Search Results (10,824)

Rising ambient temperatures and the increasing frequency of training and competition in hot climates have made heat stress a major challenge in football. Under such conditions, players experience greater cardiovascular and thermoregulatory strain, faster glycogen use, higher perceived exertion, and progressive impairment in repeated high-intensity actions and decision-making. These responses have intensified interest in nutritional strategies that might complement heat acclimation, hydration/electrolyte planning, cooling practices, and recovery management. This narrative review critically synthesizes current evidence on nutritional interventions that may be relevant to football performed in the heat, with emphasis on hydration and electrolyte replacement, carbohydrate–protein strategies, taurine, branched-chain amino acids (BCAAs), creatine, menthol, antioxidant- and nitrate-related approaches, and selected multi-ingredient products. Across the available literature, hydration/electrolyte planning and carbohydrate–protein feeding remain the practical foundation, menthol appears most consistently useful for perceptual cooling, creatine seems safe and potentially helpful for repeated-sprint support, and taurine is promising but still supported by relatively few trials. By contrast, evidence for BCAAs, antioxidants, nitrates, and caffeine as stand-alone heat strategies, as well as for many compound supplements, remains inconsistent, context-specific, or too indirect for strong football-specific endorsement. Overall, the evidence base remains heterogeneous in study quality, protocol design, exercise mode, and sport specificity. A substantial proportion of the available data is derived from cycling, endurance, or laboratory heat-exercise models rather than football-specific trials. Accordingly, any practical recommendation should be interpreted cautiously and embedded within broader heat-management strategies. Future work should prioritize ecologically valid randomized controlled trials in football or football-like intermittent protocols, with transparent reporting of dose, timing, perceptual outcomes, and match-relevant performance measures. Full article

The present study examined whether offensive play-type indicators in professional basketball reflect broader latent playing-style dimensions and whether play-type usage or efficiency is more strongly associated with competitive success. Data were obtained from the official NBA statistics website and included 6400 games across five seasons (2019–2020 to 2023–2024), comprising 5979 regular-season games and 421 playoff games. For each offensive play type, two indicators were analysed separately: usage percentage and efficiency, operationalised as points per possession (PPP). Principal component analyses were conducted independently for regular-season and playoff data, and for usage and efficiency variables. In addition, linear mixed-effects models were used to examine the relationship between play-type indicators and competitive success while accounting for games nested within teams. Only regular-season efficiency variables showed adequate sampling adequacy for factorial analysis (KMO = 0.774), yielding a four-component solution that explained 58.85% of the total variance. In the mixed-effects models, usage variables were not significantly associated with success, whereas efficiency indicators showed greater explanatory value. Specifically, pick-and-roll ball handler PPP and spot-up PPP emerged as the strongest positive predictors of success, with smaller effects observed for roll-man PPP and cut PPP. The efficiency-only model improved model fit relative to the frequency-only model (marginal R² = 0.799 vs. 0.755), whereas adding usage variables to efficiency provided only a negligible additional contribution (marginal R² = 0.803). These findings suggest that, in the NBA, competitive success is more closely related to the effectiveness with which offensive actions are executed than to the relative frequency with which they are used. From an applied perspective, play-type efficiency appears to provide more actionable information than usage-based summaries for performance analysis and tactical decision-making. Full article

The flexible job shop scheduling problem (FJSP) is a typical combinatorial optimization problem in smart manufacturing. Although existing methods have considered machine competition relationships, they lack explicit structured modeling of machine competition relationships induced by candidate operations and are not systematically integrated across state representation, representation learning, and decision-making processes. To address this, this paper proposes a competition-aware dual-attention deep reinforcement learning method. We construct a dynamic heterogeneous graph representation, where machine competition is modeled as state-dependent edges instantiated via a 3D competition tensor, transforming machine competition relationships into structured information, thereby enhancing the model’s ability to characterize complex resource competition patterns. On this basis, we have designed the Competition-Aware Dual-Attention Network (CADAN), which injects competition information into both the attention computation and representation learning processes via a dual-path mechanism, enabling more expressive modeling of machine competition relationships, and which introduces a head-wise competition bias to capture heterogeneous competition patterns. Furthermore, we have developed an adaptive decision head to refine the scores of candidate actions. Our experimental results demonstrate that the proposed method outperforms classical dispatching rules and achieves competitive or superior performance compared with representative evolutionary and learning-based methods on synthetic datasets, public benchmark datasets, and a real-world industrial machining scenario involving mechanical transmission components. Full article

Neural implicit surface representations have yielded impressive results in 3D reconstruction, yet existing methods tend to introduce noise in smooth regions or fail to capture fine details in complex areas, primarily due to a lack of explicit spatial structure modeling. To address these limitations, we propose a geometry-adaptive surface reconstruction method based on Morton codes. By mapping 3D space onto octree traversal paths, this approach provides a natural spatial structural prior for the reconstruction process. For each query point, an implicit octree generates a unique root-to-leaf trajectory, yielding spatially adaptive weights that modulate multi-resolution geometric features. Specifically, low-frequency coarse features dominate in flat regions to suppress noise, whereas high-frequency fine features are activated in edge-rich areas to recover intricate geometry. Experimental results demonstrate competitive performance across multiple datasets, particularly in reconstructing sharp features and fine-grained geometric details. Full article

Multiplication is a fundamental operation in digital signal processing, embedded computing, and nanoscale arithmetic data paths, where area, delay, and energy efficiency are critical design constraints. However, nanoscale multiplier design is challenged by high interconnect complexity, frequent wire crossings, clock-zone synchronization issues, and the rapid growth of area and latency with operand size. Quantum-dot cellular automata (QCA) technology offers a promising post-CMOS platform for compact arithmetic circuit realization through field-coupled computation and transistor-free switching. This paper presents a single-layer QCA-based Dadda Tree Multiplier (DTM) using layout-aware integration of compact half-adder, full adder, XOR, and carry-skip adder modules. The proposed design emphasizes partial-product compression, routing compactness, clock-aware organization, and area-efficient final accumulation. Functional verification is performed using QCADesigner 2.0.3, while energy-related behavior is evaluated using QCADesigner-E under the conventional QCA simulation framework. The proposed DTM consists of 4282 cells and occupies 6.14 μm². Compared with a recent compact QCA multiplier baseline, the proposed architecture reduces cell count by 59.12% and occupies area by 39.80%, while maintaining competitive clocking latency. These results indicate that layout-aware integration of arithmetic modules can substantially improve the area efficiency of QCA-based multipliers, making the proposed design a compact arithmetic core for future nanoscale embedded and signal-processing systems. Full article

Traffic flow forecasting remains challenging because raw traffic flow observations often contain mixed temporal patterns, including slowly varying trends and fast local fluctuations. To address this issue, this paper proposes a Multivariate Empirical Mode Decomposition (MEMD)-guided dual-branch recurrent framework for multistep point forecasting. Specifically, MEMD is used as an alignment-preserving multivariate decomposition mechanism to obtain frequency-aligned components, which are then reconstructed into low-frequency trend and high-frequency residual components. The trend component is modeled by a Long Short-Term Memory (LSTM) branch to capture smooth long-term evolution, while the residual component is learned by a Bidirectional Gated Recurrent Unit (Bi-GRU) branch to characterize short-term oscillatory dynamics. A lightweight fusion head is then used to integrate the two branch-specific representations for final prediction. Experiments on PeMS04 and PeMS08, two traffic datasets derived from the California Department of Transportation Performance Measurement System, show that the proposed method achieves competitive performance across mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE), reaching 19.67/31.59/12.95% on PeMS04 and 15.51/24.43/9.86% on PeMS08. Compared with representative recent baselines, the proposed method achieves competitive results, with relative gains reaching 5.89% on PeMS04 and 5.35% on PeMS08 in selected metric-wise comparisons. These results indicate that MEMD-guided trend–residual representation learning can improve multistep traffic flow forecasting. Full article

Nitrogen forms and native plant traits jointly regulate the competitive ability of invasive plants. This study investigated the invasive species Amaranthus retroflexus and the native species Portulaca oleracea and Medicago sativa. Using a pot experiment, we analyzed their competitive effects under NO₃⁻-N, NH₄⁺-N, CO(NH₂)₂-N and mixed nitrogen (Mix-N) treatments. The results showed that nitrogen addition had no significant effect on the relative yield of A. retroflexus but significantly increased the relative yield of P. oleracea, thereby weakening the competitive advantage of A. retroflexus. In contrast, nitrogen addition had no significant effect on the relative yield of M. sativa but significantly increased the relative yield of A. retroflexus, thereby enhancing the competitive advantage of A. retroflexus. The effect of NO₃⁻-N treatment varied markedly between the two mixed-culture systems: it strengthened the advantage of A. retroflexus when grown with M. sativa yet weakened the advantage when grown with P. oleracea. Further analysis revealed that the competitive advantage of A. retroflexus was associated with the optimization of its photosynthetic traits and nitrogen absorption efficiency. Specifically, it included greater leaf number, leaf area, SPAD value, and leaf biomass. In summary, the competitive performance of invasive plants is not a fixed attribute but rather a dynamic outcome jointly regulated by the interplay between native plant traits and soil nitrogen forms. This provides new insight into the invasion mechanism of alien plants and aids in formulating targeted control strategies. Full article

The purpose of this study is to provide data-driven training optimization tools for indoor rowing coaches and athletes, provide quantitative reference for training monitoring and performance analysis in a controllable environment, and help improve the scientific level of competitive performance and training management. To address the absence of quantitative analysis regarding the relationship between rowing power load and competition time during pre-competition training, this study introduces a sequential attention pooling with monotonic constraints (SAP-MC) to systematically analyze data from the rowing power sensor system. The results show that the model effectively captures the negative correlation between power output and competition time. Specifically, when the average power is increased from 230 W to 290 W, the competition time is reduced from 435.2 s to 409.6 s, resulting in a significant reduction of 25.6 s (p < 0.001). When the coefficient of variation of power output (cv_power) increased from 0.08 to 0.18, the competition time was prolonged by 14.2 s (p < 0.01). In addition, when the acute-chronic load ratio (ACWR) exceeds 1.2, compared with the optimal range (0.9–1.1), the competition time is increased by about 6.8 s (p < 0.05). The overall analysis shows that the average power output and power stability are the most critical variables affecting the change of competition time, followed by training load balance and segmented pace optimization. The research results validate the scientific significance of power monitoring and provide a reference for quantitatively analyzing the correlation between training load and race time in a controlled indoor rowing training environment. Full article

Cardiac arrhythmias are a leading cause of cardiovascular mortality worldwide, necessitating accurate automated detection systems for continuous monitoring and clinical decision support. This study addresses the critical challenge of severe class imbalance in ECG beat classification, where normal beats comprise 82.8% of samples while life-threatening ventricular arrhythmias represent only 6.5%. We propose a lightweight one-dimensional convolutional neural network (1D-CNN) trained with a two-pronged class-balancing strategy: random oversampling of minority classes to 35% of the majority class size, combined with class-weighted cross-entropy loss. Recent work has achieved accuracies approaching 99–100% on the MIT-BIH database through increasingly complex architectures, including transfer learning, attention mechanisms, and multi-channel fusion. However, these approaches often require millions of parameters, limiting deployability on resource-constrained wearables. Despite the recent trend toward complexity, our simple four-block CNN with only 398,469 parameters achieves 99.18% overall test accuracy and a 96.38% macro-averaged F1-score on the MIT-BIH Arrhythmia Database—competitive with state-of-the-art methods while using 90–96% fewer parameters. Critically, the model attains 98.32% recall on ventricular beats, demonstrating high sensitivity for detecting life-threatening arrhythmias. Ablation studies confirm that both oversampling and weighted loss are essential: removing either component causes catastrophic performance degradation. Our results challenge the assumption that architectural complexity is necessary for ECG classification and demonstrate that proper class imbalance handling enables simple models to achieve state-of-the-art performances with superior computational efficiency suitable for deployment in wearable cardiac monitoring devices. Full article

In recent years, vision–language models (VLMs) have been introduced into remote sensing semantic segmentation to provide richer semantic representations through visual–textual alignment. However, most existing VLM-based segmentation methods focus on global semantic alignment while neglecting pixel-level local neighborhood features, which are crucial for reliably understanding remote sensing imagery with high spatial resolution, complex structures, and strong spatial continuity. To address this issue, we propose LoVLANet (Localized Vision–Language Attention Network), a novel vision–language segmentation framework that integrates language-driven global semantics with local spatial context. LoVLANet consists of a text encoder, a visual encoder, and a segmentation decoder. Specifically, the text encoder is inherited from RemoteCLIP to preserve domain-adapted vision–language alignment. The visual encoder is built upon a Vision Transformer (ViT). To enhance local dependency modeling, we propose a Neighborhood Key–Key Encoder. It leverages a Gaussian-weighted neighborhood matrix for spatial correlation and uses key–key similarity to emphasize intrinsic semantic similarity over query-driven features, thus, preserving spatial consistency. Finally, the segmentation decoder fuses multi-scale visual features and aligns the image–text representations to generate accurate pixel-level segmentation results. Experiments on RGB remote sensing benchmarks, including LoveDA and GID, show that LoVLANet achieves competitive segmentation performance under the adopted experimental settings, with improved mIoU and clearer boundary delineation in qualitative visualizations. These results suggest the effectiveness of explicitly modeling local neighborhood relationships in VLM-based segmentation for supervised remote sensing scene understanding. Full article

Background: Combat sports performance is highly determined by psychological factors, and differences in pre-competitive anxiety and mood states can exist between medalist and non-medalists athletes; Methods: The present study aims to assess pre-competitive anxiety and mood state differences between medalist and non-medalists Brazilian Jiu-Jitsu (BJJ) athletes graded as white-belts. The Competitive State Anxiety Inventory–2 (CSAI-2) and the Brunel Mood Scale (BRUMS) questionnaires were applied to 26 BJJ white-belt athletes before a fight in a national-level competition.; Results: Medalists presented less cognitive anxiety (22.36 ± 3.82 vs. 25.21 ± 3.17; p < 0.05) and higher mental confusion (9.86 ± 3.01 vs. 7.43 ± 3.01; p < 0.05) than non-medalist athletes. No significant differences were found in any other variable; Conclusions: The relationship between pre-competitive anxiety and sport performance is clear; however, higher mental confusion in medalists is a confounding result. More research is needed on this topic to elucidate the psychological phenomena of higher mental confusion in less-experienced athletes. Full article

This study introduces a novel fusion deep learning framework that integrates a convolutional neural network (CNN), a bidirectional gated recurrent unit (BiGRU), and a self-attention (SA) mechanism to address the shortcomings of conventional linear models in modeling and predicting nonlinear dynamics of sea level changes. To further enhance model adaptability and performance, the Dream Optimization Algorithm (DOA) is incorporated to enable hyperparameter tuning, resulting in the DOA-CNN-BiGRU-SA framework, which significantly improves the model’s ability to predict nonlinear sea level time series. To mitigate the impact of randomness in neural network initialization, we initially employed a default random seed and conducted experiments with data from five tidal stations in Japan. The DOA-CNN-BiGRU-SA framework outperformed seven other relevant models. Subsequently, an extended evaluation was carried out using data from six additional tidal stations, with predictions generated across 30 different random seeds, confirming the model’s competitive accuracy and robustness. Finally, the proposed framework was applied to satellite altimetry data over the entire East and South China Sea region. Two distinct processing strategies yielded regional sea level rise trends of 3.96 ± 0.47 mm/year and 4.02 ± 0.47 mm/year, respectively, over the 1993–2023 period, and these results closely agree with those reported in the China Sea Level Bulletin report in 2023. This paper presents an integrated approach that enables joint optimization of deep learning architectures and investigates the effects of initialization randomness in neural networks, offering a robust technical solution for predicting short-term regional sea level changes. Full article

Background/Objectives: Clinical phenotyping from narrative electronic health records (EHRs) often relies on multi-stage pipelines involving span-level extraction, ontology mapping, and aggregation. Large language models (LLMs) may enable direct document-level abstraction of clinically meaningful phenotype features from complete notes. We evaluated whether GPT-5.2 could approximate human annotation for note-level multiple sclerosis (MS) phenotyping and compared its performance with human annotators, a locally run open-source LLM, HPO-based extraction tools, and a supervised clinical transformer encoder. Methods: We analyzed 100 de-identified MS neurology progress notes from a single academic medical center. Each note was annotated for the presence or absence of 17 predefined neurological phenotype categories. Two human annotators independently labeled all notes using a multi-label note-level framework in Prodigy, and disagreements were adjudicated to create a reference annotation set. GPT-5.2 was evaluated in a zero-shot setting using structured JSON output. Comparator methods included Llama-3.1 8B, Doc2Hpo, ClinPhen, PhenoSnap, and BioClinical ModernBERT. Performance was assessed using agreement, precision, recall, F1, Matthews correlation coefficient, and false-positive and false-negative assignments per note. Results: Human–human agreement was generally high, although lower for rare or ambiguously documented features. GPT-5.2 achieved the strongest automated performance, with macro-precision 0.734, macro-recall 0.921, macro-F1 0.801, and macro-averaged MCC 0.777, approaching human annotator performance. GPT-5.2 showed the lowest false-negative count per note but more false-positive assignments than either human annotator, reflecting a sensitive but more inclusive annotation profile. Llama-3.1 8B performed competitively among automated methods, whereas HPO-based extraction tools and BioClinical ModernBERT showed lower performance on this low-resource note-level task. Secondary review of GPT-5.2 discordant assignments found no clear hallucinations and suggested that some apparent false positives reflected phenotype evidence missed in the human-derived reference set. Conclusions: GPT-5.2 achieved near-human performance for document-level recognition of MS phenotype categories from narrative neurology notes. Direct note-level abstraction may provide a scalable approach for research and population-health phenotyping of large EHR note corpora. Full article

Biologically related time-series data characterize the dynamic evolution of biological systems, including genetic inheritance, disease diagnosis, and the biological microenvironment. However, accurate prediction of these data remains challenging due to their pronounced time-varying, non-stationary, and noisy characteristics. Existing approaches often fail to capture latent shifts of biologically related time series, limiting both predictive performance and time-varying pattern recognition capability. Thus, in this study, we first propose a time-varying neural network (TVNN) model that combines frequency-domain information with Koopman theory. TVNN-model Koopman transition matrices are used to model global dynamics and local time-varying behaviors for pattern extraction. Secondly, a time-varying pattern recognition large language model (TVPRLLM) is introduced to recognize and classify the extracted time-varying patterns, enabling the identification of potential pattern categories. Thirdly, we have developed a biology-related time-series predictive platform that can offer visualization, data analysis, and predictive services. Experimental results demonstrate that the TVNN model outperforms existing mainstream methods in predicting biology-related time-varying time series, and that it achieves competitive forecasting performance, though its behavior depends strongly on the design of the frequency-domain decomposition. Additional robustness analyses reveal that the choice of Fourier masking strategy can materially affect both RMSE and long-horizon stability. We further show that Koopman-derived time-varying representations are highly discriminative for dynamic state recognition. Full article

Recent advances in 3D Gaussian splatting (3DGS) have enabled efficient image-based scene reconstruction, but existing methods that rely heavily on multi-view photometric consistency remain sensitive to dynamic illumination and weakly constrained regions. This issue is especially evident in tunnel scenes, where limited ambient light and localized active illumination cause strong appearance variation and shadowed regions that appear weakly textured in the captured images. As a result, existing methods often suffer from appearance inconsistency, floating artifacts, and unstable Gaussian distributions. To address these challenges, we present dynamic lighting-aware Gaussian splatting (DLG-GS), a real-time framework designed primarily for tunnel-oriented reconstruction under dynamic lighting. DLG-GS includes two complementary components: a dynamic lighting-adaptive appearance modeling strategy that reduces illumination-induced artifacts while preserving local texture details, and a voxel–depth joint constraint that uses monocular depth priors to regularize the spatial distribution of voxel anchors and neural Gaussians, thereby improving optimization stability and suppressing floating artifacts in shadow-induced weak-texture regions. By jointly optimizing appearance adaptation and depth-guided spatial regularization, DLG-GS improves reconstruction stability and rendering quality while maintaining real-time performance. Experiments on a self-collected tunnel dataset show clear improvements over selected baselines, and additional evaluations on public benchmarks indicate competitive performance beyond the target tunnel setting. Full article