A bibliometric analysis and evaluation of publications published in the meat science journal from 1980 to 2025.

Shifting vehicle management to performance-based for better air quality and health.

Omnidirectional videos (ODVs) are redefining viewer experiences in virtual reality (VR) by offering an unprecedented full field-of-view (FOV). This study extends the domain of saliency prediction to 360$^\circ$∘ environments, addressing the complexities of spherical distortion and the integration of spatial audio. Contextually, ODVs have transformed user experience by adding a spatial audio dimension that aligns sound direction with the viewer's perspective in spherical scenes. Motivated by the lack of comprehensive datasets for 360$^\circ$∘ audio-visual saliency prediction, our study curates YT360-EyeTracking, a new dataset of 81 ODVs, each observed under varying audio-visual conditions. Our goal is to explore how to utilize audio-visual cues to effectively predict visual saliency in 360$^\circ$∘ videos. Towards this aim, we propose two novel saliency prediction models: SalViT360, a vision-transformer-based framework for ODVs equipped with spherical geometry-aware spatio-temporal attention layers, and SalViT360-AV, which further incorporates transformer adapters conditioned on audio input. Our results on a number of benchmark datasets, including our YT360-EyeTracking, demonstrate that SalViT360 and SalViT360-AV significantly outperform existing methods in predicting viewer attention in 360$^\circ$∘ scenes. Interpreting these results, we suggest that integrating spatial audio cues in the model architecture is crucial for accurate saliency prediction in omnidirectional videos.

The study of unstable nuclei beyond the nucleon drip line is an important method to study the nuclear interaction and structure in the extremely neutron- or proton-rich system, and various nuclides beyond the proton drip line mainly decay by spontaneous one-proton emission. Using the deformed Woods-Saxon potential, spin-orbit potential and the expanded Coulomb potential to construct the daughter-proton potential, based on the quantum tunneling model and the microscopic Gamow state theory, the half-lives data of various proton emitters are systematically calculated. By using nuclear data from different source and comparing to experiments, the dependence of proton emission on decay energy and spectroscopic factors is evaluated. Additionally, based on previous observations, the half-life of the possibly lighter proton emitter in the <i>fpg</i>-shell below has been theoretically predicted. Our results are compiled into a comprehensive dataset of half-lives for both experimentally confirmed emitters (50 < <i>Z</i> < 84) and theoretically predicted emitters (30 < <i>Z</i> < 50), providing a useful reference for future experimental investigations related to the proton drip line. The datasets presented in this paper, including our results of calculation, are openly available at https://www.doi.org/10.57760/sciencedb.27551 (Please use the private access link https://www.scidb.cn/s/zQzA3e to access the dataset during the peer review process).

Microplastic pollution in Chinese Rivers: A detailed analysis of distribution, risk factors, and ecological impact.

Unraveling the carbon fraction heterogeneity in China's Rivers: Hydrology, nutrients, and dam regulation.

Melissopalynology is the gold standard for authenticating honey but traditional microscopic analysis is time-consuming and subjective. This study evaluates a hybrid artificial intelligence approach to automate pollen classification using the comprehensive POLLEN73S dataset, which features 73 distinct pollen types from the Brazilian Savanna. To address class imbalance, the dataset was expanded to 7300 images using data augmentation. We extracted morphological features using three pre-trained deep learning models (ResNet50, EfficientNetB0, MobileNetV2) and classified them using 17 traditional machine learning algorithms. The hybrid model combining ResNet50 features with Linear Discriminant Analysis (LDA) achieved the highest accuracy of 97.00%. Error analysis indicated that misclassifications were concentrated among taxonomically similar genera, such as Serjania, due to shared exine structures. These results demonstrate that the proposed hybrid model offers a highly accurate and scalable solution for laboratory-based honey authentication, provided it is integrated with debris detection systems to handle real-world samples.

This study critically examines the environmental and social impacts of Nigeria’s fossil fuel dependency, focusing on pollution levels, health outcomes, and climate risk exposure within oil-producing regions. Utilizing a comprehensive dataset spanning 2015 to 2023, the analysis employs panel regression and instrumental variable techniques to address potential endogeneity and capture dynamic relationships. Results indicate that fossil fuel extraction and associated activities contribute to environmental degradation and increased vulnerability to climate risks, though the direct statistical significance of some effects remains limited, potentially due to data constraints. The findings underscore the complex interplay between economic development, environmental sustainability, and social wellbeing in resource-rich contexts. Policy recommendations emphasize the importance of strengthening regulatory frameworks, promoting renewable energy transitions, and enhancing community health infrastructure to mitigate adverse effects. The study contributes to the growing body of literature on sustainable development in fossil fuel-dependent economies and highlights avenues for future interdisciplinary research.

This paper provides a comprehensive review of existing literature on decision-making behaviour of retail investors in China and the US when facing the same risk events. By comparing cross-national studies, behavioral economics theories and empirical analysis to find differences in risk perception and herd behaviour. It shows that cultural background, regulatory and informational environments are key influencing factors. Which means that Chinese retail investors tend to react more quickly and emotionally to policies, while US retail investors are on the other side they based on company fundamental and more concern about long-tern returns. For example, the lack of direct comparative data undermines the persuasiveness of research conclusions to a certain extent; moreover, different studies adopt diverse modeling frameworks, making it difficult to effectively integrate and compare the research results. Based on this, future research should strive to construct clearer and more comprehensive cross-cultural datasets to provide a more solid data foundation for studies. Additionally, it is necessary to deeply explore the impact of psychological biases on the decision-making behaviors of Chinese and American retail investors, so as to further improve the relevant research system.

The aryl hydrocarbon receptor (AhR) plays a crucial role in mediating xenobiotic responses, as well as regulating broader metabolic, differentiation, and stress response programs. In this study, we present a comprehensive long-read RNA sequencing dataset that examines transcriptional changes in the HepaRG human cell line during differentiation induced by dimethyl sulfoxide (DMSO) and acute activation of the AhR with 3-methylcholanthrene (3-MC). We identified 946 genes that were differentially expressed between the NonDiff and Diff conditions (303 genes upregulated and 643 genes downregulated), and 1786 genes that showed differential expression between Diff and Ind conditions (961 genes upregulated and 825 genes downregulated). The acute induction of 3-MC produced a robust AhR signature, characterized by the robust induction of CYP1A1 and CYP1B1, along with a coordinated downregulation of several constitutive hepatic genes involved in drug metabolism (e.g., CYP3A4 and CYP2C8). To facilitate further analysis and reuse of our data, we have provided processed gene-level count matrices, transcript per million (TPM) tables, and detailed differential expression results, as well as analysis scripts. This resource supports research into AhR biology, pharmacogene regulation, and the development of methods for long-read transcriptomics in liver models.

Klebsiella pneumoniae (K. pneumoniae), a multidrug-resistant Gram-negative bacillus, represents a significant global health threat due to its role in hospital-acquired infections and the emergence of carbapenem-resistant hypervirulent strains. This study integrates the Drug Repurposing Knowledge Graph (DRKG) with molecular dynamics (MD) simulations to identify and validate stable structural segments of the KPHS_11890 gene, which encodes a membrane fusion protein of the AcrAB-TolC efflux pump that is critical for antibiotic resistance in K. pneumoniae. Using the PyKEEN framework, a knowledge graph embedding model was trained on a comprehensive dataset combining DrugBank, K. pneumoniae strain sequences, and NCBI databases, identifying KPHS_11890 as a top-ranked candidate (Hits@10 = 0.1602). The structural reliability of the target was first confirmed via rigorous quality assessment (Ramachandran plot, ERRAT, and ProSA), followed by triplicate 100-ns molecular dynamics simulations using GROMACS 2025. The integrated analysis of essential dynamics and free energy landscapes (FEL) revealed a thermodynamically stable core domain (residues 18-342) and a critical functional hinge region near residue 115. The structural rigidity of the core suggests minimized entropic penalties for inhibitor binding, while the identified hinge motion presents a specific mechanical vulnerability for allosteric locking. This integrated DRKG-MD approach not only efficiently pinpoints high-potential targets but also elucidates their biophysical mechanisms, providing a robust structural basis for designing novel inhibitors to overcome efflux pump-mediated resistance.

Synthetic Aperture Radar (SAR) images of the sea surface reveal a variety of oceanic and atmospheric phenomena. Automatically detecting and identifying these phenomena is essential for understanding ocean dynamics and ocean–atmosphere interactions. This study selected 2383 Sentinel-1 Wave (WV) mode images and 2628 Interferometric Wide swath (IW) mode sub-images to construct a semantic segmentation dataset covering 12 typical oceanic and atmospheric phenomena, with a balanced distribution of approximately 400 sub-images per category, culminating in a comprehensive dataset of 5011 samples. The images in this dataset have a resolution of 100 m and dimensions of 256 × 256 pixels. We propose Segformer-OcnP model based on Segformer for the semantic segmentation of these multiple oceanic and atmospheric phenomena. Experimental results demonstrate that Segformer-OcnP outperforms classic CNN-based models (U-Net, DeepLabV3+) and mainstream Transformer-based models (SETR, the original Segformer), achieving 80.98% mDice, 70.32% mIoU, and 86.77% Overall Accuracy, verifying its superior segmentation performance.

This study examines the impact of air pollution on firm-level import behavior in China, leveraging a comprehensive dataset spanning 2007 to 2016. By integrating city-level air quality data with firm-level production and trade records, the analysis employs a two-way fixed effects model and instrumental variable approach (using thermal inversion frequency as an instrument for PM₂.₅) to address endogeneity concerns. The results indicate that a 1% increase in PM₂.₅ concentration reduces firm imports by an average of 0.82%, primarily through two dominant mechanisms: suppressed labor productivity (mediated by health-induced efficiency losses) and liquidity crowding-out due to environmental investment demands. In contrast, the hypothesized import substitution effect—where firms might seek foreign intermediates amid rising domestic costs—is statistically insignificant, likely due to high switching costs and informational frictions. Heterogeneity analyses reveal stronger negative effects for non-state-owned enterprises, capital-intensive industries, and firms in inland regions. Notably, a countervailing increase in robot imports (5.3%) suggests strategic automation adoption to mitigate pollution-induced productivity declines. These findings underscore the need for differentiated environmental regulations and green trade incentives to balance sustainability goals with economic competitiveness

Accurately predicting the specific heat capacity of nanofluids is critical for optimizing their performance in engineering and industrial applications. This study explores twelve machine learning and deep learning models using conventional and stacking ensemble techniques. In the stacking framework, a linear regression model is employed as a meta-learner to improve base model performance. Additionally, two nature-inspired metaheuristic optimization algorithms-Particle Swarm Optimization and Grey Wolf Optimization-were used to fine-tune the hyperparameters of machine learning models. This research is based on a comprehensive dataset of 1,269 experimental nanofluid samples, with key inputs including nanofluid type (hybrid and direct), temperature, and volume concentration. To improve model generalization, data augmentation strategies inspired by polynomial/Fourier expansions and autoencoder-based methods were implemented. The results demonstrate that the stacked multi-layer perceptron model, integrated with linear regression, achieved the highest predictive accuracy, recording an R² score of 0.99927, a mean squared error of 466.06, and a root mean squared error of 21.58. Among standalone machine learning models, CatBoost was the best performer (R² score: 0.99923, MSE: 487.71, RMSE: 22.08), ranking second overall. The impact of metaheuristic optimization was significant; Grey Wolf Optimization, for instance, reduced the LightGBM model's mean squared error from 29386.43 to 6549.006. These findings underscore the efficacy of hybrid ML/DL frameworks, advanced data augmentation, and metaheuristic optimization in predictive modeling of nanofluid thermophysical properties, providing a robust foundation for future research in heat transfer applications.

A few models have been developed in recent years to predict all-cause mortality in patients with chronic kidney disease (CKD). However, many have been developed using inappropriate methods and have not been externally validated. This study aims to improve our previously validated tool for predicting 2-year all-cause mortality in stage 4-5 CKD patients by enlarging the training dataset, thereby enhancing its robustness, which is further supported by a second external validation. The Bayesian network-based 2-year all-cause mortality prediction tool was trained on a comprehensive national dataset, which was further enriched by incorporating data from a previous external validation. Internal performance was assessed using 10-fold cross-validation, while external validity was confirmed through a second validation on the CERRENE cohort. The discriminatory ability of the prediction tool was evaluated both internally and externally using accuracy, c-statistic, sensitivity, and specificity. The calibration of the external validation was visualized with a calibration curve. The prediction tool was developed using a training dataset of 1,061 patients (median age 72.3 years; 2-year mortality rate 21.2%) and externally validated using data from 409 patients (median age 77.5 years; 2-year mortality rate 17.6%) with CKD stage 4 or 5. The tool demonstrated satisfactory performance in both internal and external validation (accuracy: 77.2% and 77.8%; AUC-ROC: 0.76 and 0.74; sensitivity: 47.1% and 54.2%; specificity: 85.3% and 82.3%, respectively). The calibration curve demonstrated acceptable agreement between predicted and observed outcomes, and Brier score = 0.132. The updated prediction tool demonstrated satisfactory performance in both internal and external validation processes. Before it can be used in clinical practice, it must undergo national and international external validations, which are currently in progress.

The increasing electrification of mobility within smart cities has accelerated the need for intelligent energy management strategies that jointly address cost, emissions, and battery health. This study develops a health-aware hybrid reinforcement–predictive energy manager (H-RPEM) designed for photovoltaic–electric vehicle (PV-EV) microgrids. The proposed controller unifies model-based predictive optimisation with adaptive reinforcement learning to achieve both short-term operational efficiency and long-term asset preservation. A comprehensive dataset of solar generation, EV charging behaviour, and stochastic load profiles was employed to train and validate the hybrid control framework under realistic operating conditions. Quantitative results indicate that the proposed H-RPEM controller achieves an 18.7% reduction in total operating cost and a 22.5% decrease in carbon emissions, whilst maintaining the battery state-of-health above 0.95 throughout a 24 h operational cycle. When benchmarked against standard predictive control, the hybrid strategy converges 30-40 episodes faster and delivers a 25% improvement in reward stability, demonstrating enhanced robustness and learning efficiency. The results confirm that H-RPEM achieves robust and balanced performance across economic, environmental, and technical domains, establishing it as a scalable and health-conscious control solution for next-generation smart city microgrids.

Abstract This study examines the association between cash rewards for preventive health behaviors and subsequent health care expenditures under social health insurance schemes. We analyze a comprehensive dataset of all clients of a large German health insurance company, covering 1 year before and 5 years after the introduction of a cash bonus program. Program participation was associated with increased prevention efforts and reduced health care expenditures each year since its implementation. The estimated associations were more pronounced for individuals who participated in the program for multiple years. Statistically significant negative associations with health care expenditures are observed only in the first and fifth years. However, when dynamic selection is considered, the statistical significance of the associations in the fifth year becomes less clear. The estimated temporal effect dynamics highlight the complexity of strategies aimed at achieving long-term cost savings in health insurance.

Six soil profiles with metamorphic and weakly developed structure, formed on old and young Quaternary proluvial and colluvial deposits, were studied. A comprehensive dataset was obtained for over 50 elements (including more than two-thirds of the CRMs), across 20 genetic horizons of various types - metamorphic, gleyed, and contaminated with agglomeration dust, coke, and other materials. Due to the growing interest and knowledge of these raw materials, this study has derived statistical relationships between many of them in the context of increased values of most. The soils are geochemically and technogenically enriched, reflecting both natural and anthropogenic influences. Three genetic and one technogenic geochemical associations were identified through XRD, LA-ICP-MS, statistical and geochemical analyses (PCA, Pearson’s correlation, Concentration coefficients - CC). The plutonogenic association Hf–P–W–Nb–Ce/LREE–Sr–Bi–Ge+Ta–Th–Rb–Cs–Au, derived from the potassic-alkaline Buhovo–Seslavtsi pluton, shows strong correlations (r (P) = 0.46–0.86; r (W) = 0.50–0.98; r (Th) = 0.55–0.98). The ore-related association As–Mo–Co–Ni–Ga–Mn–Sb–Cu–Zn–U–V (r = 0.45–0.94) reflects proluvial material enriched with sulfide and uranyl-phosphate minerals. The lithophillyc association Ga–Y/HREE–Ce/LREE–Al–Ti–Sc–Si (r(Sc) = 0.63–0.88) represents a stable Al–Si–oxide-controlled system, while the technogenic Ba–Mn–Pb–Ag–Sr–Sb–Cu association (r(Ba) = 0.51–0.75; r(Pb) = 0.41–0.91) records up to 11-fold Pb and 4-fold Ba enrichment in surface horizons due to persistent metallurgical impact.

ABSTRACT This study presents a machine learning‐driven framework for the optimal design of kirigami‐based piezoelectric displacement sensors. We generated a comprehensive dataset using finite element analysis (FEA) by systematically varying key design parameters of the polyvinylidene fluoride (PVDF) film, including the number of cuts, joint configurations, and joint widths. An artificial neural network (ANN) with optimized hyperparameters is then trained to function as an inverse design model. This model accurately recommends the design parameters required to achieve user‐specified performance targets for two critical metrics: maximum displacement and displacement sensitivity. Experimental validation confirmed the model's high predictive accuracy, yielding coefficients of determination (R 2 ) of 0.980 for displacement and 0.950 for sensitivity. To compensate for the limited amount of data, k‐fold cross‐validation is conducted, allowing us to evaluate the model's robustness and generalization capability. Furthermore, as a practical demonstration, we fabricated composite sensors integrating elements with dual sensitivity settings, enabling sophisticated applications such as multi‐resolution motor control. Ultimately, this approach enables the rapid design of input devices with performance tailored to specific user objectives, offering significant advantages including improved operational efficiency and enhanced precision control.

ABSTRACT The skin, the largest organ of the human body, is vulnerable to numerous pathological conditions collectively referred to as skin lesions, encompassing a wide spectrum of dermatoses. Diagnosing these lesions remains challenging for medical practitioners due to their subtle visual differences, many of which are imperceptible to the naked eye. While not all lesions are malignant, some serve as early indicators of serious diseases such as skin cancer, emphasizing the urgent need for accurate and timely diagnostic tools. This study advances dermatological diagnostics by curating a comprehensive and balanced dataset containing 9360 dermoscopic and clinical images across 39 lesion categories, synthesized from five publicly available datasets. Five state‐of‐the‐art deep learning architectures—MobileNetV2, Xception, InceptionV3, EfficientNetB1, and Vision Transformer (ViT)—were systematically evaluated on this dataset. To enhance model precision and robustness, Efficient Channel Attention (ECA) and Convolutional Block Attention Module (CBAM) mechanisms were integrated into these architectures. Extensive evaluation across multiple performance metrics demonstrated that the Vision Transformer with CBAM achieved the best results, with 93.46% accuracy, 94% precision, 93% recall, 93% F1‐score, and 93.67% specificity. These findings highlight the effectiveness of attention‐guided Vision Transformers in addressing complex, large‐scale, multi‐class skin lesion classification. By combining dataset diversity with advanced attention mechanisms, the proposed framework provides a reliable and interpretable tool to assist medical professionals in accurate and efficient lesion diagnosis, thereby contributing to improved clinical decision‐making and patient outcomes.