Search Results (2,148)

Traffic systems are quintessential complex systems, characterized by nonlinear interactions, multiscale dynamics, and emergent spatiotemporal patterns over complex networks. These properties make traffic prediction highly challenging, as it requires jointly modeling stable global topology and time-varying local dependencies. Existing graph neural networks often rely on predefined or static learnable graphs, overlooking hidden dynamic structures, while most RNN- or CNN-based approaches struggle with long-range temporal dependencies. This paper proposes a Spatiotemporal Causal–Trend Network (SCTN) tailored to complex transportation networks. First, we introduce a dual-path adaptive graph learning scheme: a static graph that captures global, topology-aligned dependencies of the complex network, and a dynamic graph that adapts to localized, time-varying interactions. Second, we design a Gated Temporal Attention Module (GTAM) with a causal–trend attention mechanism that integrates 1D and causal convolutions to reinforce temporal causality and local trend awareness while maintaining long-range attention. Extensive experiments on two real-world PeMS traffic flow datasets demonstrate that SCTN consistently achieves superior accuracy compared to strong baselines, reducing by 3.5–4.5% over the best-performing existing methods, highlighting its effectiveness for modeling the intrinsic complexity of urban traffic systems. Full article

Dopamine (DA) is a critical catecholaminergic neurotransmitter that facilitates signal transduction across synaptic junctions and modulates essential neurophysiological processes, including motor coordination, motivational drive, and reward-motivated behaviors. The fabrication of cost-effective, miniaturized, and high-fidelity analytical platforms is imperative for real-time DA monitoring. Due to its inherent electrochemical activity, carbon-based amperometric sensors constitute the primary modality for DA quantification. In this study, graphite, multi-walled carbon nanotubes (MWCNTs), and graphene were immobilized within an ethyl 2-cyanoacrylate (ECA) polymer matrix. ECA was selected for its rapid polymerization kinetics and established biocompatibility in electrochemical frameworks. All fabricated composites demonstrated robust electrocatalytic activity toward DA; however, MWCNT- and graphene-based sensors exhibited superior analytical performance, characterized by highly competitive limits of detection (LOD) and quantification (LOQ). Specifically, MWCNT-modified electrodes achieved an interesting LOD of 0.030 ± 0.001 µM and an LOQ of 0.101 ± 0.008 µM. Discrepancies in baseline current amplitudes suggest that the spatial orientation of carbonaceous nanomaterials within the cyanoacrylate matrix significantly influences the electrochemical surface area and resulting baseline characteristics. The impact of interfering species commonly found in biological environments on the sensors’ response was systematically evaluated. The best-performing sensor, the graphene-based one, was used to measure the DA intracellular content of PC12 cells. Full article

Recent years have witnessed performance improvements in Multimodal Large Language Models (MLLMs) on downstream natural image understanding tasks. However, when applied to the function graph reasoning task, which is highly information-dense and abundant in fine-grained structural details, these models face pronounced performance degradation. The challenges are primarily characterized by several core issues: the static projection bottleneck, inadequate cross-modal interaction, and insufficient visual context in text embeddings. To address these problems, this study proposes a multimodal feature fusion enhancement method for function graph reasoning and constructs the FuncFusion-Math model. The core innovation of this model resides in its design of a dual-path feature fusion mechanism for both image and text. Specifically, the image fusion module adopts cross-attention and self-attention mechanisms to optimize visual feature representations under the guidance of textual semantics, effectively mitigating fine-grained information loss. The text fusion module, through feature concatenation and Transformer encoding layers, deeply integrates structured mathematical information from the image into the textual embedding space, significantly reducing semantic deviation. Furthermore, this study utilizes a four-stage progressive training strategy and incorporates the LoRA technique for parameter-efficient optimization. Experimental results demonstrate that the FuncFusion-Math model, with 3B parameters, achieves an accuracy of 43.58% on the FunctionQA subset of the MathVista test set, outperforming a 7B-scale baseline model by 13.15%, which validates the feasibility and effectiveness of the proposed method. Full article

The selective hydrogenolysis of glucose into 1,2-propanediol (1,2-PG) constitutes a significant yet challenging transformation in biomass valorization, as it involves a highly coupled network of isomerization, C-C bond cleavage, and hydrogenation steps. Herein, a highly efficient Ni-CeO₂ catalyst supported by basic Al₂O₃ is developed via a urea-assisted precipitation strategy. Systematic catalytic evaluation and comprehensive characterization reveal that this synthesis method markedly enhances Ni dispersion and hydrogen activation capacity, while CeO₂ modification modulates the electronic state of Ni and introduces strong Lewis basic sites associated with oxygen vacancies. The synergistic interplay between Ni and CeO₂ effectively promotes glucose isomerization and retro-aldol condensation while maintaining sufficient hydrogenation activity. As a result, the optimized catalyst achieves a 1,2-PG yield of 45.1% with over 99% glucose conversion under optimal hydrothermal reaction conditions. Moreover, the catalyst exhibits relatively stable catalytic performance over four consecutive runs. This work elucidates key structure–activity relationships in multifunctional Ni-based catalysts and provides design principles for efficient biomass-derived polyol production. Full article

Cardiovascular disease (CVD) continues to be the primary cause of morbidity and mortality worldwide, underscoring the urgent need for novel therapeutic alternatives. In recent years, marine ecosystems have garnered increasing attention as a promising source of bioactive compounds with unique structural and pharmacological properties. Marine-derived toxins and venoms, including tetrodotoxin, ω-conotoxins, anthopleurins, palytoxin, brevetoxin, aplysiatoxin, and asterosaponins, exert cardioprotective effects through diverse mechanisms such as modulation of ion channels, inhibition of sympathetic overactivity, antioxidative actions, and enhancement of myocardial contractility. These properties make them potential candidates for addressing various CVD manifestations, including arrhythmia, hypertension, ischemia–reperfusion injury, and heart failure. However, despite their therapeutic promise, the clinical application of these marine compounds remains limited due to poor tissue selectivity, narrow therapeutic indices, proinflammatory activity, and limited metabolic stability. Structural modifications, advanced drug delivery platforms, and in vivo validation studies are crucial for overcoming these challenges. This review highlights the pharmacological actions, molecular targets, and cardiovascular relevance of selected marine toxins and venoms while also addressing key translational barriers. Advances in biotechnology and peptide engineering are enabling the safer and more targeted use of these compounds. Collectively, marine-derived toxins and venoms represent a largely untapped but highly promising frontier in cardiovascular drug discovery. Strategic research focused on elucidating mechanisms, optimizing delivery, and translating clinical applications will be critical to unlocking their full therapeutic potential. Full article

Despite extensive exploration of gelatin methacryloyl (GelMA)-based hydrogels for bone tissue engineering, their clinical translation is hindered by a critical trade-off: poor precursor stability leads to rapid sedimentation of bioactive fillers like hydroxyapatite (HAp), while formulations optimized for injectability often sacrifice mechanical integrity or handling precision. To overcome this challenge, we report a rheologically engineered, injectable composite hydrogel scaffold that integrates unmodified gelatin as a thermoresponsive viscosity modulator into a GelMA/HAp matrix. The incorporation of gelatin yields a stable, paste-like precursor at physiological temperature, which effectively prevents HAp sedimentation and enables precise, filamentous extrusion. Subsequent UV crosslinking locks the homogeneous structure in place, resulting in a mechanically robust scaffold with significantly enhanced compressive modulus. In vitro studies demonstrate that this biomimetic microenvironment not only supports high viability and proliferation of bone marrow stromal cells (BMSCs) but also potently enhances their osteogenic differentiation, as evidenced by upregulated alkaline phosphatase activity, Runx2 expression, and matrix mineralization. This simple, one-step strategy successfully reconciles injectability, structural fidelity, and bioactivity, offering a highly promising and clinically translatable platform for minimally invasive bone regeneration. Full article

Accurate grading of peach ripeness is essential for determining optimal harvest timing, postharvest storage potential, and market value. However, traditional methods are often inefficient and highly subjective. Meanwhile, existing deep learning models face challenges in balancing complexity with accuracy. To address this, this paper proposes a lightweight improved model named LightConvNeXt-FCS, which centers on a novel lightweight module named LightBlock. This module drastically reduces the parameter count and computational overhead. To preserve representational capacity, auxiliary structures—including attention enhancement, cross-stage connections, and multi-scale fusion—are incorporated. Experimental results show that the model requires only 2.75 M parameters and 624.23 M FLOPs, representing a 90.1% reduction in parameters and an 86.0% decrease in computational cost compared to ConvNeXt-Tiny, with the model size compressed to 9.9% of the original, while achieving an accuracy of 94.62%, slightly outperforming the original model. This approach effectively resolves the common trade-off between model complexity and accuracy. By achieving high accuracy with a lightweight architecture, it provides a more feasible solution for deploying rapid and intelligent fruit ripeness grading systems. Full article

Reciprocating compressors, which serve as core equipment in the petrochemical and natural gas transmission sectors, operate under prolonged variable loads and high-frequency impact conditions. Critical components, such as valves and piston rings, are prone to failure. Existing fault diagnosis methods suffer from inadequate spatio-temporal feature extraction and neglect spatio-temporal correlations. To address this, this paper proposes a spatio-temporal feature fusion-based fault diagnosis method for reciprocating compressors. This method constructs a spatio-temporal feature fusion model (STFFM) comprising three principal modules: First, a spatio-temporal feature extraction module employing a multi-layered stacked bidirectional gated recurrent unit (BiGRU) with batch normalisation to uncover temporal dependencies in long-term sequence data. A graph structure is constructed via k-nearest neighbours (KNN), and an enhanced graph isomorphism network (GIN) is integrated to capture spatial domain fault information variations. Second, the spatio-temporal bidirectional attention-gated fusion module employs a bidirectional multi-head attention mechanism to enhance temporal and spatial features. It incorporates a cross-modal gated update mechanism and learnable weight parameters to dynamically retain the highly discriminative features. Third, the classification output module enhances the model’s generalisation capability through multi-layer fully connected layers and regularisation design. Research findings demonstrate that this approach effectively integrates spatio-temporal coupled fault features, achieving an average accuracy of 99.14% on an experimental dataset. This provides an effective technical pathway for the precise identification of faults in the critical components of reciprocating compressors. Full article

Free-space optical (FSO) communication is a promising alternative to radio-frequency (RF) and optical fiber systems due to its high data rates and large bandwidth. However, its performance is highly susceptible to atmospheric conditions such as fog, rain, snow, and haze. This paper analyzes fog-induced signal attenuation in the Morogoro and Mbeya regions of Tanzania using the Kim and Kruse attenuation models. To improve link performance, a quadrature amplitude modulation (QAM) multiple-input multiple-output (MIMO) FSO link was designed and analyzed using OptiSystem 22.0. In Mbeya, light fog conditions with 0.5 km visibility resulted in an attenuation of 32 dB/km, a bit error rate (BER) of 4.5 × 10⁻²³, and a quality factor of 9.79 over a 2.62 km link. In Morogoro, dense fog with 0.05 km visibility led to an attenuation of 339 dB/km, a BER of 1.12 × 10⁻¹⁵, and a maximum link range of 0.305 km. Experimental measurements were further conducted under clear, moderate, and dense fog conditions to systematically evaluate the FSO link performance. The results demonstrated that MIMO techniques significantly enhanced link performance by mitigating fog effects. Moreover, a dedicated application was developed to analyze transmission errors and evaluate system performance metrics. Additionally, a mathematical model of the FSO link was developed to describe and forecast the performance of the MIMO FSO system in atmospheric conditions impacted by fog. Full article

Type 2 diabetes mellitus (T2DM) is characterized not only by chronic hyperglycemia but also by profound adipose tissue dysfunction, including impaired lipid handling, low-grade inflammation, mitochondrial dysfunction, and extracellular matrix (ECM) remodeling. These adipose tissue alterations play a central role in the development of systemic insulin resistance, ectopic lipid accumulation, and cardiometabolic complications. Glucagon-like peptide-1 receptor agonists (GLP-1RAs), particularly semaglutide, have emerged as highly effective therapies for T2DM and obesity. While their glucose-lowering and appetite-suppressive effects are well established, accumulating evidence indicates that semaglutide exerts pleiotropic metabolic actions that extend beyond glycemic control, with adipose tissue representing a key target organ. This review synthesizes current preclinical and clinical evidence on the molecular and cellular mechanisms through which semaglutide modulates adipose tissue biology in T2DM. We discuss depot-specific effects on visceral and subcutaneous adipose tissue, regulation of adipocyte lipid metabolism and lipolysis, enhancement of mitochondrial biogenesis and oxidative capacity, induction of beige adipocyte programming, modulation of adipokine and cytokine secretion, immunometabolic remodeling, and attenuation of adipose tissue fibrosis and ECM stiffness. Collectively, available data indicate that semaglutide promotes a functional shift in adipose tissue from a pro-inflammatory, lipid-storing phenotype toward a more oxidative, insulin-sensitive, and metabolically flexible state. These adipose-centered adaptations likely contribute to improvements in systemic insulin sensitivity, reduction in ectopic fat deposition, and attenuation of cardiometabolic risk observed in patients with T2DM. Despite compelling mechanistic insights, much of the current evidence derives from animal models or in vitro systems. Human adipose tissue-focused studies integrating molecular profiling, advanced imaging, and longitudinal clinical data are therefore needed to fully elucidate the extra-glycemic actions of semaglutide and to translate these findings into adipose-targeted therapeutic strategies. Full article

Pain is an unpleasant but essential sensory experience that serves as a protective mechanism, yet it can also manifest maladaptively in a wide range of pathological conditions. Current analgesic strategies rely heavily on opioid medications and non-steroidal anti-inflammatory drugs (NSAIDs); however, concerns regarding addiction, tolerance, and dose-limiting adverse effects highlight the urgent need for safer and more effective therapeutics. Voltage-gated sodium (Na_v) channels, which govern the initiation and propagation of action potentials, have emerged as promising targets for mechanism-based analgesic development. In particular, the Na_v1.7–Na_v1.9 subtypes have attracted substantial interest owing to their enrichment in the peripheral nervous system—despite broader expression elsewhere—and their central roles in nociception, offering the potential for non-addictive, subtype-selective pain modulation. This review summarizes the physiological roles of these channels in nociception, examines how disease-associated mutations shape pain phenotypes, and highlights recent advances in drug discovery targeting Na_v1.7 and Na_v1.8. The recent FDA approval of VX-548 (suzetrigine), a first-in-class and highly selective Na_v1.8 inhibitor, marks a major milestone that validates peripheral Na_v channels as clinically actionable targets for analgesia. We also discuss the remaining challenges and emerging opportunities in the pursuit of next-generation, mechanism-informed analgesics. Full article

Background: Brain–Computer Interfaces (BCIs) based on Steady-State Visually Evoked Potentials (SSVEPs) and Extended Reality (XR) offer promising solutions for highly wearable applications, but their classification performance can be affected by systematic effects in stimulus presentation. Novelty: This study introduces a novel online compensation method to compensate for systematic effects in the Refresh Rate (RR) of XR displays, enhancing SSVEP classification without requiring additional training or invasive measurements. Methods: A non-invasive monitoring module was incorporated into the developed BCI pipeline to measure frame rate variations in the XR display, allowing deviations between nominal RR and measured values to be automatically detected and compensated for. Classification performance was evaluated using Filter Bank Canonical Correlation Analysis (FBCCA). Statistical significance was assessed using Student’s t-test. Materials: Two datasets were used: a dataset based on Moverio BT-350, including 9 subjects, and a dataset based on HoloLens 2, including 30 subjects, all collected by the authors. Results: The proposed compensation method led to significant improvements in SSVEP classification accuracy, proportional to the magnitude of fps deviations. In some cases, classification accuracy increased by up to 300% relative to its original value. Statistical analyses confirmed the reliability of the results across subjects and datasets. Conclusions: These findings show that the proposed method effectively enhances SSVEP-based BCIs in XR environments and provides a robust foundation for practical applications requiring high reliability. Full article

The coefficient of friction (COF) in superlubrication systems exhibits highly strong correlation with experimental parameters. However, previous studies mainly reported the stable superlubrication formed at small sliding radii (2–5 mm), often without clearly specifying the friction radius. This lack of critical parametric not only affects the reproducibility of the experimental results but also hinders the fundamental understanding of superlubrication formation. In this work, the effects of friction radius and linear velocity on lubrication performance were investigated by using a graphene oxide–ethylene glycol (GO-EG) model lubricant in ball-on-disk rotational sliding. The results showed that superlubricity is also achievable within a large radius range (>5 mm). The GO-EG lubricant attained a stable superlubrication state with a minimum COF of 0.0065 under the conditions of a linear velocity of 0.40 m/s and a friction radius of 6 mm. This value was approximately 50% lower than that under a 2 mm radius condition. Even when the friction radius is increased to 8 mm while maintaining the same linear velocity, stable superlubricity can still be retained. Subsequently, a systematic analysis of the tribological experimental data revealed that at different fixed velocity ranges, the average COF showed different radius dependence. At low linear velocity (<0.2 m/s), the COF decreased with increasing radius, whereas at higher velocity (>0.3 m/s), the COF first decreased and then increased with radius. Mechanistic analysis showed that friction radius governed the COF by modulating rotational speed, shear rate, running-in states, and water-film evolution, including evaporation-driven viscosity changes and hydration-layer stability. This study enhances the understanding of superlubrication formation conditions and provides guidance for the evaluation and measurement of lubrication systems. Full article

Accurate traffic flow forecasting is a fundamental component of Intelligent Transportation Systems and proactive urban mobility management. However, the inherent complexity of urban traffic flow, characterized by non-stationary dynamics and multi-scale temporal dependencies, poses significant modeling challenges. Existing spatio-temporal models often struggle to reconcile the discrepancy between static physical road constraints and highly dynamic, state-dependent spatial correlations, while their reliance on fixed temporal receptive fields limits the capacity to disentangle overlapping periodicities and stochastic fluctuations. To bridge these gaps, this study proposes a novel Multi-scale Graph-Decoupling Spatial–temporal Network (MS-GSTN). MS-GSTN leverages a Hierarchical Moving Average decomposition module to recursively partition raw traffic flow signals into constituent patterns across diverse temporal resolutions, ranging from systemic daily trends to high-frequency transients. Subsequently, a Tri-graph Spatio-temporal Fusion module synergistically models scale-specific dependencies by integrating an adaptive temporal graph, a static spatial graph, and a data-driven dynamic spatial graph within a unified architecture. Extensive experiments on four large-scale real-world benchmark datasets demonstrate that MS-GSTN consistently achieves superior forecasting accuracy compared to representative state-of-the-art models. Quantitatively, the proposed framework yields an overall reduction in Mean Absolute Error of up to 6.2% and maintains enhanced stability across multiple forecasting horizons. Visualization analysis further confirms that MS-GSTN effectively identifies scale-dependent spatial couplings, revealing that long-term traffic flow trends propagate through global network connectivity while short-term variations are governed by localized interactions. Full article

Module-level free-space optical interconnects require actuators to combine both large stroke and high stability. To address this core trade-off that plagues traditional folded-beam actuators, we have developed a millimeter-scale MEMS electromagnetic actuator integrating a Differential Motion Rejection (DMR) unit with a rigid frame. Its performance was systematically evaluated through magnetic–structural coupling modeling, finite element simulation, and experiments. The actuator achieved millimeter-scale stroke under sinusoidal drive, with a primary resonant frequency of approximately 31 Hz. The introduction of the DMR and frame proved highly effective: the out-of-plane displacement at resonance was reduced by about 97%, the static Z-direction stiffness increased by over 50 times, and the displacement crosstalk decreased to 0.265%. Optical testing yielded a stable deflection angle of approximately ±21°. These results demonstrate that this design successfully combines large stroke with high stability, significantly suppressing out-of-plane parasitic motion and crosstalk, making it suitable for module-level optical interconnect systems with stringent space and stability requirements. Full article