study systematically reviews recent advances across eight core research domains: UAV emergency delivery systems, ground–air integrated transportation coordination, emergency logistics optimization, UAV path planning and scheduling algorithms, collaborative optimization between ground vehicles and UAVs, emergency response decision support systems, low-altitude economy and urban air traffic management, and intelligent transportation system integration. Research findings indicate that UAV delivery technologies in emergency contexts have evolved from single-aircraft applications to intelligent multi-modal collaborative systems, demonstrating significant advantages in medical supply distribution, disaster relief, and search-and-rescue operations. Current technological development exhibits four major trends: hybrid optimization algorithms, multi-UAV cooperation, artificial intelligence enhancement, and real-time adaptation capabilities. However, critical challenges persist, including regulatory framework integration, adverse weather adaptability, cybersecurity protection, human–machine interface design, cost–benefit assessment, and standardization deficiencies. Future research should prioritize distributed decision architectures, robustness optimization, cross-domain collaboration mechanisms, emerging technology integration, and practical application validation. This comprehensive review provides systematic theoretical foundations and practical guidance for emergency management agencies in formulating technology development strategies, enterprises in investment planning, and research institutions in determining research priorities. Full article