Search Results (1,471)

Precise identification of early peanut leaf spot is strategically significant for safeguarding oilseed supplies and reducing pesticide reliance. However, general-purpose detectors face severe domain adaptation bottlenecks in unstructured field environments due to small feature dissipation, physical occlusion, and class imbalance. To address this, this study constructs a dataset spanning two phenological cycles and proposes POD-YOLO, a physics-aware and dynamics-optimized lightweight framework. Anchored on the YOLOv10n architecture and adhering to a “data-centric” philosophy, the framework optimizes the parameter convergence path via a synergistic “Augmentation-Loss-Optimization” mechanism: (1) Input Stage: A Physical Domain Reconstruction (PDR) module is introduced to simulate physical occlusion, blocking shortcut learning and constructing a robust feature space; (2) Loss Stage: A Loss Manifold Reshaping (LMR) mechanism is established utilizing dual-branch constraints to suppress background gradients and enhance small target localization; and (3) Optimization Stage: A Decoupled Dynamic Scheduling (DDS) strategy is implemented, integrating AdamW with cosine annealing to ensure smooth convergence on small-sample data. Experimental results demonstrate that POD-YOLO achieves a 9.7% precision gain over the baseline and 83.08% recall, all while maintaining a low computational cost of 8.4 GFLOPs. This study validates the feasibility of exploiting the potential of lightweight architectures through optimization dynamics, offering an efficient paradigm for edge-based intelligent plant protection. Full article

During tendon healing, collagen III expression precedes that of collagen I. The collagen I-to-III ratio at a certain time point post-laceration serves as an indicator of the healing status. Consequently, it is crucial to understand how different therapeutic approaches to support tendon healing affect the collagen I-to-III ratio in the extracellular matrix of a healing tendon, particularly across distinct anatomical zones. We compared the impact of a platelet-derived growth factor-BB (PDGF-BB) treatment via controlled release from coaxially electrospun DegraPol^® (Ab medica, Cerro Maggiore, Italy) hollow-fiber mesh with a treatment by the vehicle alone (no PDGF-BB) in the rabbit Achilles tendon full transection model and provide data on the collagen I-to-III ratio 3 weeks post-operation. For this purpose, we compared a dual-color Herovici staining to two single IHC labeling, for collagen I and collagen III, respectively. Herovici staining (HV) was expected to offer a more precise approach (pink-to-blue histogram) than the two separately labeled IHC stainings, both with chromogenic DAB labeling (red-to-green histogram), despite an anticipated positive correlation of the data assessed by these methods. Different zones were compared, i.e., native tendon tissue, reactive zone at interface to implant, hot zone within the core of the healing tendon and the zone within the scaffold, meaning the collagen deposited within the fibers of the implanted DegraPol^® tube, respectively. The analysis revealed that the ratios obtained via HV correlated weakly with the ratios obtained by IHC. Based on HV, PDGF-BB therapy led to higher collagen I-to-III ratios in all zones, except for the zone within the scaffold pores, while IHC did not reveal significant differences. Notably, collagen I-to-III ratios were not higher in immediate proximity, but rather distal from the PDGF-BB releasing implant, specifically in the core of the healing tendon tissue. Hence, a PDGF-BB therapy is suggestive of greater collagen maturation in specific zones of the healing tendon. Full article

High-quality code representations are fundamental to code intelligence. Achieving such representations with parameter-efficient fine-tuning (PEFT) remains a key challenge. While code pre-trained models (CodePTMs) offer a robust foundation for general-purpose embeddings, current PEFT approaches face two main obstacles when adapting them: (i) they fail to adequately capture the deep structural characteristics of programs, and (ii) they are limited by the model’s finite internal parameters, restricting their ability to overcome inherent knowledge bottlenecks. To address these challenges, we introduce a parameter-efficient code representation learning framework that combines retrieval augmentation with structure-aware priors. Our framework features three complementary, lightweight modules: first, a structure–semantic dual-channel retrieval mechanism that infuses high-quality external code knowledge as non-parametric memory to alleviate the knowledge bottleneck; second, a graph relative bias module that strengthens the attention mechanism’s capacity to model structural relationships within programs; and third, a span-discriminative contrastive objective that sharpens the distinctiveness and boundary clarity of span-level representations. Extensive experiments on three benchmarks spanning six programming languages show that our method consistently outperforms state-of-the-art parameter-efficient baselines. Notably, on structure-sensitive tasks using the PLBART backbone, RS-Rep surpasses full fine-tuning, delivering a 22.1% improvement in Exact Match for code generation and a 4.4% increase in BLEU scores for code refinement, all while utilizing only about 5% of the trainable parameters. Full article

This paper proposes a PI-Dual-STGCN fault diagnosis model based on a SHAP-LLM joint explanation framework to address issues such as the lack of transparency in the diagnostic process of deep learning models and the weak interpretability of diagnostic results. PI-Dual-STGCN enhances the interpretability of graph data by introducing physical constraints and constructs a dual-graph architecture based on physical topology graphs and signal similarity graphs. The experimental results show that the dual-graph complementary architecture enhances diagnostic accuracy to 99.22%. Second, a general-purpose SHAP-LLM explanation framework is designed: Explainable AI (XAI) technology is used to analyze the decision logic of the diagnostic model and generate visual explanations, establishing a hierarchical knowledge base that includes performance metrics, explanation reliability, and fault experience. Retrieval-Augmented Generation (RAG) technology is innovatively combined to integrate model performance and Shapley Additive Explanations (SHAP) reliability assessment through the main report prompt, while the sub-report prompt enables detailed fault analysis and repair decision generation. Finally, experiments demonstrate that this approach avoids the uncertainty of directly using large models for fault diagnosis: we delegate all fault diagnosis tasks and core explainability tasks to more mature deep learning algorithms and XAI technology and only leverage the powerful textual reasoning capabilities of large models to process pre-quantified, fact-based textual information (e.g., model performance metrics, SHAP explanation results). This method enhances diagnostic transparency through XAI-generated visual and quantitative explanations of model decision logic while reducing the risk of large model hallucinations by restricting large models to reasoning over grounded, fact-based textual content rather than direct fault diagnosis, providing verifiable intelligent decision support for industrial fault diagnosis. Full article

Objectives: The purpose of the present study was to investigate the acute effects of cold-water immersion (CWI), cryotherapy (CRT), and intermittent pneumatic compression (IPC) on lower-body neuromuscular performance in female basketball players. Methods: Eighteen athletes volunteered to participate (body mass = 63.0 ± 7.2 kg; height = 171.4 ± 6.5 cm; age = 16.4 ± 1.2 years), completing testing at three time points: (i) pre-practice, (ii) post-practice, and (iii) 45–60 min following a randomly assigned recovery intervention. At each time point, athletes performed three countermovement vertical jumps on a dual force plate system sampling at 1000 Hz (VALD Performance). To standardize external load across groups, all players wore inertial measurement units (Kinexon). Results: The two-way repeated measures ANOVA showed no statistically significant interaction (p > 0.05) between the three testing time points and recovery modalities for any of the analyzed variables. However, a significant main effect of time was observed, with 13 of 20 force-time metrics (65%), including jump height, reactive strength index-modified, contraction time, and concentric peak and mean force, declining post-recovery compared with pre-practice values, regardless of the recovery intervention applied. External load measures (e.g., total distance, number of jumps) remained consistent across groups. Conclusions: Overall, these findings suggest that CWI, CRT, and IPC were no more effective than passive recovery (i.e., control group) in mitigating post-practice declines in lower-body force and power-producing capacities. Full article

Viral RNA structure plays a critical regulatory role in viral replication, serving as a dual-purpose mechanism for encoding genetic information and controlling biological processes. However, these structural elements also serve as pathogen-associated molecular patterns (PAMPs), which are recognized by pattern recognition receptors (PRRs) of the host innate immune system. This review discusses the complex and poorly understood relationship between viral RNA structure and recognition of RNA by PRRs, specifically focusing on Toll-like receptor 3 (TLR3) and Retinoic acid-inducible gene I (RIG-I). While current interaction models rely upon data generated from use of synthetic ligands such as poly(I:C) or perfectly base-paired double-stranded RNA stems, this review highlights significant gaps in our understanding of how PRRs recognize naturally occurring viral RNAs that fold into highly complex three-dimensional structures. Furthermore, we explore how viral evolution and nucleotide variations, such as those observed in influenza viruses, can drastically alter local and distal RNA structure, potentially impacting immune detection. We conclude that moving beyond synthetic models to understand natural RNA structural dynamics is essential for elucidating the mechanisms of viral immune evasion and pathogenesis. Full article

This study presents a multiscale, uncertainty-aware hybrid deep learning approach addressing the short-term wind speed prediction problem, which is critical for the reliable planning and operation of wind energy systems. Wind signals are decomposed using adaptive variational mode decomposition (VMD), and the resulting wind components are processed together with meteorological data through a dual-stream CNN–BiLSTM architecture. Based on this multiscale representation, probabilistic forecasts are generated using quantile regression to capture best- and worst-case scenarios for decision-making purposes. Unlike fixed prediction intervals, the proposed approach produces adaptive prediction bands that expand during unstable wind conditions and contract during calm periods. The developed model is evaluated using four years of meteorological data from the Afyonkarahisar region of Türkiye. While the proposed model achieves competitive point forecasting performance (RMSE = 0.700 m/s and MAE = 0.54 m/s), its main contribution lies in providing reliable probabilistic forecasts through well-calibrated uncertainty quantification, offering decision-relevant information beyond single-point predictions. The proposed method is compared with a classical CNN–LSTM and several structural variants. Furthermore, SHAP-based explainability analysis indicates that seasonal and solar-related variables play a dominant role in the forecasting process. Full article

Eutrophication is a major threat to freshwater ecosystems, leading to harmful algal blooms, biodiversity loss, and hypoxia. Excessive nutrient loading, primarily from nitrates and phosphates, is driven by fertilizer runoff, sewage discharge, and agricultural practices. Sediment microbial fuel cells (sMFCs) have emerged as a potential bioremediation strategy for nutrient removal while generating electricity. Although various studies have explored ways to enhance sMFC performance, limited research has examined the relationship between external resistance, electricity generation, and nutrient removal efficiency. This study demonstrated effective nutrient removal from overlying water, with 1200 Ω achieving the highest nitrate and phosphate removal efficiency at 59.0% and 32.2%, respectively. The impact of external resistances (510 Ω and 1200 Ω) on sMFC performance was evaluated, with the 1200 Ω configuration generating a maximum voltage of 466.7 mV and the 510 Ω configuration generating a maximum current of 0.56 mA. These findings show that external resistance plays a major role in both electrochemical performance and nutrient-removal efficiency. Higher external resistance consistently resulted in greater voltage output and improved removal of nitrate and phosphate. The findings also indicate that sMFCs can serve as a dual-purpose technology for nutrient removal and electricity generation. The power output may be sufficient to support small, eco-friendly biosensing devices in remote aquatic environments while mitigating eutrophication. Full article

Recycling industrial CO₂ into agricultural systems offers a dual-purpose strategy for achieving carbon neutrality and enhancing sustainable crop production. Although elevated CO₂ is known to influence plant growth, the directed delivery of industrially sourced CO₂ via drip irrigation to modulate rhizosphere processes in arid soils remains underexplored. We conducted a two-year field experiment in a Xinjiang cotton field to evaluate the effects of five concentrations of industrial CO₂ solution (0.00–0.16 kg·m⁻³) on soil properties, nutrient dynamics, and crop performance. The optimal CO₂ treatment (0.08 kg·m⁻³) significantly reduced soil pH by up to 0.3 units and electrical conductivity by up to 27.9%, while enhancing the availability of ammonium-N (51.1%), available P (8.1%), and available K (32.65%). These improved soil conditions subsequently enhanced plant N, P, and K accumulation (56.2%, 41.9%, and 53.2%, respectively), total biomass (31.8%), and seed cotton yield (5.76–6.06%). Our findings demonstrate that CO₂-enriched irrigation enhances the rhizosphere microenvironment and nutrient availability, providing a novel pathway for carbon recycling and high-efficiency cotton production in arid regions. Full article

Background: Adolescence is a sensitive period for psychological, academic, and social development, and sports participation has been described as a potential protective factor for academic performance and psychological well-being. However, limited research has examined the combined influence of sports involvement, sport type, and family background on adolescents’ academic and psychological outcomes. This study aimed to investigate the associations between organized sport participation, sport type (football vs. judo), psychological well-being, psychosomatic symptoms, academic performance, and family socioeconomic background among adolescent boys. Methods: The sample consisted of 52 boys aged 11–14 years from a rural school, divided into football players (n = 13), judo athletes (n = 13), non-athletes (n = 13), and a contextual subgroup of students with special educational needs (SEN; n = 13), with the latter included for exploratory purposes only. Data included school-record-based academic performance and validated self-report measures of life satisfaction, depressive symptoms, psychosomatic complaints, perceived physical fitness, and socioeconomic background. Results: Athletes demonstrated significantly higher academic achievement than non-athletes in overall grade point average (p < 0.001), mathematics (p < 0.001), Romanian (p < 0.001), English (p = 0.03), and Hungarian (p < 0.001). They also reported higher life satisfaction (p < 0.001) but simultaneously showed slightly elevated depressive symptom scores (p < 0.001), indicating a paradoxical pattern of concurrent psychosocial benefits and psychological strain. Parental education (p < 0.001), parental occupational status (p = 0.01), and fathers’ occupational position (p = 0.02) were significantly higher among athletes’ families. Perceived physical fitness was also rated higher by athletes (p < 0.001). No significant differences were found in body mass index, family structure, or most psychosomatic symptoms. Conclusions: Sport participation was associated with more favorable academic and psychological indicators, yet also with elevated depressive symptoms, highlighting the dual nature of organized sport during adolescence. Future research should apply longitudinal designs, include female participants, and incorporate objective indicators of training load. Full article

This study focused on comparing broiler meat quality across different production systems and seasons. Chicken carcasses from intensive, free-range, and dual-purpose poultry systems were analyzed for intrinsic and extrinsic quality traits. The results revealed significant effects of the production system and season. Carcasses from dual-purpose and intensive systems were heavier. Greater carcass weight was recorded in autumn and winter. The mean post-mortem pH of breast and thigh was lower in extensive and dual-purpose systems and significantly lower in winter and spring. Colorimetric parameters varied by system, as higher means of redness (intensive), yellowness (free-range), and lightness (dual-purpose) were observed. Meat from intensive systems was less firm, showed higher levels of unsaturated fatty acids and better oxidation stability. Dual-purpose displayed higher levels of polyunsaturated fatty acids. The interaction effect was significant for most quality parameters. Full article

The valorization of nutrient-rich institutional effluents represents a promising route for sustainable algal biotechnology. This study investigates the potential of canteen wastewater (CW) as an alternative culture medium for Spirulina platensis, integrating wastewater treatment with high-value carotenoid and lipid production. Growth performance, biochemical composition, and nutrient removal efficiencies were systematically evaluated in 2 L photobioreactors under optimized conditions. Spirulina cultured in 75% CW under 180 μmol photons m⁻² s⁻¹ achieved a biomass productivity of 0.071 g L⁻¹ day⁻¹, nearly three-fold higher than the synthetic BG-11 control (0.023 g L⁻¹ day⁻¹). Nutrient remediation was highly efficient, with 92.12% nitrate and 90.05% phosphate removal, effectively reducing effluent concentrations below discharge limits. Biochemical profiling revealed that wastewater-grown biomass contained 54.3% protein and 7.85% lipids, with a remarkable carotenoid yield of 21.81 mg g⁻¹ DW—significantly higher than the control (6.85 mg g⁻¹ DW). Mechanistic analysis suggests that the balanced nutrient stoichiometry (C:N:P ≈ 30:4:1) and mixotrophic conditions enhanced biomass quality while mitigating ammonia toxicity. This study demonstrates the first integrated application of canteen wastewater for dual-purpose bioremediation and pigment-rich biomass production, establishing a scalable circular bioeconomy framework for institutional waste management. Full article

The extensive mining area and multitude of working sites in Maping Phosphate Mine result in a complex ventilation system. This complexity manifests as uneven airflow distribution at working faces, posing considerable challenges for efficient ventilation management. An intelligent ventilation management system based on the Python PyQt5 library was developed for Maping Phosphate Mine to improve ventilation efficiency, lower dust concentration at the working face, and enhance safety by addressing uneven air volume distribution. The implementation of an integrated system, comprising a 3D ventilation network model, remote control capabilities, and smart algorithms, has successfully realized zonal planning and on-demand ventilation in the mine’s underground workings. To adapt to the fluctuating air demand at the tunneling face, a remote intelligent control scheme for louvered dampers was implemented. This dynamic demand-based strategy achieves precise distribution of air volume throughout the ventilation network. The research results demonstrate that the system effectively addresses the uneven distribution of air volume, thereby improving the overall ventilation environment and reducing the risk of ventilation-related accidents. The system serves dual purposes: it provides an intelligent ventilation control mechanism and integrates seamlessly with the key subsystems for underground safety production. This synergy is instrumental in advancing the mine’s digitalization and intelligent transformation initiatives. Field test results indicate that the system achieved a 30% reduction in energy consumption and a 70% decrease in dust concentration at the working face, respectively. Full article

This paper proposes a low-cost measurement system that can be used to perform simultaneous blood pressure (BP) and breathing (BR) measurements. Regarding BP measurements, the main parameters that are accessed include systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure blood pressure (MAP), and heartbeat rate (HR). Concerning BR measurements, the main parameters that are accessed include the inspiration period and amplitude (IPA), the expiration period and amplitude (EPA), and the breathing rate (BR), as well as the statistical and standard deviation of all these parameters. The dual measurement capability of the proposed measurement system is very important since blood pressure and breathing parameters are not statistically independent and it is possible to obtain additional and valuable clinical information from the information provided by both biomedical variables when measured simultaneously. The analysis of the correlation between these variables is particularly important after performing intensive physical exercises, since it enables cardiac rehabilitation assessment, pre-surgical risk evaluation, detection of silent ischemia, and monitoring of chronic diseases recovery, among others. Regarding the performance evaluation of the proposed biomedical device, a prototype of the measurement system was developed, tested, and calibrated. Several experimental tests were carried out to evaluate the performance of the proposed measurement system and to obtain the correlation coefficients between different blood pressure and breathing parameters. The tests were based on a statistically significant number of measurements that were performed with a population that integrated twenty students in two groups with different habits of physical exercise practice but subjected to a set of common physical exercises, with graduated intensity levels. Full article

Introduction: Overweight and obesity are becoming increasingly prevalent. Incretin-based obesity treatments—glucagon-like peptide-1 receptor agonists (GLP-1 RAs) and dual glucagon-like peptide-1 receptor/glucose-dependent insulinotropic polypeptide receptor agonists (GIP/GLP-1 RAs or dual agonists)—are a major stride in the evolution of obesity management. However, like weight loss with other means, they are associated with an inadvertent significant loss of lean body mass, including muscle. This has led to a resurgence in research for the preservation of lean body mass, the loss of which occurs with weight loss. The purpose of this narrative review is to discuss the mechanisms involved with lean body loss and capture the research landscape of the different classes of pharmacological agents being developed to address this problem. Methodology: We queried PubMed, Medline, and Scopus for randomized controlled trials and phase II or phase III trials using key words to capture the breath of this topic—obesity, weight loss, muscle loss, lean mass, and muscle preservation. Animal studies were excluded. We analyzed the studies conducted to date. Results: Weight loss, regardless of the method used to achieve it, is inadvertently accompanied by lean body mass loss, to varying degrees. There are several mechanisms that govern the loss of lean body mass and, more specifically, the loss of muscle mass; as such, several classes of medications have been explored, targeting different pathways and receptors—including bimagrumab (activin receptor agonist), tesamorelin (growth hormone releasing hormone agonists), and enobosarm (selective androgen receptor modulator). Most of these drugs are in the early phases of research development, but some show great promise. Conclusion: This narrative review attempts to detail the physiology of muscle mass loss when accompanied by weight loss and identify pharmacological targets that can be utilized to minimize it with mechanisms, effects, side effects, and research developmental progress. Full article