Search Results (4,145)

Feed shortage and water scarcity critically constrain livestock production in semiarid regions, yet the interactive mechanism of growth by which planting density and mowing shape the trade-off between biomass yield and nutritional quality remain poorly quantified for rainfed sweet sorghum (Sorghum bicolor L. Moench). The Loess Plateau is located in the north-central part of China. It belongs to the semiarid continental monsoon climate and is mainly rainfed by agriculture. A two-year field experiment (2023–2024) was conducted in this region to investigate how four planting densities (50,000, 70,000, 90,000 and 110,000 plants·ha⁻¹) and two mowing regimes regulate plant morphology, root development and comprehensive nutritional traits. Compared with a mowing treatment, a non-mowing treatment significantly increased dry matter yield from 8.54 t·ha⁻¹ to 21.59 t·ha⁻¹ (an increase of 152.7%) in 2023 and from 10.32 t·ha⁻¹ to 20.91 t·ha⁻¹ (an increase of 102.6%) in 2024. Increasing planting density elevated population biomass, with the 90,000 plants·ha⁻¹ treatment optimally balancing individual growth and interplant resource competition. Mowing reduced stem diameter, leaf area index, and root biomass, thereby lowering fiber concentration but suppressing total dry matter accumulation. Structural equation modeling (GFI = 0.937) quantified this dual effect: mowing exerted a strong negative direct effect on dry matter yield (path coefficient = −4.211, explaining 94.9% of the yield variance) but indirectly improved nutritional quality by thinning stems to optimize leaf allocation. Integrated random forest and radar chart multi-index evaluation identified non-mowing at 90,000 plants·ha⁻¹ as the optimal cultivation regime (comprehensive score Y = 0.917), which coordinated population biomass and nutritional performance while delivering a 23.3% higher net profit than conventional mowing treatments. This study revealed the physiological trade-off path between biomass productivity of sorghum and its quality under drought stress under rainfed conditions on the Loess Plateau and provides a quantitative, replicable evaluation framework to optimize agronomic management for sweet sorghum and analogous C4 crops in global semiarid rainfed ecosystems, offering practical strategies for sustainable forage production. Full article

High levels of global water pollution have driven the development of effective strategies for removing priority contaminants; this is a worldwide issue that calls for innovation in traditional methodologies using affordable materials with high removal capacities. Experimental and DFT studies were combined to elucidate the adsorption, selectivity, and competitive mechanisms of Hg²⁺, Ni²⁺, and Cu²⁺ onto a chitosan-based hydrogel, linking experimental behavior with molecular-level interactions. FTIR and XRD analysis allowed the determination of the hydrogel composition, the verification of the cross-linking of the polymer with the cross-linking agent, and the detection of heavy metals analyzed on the adsorbent. After the characterization, the isotherms were experimentally obtained in single and binary systems at 298.15, 303.15, and 313.15 K and pH 4. In binary experiments, antagonistic adsorption occurred due to competitive binding of heavy metals to the hydrogel’s active sites. The highest adsorption capacities achieved from individual solutions were 0.723, 0.187, and 0.420 mmol/g at 313.15 K for Hg²⁺, Ni²⁺, and Cu²⁺, respectively, and those for binary solutions were 0.633, 0.080, and 0.238 mmol/g at 313.15 K for Hg²⁺, Ni²⁺, and Cu²⁺, respectively. Adsorption kinetics in both single- and binary systems were well fitted by a pseudo-second-order model, consistent with a chemisorption-controlled mechanism. The adsorption process was endothermic and spontaneous, and it was determined that the adsorption was attributed to primary amine and hydroxyl groups available on the surfaces of the cross-linked hydrogels. DFT calculations revealed preferential interactions of Hg²⁺, Ni²⁺, and Cu²⁺ with specific hydrogel functional groups, supporting the experimental adsorption selectivity. Full article

This study aimed to evaluate the effects of dietary supplementation with Melissa officinalis on the haematological, biochemical, neuroendocrine, and behavioural parameters of weaned pigs, as well as on behaviours associated with the establishment of social hierarchy following transfer to the fattening stage under conditions of social competition. The study included 48 pigs allocated to control and experimental treatments. Over the 55-day experimental period, pigs in the experimental treatment received dried Melissa officinalis leaf material incorporated into the diet, with the daily amount gradually increasing from 3.20 to 9.60 g/animal/day. The mean daily amount administered throughout the experimental period was 6.99 g/animal/day, approximately 7.0 g/animal/day. During the growing period, behavioural responses, serum serotonin and cortisol concentrations, and haematological and biochemical parameters were assessed. Following transfer to the fattening stage, agonistic behaviour associated with the establishment of social hierarchy was monitored. The data were analysed using one-way analysis of variance and correlation analysis. Dietary supplementation with Melissa officinalis significantly increased white blood cell (WBC; p = 0.048), lymphocyte (LYM; p = 0.008), and monocyte counts (MON; p = 0.004), as well as serum serotonin concentrations (p < 0.001). Serum cortisol concentrations were 37.73% lower in the experimental treatment than in the control treatment, although the difference was not statistically significant (p = 0.304). Pigs in the experimental treatment exhibited lower locomotor activity (p = 0.001), a tendency towards less climbing behaviour (p = 0.059), and a longer duration of play fighting (p = 0.001). Following transfer to the fattening phase, pigs in the experimental treatment exhibited a lower frequency of agonistic interactions during the establishment of social hierarchy (p = 0.008). In the control treatment, cortisol was negatively correlated with active behaviour (r = −0.845; p < 0.01) and positively correlated with inactive behaviour (r = 0.845; p < 0.01), whereas serotonin was positively correlated with play fighting (r = 0.462; p < 0.05). These findings indicate that the inclusion of dried Melissa officinalis leaf material in the diet improves immune status, beneficially modulates neuroendocrine regulation, and facilitates the social adaptation of pigs by reducing agonistic behaviour. Therefore, Melissa officinalis may represent a promising phytogenic feed additive for mitigating social stress and improving pig welfare during critical production stages. Full article

Kuzushiji, a classical Japanese cursive script used for more than one thousand years, is preserved in a vast number of historical documents that constitute an important part of Japan’s cultural heritage. However, the complexity and variability of Kuzushiji characters make these documents difficult to access for non-specialists, creating challenges for the digitization, preservation, and dissemination of historical knowledge. To address this issue, this study proposes the Lightweight Kuzushiji Recognition System (LKRS), a lightweight recognition framework designed for 1120-class Kuzushiji recognition and practical online deployment. LKRS integrates modern lightweight neural network architectures with a one-shot pruning framework to reduce model complexity while maintaining recognition performance. In addition, a web-based interface is developed to support interactive character selection and repeated recognition, providing convenient access for researchers and the general public. To evaluate the effectiveness of the proposed framework, extensive experiments are conducted on multiple public datasets and different backbone networks. The optimized EfficientNet-B0 model achieves a recognition accuracy of 94.08% on the Kuzushiji dataset while reducing the number of parameters and floating-point operations (FLOPs) by 62.50% and 86.91%, respectively. Experimental results on Fashion-MNIST, CIFAR-10, and Kuzushiji further demonstrate the effectiveness of the proposed framework in reducing model complexity while maintaining competitive classification performance across different datasets and network architectures. The proposed LKRS provides an efficient and accessible tool for Kuzushiji recognition and contributes to the digitization, preservation, and dissemination of Japanese cultural heritage. Furthermore, the framework offers a reusable reference for developing lightweight recognition systems for other historical scripts. Full article

Higher-dimensional fourth-order nonlinear evolution equations model the competition among anisotropic dispersion, derivative coupling, and nonlinear wave steepening, but exact reductions can become unreliable when the governing equation, invariants, or parameter branches are not checked consistently. This study analyzes a $(2+1)$-dimensional fourth-order nonlinear evolution equation because a verified analytical description of its conservation structure, invariant waves, and spectral sideband behavior is useful both for qualitative wave interpretation and for benchmarking numerical calculations. The equation admits an exact local conservation-law family and a verified Lie point-symmetry subalgebra. Its invariant and traveling-wave reductions produce a non-degenerate Jacobi-elliptic gradient family, a bounded hyperbolic front, and one- and two-exponential waves subject to explicit dispersion and nonresonance conditions. Linearization about an affine exact background gives a closed quadratic sideband-dispersion relation, an explicit discriminant, and the corresponding growth rate. The analysis shows that genuine wave branches must be separated from parameter choices that collapse the reduced equation to an identity. The main contribution is therefore a consistency-first framework that combines symmetry reduction, conservation laws, exact-wave construction, admissibility conditions, and direct residual verification, thereby going beyond earlier treatments centered mainly on isolated lump or interaction formulas. Full article

Hyperspectral and multispectral image fusion requires enhancing spatial details while preserving the pixel-wise spectral fidelity of hyperspectral observations. Transformer-based approaches provide effective contextual modeling, yet hierarchical token merging may cause spectral mixing, whereas pixel-preserving tokenization restricts the spatial context captured by window attention. To resolve this trade-off, we propose a Lightweight Multiscale Contextual Attention Network (LMCAN) that reconstructs spatial context without altering the pixel-level representation. The proposed network progressively broadens contextual perception and coordinates complementary spatial dependencies within the attention process, enabling information at different scales to interact adaptively rather than being modeled in isolation. It further recovers local correlations omitted by fixed window partitioning, improving the reconstruction of boundaries and fine structures. Through this unified design, spectral preservation and spatial context restoration are jointly achieved within a shallow and efficient architecture. Experiments on four benchmark datasets demonstrate competitive accuracy with substantially lower complexity. On CAVE, LMCAN achieves 49.04 dB PSNR and 2.47 SAM with only 0.157 M parameters and 12.06 G FLOPs. Full article

Early stroke risk prediction offers an opportunity for timely interventions and may help reduce the clinical burden associated with stroke. Artificial intelligence (AI) provides medical practitioners with tools to analyse clinical biomarkers and predict a patient’s stroke risk. However, existing models lack interpretability and explainable decision support, limiting their adoption in clinical settings. This paper proposes a hybrid Gated Recurrent Unit (GRU)-TabTransformer architecture using cross-attention for stroke-status prediction. The proposed architecture comprises two stages. In the first stage (Model A), ordered feature-sequence representations from a GRU encoder are combined with concatenated categorical and numerical tabular features from a TabTransformer encoder. The model passes these distinct learned representations through cross-attention and linear projection layers before the final prediction. In the second stage (Model B), we augment our model with Large Language Models (LLMs) and Local Interpretable Model-agnostic Explanations (LIME) to provide per-sample, post hoc, human-interpretable explanations based on predicted probabilities. Experimental results show that both models achieve competitive sensitivity and accuracy values. In particular, the Synthetic Minority Over-sampling Technique (SMOTE) yielded a more balanced sensitivity–specificity trade-off, with a sensitivity of 74.00% and a specificity of 75.10%. Moreover, the model achieved an accuracy of 75.05% with SMOTE. An ablation study further shows that Cross-Attention offers better sensitivity and Receiver Operating Characteristic-Area Under the Curve (ROC-AUC), while Gated Fusion performs better on several other metrics. Additionally, age and average glucose level were the most influential stroke risk indicators, while Body Mass Index (BMI) and ever-married status were secondary model-attributed features. A two-factor repeated-measures Analysis of Variance (ANOVA) confirmed an interaction between model choice and the class-balancing technique used in stroke risk prediction systems. The Mistral + Hybrid GRU-TabTransformer architecture also recorded a mean inference time of 57.59s using few-shot prompting. Overall, the results provide a proof of concept for integrating hybrid GRU-TabTransformer with cross-attention and LLM-based explainability to support interpretable stroke risk prediction systems, pending robust external validation before deployment. Full article

Aspect-based sentiment analysis (ABSA) aims to identify the sentiment polarity expressed toward a specific aspect in a sentence. Existing sequential and Transformer-based methods can effectively capture contextual semantics, but they often lack explicit modeling of aspect–opinion relations. Graph-based approaches partially address this limitation by incorporating syntactic dependency structures; however, most rely on deterministic parser-derived graphs or fixed relation weights, which may introduce noisy edges and unstable message propagation for ambiguous, informal, or domain-shifted text. To address these issues, this paper proposes a Bayesian Biaffine Variational Graph Convolutional Network (BBV-GCN) for ABSA. Specifically, a contextual encoder first generates token-level representations for each sentence–aspect pair. A biaffine relation scorer then estimates aspect-aware pairwise token interactions and constructs a soft latent relation graph. Rather than treating relation weights as deterministic values, BBV-GCN introduces latent relation variables and learns their posterior distributions through variational inference, thereby enabling uncertainty-aware graph construction. Based on the learned graph, a variational graph convolutional network performs multi-hop message passing to aggregate opinion cues, modifiers, negation patterns, and contrastive signals toward the target aspect representation. Experiments on five benchmark datasets, including Twitter, Laptop14, Restaurant14, Restaurant15, and Restaurant16, demonstrate that BBV-GCN achieves competitive and well-balanced performance relative to representative attention-based, Transformer-based, and graph-based baselines. Ablation studies further confirm the contributions of Bayesian relation modeling, KL regularization, and variational graph propagation. Visualization results illustrate how uncertainty-aware weighting can attenuate spurious relations and produce more interpretable aspect-specific latent graphs. Overall, BBV-GCN provides a robust and uncertainty-aware graph reasoning framework for fine-grained sentiment analysis. Full article

A red-emissive, derivatization-assisted aptasensor was developed for acrylamide analysis in food. Acrylamide was first converted into the xanthydrol–acrylamide adduct (XAA), which was recognized by the DNA aptamer Apt_XAA. A π-conjugation-extended thioflavin T analogue, ThT-Red, was synthesized as the fluorescent signal molecule. ThT-Red exhibited an emission maximum at 600 nm and showed enhanced fluorescence upon binding to Apt_XAA, attributable to restricted intramolecular motion. XAA binding perturbed the ThT-Red–Apt_XAA interaction, producing a concentration-dependent fluorescence decrease consistent with competitive displacement. Following a 30 min xanthydrol derivatization step, the fluorescence sensing response was obtained within 3 min. Under optimized conditions, the sensor showed a linear response to XAA over 0–2.5 μM, with a limit of detection of 0.23 μM. Recoveries of 90.14–119.72% were obtained in spiked potato-chip extracts. The sensing response was also measurable using a portable fluorometer, supporting its potential for rapid on-site analysis. This work demonstrates that structural engineering of an aptamer-responsive molecular rotor can provide an effective red-fluorescence transduction strategy for derivatization-assisted detection of low-epitope food contaminants. Full article

Carbon-based materials exhibit diverse structural responses to irradiation owing to their distinct dimensionality, degree of graphitization, surface chemistry, and pore architecture. Although irradiation has traditionally been regarded as a source of structural damage, increasing evidence demonstrates that controlled irradiation can be deliberately utilized to tailor defects, surfaces, interfaces, and pore structures, thereby enabling desirable functional properties. This review summarizes recent progress in the irradiation-induced structural evolution and functional applications of four representative carbon-based materials, including graphene-based materials, carbon nanotubes, carbon fibers, and activated carbon/biochar. Particular attention is given to the characteristic irradiation responses of different carbon architectures. In graphene, irradiation predominantly induces vacancies, reconstructed defects, and surface functionalization, providing active sites for environmental remediation. Carbon nanotubes additionally undergo inter-tube cross-linking and welding, enabling enhanced mechanical performance and tunable electronic properties. For carbon fibers, irradiation mainly regulates surface chemistry and fiber matrix interactions, facilitating interface engineering in high-performance composites. In activated carbon and biochar, irradiation modifies pore accessibility, structural disorder, and surface functional groups, thereby influencing adsorption and electrochemical performance. These distinct responses demonstrate that irradiation can evolve from a conventional damage process into a controllable materials-engineering strategy when appropriate irradiation conditions are employed. Finally, current challenges associated with optimal irradiation conditions, quantitative defect identification, and irradiation structure–property relationships are discussed. Based on these distinct responses, we propose an architecture-dependent irradiation–structure–function (A-ISF) framework that links the initial carbon architecture and irradiation conditions to dominant energy-deposition mechanisms, structural evolution pathways, property modulation, and ultimately functional applications. Within this framework, irradiation engineering is interpreted as a competition between beneficial structural modification and excessive radiation damage, giving rise to an application-dependent optimal irradiation window. Full article

Electroencephalography (EEG) provides a non-invasive and cost-effective tool for supporting the assessment of attention-deficit/hyperactivity disorder (ADHD). However, the nonstationary nature of EEG produces substantial variability among signal windows recorded from the same participant, which can obscure diagnostic structure and lead to inconsistent predictions. To address this problem, we propose the Contextualized Transfer Entropy Network (CTE-Net), an end-to-end deep-learning architecture that combines global content-based contextualization with nonlinear and directed EEG connectivity estimation. CTE-Net first employs a Transformer encoder to contextualize the multichannel representations within each EEG window. The resulting signals are processed using channel-wise nonlinear temporal filters and Takens delay-coordinate embeddings. A differentiable matrix-based Transfer Entropy module, formulated using Rényi’s $\alpha$-entropy and a rational quadratic kernel, then estimates directed predictive information dependencies between all ordered electrode pairs. The resulting connectivity coefficients are used for ADHD-versus-control classification. The model was evaluated on a publicly available pediatric EEG dataset comprising 120 participants, equally divided between ADHD and control groups, using five fixed subject-wise folds and ten random training repetitions. At the window level, CTE-Net achieved an accuracy of $80.9\pm 1.7\%$, precision of $82.7\pm 2.1\%$, and sensitivity of $84.2\pm 2.3\%$. At the participant level, it achieved an accuracy of $83.4\%$ (95% CI: $78.2$–$88.2$) and an ROC-AUC of $90.2\%$ (95% CI: $85.1$–$94.6$), demonstrating competitive and comparatively balanced classification performance. Beyond classification performance, the directed Transfer Entropy representation exhibited the lowest within-subject dispersion among the analyzed representation stages, with a median reduction of $38.35\%$ relative to raw EEG. This reduction remained consistent across different PCA dimensionalities and distance definitions. These single-dataset findings support CTE-Net as a compact and interpretable methodological framework for representing directed EEG interactions while attenuating window-specific variability within individual participants. Full article

Biomarker discovery from high-dimensional transcriptomic data is frequently hindered by the “curse of dimensionality” and model selection bias. To address this, we propose the Grouping–Scoring–Modeling (G-S-M) framework, a knowledge-driven pipeline that anchors feature selection in established disease–gene associations. G-S-M operates within a 100-iteration Monte Carlo ensemble architecture utilizing internal cross-validation to ensure unbiased evaluation. We evaluated this framework on seven cancer datasets, where it demonstrated robust discrimination with an overall mean F1-score of 0.84 across all datasets. The framework achieved the strongest performance on Acute Myeloid Leukemia (mean F1 = 0.99, AUC-ROC = 1.00) and maintained competitive accuracy even on challenging cohorts, while producing biologically interpretable gene panels traceable to named disease associations. Permutation tests (10,000 iterations) confirmed statistically significant disease–gene enrichment (p < 0.0001) in five of seven datasets, and independent protein interaction network analyses demonstrated significant enrichment of the selected features. Released as an open-source software suite with interactive interfaces, G-S-M provides a reproducible computational framework for candidate biomarker discovery. Full article

Mirror segmentation aims to identify mirror pixels from a single RGB image, yet remains challenging because mirrors provide weak intrinsic texture cues and their appearance is dominated by scene-dependent reflections under varying illumination and viewpoints. While recent models improve performance by leveraging contextual contrast, symmetry priors, frequency/spectral cues, or additional modalities (e.g., depth), many cross-cue or symmetry-aware designs still rely on direct spatial-domain fusion, such as concatenation, addition, or attention. Such fusion can amplify reflection-induced high-frequency variations and lead to leakage, shape distortion, and unstable boundaries. In this paper, we propose a symmetry-aware mirror segmentation framework that stabilizes cross-branch interaction via a frequency-domain cross-conditioned fusion mechanism. We build a dual-path Siamese encoder using the original image and its horizontally flipped counterpart, and introduce Heat Conduction Operator-based Cross Fusion (HCOCF), which performs heat-conduction-inspired spectral attenuation in the DCT domain. Unlike conventional fusion, HCOCF generates a nonnegative cross-conditioned attenuation coefficient map from the opposite branch and applies it to the DCT coefficient grid of the target branch. This produces a DCT-domain attenuation mask that controls the spectral refinement strength of each target feature stream, enabling global context propagation while suppressing unstable reflection-induced high-frequency responses without aggressive direct feature mixing. For multi-scale decoding, we adapt the cross-scale decoder of the baseline symmetry-aware architecture by replacing simple addition with conditional feature aggregation, which refines the HCOCF-enhanced features and improves boundary recovery. Extensive experiments on MSD, PMD, and RGBD-Mirror demonstrate competitive performance against representative supervised mirror segmentation methods. In particular, our RGB-only model achieves 88.47% IoU on MSD and 73.72% IoU on PMD, and remains competitive on RGBD-Mirror without using depth input. Full article

Persistent aldehydic off-notes in fish products are influenced not only by lipid oxidation but also by protein-associated retention of odor-active aldehydes. A remaining question is how gingerol side-chain length and aldehyde structure jointly alter this apparent retention. Mackerel myofibrillar protein (MP) was therefore combined with (E)-2-decenal (T2D), (E,E)-2,4-decadienal (DDE), or trans-4,5-epoxy-(E)-2-decenal (E2D), with or without 6-, 8-, or 10-gingerol. Headspace solid-phase microextraction–gas chromatography–mass spectrometry, interaction disruptors, spectroscopy, molecular docking, and 100 ns molecular dynamics simulations were integrated. All three gingerols decreased the apparent aldehyde-binding ratio of MP. At 125 µmol/g protein, 10-gingerol reduced T2D, DDE, and E2D binding from approximately 81%, 72%, and 85% to 68%, 65%, and 68%, respectively. Spectroscopic responses were consistent with changes in the optical and conformational environment; fluorescence was interpreted as apparent quenching without numerical inner-filter correction, and CD band changes provided evidence of a treatment-related conformational response without assigning exact secondary-structure fractions. Docking ranked 8-gingerol most favorably, whereas the selected binary MYH7-10-gingerol trajectory showed greater pocket residence and a more favorable MM/GBSA estimate than the corresponding MYH7-DDE trajectory. The combined evidence supports a three-layer working model involving matrix partitioning, protein-level site accessibility, and pocket-level competition. These findings provide a mechanistic basis for sensory validation rather than direct proof of deodorization. Full article

Understanding the ecological drivers shaping animal foraging and interactions remains a central question in ecology. Among the most studied systems in this field are the ant–plant interactions, although disentangling the complexity of factors acting in different contexts remains a relevant question. This study investigated how habitat structure, liquid food rewards, and interspecific competition interact to modulate the foraging patterns of the abundant ant Camponotus termitarius on the rosette-shaped plant Eryngium chamissonis in the Brazilian Pampa, where ant–plant interactions are still poorly studied. We monitored 115 plants across three sampling events, measuring ant foraging, trophobiont abundance, vegetation density, plant size, and local nest distributions, and analyzed the relationships using Piecewise Structural Equation Modeling (pSEM). The pSEM revealed that surrounding vegetation density negatively affected C. termitarius nest density, nest extensions, and hemipteran trophobionts. Conversely, denser vegetation and larger plants favored the aggressive competitor Camponotus rufipes. While trophobiont presence and proximal nesting infrastructure directly facilitated C. termitarius activity, hostplant inflorescences promoted the construction of nest extensions on plants. We conclude that C. termitarius foraging is regulated by a multidimensional network where microhabitat complexity mediates spatial niche partitioning and competitive dynamics between sympatric ants. Full article