Search Results (1,670)

Artificial intelligence (AI) is rapidly transforming anesthesiology practice across perioperative settings. This review explores the evolution and implementation of hybrid decision-making models that integrate AI capabilities with human clinical expertise. From historical foundations to current applications, we examine how machine learning algorithms, deep learning networks, and big data analytics are enhancing anesthetic care. Key applications include perioperative risk prediction, AI-assisted patient education, automated analysis of clinical records, airway management support, predictive hemodynamic monitoring, closed-loop anesthetic delivery systems, and pain management optimization. In procedural contexts, AI demonstrates promising utility in regional anesthesia through anatomical structure identification and needle navigation, monitoring anesthetic depth via EEG analysis, and improving quality control in endoscopic sedation. Educational applications include intelligent simulators for procedural training and academic productivity tools. Despite significant advances, implementation challenges persist, including algorithmic bias, data security concerns, clinical validation requirements, and ethical considerations regarding AI-generated content. The optimal integration model emphasizes a complementary approach where AI augments rather than replaces clinical judgment—combining computational efficiency with the irreplaceable contextual understanding and ethical reasoning of the anesthesiologist. This hybrid paradigm reinforces the anesthesiologist’s leadership role in perioperative care while enhancing safety, precision, and efficiency through technological innovation. As AI integration advances, continued emphasis on algorithmic transparency, rigorous clinical validation, and human oversight remains essential to ensure that these technologies enhance rather than compromise patient-centered anesthetic care. Full article

Topic modeling of large news streams is widely used to reconstruct economic and political narratives, which requires coherent topics with low lexical overlap while remaining interpretable to domain experts. We propose TF-SYN-NER-Rel, a structural–semantic term weighting scheme that extends classical TF-IDF by integrating positional, syntactic, factual, and named-entity coefficients derived from morphosyntactic and dependency parses of Russian news texts. The method is embedded into a standard Latent Dirichlet Allocation (LDA) pipeline and evaluated on a large Russian-language news corpus from the online archive of Moskovsky Komsomolets (over 600,000 documents), with political, financial, and sports subsets obtained via dictionary-based expert labeling. For each subset, TF-SYN-NER-Rel is compared with standard TF-IDF under identical LDA settings, and topic quality is assessed using the C_v coherence metric. To assess robustness, we repeat model training across multiple random initializations and report aggregate coherence statistics. Quantitative results show that TF-SYN-NER-Rel improves coherence and yields smoother, more stable coherence curves across the number of topics. Qualitative analysis indicates reduced lexical overlap between topics and clearer separation of event-centered and institutional themes, especially in political and financial news. Overall, the proposed pipeline relies on CPU-based NLP tools and sparse linear algebra, providing a computationally lightweight and interpretable complement to embedding- and LLM-based topic modeling in large-scale news monitoring. Full article

Heavy metal enrichment in agricultural soils can affect crop safety, ecosystem functioning, and long-term land productivity, yet farm-scale screening is often constrained by limited routine monitoring data. This study develops a GIS-based framework that combines field-scale spatial analysis with explainable machine learning to characterize and predict heavy metal enrichment on an intensively managed cereal farm in eastern Kazakhstan. Topsoil samples (0 to 20 cm) were collected from 34 fields across eight campaigns between 2020 and 2023, yielding 241 composite field–campaign observations for eight metals (Pb, Cu, Zn, Ni, Cr, Mo, Fe, and Mn) and routine soil properties (humus, pH in H₂O, and pH in KCl). Concentrations were generally low but spatially heterogeneous, with wide observed ranges for several elements (for example, Pb 0.06 to 2.20 mg kg⁻¹, Zn 0.38 to 7.00 mg kg⁻¹, and Mn 0.20 to 38.0 mg kg⁻¹). We synthesized multi-metal structure using an HMI defined as the unweighted mean of z-standardized metal concentrations, which supported field-level screening of persistent enrichment and emerging hot spots. We then trained Extreme Gradient Boosting models using only humus and pH predictors and evaluated performance with field-based spatial block cross-validation. Predictive skill was modest but nonzero for several targets, including HMI (mean R² = 0.20), indicating partial spatial transferability under conservative validation. SHAP analysis identified humus content and soil acidity as dominant contributors to HMI prediction. Overall, the workflow provides a transparent approach for field-scale screening of heavy metal enrichment and establishes a foundation for future integration with satellite-derived covariates for broader monitoring applications. Full article

This paper proposes an integrated, stress-based framework for fatigue diagnosis of wind turbine blades that is tailored to field deployments where detailed structural design information is unavailable. The approach combines a data-driven reduced-order model (ROM) for directional damage equivalent loads (DELs) with a physics-based Soderberg index and a one-class support vector machine (SVM) anomaly detector. The framework is implemented and evaluated using measurements from a 2 MW onshore turbine equipped with blade-root strain gauges and standard SCADA monitoring. Ten-minute operating windows are formed by synchronizing SCADA records with high-frequency strain data, converting strain to stress, and computing DELs via Rainflow counting for flapwise, edgewise, and torsional blade root directions. SCADA inputs are summarized by their 10 min statistics and augmented with yaw misalignment features; these are used to train LightGBM-based ROMs that map operating conditions to directional DELs. On an independent test set, the DEL-ROM achieves coefficients of determination of approximately 0.87, 0.99, and 0.99 for flapwise, edgewise, and torsional directions, respectively, with small absolute errors relative to the measured DELs. The Soderberg index is then used to define conservative Normal/Alert/Alarm classes based on representative material parameters, while a one-class SVM is trained on DEL- and stress-based fatigue features to learn the distribution of normal operation. A simple AND-normal/OR-abnormal rule combines the Soderberg class and SVM label into a hybrid diagnostic decision. Application to the field dataset shows that the proposed framework provides interpretable fatigue-safety margins and reliably highlights operating periods with elevated flapwise fatigue usage, demonstrating its suitability as a scalable building block for digital-twin-enabled condition monitoring and life-extension assessment of wind turbine blades. Full article

Elevators are critical vertical transportation systems in modern urban infrastructure; however, their intricate mechanical and electrical configurations render them highly susceptible to safety-critical failures. Although various automated diagnostic techniques have been proposed, many data-driven approaches exhibit limited generalizability due to their insufficient consideration of physical fault mechanisms and strong dependence on facility-specific training data. To overcome these limitations, this study presents a rule-based automated diagnostic framework for elevator state recognition that prioritizes reliability, real-time performance, and interpretability. The proposed approach explicitly integrates physically meaningful fault characteristics and dominant frequency components into the diagnostic process, and employs predefined expert rules derived from established standards to classify fault states in an automated manner. The effectiveness of the proposed method is verified using real operational data collected from an in-service elevator, demonstrating improved diagnostic accuracy and computational efficiency compared to conventional manual inspection procedures. The proposed framework provides a practical and scalable solution for intelligent elevator condition monitoring and is expected to serve as a foundational technology for future smart maintenance and preventive safety systems. Full article

Background: Due to the lack of curative treatments for inflammatory bowel disease (IBD), patients need lifelong follow-up care. Telehealth offers a valuable solution to balance routine visits with necessary monitoring. Objectives: To map what is known about the benefits and barriers encountered by nurses in their use of telehealth for the follow-up care of patients with IBD. Methods: Following the methodology from the Joanna Briggs Institute, we conducted a scoping review across four electronic databases from June 2024 to September 2025. Key search terms included “inflammatory bowel disease,” “nurse experience,” and “telehealth.” A content analysis was employed to summarize the key findings. Results: We screened 1551 records, ultimately including four original research articles from four countries. Benefits identified were as follows: (1) the vital contributions of IBD telenursing in empowering patients by bridging health literacy and self-care skills; (2) optimal use of staffing time supports patient-centred care; and (3) ease of use. Barriers included the following: (1) increased workload and task imbalances; (2) the need for customized interventions; (3) technical issues and concerns regarding the security of digital systems; (4) telehealth as a supplementary option or a standard procedure; and (5) concerns related to the patient–nurse relationship. Conclusions: Nurses view telehealth as a promising approach that enhances patients’ health literacy and self-care skills and improves patient outcomes through effective monitoring. To fully realize telehealth’s potential, implementing strategies like triage protocols, algorithmic alerts, electronic health record integration, and comprehensive nurse training to enhance patient care and engagement may be beneficial. This scoping review highlights the need for more research on nurses’ experiences with telehealth in IBD due to limited publications. Full article

Effective spatial and structural monitoring of World Heritage sites often relies on continuous high-spatial-resolution remote sensing imagery, which is often unavailable for specific years due to sensor, atmospheric, and revisit constraints. Super-resolution reconstruction thus becomes crucial for maintaining data continuity for such analyses. Traditional methods are trained on temporally aligned LR-HR pairs; however, their performance significantly declines when applied to unseen years due to temporal distribution shifts. To address this, we propose a temporal invariance-aware training strategy combined with an improved Residual Dense Network (RDN_2_M). We introduce a cross-year masked sample generation algorithm that identifies temporally stable regions via local structural similarity. This constructs explicit invariance-guided training pairs, which helps guide the model to focus on persistent structural features rather than transient appearances and to learn robust representations against inter-annual variations. Experiments on the Bin County Cave Temple (BCCT) Heritage Site dataset show our method, integrating the proposed strategy with the enhanced RDN model (RDN_2_M), significantly improves both the objective metrics and visual quality of reconstructed images. This offers a practical solution to filling temporal data gaps, thereby supporting long-term spatial and structural heritage monitoring. Full article

To develop a high-performance detection system for automated harvesting on resource-limited edge devices, we introduce FSA-DETR-P, a lightweight detection framework that addresses challenges such as illumination inconsistency, occlusion, and scale variation in complex orchard environments. Unlike traditional computationally intensive architectures, this model optimizes real-time detection transformers by integrating an efficient backbone for fast feature extraction, a simplified aggregation structure to minimize complexity, and an adaptive mechanism for multi-scale feature fusion. The optimized backbone improves early-stage texture extraction while reducing computational demands. The streamlined aggregation design enhances multi-level interactions without losing spatial detail, and the adaptive fusion module strengthens the detection of small, partially occluded, or ambiguous fruits. We created a domain-specific pomegranate dataset, expanded to 13,840 images with a rigorous 8:1:1 split for training, validation, and testing. The results show that the pruned and optimized model achieves a Mean Average Precision (mAP50) of 0.928 and mAP50–95 of 0.632 with reduced parameters (13.73 M) and lower computational costs (34.6 GFLOPs). It operates at 24.6 FPS on an NVIDIA Jetson Orin Nano, indicating a strong balance between accuracy and deployability, making it well-suited for orchard monitoring and robotic harvesting in real-world applications. Full article

Accurate recognition of lower limb activities is essential for wearable rehabilitation systems and assistive robotics like exoskeletons and prosthetics. This study introduces SDALLE, a custom hardware data acquisition system that integrates surface electromyography sensors (EMGs) and inertial measurement sensors (IMUs) into a wireless, portable platform for locomotor monitoring. Using this system, data were collected from nine healthy subjects performing four fundamental locomotor activities: walking, jogging, stair ascent, and stair descent. The recorded signals underwent an offline structured preprocessing pipeline consisting of time-series augmentation (jittering and scaling) to increase data diversity, followed by wavelet-based denoising to suppress high-frequency noise and enhance signal quality. A temporal one-dimensional convolutional neural network (1D-TCNN) with three convolutional blocks and fully connected layers was trained on the prepared dataset to classify the four activities. Classification using IMU sensors achieved the highest performance, with accuracies ranging from 0.81 to 0.95. The gyroscope X-axis of the left Rectus Femoris achieved the best performance (0.95), while accelerometer signals also performed strongly, reaching 0.93 for the Vastus Medialis in the Y direction. In contrast, electromyography channels showed lower discriminative capability. These results demonstrate that the combination of SDALLE hardware, appropriate data preprocessing, and a temporal CNN provides an effective offline sensing and activity classification pipeline for lower limb activity recognition and offers an open-source dataset that supports further research in human activity recognition, rehabilitation, and assistive robotics. Full article

This paper proposes a privacy-preserving framework that couples probabilistic wind power forecasting with decentralized anomaly detection in complex power networks. We first design an adaptive federated learning (FL) scheme to produce probabilistic forecasts for multiple geographically distributed wind farms while keeping their raw data local. In this scheme, an artificial neural network with quantile regression is trained collaboratively across sites to provide calibrated prediction intervals for wind power outputs. These forecasts are then embedded into an alternating direction method of multipliers (ADMM)-based load-side dispatch and anomaly detection model for decentralized power systems with plug-and-play industrial users. Each monitoring node uses local measurements and neighbor communication to solve a distributed economic dispatch problem, detect abnormal load behaviors, and maintain network consistency without a central coordinator. Experiments on the GEFCom 2014 wind power dataset show that the proposed FL-based probabilistic forecasting method outperforms persistence, local training, and standard FL in RMSE and MAE across multiple horizons. Simulations on IEEE 14-bus and 30-bus systems further verify fast convergence, accurate anomaly localization, and robust operation, indicating the effectiveness of the integrated forecasting–dispatch framework for smart industrial grids with high wind penetration. Full article

Irregular wheel wear can significantly degrade wheel–rail interaction performance and, in severe cases, compromise the safety of high-speed trains. Accurate and timely monitoring of wheel wear is crucial for maintaining operational reliability. Existing monitoring methods often rely on high-end sensors or are sensitive to environmental disturbances, limiting their practical deployment. This study proposes a novel method for monitoring irregular wheel wear by analyzing the stator current spectrum of traction motors. Firstly, an electromechanical coupled model is developed by integrating the electric drive system with the vehicle–track dynamic model to capture the propagation of wear-induced excitation. The effect of polygonal wear on the stator current is investigated, revealing the presence of harmonic components coupled with the wear excitation frequency. To extract these features, a comb filter based on Variational Mode Decomposition (VMD) is introduced. The method effectively isolates wheel wear-related harmonics from existing electrical harmonics in the stator current signal. Simulation results demonstrate that the proposed approach can accurately detect harmonic features caused by polygonal wear, validating its applicability. This method provides a feasible and non-intrusive solution for wheel wear monitoring, offering theoretical support for condition-based maintenance of high-speed rail systems. Full article

The widespread circulation of Eurasian avian-like H1N1 (EA H1N1) swine influenza virus poses significant zoonotic and pandemic risks worldwide. However, current diagnostic methods are difficult to deploy in the field, as they generally require specialized laboratory infrastructure and trained personnel. Here, we present a novel dual-signal detection platform that combines reverse transcription recombinase polymerase amplification (RT-RPA) with CRISPR/Cas12a technology for rapid, on-site EA H1N1 detection. We established an integrated one-tube assay by designing and optimizing RT-RPA primers targeting a conserved region of the hemagglutinin (HA) gene, together with engineered CRISPR/Cas12a guide RNAs exhibiting high specificity. The platform incorporates two complementary readout modes: real-time fluorescence monitoring and visual colorimetric detection using a smartphone. The assay shows excellent analytical specificity, with no cross-reactivity observed against other swine influenza virus subtypes or common swine pathogens, (including CSFV, PRRSV, PEDV, PCV, TGEV, and RV). The detection limit is 2 copies/μL, and the entire procedure can be completed within 30 mins using simple portable equipment. When evaluated on 86 clinical samples, the assay demonstrated 94.18% concordance with RT-qPCR. Compared with conventional diagnostic methods, this RT-RPA–CRISPR/Cas12a assay offers greater convenience and cost-effectiveness. Its strong potential for field-based rapid testing underscores promising application prospects in swine influenza surveillance and control programs. Full article

Accurate distance perception and collision reasoning are crucial for robotic manipulation in the confined interior of tokamak vacuum vessels. Traditional mesh- or voxel-based methods suffer from discretization artifacts, discontinuities, and heavy memory requirements, making them unsuitable for continuous geometric reasoning and optimization-based planning. This paper presents an Occupancy-Aware Neural Distance Perception (ONDP) framework that serves as a compact and differentiable geometric sensor for manipulator obstacle avoidance in reactor-like environments. To address the inadequacy of conventional sampling methods in such constrained environments, we introduce a Physically-Stratified Sampling strategy. This approach moves beyond heuristic adaptation to explicitly dictate data distribution based on specific engineering constraints. By injecting weighted quotas into critical safety buffers and enforcing symmetric boundary constraints, we ensure robust gradient learning in high-risk regions. A lightweight neural network is trained directly in physical units (millimeters) using a mean absolute error loss, ensuring strict adherence to engineering tolerances. The resulting model achieves approximately 2–3 mm near-surface accuracy and supports high-frequency distance and normal queries for real-time perception, monitoring, and motion planning. Experiments on a tokamak vessel model demonstrate that ONDP provides continuous, sub-centimeter geometric fidelity. Crucially, benchmark results confirm that the proposed method achieves a query frequency exceeding 15 kHz for large-scale batches, representing a 5911× speed-up over mesh-based queries. This breakthrough performance enables its seamless integration with trajectory optimization and model-predictive control frameworks for confined-space robotic manipulation. Full article

Background: Artificial Intelligence (AI) has emerged as a transformative tool in in vitro fertilization (IVF) as it has done in other sectors. In IVF, AI offers advancements in embryo selection, treatment personalization, and outcome prediction. It does so by leveraging deep learning and computer vision, as well as AI-driven platforms such as ERICA, iDAScore, and IVY where the goal is to address the limitations of traditional embryo assessment. Key amongst them are the issues of subjectivity, labor intensity, and limited predictive power. Despite rapid technological progress, the integration of AI into routine IVF practice faces key challenges. These are issues related to clinical validation, ethical dilemmas, and workflow adaptation. Rationale/Objectives: This review synthesizes current evidence to evaluate the role of AI in IVF, focusing on six critical dimensions: (1) the evolution of AI from traditional embryology to algorithmic assessment, (2) clinical validation and regulatory considerations, (3) limitations and ethical challenges, (4) pathways for clinical integration, (5) real-world applications and outcomes, and (6) future directions and policy recommendations. The objective is to provide a comprehensive roadmap for the responsible adoption of AI in reproductive medicine. Outcomes: AI demonstrates significant potential to improve the precision and efficiency of IVF. Studies report that AI models can achieve 10 to 25% higher accuracy in predicting embryo viability and implantation potential compared to traditional morphological assessment by embryologists. This enhanced predictive power supports more consistent embryo ranking, facilitates elective single-embryo transfer (eSET) strategies, and is associated with 30 to 50% reductions in embryologist workload per embryo cohort. Early adopters report promising trends. However, large-scale randomized controlled trials have yet to conclusively demonstrate a statistically significant increase in live birth rates per transfer compared to expert embryologist selection. The most immediate and evidenced value of AI lies in hybrid decision-making models. This is where it augments embryologists by providing data-driven, objective support, thereby standardizing workflows and reducing subjectivity. Wider Implications: The sustainable integration of AI into IVF banks on three key aspects: robust evidence generation, interdisciplinary collaboration, and global standardization. To foster these, policymakers ought to establish regulatory frameworks for transparency and bias mitigation. On their part, clinicians need training to interpret AI outputs critically. Ethically, safeguarding patient trust and equity is non-negotiable. Future innovations, mainly AI-enhanced genomics and real-time monitoring, could further personalize care. However, their success depends on addressing current limitations. By balancing innovation with ethical vigilance, AI holds the potential to revolutionize IVF while upholding the highest standards of patient care. Full article

Fishermen operating in pelagic fisheries often experience significant barriers to medical care due to geographic isolation, harsh environmental conditions, and the absence of onboard healthcare personnel. Telemedicine offers an effective approach to overcome these limitations by enabling remote diagnosis, monitoring, and treatment through satellite-based communication systems. This review summarizes the progress and applications of telemedicine in maritime and other austere environments, focusing on technological advancements, clinical implementations, and emerging trends in artificial intelligence-driven healthcare. Evidence from pilot and retrospective studies highlights the growing use of wearable devices, telementored ultrasound, digital photography, and cloud-based monitoring systems for managing acute and chronic medical conditions at sea. The integration of machine learning and deep learning algorithms has further improved fatigue, stress, and motion detection, enhancing early risk assessment among seafarers. Despite challenges such as limited connectivity, data privacy concerns, and training requirements, the adoption of telemedicine significantly improves health outcomes, reduces emergency evacuations, and promotes occupational safety. Future directions emphasize the development of 5G-enabled Internet of Medical Things networks and predictive AI tools to establish comprehensive maritime telehealth ecosystems for fishermen in pelagic operations. Full article