Browse Articles
Discover research articles across all indexed journals
Hybrid deep learning architecture for scalable and high-quality image compression
The small molecule LOXL2 inhibitor SNT-5382 reduces cardiac fibrosis and achieves strong clinical target engagement
Application of virtual monochromatic images in bone mineral density measurements: a phantom study
Donor liver natural killer cells ameliorate liver allograft rejection via inducing apoptosis of alloreactive CD8+ T cells generating CD4+CD25+ regulatory T cells
Transforming language research from classic desktops to virtual environments
Abstract Virtual Reality (VR) offers novel opportunities for investigating human perception beyond conventional laboratory settings, facilitating the study of naturalistic behavior with controlled virtual environments. To benefit from this technology, the foundational aspects must be compared to traditional personal computer (PC) monitor setups. The validity and reliability of stimuli presentation and response collection must be established to ensure that any findings can be attributed to experimental variables, not the method. To address this, we designed a single-word recognition (lexical decision) task administered in both VR and PC monitor setups. Stimulus presentation was controlled across tasks, visual angles were matched, and responses were gathered via VR controllers in both settings. Results replicated the lexicality effect (i.e., faster word than pseudoword reading) in both setups. Reaction times and error rates showed no significant differences between VR and PC monitor setups, underscoring VR’s utility for collecting reliable behavioral data in language studies. These results demonstrate that VR-derived mental chronometry measures yield findings comparable to conventional methods, establishing a benchmark for future immersive research.
N-acetylcysteine reduces the hepatic complications of social isolation stress through modulation of interleukin 1 and 6 gene expression and liver enzymes in mice
3D-printed fake wasps help explain bad animal mimicry
Analysis and implementation of variable frequency controlled dynamic wireless charging system with half-bridge multi-leg converter topology
Abstract Resonant Inductive Power Transmission (RIPT) represents a cutting-edge Wireless Power Transfer (WPT) technology, emerging as a secure and practical solution for charging electric vehicles (EVs). While Dynamic Wireless Charging Systems (DWCS) reduce the need for large batteries compared to static charging, they entail higher initial investments. This study introduces an innovative approach to DWCS utilizing a half-bridge-based multi-legged inverter configuration. Each leg of the inverter functions independently for a transmitter coil, effectively reducing overall system costs. In this method, the Variable Frequency Control Technique (VFCT) is introduced in half-bridge DWCS with S-S and LCC-S compensations. Analyzed its VFCT on the DWCS. In addition, explores the impact of square and rectangular coils on the proposed approach, coupled with an analysis of coil gaps’ effects on receiving power. Through an exploration of the half-bridge multi-legged DWCS and a comprehensive evaluation of coil gap and size influences, this study provides valuable insights for optimizing RIPT technology to achieve efficient and cost-effective EV charging.
Microbial communities associated with plastic fishing nets: diversity, potentially pathogenic and hydrocarbon degrading bacteria
Abstract Abandoned, lost or otherwise discarded fishing gear (ALDFG) represent a major source of marine plastic litter pollution. Similar to other plastic litter, these items can provide a new surface for the growth of biofilms harboring distinct microbial communities, containing potential opportunistic pathogens or pollutant-degrading microorganisms. While knowledge is increasing for marine plastic litter and microplastic-associated biofilms, there is a gap on the plastisphere research for fishing gear. This study aimed to comprehend the structure and dynamics of the microbial communities attached to plastic fishing nets, mimicking a scenario when lost at sea, but also to assess if polymer type can influence these communities. For that, a one-year in situ experiment was employed inside a recreational marina (port of Leixões, Portugal), using 3 types of plastic fishing nets (Braided Polyethylene (PE), Braided Nylon and Thin Nylon) submersed in the seawater. Seasonal samplings of nets and surrounding seawater were performed for microbial community analysis by 16 S rRNA metabarcoding. One month-old-nets samples were additionally collected for cultivation of bacterial strains in the laboratory. In general, microbial communities found in the biofilms attached to fishing nets were taxonomically distinct and more diverse, when compared to the surrounding seawater. Biofilm communities were not shaped by the polymer type, instead, they displayed a succession pattern over time. Biofilm communities were predominantly composed of the phyla Proteobacteria, Bacteroidetes and Verrucomicrobiota. Additionally, the families Sphingomonadaceae, Rubritaleaceae, Rhizobiaceae and Saprospiraceae were specifically associated with fishing net biofilms. From the 3 nets, a total of 123 bacterial strains from 46 bacterial genera were recovered. The genera Acinetobacter, Bacillus, Rhodococcus, Shewanella, Streptomyces and Vibrio were common to all nets. Commonly associated hydrocarbon and plastic - degrading taxa were highly abundant in the biofilm communities (> 2% abundance) and some were even possible to cultivate in laboratory. In addition, biofilm communities presented as well, potentially pathogenic genera, such as Clostridium and Mycobacterium, but in low abundances (< 1%). With this work, a deeper knowledge on the plastisphere associated with different plastic fishing gear was obtained, along with the isolation of bacterial strains with potential for future exploration of plastic biodegradation.
Survival outcomes and prognostic factors of early-onset and late-onset metastatic esophageal cancer: a population-based study
Identifying individuals from their brain natural frequency fingerprints
Abstract Neural oscillations are critical for brain function and cognition. Thus, identifying the typical or natural frequencies of the brain is an important step in understanding its functional architecture. Recently, a data-driven algorithm has been developed for mapping these frequencies throughout the cortex, free of anatomical and frequency-band constraints, but its robustness is limited to group-level analyses. Here, we adapt this algorithm to improve the single-subject maps of natural frequencies derived from magnetoencephalography and validate them using the fingerprinting technique. Modifications to the original method included (1) increasing the number of power spectra assigned to each k-means cluster, and (2) smoothing across neighboring voxels. Our results show high accuracy in individual identification within single sessions and across sessions separated by over four years. This demonstrates the stability and reliability of the single-subject mapping of natural frequencies, enhancing opportunities for identification of pathological variations in the intrinsic oscillatory activity of individuals.
Mechanisms of delayed ischemia/reperfusion evoked ROS generation in the hippocampal CA1 zone of adult mouse brain slices
Abstract ROS overproduction is an important contributor to delayed ischemia/reperfusion induced neuronal injury, but relevant mechanisms remain poorly understood. We used oxygen–glucose deprivation (OGD)/reperfusion in mouse hippocampal slices to investigate ROS production in the CA1 pyramidal cell layer during and after transient ischemia. OGD evoked a 2-stage increase in ROS production: 1st—an abrupt increase in ROS generation starting during OGD followed by a marked slowing; and 2nd—a sharp ROS burst starting ~ 40 min after reperfusion. We further found that a slight mitochondrial hyperpolarization occurs shortly after OGD termination. Consequently, we showed that administration of low dose FCCP or of FTY720 (both of which cause mild, ~ 10%, mitochondrial depolarization), markedly diminished the delayed ROS burst, suggesting that mitochondrial hyperpolarization contributes to ROS production after reperfusion. Zn 2+ chelation after OGD withdrawal also substantially decreased the late surge of ROS generation—in line with our prior studies indicating a critical contribution of Zn 2+ entry into mitochondria via the mitochondrial Ca 2+ uniporter (MCU) to mitochondrial damage after OGD. Thus, reperfusion-induced mitochondria hyperpolarization and mitochondrial Zn 2+ accumulation both contribute to mitochondrial ROS overproduction after ischemia. As these events occur after reperfusion, they may be amenable to therapeutic interventions.
Entanglement and Bell inequality violation in vector diboson systems produced in decays of spin-0 particles
Targeting three United States priority populations of people who smoke with educational nicotine messages using curiosity-eliciting strategies
Reflow solder flux residue and humidity interaction: investigation using real PCBA component designs
Exploring the link between the ZJU index and sarcopenia in adults aged 20–59 using NHANES and machine learning
Scalable evaluation framework for retrieval augmented generation in tobacco research using large Language models
Abstract Retrieval-augmented generation (RAG) systems show promise in specialized knowledge domains, but the tobacco research field lacks standardized assessment frameworks for comparing different large language models (LLMs). This gap impacts public health decisions that require accurate, domain-specific information retrieval from complex tobacco industry documentation. To develop and validate a tobacco domain-specific evaluation framework for assessing various LLMs in RAG systems that combines automated metrics with expert validation. Using a Goal-Question-Metric paradigm, we evaluated two distinct LLM architectures in RAG configurations: Mixtral 8 × 7B and Llama 3.1 70B. The framework incorporated automated assessments via GPT-4o alongside validation by three tobacco research specialists. A domain-specific dataset of 20 curated queries assessed model performance across nine metrics including accuracy, domain specificity, completeness, and clarity. Our framework successfully differentiated performance between models, with Mixtral 8 × 7B significantly outperformed Llama 3.1 70B in accuracy (8.8/10 vs. 7.55/10, p < 0.05) and domain specificity (8.65/10 vs. 7.6/10, p < 0.05). Case analysis revealed Mixtral’s superior handling of industry-specific terminology and contextual relationships. Hyperparameter optimization further improved Mixtral’s completeness from 7.1/10 to 7.9/10, demonstrating the framework’s utility for model refinement. This study establishes a robust framework specifically for evaluating LLMs in tobacco research RAG systems, with demonstrated potential for extension to other specialized domains. The significant performance differences between models highlight the importance of domain-specific evaluation for public health applications. Future research should extend this framework to broader document corpora and additional LLMs, including commercial models.