Browse Articles
Discover research articles across all indexed journals
Nonlinear relationship between blood urea nitrogen to albumin ratio and mortality risk in older patients with cerebrovascular and cardiovascular diseases: An NHANES analysis
Background The blood urea nitrogen to albumin ratio (BAR) has emerged as a potential prognostic biomarker in elderly patients with cardiovascular and cerebrovascular diseases (CVDs). This study investigates the association between BAR and all-cause as well as cardiac mortality in this population. Methods We analyzed data from 4,113 elderly CVDs patients derived from the National Health and Nutrition Examination Survey (NHANES), with a mean follow-up of 82.4 months. Participants were categorized into three BAR groups: T1 (<3.55), T2 (3.55–5.00), and T3 (≥5.00). Weighted multivariable Cox regression assessed the association between BAR and all-cause mortality. The Fine and Gray competing risks model evaluated cardiac mortality, accounting for competing events. Hazard ratios (HRs) were calculated for continuous and categorical BAR. Subgroup, threshold effect, and sensitivity analyses were performed to confirm the robustness and explore nonlinear relationships. Results During follow-up, 2,178 all-cause and 752 cardiac deaths occurred. Continuous BAR was significantly associated with increased all-cause mortality (HR = 1.10, 95% CI: 1.07–1.13, p < 0.001). Compared to T1, the highest BAR group (T3) showed elevated all-cause mortality risk (HR = 1.33, 95% CI: 1.16–1.53, p < 0.001). Each unit increase in BAR corresponded to a 9% increase in all-cause mortality and a 14% increase in cardiac mortality. Threshold analysis revealed a nonlinear association with increased risk above specific BAR levels. Subgroup and sensitivity analyses further validated these findings. Conclusion BAR is a significant and independent predictor of all-cause and cardiac mortality in elderly patients with CVDs. Incorporation of BAR into clinical risk assessment may help improve identification of high-risk patients and support targeted interventions.
Applying machine learning to predict quality ANC determinants in Bangladesh: a BDHS-2022 cross-sectional study
Multimodal emotion recognition via adaptive high-order transforme network
Multimodal emotion recognition leverages multiple modalities to capture emotional cues more comprehensively, thereby improving the accuracy and robustness of emotion recognition. From the perspective of multimodal data and feature learning, reducing information redundancy in multimodal data and enhancing the discriminability of deep feature co-learning can effectively boost recognition performance. Based on this, this paper proposes a multimodal emotion recognition method based on an Adaptive High-order Transformer Network (AHOT). This method constructs Adaptive Selection Transformer block (AST) and Cross-modal Feature Fusion block (CMFF) for each modality branch, aiming to fully capture non-redundant feature representations from each modality and the interactions between modalities. In addition, a sparse high-order feature learning module is designed to enable the learning of highly discriminative high-order features across modalities. Experimental results on two multimodal emotion recognition datasets (IEMOCAP and CMU-MOSEI) demonstrate that, compared with several related methods, the proposed AHOT effectively improves emotion recognition accuracy. Moreover, ablation studies and parameter analyses further validate the effectiveness of AHOT.
Eco-friendly synthesis of silver nanoparticles: multifaceted antioxidant, antidiabetic, anticancer, and antimicrobial activities
Abstract Diabetes, cancer, and multidrug-resistant bacteria are major global health threats, driving the search for novel, safe, and affordable therapeutics. Here, silver nanoparticles (Ag-NPs) were biosynthesized using the aqueous extract of Rosmarinus officinalis L. via a sustainable, eco-friendly, and cost-effective green approach. Characterization by UV–Vis, FT-IR, EDX, XRD, TEM, SEM, TGA, DLS, and zeta potential confirmed the formation of well-dispersed, spherical, crystalline nanoparticles with an average size of 60.5 nm, uniform morphology, and high thermal stability. The Ag-NPs displayed concentration-dependent multifunctional activities. Antibacterial assays revealed strong effects against both standard and MDR strains, including Bacillus subtilis , Staphylococcus aureus , Pseudomonas aeruginosa , Escherichia coli , and Klebsiella pneumoniae (inhibition zones: 11.7–29.7 mm). Potent antioxidant activity was observed with an EC₅₀ of 7.81 µg mL⁻ 1 , close to ascorbic acid (3.27 µg mL⁻ 1 ). Antidiabetic activity reached 85.5% (α-amylase) and 82.6% (α-glucosidase) inhibition at 1000 µg mL⁻ 1 , comparable to acarbose (97.5% and 96.3%). Moreover, Ag-NPs showed selective cytotoxicity against MDA and PANC-1 cells (IC₅₀: 177.2 and 115.3 µg mL⁻ 1 ), with lower toxicity toward Vero and Wi38 normal cells (IC₅₀: 233 and 207 µg mL⁻ 1 ). These findings highlight the promise of R. officinalis –mediated Ag-NPs as multifunctional nanomaterials for biomedical applications.
Correction: Validation of the person-centered maternity care scale at governmental health facilities in Cambodia
Job Stress and Ethical Decision-Making Among Healthcare Workers in Post-Pandemic Infectious Disease Control
Correction: Association between self-administrated prophylactics and SARS-CoV-2 infection among traditional market vendors from the Central Highlands of Peru: A nested case-control study
Genome of an early Okhotsk individual reveals ancient admixture between Jomon and Kamchatka lineages
Abstract The prehistoric Okhotsk culture was distributed along the southern coastal regions of the Sea of Okhotsk during the late first millennium AD. A previous study that performed whole-genome sequencing of a late Okhotsk individual suggested two migration waves from the Russian Far East to northern Japan. The first wave is estimated to have originated from the Kamchatka Peninsula around 2000 years before present (BP), and the second from the Amur Basin around 1600 BP. These findings suggest the past existence of an admixed hypothetical population between the Kamchatka and indigenous Jomon lineages in northern Japan between 2000 and 1600 BP, although direct genetic evidence has not yet been obtained. Here, we present the genome data of an early Okhotsk individual (NAT004) excavated from northern Japan. Admixture modelling reveals that the genome can be explained as a mixture of Kamchatka and Jomon ancestries, providing direct support for the existence of the previously hypothesized hypothetical population. This result offers new insights into the prehistoric population dynamics of northern Japan and contributes to the broader understanding of its archaeological and anthropological history.
Novel insights into neuropathy: The impact of prolonged hyperglycemia on long non-coding RNA expression
Multiple evidence suggests that type 1 diabetes triggers perturbations in the nervous system both in human patients as well as in animal models of the disease. These perturbations are likely controlled by the expression of long non-coding RNAs (lncRNAs) and are present both in peripheral and central nervous system. To dissect the role of lncRNAs in diabetes-affected nervous system malfunctions, we conducted a comparative analysis of spinal cord transcriptome profiles between long-term (six months of duration) diabetic versus non-diabetic mice. The analysis of RNA sequencing data revealed that of 277 unique differentially expressed transcripts, 201 were up-regulated and 76 were down-regulated in the diabetic lumbar spinal cord. We also observed elevated expression of Snhg15 lncRNA in diabetic spinal cord. The in-depth data analysis revealed differential expression of lncRNAs involved in the PI3K-Akt signaling pathway (KEGG: mmu04151) as well as substantial differences in several biological processes such as developmental process, cell communication, anatomical structure development and multicellular organismal process. Our analysis verified the role of lncRNAs in mouse spinal cord during the progression of type 1 diabetes and confirmed molecular alternations in the spinal cord occurring in the course of diabetic neuropathy.
Instability mapping of Dhaka-Kasiani-Gopalganj railway line in Bangladesh with InSAR time series analysis
Extended Directed Fuzzy Social Network Analysis: A framework and application to curriculum networks in Chinese vocational education
Due to the differences in node types and the diversity of network relationships, Fuzzy Social Network Analysis (FSNA) needs to specifically address the issues of network heterogeneity and relationship ambiguity. To address this challenge, we propose a new analytical framework called Extended Directed Fuzzy Social Network Analysis Framework (EFDSNAF), which establishes the Typical Connections to assist in evaluating the fuzzy network. Meanwhile, in the area of fuzzy centrality measures, we enhance the variability of the Fuzzy Intensity of Path and propose the term “Total Fuzzy Intensity of Path” (TFIP), considering the distinct characteristics of different networks may lead to variations in path intensity expressions and differences in closeness relationships. Based on this, we optimize the computational methods for fuzzy betweenness centrality and fuzzy closeness centrality, with the efficacy of the method being demonstrated through two examples. Then we applied EDFSNAF to analyze Chinese vocational education curriculum network, with empirical investigation on the Urban Rail Transit Operation and Management Major (URTOMM) and Urban Rail Transit Communication and Signaling Technology Major (URTCSTM). Through EDFSNAF, core courses were identified, and network metrics for different majors effectively captured essential disciplinary differences between the two fields, clearly demonstrating the effectiveness of EDFSNAF.
Intelligent monitoring system for quality of life of colostomy patients based on deep learning and AR
The impact of patent activity on idiosyncratic volatility in U.S. pharmaceutical companies
This study examines the impact of patent activity on the idiosyncratic volatility (IVOL) of U.S. pharmaceutical companies, addressing a critical gap in the literature on the relationship between innovation and firm-specific risk. Using panel data from Thomson Reuters/Refinitiv covering 2,910 firms over 2005−2024, we employ the Fama-French 5-factor model to isolate firm-specific volatility and analyze how patent events and pharmaceutical development activities affect stock price risk. Our findings reveal a complex relationship between innovation and volatility that varies by development stage. While patent activity overall reduces idiosyncratic volatility, early and mid-stage development projects (Phase I and II) initially increase firm-specific risk, reflecting inherent uncertainties in drug development. Conversely, newly launched products significantly reduce volatility, indicating that risk mitigation occurs primarily at commercialization. These relationships remain robust during crisis periods, including the 2008−09 financial crisis and COVID-19 pandemic. The results provide valuable insights for investors seeking to assess pharmaceutical investment risks, managers optimizing innovation portfolios, and policymakers designing intellectual property frameworks. The study’s focus on the U.S. market and reliance on patent counts rather than quality measures suggest important avenues for future research across different regulatory environments and innovation metrics.
Instanton-like effect caused by qubit-boson interaction in light of quantum simulation
Citation proximus: The role of social and semantic ties on citations
Despite being considered as key indicators of research impact, citations are shaped by factors beyond intrinsic research quality—such as including prestige, social networks, and research topics. While the Matthew Effect explains how prestige accumulates, our study contextualizes this by showing that other mechanisms also play a role in citation accumulation. Analyzing a large dataset of U.S. economic (N = 43,467) and their citation linkages (N = 264,436), we find that close ties in the collaboration network are the strongest predictor of citations, closely followed by semantic similarity between citing and cited papers. This suggests that citations are not only driven by prestige but are strongly affected by f social networks and intellectual proximity. Prestige remains an important factor affecting citations for highly cited papers, but for most papers, proximity—both social and semantic—plays a more significant role. These findings redirect focus from extreme cases of highly cited research to the overall citation distribution, which influences most scientists’ career paths and knowledge production. Recognizing the diverse factors influencing citations is critical for science policy and for developing a reward system of science that is fairer and reflects a diversity of contributions to science.
Impact of aging on gut-lung-adipose tissue interactions and lipid metabolism during influenza infection in mice
Abstract Influenza remains a major threat to human health, especially for the elderly. Aging leads to substantial changes to lung function, gut microbiota, and white adipose tissue (WAT)—a key endocrine organ regulating energy balance and lipid metabolism. In the current study, we performed a multi-omics analysis to investigate how influenza impacts the gut-lung-adipose tissue axis differently with age at days 2, 4, 7, 14, and 28 post-infection (dpi). Compared to young-adult mice, aged mice experienced worse disease outcomes following infection, along with distinct WAT alterations, including impaired browning, heightened inflammation, and reduced innate immune cell recruitment. Age-related differences were also evidenced in infection-driven shifts in gut microbiota. Akkermansia levels rose only in young mice from 4 dpi, while Faecalibaculum and Muribaculum expanded exclusively in aged mice at 7 dpi, the only timepoint at which their abundance correlated with lung pathology. Serum metabolomics at 7 dpi also revealed age-dependent metabolic responses to infection. Compared to their non-infected counterparts, young mice had lower levels of p-Cresol-sulfate and Indoxyl-sulfate alongside higher triglycerides, whereas aged mice showed disrupted glycerophospholipid metabolism. By pinpointing specific gut bacteria as potential probiotics and identifying lipid pathways associated with disease progression, these findings could lead to the development of targeted, age-specific strategies to mitigate influenza severity in the elderly.
Dual-branch differential channel hypergraph convolutional network for human skeleton based action recognition
Graph Convolutional Networks (GCNs) perform well in skeleton action recognition tasks, but their pairwise node connections make it difficult to effectively model high-order dependencies between non-adjacent joints. To address this issue, hypergraph methods have emerged with the aim of capturing complex associations between multiple joints. However, existing methods either rely on static hypergraph structures or fail to fully exploit feature interactions between channels, limiting their ability to adapt to complex action patterns. Therefore, we propose the Dual-Branch Differential Channel Hypergraph Convolutional Network (DBC-HCN), which leverages hypergraphs’ ability to represent a priori non-natural dependencies in skeletal structures. It extracts spatio-temporal topological information and higher-order correlations by integrating static and dynamic hypergraphs, leveraging channel optimization and inter-hypergraph feature interactions. Our network comprises two parallel streams: a Spatio-Temporal Dynamic Hypergraph Convolutional Network (ST-HCN) and a Channel-Differential Hypergraph Convolutional Network (CD-HCN). The Spatio-Temporal Dynamic Hypergraph Convolutional stream is mainly based on the natural topology of the human skeleton, and uses dynamic hypergraphs to model the dependencies of skeletal points in spatio-temporal dimensions, so as to accurately capture the spatio-temporal characteristics of the movements. In contrast, Channel-Differential Hypergraph Convolutional stream focuses on the feature differences between different channels and extracts the characteristics of motion changes between individual skeletal points during action execution to enhance the portrayal of action details. In order to enhance the network’s representational capability, we fuse the dual streams with different action feature representations, so that the Spatio-Temporal Dynamic Hypergraph Convolutional stream and the Channel-Differential Hypergraph Convolutional stream learn from each other’s representations to better enrich the action feature representations. We experiment the model on three datasets, Kinetics-Skeleton 400, NTU RGB + D 60 and NTU RGB + D 120, and the results show that our proposed network is more competitive. The accuracy reaches 96.9% and 92.7% for the cross X-View and X-Sub benchmarks of the NTU RGB + D 60 dataset, respectively. Our code is publicly available at: https://github.com/hhh1234hhh/DBC-HCN .
Actin cytoskeleton dynamics affect replication of human Metapneumovirus
Sequencing and analysis of 131 SARS-CoV-2 isolates in previously sampled and unsampled regions of Jordan from 2020 to 2023
The Hashemite Kingdom of Jordan remains an understudied country for next generation sequencing analysis of SARS-CoV-2 genomes collected during the 2019 pandemic. Here we provide 131 additional reference genomes collected between 2020–2023 from SARS-CoV-2-positive patients across Jordan. Phylogenetic analysis supports existing pandemic narratives of changing clade dominance over time and adds genomes in novel Jordanian locations and timepoints to make Jordan SARS-CoV-2 databases more comprehensive. Samples from the less-sequenced cities of Ajloun, Jaresh, Karak, and Madaba identified previously unreported lineages while Amman, Irbid, and Zarqa have existing sequencing efforts bolstered. Despite many incomplete patient records and a relatively small sample size, we observe interesting symptom patterns that support existing global and Jordanian pandemic narratives. We note how in-country COVID-19 pandemic genomic studies showcase Jordan’s efforts to expand next generation sequencing capabilities, especially through the leveraging of EDGE COVID-19, a bioinformatics platform for performing rapid, batched analysis of SARS-CoV-2 sequencing that streamlines sample processing prepared from a network of hospital locations.