Search Results (168,829)

Aiming to reduce muscle fatigue and prevent occupational injuries caused by prolonged lifting in insulator replacement operations, this study presents the design of an upper-limb exoskeleton. Firstly, this study performs kinematic analysis and phase segmentation of the lifting motion in the insulator replacement operation. Based on the analysis, in terms of mechanical structure, the proposed upper-limb exoskeleton adopts a unilateral three-degree-of-freedom shoulder mechanism that biomimics the human glenohumeral joint, which reduces the misalignment between the exoskeleton and the human body. Meanwhile, a waist–back support structure is integrated into the exoskeleton to realize a more reasonable torque transmission path. In terms of the control strategy, based on the operation’s phase segmentation and dynamic modeling of the human upper limb, this study develops a neural network-based assistive control algorithm for insulator replacement operations, enabling the exoskeleton to provide phase-specific torque output. Experimental results demonstrate that, under a simulated insulator replacement operation with a 20 kg load, the exoskeleton significantly reduces the subject’s sEMG activity of the biceps brachii and triceps brachii, effectively alleviating muscle fatigue. Full article

Transportation agencies receive large volumes of free-form public comments describing infrastructure conditions, safety concerns, and service issues. These comments are often processed manually for downstream operational actions, which is time-consuming, inconsistent across reviewers, and difficult to scale, thereby limiting their value for operational decision-making. This study presents a machine learning and Large Language Model (LLM) framework for automated triage of free-form public comments, assigning each report to a three-level hierarchical taxonomy consisting of Category, Subcategory, and Final Decision. The proposed framework uses agency historical data together with retrieval-based evidence, where semantically similar past comments are provided to the LLM as contextual support to better align predictions with agency-specific labeling practices. The framework was evaluated using TF-IDF with Logistic Regression, TF-IDF with Linear SVM, embedding-based kNN with cosine similarity, few-shot LLM prompting, and retrieval-based LLM prompting. Results show that retrieval-based prompting achieved the best overall performance, with the highest accuracy at both the Category and Subcategory levels. At the Final Decision level, retrieval-based prompting slightly outperformed kNN, while few-shot prompting performed worse. Error analysis showed that many misclassifications were semantically plausible alternatives, reflecting the overlap across infrastructure-related complaint categories. When a second candidate label was allowed, further improving performance. Latency analysis also indicated that the framework can process more than 2000 comments in under 30 min, supporting faster and more consistent agency workflows. Full article

Traditional homogeneous materials often face an inherent trade-off between strength and toughness, restricting their application in high-performance impact protection. Mechanical metamaterials overcome this fundamental limitation by integrating structure and material. The 3D-printed Bouligand laminates (3DPBLs), a type of mechanical metamaterial, are renowned for their exceptional impact resistance. While the 3DPBLs have been proven to provide superior resistance under normal impact, actual service conditions inevitably involve complex, multi-directional loading. We aimed to investigate the 3DPBLs’ oblique impact resistance here. To this purpose, samples of 3DPBLs with varying helical angles (0°, 7°, 15°, 60°, 90°) were fabricated and subjected to low-velocity drop-weight impact tests at impact angles of 0°, 30°, 45°, and 60° to evaluate their damage evolution and energy dissipation. The experimental investigation exhibited distinct temporal evolutions of contact forces, with the 15° helical configuration identified as the optimal design. Further numerical analysis using a finite element model (validated with a deviation < 10%) is conducted to simulate performance under diverse impact angles in order to validate the reasonability of the experimental investigation. Mechanistically, 3DPBLs enhance impact resistance by increasing fracture tortuosity through their periodically rotated layered structure. These findings establish a theoretical foundation for developing high-performance, lightweight, and toughened protective materials. Full article

Resilient wheels are widely employed in metro vehicles to mitigate vibration and noise, in which rubber damping components play a critical role in load transmission and fatigue resistance. However, stress concentration and cyclic loading can significantly compromise their durability and service life. In this study, the structural optimization and fatigue life of rubber damping components in resilient wheels are systematically investigated based on finite element analysis and in-service metro operational data. A three-dimensional finite element model incorporating hyperelastic material behavior is developed to evaluate stress distributions under three representative conditions: press-fit assembly, straight-line operation, and curved-track operation. Based on the resulting stress fields, critical high-stress regions within the rubber component are identified and selected as targets for structural optimization. The Design of Experiments (DOE) methodology, integrated with the Isight 2022 optimization platform, is employed to determine the optimal geometric parameters that minimize the von Mises equivalent stress. Furthermore, a fatigue life prediction framework is established using actual metro service mileage data. Fatigue performance is assessed using Fe-safe 2022 software in conjunction with rubber fatigue crack propagation theory, and the results before and after optimization are systematically compared. This study demonstrates that stress concentrations in resilient wheel rubber damping components predominantly occur at fillet transition regions, governed by load transfer characteristics under press-fitting and service conditions. Through DOE-based structural optimization, the critical geometric parameters are effectively refined, leading to a significant reduction in stress levels in key regions. As a result, the proposed approach markedly improves fatigue performance, extending the minimum fatigue life from 1300 days to 24,322 days, thereby substantially enhancing the durability and reliability of the resilient wheel system. Full article

Accurate, timely, and resource-efficient decision-making is critical for sustainable precision agriculture. This paper proposes an agentic AI-based Internet of Things (IoT) framework that enables coordinated, closed-loop perception–decision–action processes across heterogeneous sensing and actuation components. The framework models agricultural systems as distributed collections of goal-driven agents responsible for multimodal sensing, uncertainty-aware reasoning, and adaptive decision-making. To provide a structured foundation, the proposed architecture is formalized within a Multi-Agent Partially Observable Markov Decision Process (MPOMDP) perspective, enabling systematic treatment of coordination, uncertainty, and decision policies. The framework integrates multimodal information sources, including vision-based perception and environmental sensing, and defines mechanisms for their fusion and use in system-level decision-making. A proof-of-concept instantiation is presented using publicly available datasets, combining visual perception models and tabular reasoning models within the proposed agentic workflow. The experiments are designed to demonstrate the feasibility, modularity, and coordination capabilities of the framework, rather than to benchmark predictive performance or provide field-validated evaluation. The results illustrate how multimodal information can be integrated to support adaptive and resource-aware decision processes. Finally, the paper discusses key challenges and outlines directions for future work, including real-world deployment, integration with physical actuation systems, and validation under operational conditions. Full article

Modeling and forecasting nonstationary and nonlinear economic time series remain fundamentally challenging due to structural breaks, volatility clustering, and noise contamination that distort the intrinsic stochastic structure. To address these limitations, this study proposes a novel three-stage hybrid statistical framework that systematically integrates multi-level signal decomposition with structured parametric modeling to enhance predictive accuracy. The proposed hybrid architectures—EMD–EEMD–ARIMA, EMD–EEMD–GMDH, and EMD–EEMD–ETS—employ a hierarchical decomposition–reconstruction strategy before forecasting. In the first stage, Empirical Mode Decomposition (EMD) decomposes the observed series into intrinsic mode functions (IMFs) and a residual component. In the second stage, Ensemble Empirical Mode Decomposition (EEMD) is applied to further refine the extracted components, mitigating mode mixing and improving signal separability. In the final stage, each reconstructed component is modeled using ARIMA, Exponential Smoothing State Space (ETS), and Group Method of Data Handling (GMDH) frameworks, and the individual forecasts are aggregated to obtain the final prediction. Empirical evaluation based on a recursive one-step-ahead forecasting scheme demonstrates consistent numerical improvements across all standard accuracy measures. In particular, the proposed EMD–EEMD–ARIMA model achieves the lowest forecasting error, reducing the root-mean-square error (RMSE) by approximately 6–7% relative to the best-performing single-stage model and by about 3–4% relative to the two-stage EMD-based hybrids. Similar improvements are observed in mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), indicating enhanced stability and robustness of the three-stage architecture. The results provide strong numerical evidence that multi-level decomposition combined with structured statistical modeling yields superior predictive performance for complex nonlinear and nonstationary time series. The proposed framework offers a mathematically coherent, computationally tractable, and systematically structured hybrid modeling strategy that effectively integrates noise-assisted decomposition with parametric and data-driven forecasting techniques. Full article

Cardiac point-of-care ultrasound (cPOCUS) is used to obtain key information about the heart’s structure and function when an echocardiogram is not available. This prospective, cross-sectional study aimed to compare fractional shortening (FS%) and left-atrium-to-aorta ratio (LA:Ao) obtained by cPOCUS in different body positions to echocardiography. Thirty-nine dogs had cPOCUS performed in three different body positions: left lateral recumbency (RT), right lateral recumbency (LT), and standing or sternal recumbency (RST). The cPOCUS values from each body position for FS% and LA:Ao were tested for agreement with the echocardiogram by Bland-Altman plots, correct clinical assessment by generalized estimated equation models, and quality score of the cPOCUS images as a percentage. Bland-Altman analysis showed a positive bias for FS% (0.9% to 9.8%) and both positive and negative bias for LA:Ao (within 0.2) for the cPOCUS values. The correct clinical assessment was made in the majority of cases for FS% in RST (67%) and RT (67%), and for LA:Ao in RST (55%). The clinical assessment was more often correct when the image quality score was higher. Intraclass correlation showed good agreement (≥0.61) between investigators for FS% in all body positions and LA:Ao in RT and RST. This study showed that cPOCUS performed from the right hemithorax can provide estimates and correct clinical assessment of FS% and LA:Ao. Obtaining measurements in the LT position is not recommended. Full article

This study presents a comparative analysis of five gender equality certification schemes alongside the ISO 53800 standard with the aim of distilling shared conceptual foundations and design principles that can inform progress toward Sustainable Development Goal (SDG) 5 on gender equality. The comparative analysis reveals marked heterogeneity in scope, design architecture, indicators, and transparency. Methodologically, the study draws on the relevant literature, documentary evidence, and semi-structured consultations with five experts in gender equality, diversity management, auditing, and ESG reporting. Building on the most effective and robust features across gender equality schemes, the study proposes a universal framework for gender equality certification. Under this framework, an ideal universal certification model should apply the same core requirements to both public and private organizations, while including simplified procedures tailored to small- and medium-sized enterprises (SMEs). Moreover, the model should rely on a limited set of key performance indicators (KPIs), focusing on the most material dimensions and prioritizing quantitative measures. It should also strengthen employee feedback mechanisms and enhance accountability in corporate governance. The framework should also pay attention to intersectional dimensions, extend responsibility across the value chain, and address the gender-related implications of artificial intelligence (AI). Importantly, an ideal universal gender equality certification should ensure a high level of transparency through the public disclosure of certified organizations, assessment criteria, KPIs, and levels or scores achieved. Furthermore, it should be supported by a free digital self-assessment tool and robust auditing arrangements, underpinned by a sufficiently large pool of accredited certification bodies and gender-balanced audit teams. Finally, it should undergo periodic review and align with Environmental, Social, and Governance (ESG) principles and other related SDGs. Full article

Background/Objectives: Resin 3D printing is widely used to fabricate orthodontic diagnostic models, but the practical performance of commercial desktop workflows compared to dental-certified workflows is still debated. This study compared the dimensional accuracy and 7-day stability of maxillary orthodontic models printed from the same master STL file using a dental-certified workflow versus two commercial desktop workflows. Methods: An ISO 20896-1:2019-based reference cast with four 6 mm calibration spheres was used to generate a master STL file. Fifteen models were printed (n = 5 per workflow) using Primeprint™ (dental-certified workflow) and two commercial desktop printers (Anycubic Photon Mono M5s; Phrozen Sonic Mighty 14K REVO). The models were digitized at baseline (T0, ≤48 h) and after 7 days (T7) using a laboratory scanner. Surface superimposition in CloudCompare^® calculated the RMS (root mean square) surface deviation and mean signed deviation, and two calibrated operators performed independent extractions. Results: The mean RMS deviations were <0.10 mm for all workflows at both time points. No between-workflow differences were detected at T0 (H = 2.000; p = 0.368) or T7 (H = 1.520; p = 0.468), no within-workflow T0–T7 changes were significant (all p > 0.05), and the inter-operator agreement was excellent (ICC 0.991–0.999). Conclusions: Under the tested workflows, dental-certified and commercial desktop resin printing produced orthodontic models with a comparable global surface accuracy and short-term dimensional stability. Full article

Vision–language models (VLMs), such as Contrastive Language–Image Pretraining (CLIP), are increasingly deployed in real-world applications, including content moderation, misinformation detection, and fraud analysis, making their robustness to adversarial attacks a critical concern. While adversarial robustness has been widely studied in unimodal models, modality-specific vulnerabilities in multimodal models remain underexplored. In this work, we analyze CLIP by applying gradient-based adversarial attacks to its vision and language modalities, both independently and jointly, and evaluating performance on two multimodal classification benchmarks: the Facebook Hateful Memes dataset and a large-scale Suspicious Car Parts dataset. Using Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks along with multiple adversarial retraining strategies, we show that adversarial perturbations on the image modality consistently cause the most severe and unstable performance degradation. These results demonstrate that the vision modality is the primary vulnerability in CLIP, highlighting the need for modality-specific defense strategies that focus more on the weaker modality in multimodal systems. Full article

Earthquakes remain one of the greatest threats to urban resilience, demanding innovative strategies that go beyond traditional earthquake-resistant design. Among emerging solutions, viscous fluid dampers stand out as one of the most effective mechanisms for controlling structural responses and reducing damage. This research analyzes the seismic performance of a 12-story multifamily building equipped with viscous fluid dampers, developed using a comprehensive Building Information Modeling (BIM) methodology. The architectural model was integrated into a BIM environment, ensuring precision, coordination, and digital consistency. A time-history analysis was conducted in ETABS comparing two configurations—with and without dampers—subjected to seismic records from Lima-Perú, Ica-Perú, and Tarapacá-Chile. The results show that incorporating dampers significantly improves structural behavior, reducing maximum displacements by 52.25% and inter-story drifts by 47.37%. These findings confirm the ability of dampers to effectively dissipate seismic energy. Likewise, BIM integration establishes a robust digital framework for sustainable, coordinated, and resilient seismic design in high-rise buildings. Full article

The early-life rumen microbiome is highly dynamic, shaped by dietary transitions and maternal influences. Several dietary additives have been studied during the pre- and post-weaning periods to improve animal welfare, growth performance, and farming efficiencies. This study investigated microbial community assembly and growth performance of lambs provided with a mannan-rich fraction (MRF) supplement, either through maternal supplementation, directly, or via a combination of both. Using metagenomic sequencing and gas chromatography, we found differences in rumen microbial alpha and beta diversity related to both sampling time point and MRF supplementation (p < 0.05). At week 8, lamb microbiomes showed greater variance in their Shannon alpha diversity, with direct MRF supplementation only to the lamb resulting in a significantly greater diversity (p < 0.05). At week 20, combined maternal and lamb supplementation resulted in the highest Shannon diversity and was different compared to all other groups (p < 0.05). Beta diversity analyses combined with differential abundance analyses revealed that microbial community structures are driven by both diet and time, with maternal MRF supplementation associated with enrichment of taxa involved in carbohydrate fermentation and succinate metabolism, including Succiniclasticum ruminis, Succinovibrio dextrinosolvens, and Fibrobacter succinogenes. Generalized linear modeling identified significant associations between microbial alpha diversity metrics and total volatile fatty acids in lambs, particularly butyrate and valerate. Furthermore, at week 8, there was a significant positive correlation between alpha diversity metrics and propionate and valerate. In this study, lambs receiving MRF through maternal and direct supplementation had the highest growth performance, measured as the median average daily gains (kg) and final weights (kg) of lambs. These findings suggest that MRF supplementation, especially when provided both maternally and directly, may influence the lamb rumen microbiome and alter its metabolic potential with potential implications for optimizing early-life nutrition strategies in ruminant production systems. Full article

Archaeological reports show that about 70% of Han dynasty pictorial stone sites feature historical figures, revealing a significant yet understudied aspect of tomb ritual practice (muji yishi). This study examines how these depictions may reflect ritual characteristics and their relationship to temple ritual practice (miaoji yishi). From the Qin to Han period (221 BCE–220 CE), tomb and temple rituals increasingly converged; temple rituals were sometimes performed by tombs, and the imagery incorporated cosmological models alongside representations of daily life, including clothing, diet, dwellings, and mobility. The historical figures depicted can be grouped into three categories: emperors and sages, loyal ministers and righteous heroes, and filial sons and chaste women. These figures were closely associated with ideals of transcendence and immortality, suggesting a ritual framework that connected temple and tomb practices, with emperors and sages appearing most frequently, accounting for about 80% of the depictions. Notably, these images occur predominantly in commoners’ tombs (approximately 95%), where fewer social restrictions may have allowed greater creative freedom. While research on tomb ritual practices has traditionally relied on textual sources, the present study emphasizes archaeological evidence, offering an analytical perspective on the relationship between temple and tomb rituals in Han funeral art and highlighting their potential role in shaping Han ritual logic and religious expression. Full article

Eucalyptus plantations are expanding rapidly in southern China, delivering economic benefits but also posing ecological risks, which creates a pressing need for precise, municipal-scale monitoring. Mapping eucalyptus with sub-meter resolution imagery, however, is confronted by two main challenges: (1) the pronounced multi-scale heterogeneity of fragmented stands, and (2) the difficulty in achieving precise boundary delineation due to shadowed and complex canopy edges. To address these, this study makes two primary contributions. First, we present the Eucalyptus Semantic Segmentation Dataset (ESSD)—a high-quality, pixel-level annotated dataset that includes geographic coordinates to support reproducible research. Second, we propose SDCNet, a task-specific deep learning network optimized for eucalyptus mapping. SDCNet incorporates a redesigned SD-ASPP module that leverages Deep Over-parameterized Convolution (DO-Conv) to capture multi-scale features, alongside a novel Coordinated Self-Attention Mechanism (CSAM) to enhance the accuracy of canopy boundary detection. Ablation studies confirm the effectiveness of each component. In benchmark tests against seven state-of-the-art semantic segmentation models, SDCNet achieves superior performance, obtaining a per-class Intersection over Union (IoU) of 88.83% and an F1-score of 93.81% for eucalyptus—an improvement of +2.24% in IoU and +1.71% in F1-score over the strongest baseline. Applied to Nanning City, SDCNet produces the first 0.3 m resolution eucalyptus distribution map for the region. This map reveals a critical finding: within the watershed of the Xiyunjiang Reservoir—Nanning’s primary drinking water source—eucalyptus plantations cover more than 50% of the forested area. This result provides the first quantitative, high-resolution evidence of potential hydrological risk at a municipal scale. Our work establishes an integrated framework that bridges advanced remote sensing with actionable forest governance, offering scientifically grounded support for ecological risk assessment and sustainable land-use policy. Full article

Objective: To evaluate clinical factors associated with termination of pregnancy (TOP) recommendations following prenatal diagnosis of congenital heart disease (CHD) within a multidisciplinary fetal council model. Methods: This retrospective cohort study included 146 fetuses with prenatally diagnosed CHD discussed in a tertiary referral center fetal council between October 2023 and December 2025. The primary outcome was council-issued recommendation for TOP (yes/no). Variables included gestational age (GA) at diagnosis, cardiac severity (Davey scale grouped as low, moderate, high), extracardiac anomalies, fetal growth restriction (FGR), and genetic evaluation/results. Group comparisons were performed using Mann–Whitney U and χ² tests. Independent associations were assessed using binary logistic regression. A subgroup analysis was conducted in isolated CHD cases (no extracardiac structural anomalies). Results: A total of 146 fetuses with prenatal CHD were included in the analysis. TOP was recommended in 71 cases (48.6%). GA at diagnosis did not differ between groups when analyzed continuously; however, categorical GA showed significant differences, with earlier diagnoses more frequent among TOP-recommended cases. Cardiac severity distribution differed significantly between groups. In multivariable analysis, GA at diagnosis and cardiac severity were independently associated with TOP recommendation. Compared with <20 weeks, diagnosis at 20–23+6 weeks (OR 17.96, 95% CI 3.50–92.22) and ≥24 weeks (OR 3.92, 95% CI 1.53–10.06) increased the odds of TOP recommendation. Relative to high severity, moderate (OR 0.23, 95% CI 0.07–0.72) and low severity (OR 0.20, 95% CI 0.08–0.50) were associated with lower odds of TOP recommendation. Extracardiac anomalies, genetic findings, and FGR were not independently associated after adjustment. Similar patterns were observed in isolated CHD cases. Conclusions: In a multidisciplinary prenatal counseling setting, TOP recommendations after prenatal CHD diagnosis were primarily driven by cardiac severity and GA at diagnosis, rather than extracardiac or genetic findings alone. Full article