Search Results (468)

Intention prediction is essential for enabling intuitive and adaptive control in upper-limb exoskeletons, especially in dynamic industrial environments. However, the suitability of different cues, sensors, and computational models for real-world industrial applications remains unclear. This systematic review, conducted according to PRISMA guidelines, analyzes 29 studies published between 2007 and 2024 that investigate intention prediction in active exoskeletons. Most studies rely on motion capture (14) and electromyography (14) to estimate joint torque or trajectories, predicting from 450 ms before to 660 ms after motion onset. Approaches include model-based and model-free regression, as well as classification methods, but vary significantly in complexity, sensor setups, and evaluation procedures. Only a subset evaluates usability or support effectiveness, often under laboratory conditions with small, non-representative participant groups. Based on these insights, we outline recommendations for robust and adaptable intention prediction tailored to industrial task requirements. We propose four generalized support modes to guide sensor selection and control strategies in practical applications. Future research should leverage wearable sensors, integrate cognitive and contextual cues, and adopt transfer learning, federated learning, or LLM-based feedback mechanisms. Additionally, studies should prioritize real-world validation, diverse participant samples, and comprehensive evaluation metrics to support scalable, acceptable deployment of exoskeletons in industrial settings. Full article

The COVID-19 pandemic highlighted the need for a scientific framework that enables quantitative assessment and control of airborne infection risks in indoor environments. This study identifies limitations in the traditional Wells–Riley model—specifically its assumptions of perfect mixing and steady-state conditions—and addresses these shortcomings by adopting the REHVA (Federation of European Heating, Ventilation and Air Conditioning Associations) infection risk assessment model. We propose a five-tier risk classification system (Monitor, Caution, Alert, High Risk, Critical) based on two key metrics: the probability of infection (Pₙ) and the event reproduction number (R_event). Unlike the classical model, our approach integrates airborne virus removal mechanisms—such as natural decay, gravitational settling, and filtration—with occupant dynamics to reflect realistic contagion scenarios. Simulations were conducted across 10 representative indoor settings—such as classrooms, hospital waiting rooms, public transit, and restaurants—considering ventilation rates and activity-specific viral emission patterns. The results quantify how environmental variables (ventilation, occupancy, time) impact each setting’s infection risk level. Our findings indicate that static mitigation measures such as mask-wearing or physical distancing are insufficient without dynamic, model-based risk evaluation. We emphasize the importance of incorporating real-time crowd density, occupancy duration, and movement trajectories into risk scoring. To support this, we propose integrating computer vision (CCTV-based crowd detection) and entry/exit counting sensors within a live airborne risk assessment framework. This integrated system would enable proactive, science-driven epidemic control strategies, supporting real-time adaptive interventions in indoor spaces. The proposed platform could serve as a practical tool for early warning and management during future airborne disease outbreaks. Full article

Background/Objectives: The relationship between systemic health and ocular diseases is well-documented, with various body metrics potentially playing significant roles in the pathogenesis of cataracts, glaucoma, and age-related macular degeneration (AMD). However, comprehensive studies linking these metrics with ocular health are sparse. This study aims to explore the associations between height, weight, waist circumference, and BMI with the prevalence and current status of cataracts, glaucoma, and AMD in a large cohort. Methods: We used data from Korean National Health and Nutrition Survey (KNHANES 2015–2021), a national, cross-sectional health examination and survey, for which representative data on the health, nutritional status, and physical activities of the Korean general population are collected by the Korea Centers for Disease Control and Prevention (KCDC). We compared height, weight, waist circumference, and BMI among patients with diagnosed and current cataracts, glaucoma, and AMD versus those without these conditions. Statistical analyses included t-tests and Pearson correlation analyses to examine the relationships between body metrics and ocular diseases. Results: Our findings indicate that shorter height and lower weight are associated with diagnosed cataracts and glaucoma but not with their current status. A greater waist circumference was observed in patients with diagnosed cataracts, glaucoma, and AMD compared to controls, suggesting central obesity as a potential associated factor. No significant differences in BMI were found in patients with current ocular diseases. Additionally, certain body metrics were correlated with refractive errors and visual acuity, suggesting broader implications for ocular health. Conclusions: The study highlights significant associations between body metrics and the risk of developing cataracts, glaucoma, and AMD. AMD was found to be more closely related to systemic diseases, such as diabetes and hypertension, than to body metrics. These findings suggest that interventions targeting obesity and metabolic health could potentially reduce the risk or severity of these common ocular conditions. Further research is needed to confirm these relationships and explore underlying mechanisms. Full article

Air pollution is steadily increasing due to industrialization, economic activities, and transportation. High levels pose a significant threat to human health and well-being worldwide. Saudi Arabia is a growing country with air quality indices ranging from moderate to unhealthy. Although there are many monitoring stations distributed throughout the country, mathematical modeling of air pollution is still crucial for health and environmental decision-making. From this perspective, in this study, a data-driven approach based on pollutant records and a Deep Learning (DL) Long Short-Term Memory (LSTM) algorithm is carried out to perform temporal modeling of selected pollutants (PM₁₀, PM_2.5, CO and O₃) based on time series combined with a spatial modeling focused on selected cities (Riyadh, Jeddah, Mecca, Rabigh, Abha, Dammam and Taif), covering ~48% of the total population of the country. The best forecasts were provided by LSTM in cases where the datasets used were of relatively large size. Numerically, the obtained performance metrics such as the coefficient of determination (R²) ranged from 0.2425 to 0.8073. The best LSTM results were compared to those provided by two ensemble methods, Random Forest (RF) and eXtreme Gradient Boosting (XGBoost), where the merits of LSTM were confirmed mainly in terms of its ability to capture hidden relationships. We also found that overall, meteorological factors showed a weak association with pollutant concentrations, with ambient temperature exerting a moderate influence. However, incorporating ambient temperature into LSTM models did not lead to a significant improvement in predictive accuracy. The developed approach can be used to support decision-making in environmental and health domains, as well as to monitor pollutant concentrations based on historical time series records. Full article

Monitoring optically-active water quality (OAWQ) parameters faces key challenges, primarily due to limited in situ measurements and the restricted availability of high-resolution multispectral remote sensing imagery. While deep learning has shown promise for OAWQ estimation, existing approaches such as GeoTile2Vec, which relies on geographic proximity, and SimCLR, a domain-agnostic contrastive learning method, fail to capture land cover-driven water quality patterns, limiting their generalizability. To address this, we present deep meta-connectivity representation (DMCR), which integrates multispectral remote sensing imagery with limited in situ measurements to estimate OAWQ parameters. Our approach constructs meta-feature vectors from land cover images to represent the water quality characteristics of each multispectral remote sensing image tile. We introduce the meta-connectivity concept to quantify the OAWQ similarity between different tiles. Building on this concept, we design a contrastive self-supervised learning framework that uses sets of quadruple tiles extracted from Sentinel-2 imagery based on their meta-connectivity to learn DMCR vectors. After the core neural network is trained, we apply a random forest model to estimate parameters such as chlorophyll-a (Chl-a) and turbidity using matched in situ measurements and DMCR vectors across time and space. We evaluate DMCR on Lake Erie and Lake Ontario, generating a series of Chl-a and turbidity distribution maps. Performance is assessed using the $R^2$ and $RMSE$ metrics. Results show that meta-connectivity more effectively captures water quality similarities between tiles than widely utilized geographic proximity approaches such as those used in GeoTile2Vec. Furthermore, DMCR outperforms baseline models such as SimCLR with randomly cropped tiles. The resulting distribution maps align well with known factors influencing Chl-a and turbidity levels, confirming the method’s reliability. Overall, DMCR demonstrates strong potential for large-scale OAWQ estimation and contributes to improved monitoring of inland water bodies with limited in situ measurements through meta-connectivity-informed deep learning. The temporal-spatial water quality maps can support large-scale inland water monitoring, early warning of harmful algal blooms. Full article

Despite cognitive workload (CW) being a critical metric in several applications, no technology exists to seamlessly and reliably quantify CW. Previously, we demonstrated the feasibility of a wearable MagnetoCardioGraphy (MCG) sensor to classify high vs. low CW based on MCG-derived heart rate variability (mHRV). However, our sensor was unable to address certain critical operational requirements, resulting in noisy signals, often to the point of being unusable. In addition, test conditions for the participants were not decoupled from motion (i.e., physical activity (PA)), raising questions as to whether the noted changes in mHRV were attributed to CW, PA, or both. This study reports software and hardware advancements to optimize the MCG data quality, and investigates whether changes in CW (in the absence of PA) can be reliably detected. Performance is validated for healthy adults (n = 10) performing three types of CW tasks (one for low CW and two for high CW to eliminate the memory effect). Results demonstrate the ability to retrieve MCG R-peaks throughout the recordings, as well as the ability to differentiate high vs. low CW in all cases, confirming that CW does modulate the mHRV. A paired Bonferroni t-test with significance $\alpha =0.01$ confirms the hypothesis that an increase in CW decreases mHRV. Our findings lay the groundwork toward a seamless, practical, and low-cost sensor for monitoring CW. Full article

Objectives: Shoulder injury prevalence appears to be the highest among all injuries in CrossFit (CF) athletes. Nevertheless, there is no evidence deriving from prospective studies to explain this phenomenon. The purpose of this study was to document shoulder injury incidence in CF participants over a 12-month period and prospectively investigate the risk factors associated with their demographic, epidemiological, and functional characteristics. Methods: The sample comprised 109 CF athletes in various levels. Participants’ data were collected during the baseline assessment, using a specially designed questionnaire, as well as active range of motion, muscle strength, muscle endurance, and sport-specific tests. Non-parametric statistical tests and inferential statistics were employed, and in addition, linear and regression models were created. Logistic regression models incorporating the study’s continuous predictors to classify injury occurrence in CF athletes were developed and evaluated using the Area Under the ROC Curve (AUC) as the performance metric. Results: A shoulder injury incidence rate of 0.79 per 1000 training hours was recorded. Olympic weightlifting (45%) and gymnastics (35%) exercises were associated with shoulder injury occurrence. The most frequent injury concerned rotator cuff tendons (45%), including lesions and tendinopathies, exhibiting various severity levels. None of the examined variables individually showed a statistically significant correlation with shoulder injuries. Conclusions: This is the first study that has investigated prospectively shoulder injuries in CrossFit, creating a realistic profile of these athletes. Despite the broad spectrum of collected data, the traditional statistical approach failed to identify shoulder injury predictors. This indicates the necessity to explore this topic using more sophisticated techniques, such as advanced machine learning approaches. Full article

Emerging digital technologies are transforming how consumers participate in financial markets, yet their benefits depend critically on the speed, reliability, and transparency of the underlying platforms. Online stock trading platforms must maintain high efficiency underload to ensure a good user experience. This paper presents performance analysis under various load conditions based on the containerized stock exchange system. A comprehensive data logging pipeline was implemented, capturing metrics such as API response times, database query times, and resource utilization. We analyze the collected data to identify performance patterns, using both statistical analysis and machine learning techniques. Preliminary analysis reveals correlations between application processing time and database load, as well as the impact of user behavior on system performance. Association rule mining is applied to uncover relationships among performance metrics, and multiple classification algorithms are evaluated for their ability to predict user activity class patterns from system metrics. The insights from this work can guide optimizations in similar distributed web applications to improve scalability and reliability under a heavy load. By framing performance not merely as a technical property but as a determinant of financial decision-making and well-being, the study contributes actionable insights for designers of consumer-facing fintech services seeking to meet sustainable development goals through trustworthy, resilient digital infrastructure. Full article

The integration of advanced computer vision and artificial intelligence (AI) techniques into collaborative robotic systems holds the potential to revolutionize human–robot interaction, productivity, and safety. Despite substantial research activity, a systematic synthesis of how vision and AI are jointly enabling context-aware, adaptive cobot capabilities across perception, planning, and decision-making remains lacking (especially in recent years). Addressing this gap, our review unifies the latest advances in visual recognition, deep learning, and semantic mapping within a structured taxonomy tailored to collaborative robotics. We examine foundational technologies such as object detection, human pose estimation, and environmental modeling, as well as emerging trends including multimodal sensor fusion, explainable AI, and ethically guided autonomy. Unlike prior surveys that focus narrowly on either vision or AI, this review uniquely analyzes their integrated use for real-world human–robot collaboration. Highlighting industrial and service applications, we distill the best practices, identify critical challenges, and present key performance metrics to guide future research. We conclude by proposing strategic directions—from scalable training methods to interoperability standards—to foster safe, robust, and proactive human–robot partnerships in the years ahead. Full article

Objectives: The early detection of low bone mineral density (BMD) is essential for preventing osteoporosis and related complications. While dual-energy X-ray absorptiometry (DXA) remains the gold standard for diagnosis, its cost and limited availability restrict its use in large-scale screening. This study investigated whether raw bioelectrical impedance analysis (BIA) data combined with explainable machine learning (ML) models could accurately classify osteopenia in women aged 40 to 55. Methods: In a cross-sectional design, 138 women underwent same-day BIA and DXA assessments. Participants were categorized as osteopenic (T-score between −1.0 and −2.5; n = 33) or normal (T-score ≥ −1.0) based on DXA results. Overall, 24.1% of the sample were classified as osteopenic, and 32.85% were postmenopausal. Raw BIA outputs were used as input features, including impedance values, phase angles, and segmental tissue parameters. A sequential forward feature selection (SFFS) algorithm was employed to optimize input dimensionality. Four ML classifiers were trained using stratified five-fold cross-validation, and SHapley Additive exPlanations (SHAP) were applied to interpret feature contributions. Results: The neural network (NN) model achieved the highest classification accuracy (92.12%) using 34 selected features, including raw impedance measurements, derived body composition indices such as regional lean mass estimates and the edema index, as well as a limited number of categorical variables, including self-reported physical activity status. SHAP analysis identified muscle mass indices and fluid distribution metrics, features previously associated with bone health, as the most influential predictors in the current model. Other classifiers performed comparably but with lower precision or interpretability. Conclusions: ML models based on raw BIA data can classify osteopenia with high accuracy and clinical transparency. This approach provides a cost-effective and interpretable alternative for the early identification of individuals at risk for low BMD in resource-limited or primary care settings. Full article

Every day humans interact with smartphones that have embedded sensors that enable the tracking of changing physical activities of the device owner. However, several problems arise with the recognition of multiple activities (such as walking, sitting, running, and other) on smartphones. Firstly, most of the devices do not recognize some activities well, such as walking upstairs or downstairs. Secondly, recognition algorithms are embedded into smartphone software and are static, unless updated. In this case, a recognition algorithm must be re-trained with training data of a specific size. Thus, an adaptive (also known as, online or incremental) learning algorithm would be useful in this situation. In this work, an adaptive learning and classification algorithm based on hidden Markov models (HMMs) is applied to human activity recognition, and an architecture model for smartphones is proposed. To create a self-learning method, a technique that involves building an incremental algorithm in a maximal likelihood framework has been developed. The adaptive algorithms created enable fast self-learning of the model parameters without requiring the device to store data obtained from sensors. It also does not require sending gathered data to a server over the network for additional processing, making them autonomous and independent from outside systems. Experiments involving the modeling of various activities as separate HMMs with different numbers of states, as well as modeling several activities connected to one HMM, were performed. A public dataset called the Activity Recognition Dataset was considered for this study. To generalize the results, different performance metrics were used in the validation of the proposed algorithm. Full article

Call Detail Records (CDRs) from mobile networks offer valuable insights into both network performance and user behavior. With the growing importance of data analytics, analyzing CDRs has become critical for optimizing network resources by forecasting demand across spatial and temporal dimensions. In this study, we examine publicly available CDR data from Telecom Italia to explore the spatiotemporal dynamics of mobile network activity in Milan. This analysis reveals key patterns in traffic distribution highlighting both high- and low-demand regions as well as notable variations in usage over time. To anticipate future network usage, we employ both Automated Machine Learning (AutoML) and the transformer-based TimeGPT model, comparing their performance against traditional forecasting methods such as Long Short-Term Memory (LSTM), ARIMA and SARIMA. Model accuracy is assessed using standard evaluation metrics, including root mean square error (RMSE), mean absolute error (MAE) and the coefficient of determination (R²). Results show that AutoML delivers the most accurate forecasts, with significantly lower RMSE (2.4990 vs. 14.8226) and MAE (1.0284 vs. 7.7789) compared to TimeGPT and a higher R² score (99.96% vs. 98.62%). Our findings underscore the strengths of modern predictive models in capturing complex traffic behaviors and demonstrate their value in resource planning, congestion management and overall network optimization. Importantly, this study is one of the first to Comprehensively assess AutoML and TimeGPT on a high-resolution, real-world CDR dataset from a major European city. By merging machine learning techniques with advanced temporal modeling, this study provides a strong framework for scalable and intelligent mobile traffic prediction. It thus highlights the functionality of AutoML in simplifying model development and the possibility of TimeGPT to extend transformer-based prediction to the telecommunications domain. Full article

Explainable Artificial Intelligence (XAI) is increasingly important in computer vision, aiming to connect complex model outputs with human understanding. This review provides a focused comparative analysis of representative XAI methods in four main categories, attribution-based, activation-based, perturbation-based, and transformer-based approaches, selected from a broader literature landscape. Attribution-based methods like Grad-CAM highlight key input regions using gradients and feature activation. Activation-based methods analyze the responses of internal neurons or feature maps to identify which parts of the input activate specific layers or units, helping to reveal hierarchical feature representations. Perturbation-based techniques, such as RISE, assess feature importance through input modifications without accessing internal model details. Transformer-based methods, which use self-attention, offer global interpretability by tracing information flow across layers. We evaluate these methods using metrics such as faithfulness, localization accuracy, efficiency, and overlap with medical annotations. We also propose a hierarchical taxonomy to classify these methods, reflecting the diversity of XAI techniques. Results show that RISE has the highest faithfulness but is computationally expensive, limiting its use in real-time scenarios. Transformer-based methods perform well in medical imaging, with high IoU scores, though interpreting attention maps requires care. These findings emphasize the need for context-aware evaluation and hybrid XAI methods balancing interpretability and efficiency. The review ends by discussing ethical and practical challenges, stressing the need for standard benchmarks and domain-specific tuning. Full article

This study presents a sustainable approach for synthesizing high-performance activated carbon from Spirulina Alga through hydrothermal carbonization followed by chemical activation using potassium hydroxide. The resulting activated carbon exhibited a high Brunauer–Emmett–Teller (BET) surface area of 1747 m²/g and a total pore volume of 1.147 cm³/g, with micropore volume accounting for 0.4 cm³/g. Characterization using Scanning Electron Microscopy-Energy Dispersive X-ray Spectroscopy (SEM-EDS), X-ray Photoelectron Spectroscopy (XPS), and gas adsorption analyses confirmed the presence of hierarchical micro- and mesoporosity as well as favorable surface functional groups. The synthesized carbon was used to fabricate electrodes for membrane capacitive deionization (MCDI) along with cation and anion-selective membranes, which were then tested with saline water (500–5000 ppm) and synthetic hard water (898 ppm of total salts). The salt adsorption capacity (SAC) reached 25 (batch) to 40 (continuous) mg/g, while rapid adsorption rates with average salt adsorption rates (ASARs) values exceeding 10 (batch) to 30 (continuous) mg·g⁻¹·min⁻¹ during early stages were obtained. Batch MCDI experiments demonstrated a higher SAC compared to continuous operation, with non-monotonic trends in SAC observed as a function of feed concentration. Ion adsorption kinetics were influenced by ion valency, membrane selectivity, and pore structure. The specific energy consumption (SEC) was calculated as 8–21 kJ/mol for batch and 0.1–0.5 kJ/mol for continuous process. These performance metrics are on par with or surpass those reported in the recent literature for similar single-electrode CDI configurations. The results demonstrate the viability of using Alga-derived carbon as an efficient and eco-friendly electrode material for water desalination technologies. Full article

Elevated thermal conditions, rock formations exhibit distinct mechanical behaviors that significantly deviate from their characteristics under ambient temperature environments. This phenomenon raises critical questions regarding the applicability of conventional failure criteria in accurately assessing wellbore stability and maintaining the structural integrity of subsurface infrastructure within geothermal environments. Based on the least absolute deviation method, this paper studies the response characteristics of rock strength at different temperatures and evaluates the prediction performance of six commonly used strength criteria under various temperature and stress environments. The experimental findings reveal a pronounced nonlinear dependence of rock strength on confining pressure elevation. A comparative analysis of failure criteria demonstrates hierarchical predictive performance: the Hoek–Brown (HB) criterion achieves superior temperature-dependent strength prediction fidelity, outperforming the modified Griffith (MGC), Mohr–Lade (ML), and modified Wiebols–Cook (MWC) criteria by 12–18% in accuracy metrics. Notably, the Zhao–Zheng (ZZ) and conventional Mohr–Coulomb (MC) criteria exhibit statistically significant deviations across the tested thermal range. The HB criterion’s exceptional performance in high-temperature regimes is attributed to its dual incorporation of nonlinear confinement effects and thermally activated microcrack propagation mechanisms. The implementation of this optimized model in Well X’s borehole stability analysis yielded 89% alignment between predictions and field observations, with principal stress variations remaining within 7% of critical failure thresholds. These mechanistic insights offer critical theoretical and practical references for thermo-hydro-mechanical coupling analysis in enhanced geothermal systems and deep subsurface containment structures. Full article