Search Results (20,897)

Liver fibrosis (LF) poses significant challenges in diagnosis and treatment. This study aimed to identify effective biomarkers for diagnosis and therapy, as well as to gain deeper insights into the immunological features associated with LF. LF-related datasets were retrieved from the Gene Expression Omnibus (GEO) database. Two datasets were merged to generate a metadata cohort for bioinformatics analysis and machine learning, while another dataset was reserved for external validation. Seventy-eight machine learning algorithms were employed to screen signature genes. The diagnostic performance of these genes was evaluated using receiver operating characteristic (ROC) curves, and their expression levels were validated via qRT-PCR experiments. The R language was utilized to delineate the immune landscape. Finally, correlation analysis was conducted to investigate the relationship between the signature genes and immune infiltration. Through the intersection of GEO datasets and Weighted Gene Co-expression Network Analysis (WGCNA), 42 genes were identified. Machine learning methods further narrowed down 13 signature genes (alpha-2-macroglobulin (A2M), ankyrin-3 (ANK3), complement component 7 (C7), cadherin 6 (CDH6), cysteine-rich motor neuron protein 1 (CRIM1), dihydropyrimidinase-like 3 (DPYSL3), F3, gamma-aminobutyric acid (GABA) receptor subunit epsilon (GABRE), membrane metalloendopeptidase (MME), solute carrier family 38 member 1 (SLC38A1), tropomyosin alpha-1 chain (TPM1), von Willebrand factor (VWF), and zinc finger protein 83 (ZNF83)), and qRT-PCR confirmed these genes’ expression patterns. Furthermore, these signature genes demonstrated strong correlations with multiple immune cell populations. In conclusion, the 13 genes (A2M, ANK3, C7, CDH6, CRIM1, DPYSL3, F3, GABRE, MME, SLC38A1, TPM1, VWF, and ZNF83) represent robust potential biomarkers for the diagnosis and treatment of LF. Among these genes, we first identified Gabre as related to LF and expressed in hepatocytes and cholangiocytes. The immune response mediated by these signature biomarkers plays a pivotal role in the pathogenesis and progression of LF through dynamic interactions between the biomarkers and immune-infiltrating cells. Full article

Technical losses (TLs) in power systems are an inevitable outcome of energy dissipation in components such as conductors, transformers, and transmission lines. These losses arise from the combined effects of material properties, operational conditions, and environmental factors, creating ongoing challenges for energy efficiency and grid sustainability. Their reduction requires a coordinated approach that integrates material improvements, smart grid technologies, and optimized operational practices. Reducing TLs not only improves economic efficiency but also contributes significantly to global sustainability efforts by enabling more efficient energy use and reducing carbon emissions associated with power generation. A review of recent publications shows that the literature on network losses is heavily skewed toward non-technical losses (NTLs), with TL-focused studies being fewer, often dated, and lacking comprehensive scope. This paper addresses the existing research gap by presenting a comprehensive, section-oriented taxonomy of TL mechanisms in power systems, accompanied by precise definitions for each category and a direct linkage between these categories and applicable loss mitigation measures. In addition, selected real-world projects and global initiatives aimed at reducing TLs, together with current regulatory approaches, emerging trends in this domain, and an assessment of the maturity level of technologies employed for TL reduction, are analyzed. This study aims to serve as a scientific reference to support future research and to guide policymakers, regulators, and utilities in developing more effective strategies for minimizing TLs. Full article

Achieving the full potential of wind energy in the global renewable transition depends critically on enhancing the performance and reliability of polymer composite components. This review synthesizes recent advances from 2022 to 2025, including the development of next-generation hybrid composites and the application of high-fidelity computational methods—finite element analysis (FEA), computational fluid dynamics (CFD), and fluid–structure interaction (FSI)—to optimize structural integrity and aerodynamic performance. It also explores the transformative role of artificial intelligence (AI) in structural health monitoring (SHM) and the integration of Internet of Things (IoT) systems, which are becoming essential for predictive maintenance and lifecycle management. Special focus is given to harsh offshore environments, where polymer composites must withstand extreme wind and wave conditions. This review further addresses the growing importance of circular economy strategies for managing end-of-life composite blades. While innovations such as the geometric redesign of floating platforms and the aerodynamic refinement of blade components have yielded substantial gains—achieving up to a 30% mass reduction in PLA prototypes—more conservative optimizations of internal geometry configurations in GFRP blades provide only around 7% mass reduction. Nevertheless, persistent challenges related to polymer composite degradation and fatigue under severe weather conditions are driving the adoption of real-time hybrid predictive models. A bibliometric analysis of over 1000 publications confirms more than 25 percent annual growth in research across these interconnected areas. This review serves as a comprehensive reference for engineers and researchers, identifying three strategic frontiers that will shape the future of wind turbine blade technology: advanced composite materials, integrated computational modeling, and scalable recycling solutions. Full article

Oxygenated volatile organic compounds (OVOCs) are key precursors to atmospheric ozone (O₃) and secondary organic aerosols (SOA). However, research on the sources of OVOCs is still limited, particularly in terms of multi-point observations at urban sites. This study conducted a one month continuous enhanced observation at an urban site (BA) and a suburban site (DP) in December 2024. During the study period, the average total VOCs concentration at the BA site was 29.9 ± 6.5 ppbv, significantly higher than that at the DP site (6.4 ± 1.3 ppbv). To enhance the representation of the biogenic fraction in OVOCs, isoprene was employed as a biogenic tracer; prior to source apportionment, its anthropogenic components were subtracted based on local emission ratio coefficients, thereby providing a more representative basis for biogenic source attribution. The optimized source apportionment results show that the contribution ratio of biogenic sources had decreased significantly, with a particularly noticeable decline at the urban site. Among these, the contribution rates of acetaldehyde and acetone had decreased significantly: by 14.7% and 12.2%, respectively. Based on the improved source apportionment method, the source apportionment of OVOCs at the urban site showed that methanol, acetone, and MEK were primarily dominated by anthropogenic primary sources (accounting for 44.5% to 68.5%), while acetaldehyde was primarily dominated by secondary anthropogenic generation (37.1%), indicating its key role as a photochemical product. In contrast, at the suburban site, the biogenic source contribution to acetaldehyde (37.8%) was significant. This difference highlights the necessity of optimizing biogenic source tracers and conducting OVOC source apportionment studies at multiple locations. Full article

Wood-derived ceramics represent a novel class of bio-based composite materials that integrate the hierarchical porous architecture of natural wood with high-performance ceramic phases such as silicon carbide (SiC). This review systematically summarizes recent advances in the fabrication of SiC woodceramics via two predominant sintering routes—reactive infiltration sintering and hot-press sintering—and elucidates their effects on the resulting microstructure and mechanical properties. This review leverages the intrinsic anisotropic vascular network and multiscale porosity and mechanical strength, achieving ultralightweight yet mechanically robust ceramics with tunable anisotropy and dynamic energy dissipation capabilities. Critical process–structure–property relationships are highlighted, including the role of ceramic reinforcement phases, interfacial engineering, and multiscale toughening mechanisms. The review further explores emerging applications spanning extreme protection (e.g., ballistic armor and aerospace thermal shields), multifunctional devices (such as electromagnetic shielding and tribological components), and architectural innovations including seismic-resistant composites and energy-efficient building materials. Finally, key challenges such as sintering-induced deformation, interfacial bonding limitations, and scalability are discussed alongside future prospects involving low-temperature sintering, nanoscale interface reinforcement, and additive manufacturing. This mini overview provides essential insights into the design and optimization of wood-derived ceramics, advancing their transition from sustainable biomimetic materials to next-generation high-performance structural components. This review synthesizes data from over 50 recent studies (2011–2025) indexed in Scopus and Web of Science, highlighting three key advancements: (1) bio-templated anisotropy breaking the porosity–strength trade-off, (2) reactive vs. hot-press sintering mechanisms, and (3) multifunctional applications in extreme environments. Full article

The nurse rostering problem (NRP) has attracted significant research interest in recent decades due to both its practical relevance and computational complexity. While the branch-and-price algorithm has demonstrated effectiveness in solving NRPs, its column generation component frequently produces weak lower bounds for some problem instances, which consequently degrades overall computational performance. To strengthen the lower bound quality, we propose three classes of cutting planes derived from the column generation master problem formulation: SRCs, CG rank-1 cuts, and {0, ½}-cuts. For each cut type, the separation approaches enhanced with acceleration strategies are described. These cuts are typically classified as non-robust, meaning each cut added to the master problem requires introducing a new resource in the pricing subproblem’s labeling algorithm. We therefore developed problem-specific methods to update these resources and integrate them into the NRP dominance rules. Computational experiments were conducted on benchmark instances from two international nurse rostering competitions (INRC-I and INRC-II). The results indicate that SRCs are highly effective for two challenging INRC-I instances, including one where a tighter lower bound was identified. In contrast, the {0, ½}-cuts yield the strongest performance for most selected INRC-II instances. These findings demonstrate that the cutting plane method can be used to improve lower bounds for NRPs, and that the effectiveness of different cut types in improving lower bounds is closely tied to the problem formulation. Full article

This paper analyzes the seismic performance and damage characteristics of high-rise frame-core-tube structures on soft soil, explicitly incorporating dynamic soil–pile–structure interaction (SSI). A refined 3D finite element model of a 52-storey soil–pile–structure system was developed in ABAQUS, utilizing viscous-spring boundaries and the equivalent nodal force method for seismic input. Nonlinear analyses under six seismic waves were compared to a fixed-base model neglecting SSI. Key findings demonstrate that SSI significantly alters structural response; it amplifies lateral displacements and inter-storey drift ratios throughout the structure, particularly at the top level. While total base shear decreased, frame column base shear forces substantially increased. SSI also reduced peak top-storey accelerations, diminished short-period spectral components, and prolonged the predominant period of response spectra. Analysis of member damage revealed SSI generally reduced compressive and tensile damage in core walls, floor slabs, and frame beams. Principal compressive stresses at the base of frame columns increased under SSI. These results highlight the necessity of including dynamic SSI in seismic analysis for high-rises on soft soil, specifically due to its detrimental amplification of forces in frame columns. Full article

This work introduces a comprehensive vision-based framework for autonomous space debris removal using robotic manipulators. A real-time debris detection module is built upon the YOLOv8 architecture, ensuring reliable target localization under varying illumination and occlusion conditions. Following detection, object motion states are estimated through a calibrated binocular vision system coupled with a physics-based collision model. Smooth interception trajectories are generated via a particle swarm optimization strategy integrated with a 5–5–5 polynomial interpolation scheme, enabling continuous and time-optimal end-effector motions. To anticipate future arm movements, a Transformer-based sequence predictor is enhanced by replacing conventional multilayer perceptrons with Kolmogorov–Arnold networks (KANs), improving both parameter efficiency and interpretability. In practice, the Transformer+KAN model compensates the manipulator’s trajectory planner to adapt to more complex scenarios. Each component is then evaluated separately in simulation, demonstrating stable tracking performance, precise trajectory execution, and robust motion prediction for intelligent on-orbit servicing. Full article

Domain shift poses a major challenge in many Machine Learning applications due to variations in data acquisition protocols, particularly in the medical field. Test-time augmentation (TTA) can solve the domain shift issue and improve robustness by aggregating predictions from multiple augmented versions of the same input. However, TTA may inadvertently generate unrealistic or Out-of-Distribution (OOD) samples that negatively affect prediction quality. In this work, we introduce a filtering procedure that removes from the TTA images all the OOD samples whose representations lie far from the training data distribution. Moreover, all the retained TTA images are weighted inversely to their distance from the training data. The final prediction is provided by a Self-Ensemble with Confidence, which is a lightweight ensemble strategy that fuses predictions from the original and retained TTA samples using a weighted soft voting scheme, without requiring multiple models or retraining. This method is model-agnostic and can be integrated with any deep learning architecture, making it broadly applicable across various domains. Experiments on cross-domain leukocyte classification benchmarks demonstrate that our method consistently improves over standard TTA and Baseline inference, particularly when strong domain shifts are present. Ablation studies and statistical tests confirm the effectiveness and significance of each component. Full article

Deep learning has significantly advanced medical image analysis by enabling accurate, automated diagnosis across diverse clinical tasks such as lesion classification and disease detection. However, the practical deployment of these systems is still hindered by two major challenges: the limited availability of expert-annotated data and substantial domain shifts caused by variations in imaging devices, acquisition protocols, and patient populations. Although recent semi-supervised domain generalization (SSDG) approaches attempt to address these challenges, they often suffer from two key limitations: (i) reliance on computationally expensive uncertainty modeling techniques such as Monte Carlo dropout, and (ii) inflexible shared-head classifiers that fail to capture domain-specific variability across heterogeneous imaging styles. To overcome these limitations, we propose MultiStyle-SSDG, a unified semi-supervised domain generalization framework designed to improve model generalization in low-label scenarios. Our method introduces a multi-style ensemble pseudo-labeling strategy guided by entropy-based filtering, incorporates prototype-based conformity and semantic alignment to regularize the feature space, and employs a domain-specific multi-head classifier fused through attention-weighted prediction. Additionally, we introduce a dual-level neural-style transfer pipeline that simulates realistic domain shifts while preserving diagnostic semantics. We validated our framework on the ISIC2019 skin lesion classification benchmark using 5% and 10% labeled data. MultiStyle-SSDG consistently outperformed recent state-of-the-art methods such as FixMatch, StyleMatch, and UPLM, achieving statistically significant improvements in classification accuracy under simulated domain shifts including style, background, and corruption. Specifically, our method achieved 78.6% accuracy with 5% labeled data and 80.3% with 10% labeled data on ISIC2019, surpassing FixMatch by 4.9–5.3 percentage points and UPLM by 2.1–2.4 points. Ablation studies further confirmed the individual contributions of each component, and t-SNE visualizations illustrate enhanced intra-class compactness and cross-domain feature consistency. These results demonstrate that our style-aware, modular framework offers a robust and scalable solution for generalizable computer-aided diagnosis in real-world medical imaging settings. Full article

This paper presents a hybrid diagnostic framework that integrates Knowledge Graphs (KGs) with Large Language Models (LLMs) to support fault diagnosis in complex, high-reliability systems such as nuclear power plants. The framework is based on the Dynamic Master Logic (DML) model, which organizes system functions, components, and dependencies into a hierarchical KG for logic-based reasoning. LLMs act as high-level facilitators by automating the extraction of DML logic from unstructured technical documentation, linking functional models with language-based reasoning, and interpreting user queries in natural language. For diagnostic queries, the LLM agent selects and invokes predefined tools that perform upward or downward propagation in the KG using DML logic, while explanatory queries retrieve and contextualize relevant KG segments to generate user-friendly interpretations. This ensures that reasoning remains transparent and grounded in the system structure. This approach reduces the manual effort needed to construct functional models and enables natural language queries to deliver diagnostic insights. In a case study on an auxiliary feedwater system used in the nuclear pressurized water reactors, the framework achieved over 90 percent accuracy in model element extraction and consistently interpreted both diagnostic and explanatory queries. The results validate the effectiveness of LLMs in automating model construction and delivering explainable AI-assisted health monitoring. Full article

The 1,3-dipolar cycloaddition reaction was applied to 9α-hydroxyparthenolide, an important sesquiterpene component of Anvillea radiata that was extracted directly from plant material collected in Morocco. Several new spiro-isoxazolidine derivatives were generated on the B-ring of 9α-hydroxyparthenolide (α-methylene-γ-butyrolactone (1)) by 1,3-dipolar cycloaddition of its exocyclic double bond with various nitrones. These compounds were fully characterized by spectroscopic methods. Full article

With the rapid expansion of variable renewable energy, coal-fired units are increasingly operated at low load, where non-convex cost characteristics pose challenges for spot market clearing. This study reviews and improves existing low-load generation cost models, introducing three key enhancements: (1) integrating piecewise linearization with the marginal cost approach to reduce computational burden; (2) removing redundant binary variables and incorporating previously omitted cost components to improve clearing efficiency; and (3) developing a fuel cost model that combines quasi-fixed and marginal costs for low-load generation with firing and combustion support (FCS), enabling the joint optimization of low-load and normal operations. Applied to 6-bus and provincial systems, the proposed approach achieves speed-ups of 11.3× and 6.3× over the benchmark model (Model I) while maintaining accuracy, demonstrating both its efficiency and practical applicability. Full article

Task planning for a construction robot requires systematically integrating diverse elements, such as building components, construction processes, user input, and robot software. Conventional robot programming complicates this by requiring precise entity naming, relationship definitions, unstructured language interpretation, and accurate action selection. Existing research has focused on isolated components, such as natural language processing, hardcoded data linkages, or BIM data extraction. We introduce a novel framework using an LLM as the cognitive core for autonomous construction robots, encompassing both data preparation and task planning phases. Leveraging OpenAI’s ChatGPT-4, we demonstrate how LLMs can process structured BIM data and unstructured human inputs to generate robot instructions. A prototype tested in a simulated environment with a mobile painting robot adaptively executed tasks through real-time dialogues with ChatGPT-4, reducing reliance on hardcoded logic. Results suggest that LLMs can serve as the cognitive core for construction robots, with potential for extension to more complex operations. Full article

In confronting escalating economic uncertainty, achieving a win–win situation for low-carbon transition and improved employment structure will contribute to economic recovery and sustainable growth but also contribute to building a community with a shared future for mankind. A critical issue for China’s economy and societal welfare, as well as a core component of sustainable development, concerns whether low-carbon economic transition influences employment skill structure. This study utilizes data from 30 provinces (municipalities and autonomous regions) in China from 2006 to 2021. Employing the entropy method, a low-carbon economic development level indicator system was constructed from four aspects: low-carbon output, low-carbon consumption, low-carbon resources, and low-carbon environment to measure the low-carbon economy and explore its direct and indirect effects on employment skill structure and spatial effects. The research findings indicate that low-carbon economies not only directly and significantly promote employment skill structure optimization but also indirectly generate promotional effects through pathways such as industrial structure adjustment, green innovation’s innovative effects, and factor substitution effects of increased pollution control investment. Among these, the indirect impact of industrial structure adjustment contributes most substantially. Low-carbon economies’ influence on employment skill structures exhibits spatial spillover effects, with neighboring regions’ low-carbon economies exerting positive spillover effects on local skill structures. Additionally, significant negative interdependence exists among regional employment skill structures. Based on the aforementioned research conclusions, the following recommendations are proposed: accelerate low-carbon economy development and employment skill structure enhancement in central and western regions to diminish regional disparities; encourage green innovation and promote traditional industry upgrading and transformation; formulate regional coordinated development plans, thereby strengthening the low-carbon economy’s optimizing role upon employment skills structure; and increase educational investment and strengthen labor skill training. Full article