Browse Articles

Models for predicting the risk of bloodstream infections associated with peripherally inserted central venous catheters: A scoping review

Background Peripherally inserted central venous catheters (PICC) associated bloodstream infections is a type of central line-associated bloodstream infection (CLABSI), the purpose of this scoping review was to analyse and summarize the risk prediction models for PICC-CLABSI to provide insights for clinical nursing practice. Methods A scoping review was conducted from six bibliographic databases including CNKI, Wanfang Database, VIP Chinese Journal Database, PubMed, Embase, and Web of Science, from inception to November 5, 2024. Screening was performed and relevant data was extracted independently by two researchers, the risk of bias was assessed using the Prediction model Risk Of Bias Assessment Tool (PROBAST). Results Eight studies met the inclusion criteria, which included two score models, six nomograms, and one ELM model. The included studies exhibited a high risk of bias, mainly due to methodological heterogeneity. Four models underwent external validation, two were assessed for goodness of fit. The most frequently identified predictors including catheter indwelling time, maintenance cycle/frequency, multilumen catheters, PICC parenteral nutrition, diabetes, and malignant tumors. Conclusion The risk prediction models for PICC-CLABSI demonstrated strong predictive performance. Future research should carefully address all elements of PROBAST framework during study design phase. This will facilitate both internal and external validation, as well as differentiation, calibration, and evaluation of clinical practicality. The ultimate objective is to develop a PICC-CLABSI risk prediction model that exhibits low bias risk, robust predictive performance, and high clinical applicability.

Ferroelectricity of wurtzite Be1−<i>x</i>Mg<i>x</i>O from first-principles calculation

Nitride ferroelectrics such as Al1−xScxN are emerging as promising ferroelectrics for nonvolatile memory technologies due to their high thermal stability and large remnant polarization. However, their high coercive field close to the breakdown limit causes high leakage current, which limits their device applications. Here, we use first-principles calculations to investigate wurtzite Be1−xMgxO as a promising oxide-based ferroelectric alternative. We show that Be1−xMgxO alloys maintain structural stability in the wurtzite phase up to x = 0.83 and exhibit an ultrawide bandgap over 6.2 eV across all compositions, which provides high intrinsic breakdown fields (12–71 MV/cm). The polarization switching barrier decreases significantly with increasing Mg while the coercive field remains well below the intrinsic dielectric breakdown field. Our findings predict that wurtzite Be1−xMgxO is an alternative ferroelectric oxide that has the potential to offer low leakage current and a widely tunable range of polarization and switching characteristics while preserving high dielectric strength.

Tailoring the Oxygen Vacancy Distribution in Se‐Doped RuO_x to Enhance Its Stability in Acidic Water Electrolysis

AbstractDeveloping durable ruthenium (Ru)‐based catalysts for proton exchange membrane water electrolyzer (PEMWE) remains challenging due to irreversible Ru dissolution and lattice oxygen instability. Although elemental doping is a general method to improve stability, it inadvertently induces oxygen vacancies (VOs), which are randomly distributed in the nanocatalyst. Notably, the impact of VO distribution on the stability of Ru‐based catalysts remains unresolved. Herein, we synthesized the Se‐doped Ru oxide via annealing the mixture of ruthenium (III) chloride and selenium (Se) in the air (Ur‐Se‐RuOx) with the presence of urea, showing the VOs distributed away from Se dopants, which is significantly different from the Se‐doped Ru oxide synthesized without urea (Se‐RuOx), showing VOs distributed relatively close to the Se dopants. The Ur‐Se‐RuOx demonstrates superior oxygen evolution reaction performance over Se‐RuOx. Particularly, Ur‐Se‐RuOx delivers a low working voltage (1.62 V@1 A cm−2) and excellent durability (&gt;1000 h@200 mA cm−2) in PEMWE tests. Experimental and theoretical results reveal that VOs engage in long‐range cooperation with spatially decoupled Se dopants in Ur‐Se‐RuOx, synergistically enhancing reaction kinetics via an intramolecular oxygen coupling mechanism, while inhibiting the lattice oxygen mechanism and suppressing Ru dissolution, which demonstrates a new strategy to break the activity–stability trade‐off in promising Ru‐based catalysts.

Correlation between stress hyperglycemia ratio and risk of acute respiratory distress syndrome in patients with sepsis: a MIMIC-IV cohort study

Religion or class? Measuring voting clustering on religious and socioeconomic lines in US presidential elections

Electoral behavior in the United States is shaped by more than geography and economics as it is deeply intertwined with cultural identity. Here, we quantify how voting patterns in the 2016, 2020 and 2024 presidential elections cluster not only across neighboring counties but also along shared religious and socioeconomic lines. By computing Moran’s I, a standard measure of spatial autocorrelation, under four distinct “neighborhood” definitions (physical borders, dominant religion, income bracket and urbanization level), we show that counties sharing a majority faith vote in strikingly similar ways, second only to contiguous geography. In contrast, grouping by household income or urban status yields markedly weaker clustering. These findings reveal that cultural networks, embodied by religious affiliation, exert a stronger influence on aggregate voting behavior than class differences or the urban–rural divide. Our approach highlights the power of simple network models grounded in social traits to illuminate the dynamics of political polarization and suggests new pathways for understanding how cultural identity shapes large scale electoral outcomes.

Surface-enhanced Raman scattering for biosensing and early diagnosis

Biosensing plays a crucial role in precision biomarker detection, enabling real-time monitoring and physiological health analysis, particularly in early diagnosis. Traditional biosensing techniques, such as electrochemical, fluorescence, and colorimetric methods, have shown remarkable performance. However, their reliability is often susceptible to nonspecific interference, may produce false-positive results, and is typically limited to the detection of known targets, restricting their applicability in addressing ambiguous or unknown biological problems. Surface-enhanced Raman scattering (SERS) offers a powerful alternative, combining ultra-high sensitivity and molecular specificity. By supporting both label-free and label-based strategies, SERS broadens the detection scope, making it a promising platform for next-generation biosensing. In this Perspective, we aim at specific problems, introduce why SERS can do biosensing, how to advance SERS-based biosensing, and what SERS can do for early diagnosis. Finally, we discuss strategies to bridge the gap between laboratory research and clinical applications, emphasizing the important role of standardization and artificial intelligence in driving the practical transformation of SERS-based biosensing. Through these advancements, we aim to address existing limitations and further propel the evolution of SERS in biosensing and early diagnosis.

Bionic Design of Ni⁴⁺ Lewis Acid Site Based on Selective Seawater Oxidation

AbstractThe side reaction caused by chloride ions and the toxicity to the active site have always been the hindrance to the electrocatalyst of the oxygen evolution reaction (OER) for seawater splitting. Herein, inspired by the early flowering of damaged plants, we designed a catalyst rich in oxygen vacancies (Ovac) and proved that Ovac can accelerate the formation of Ni3+ and further oxidize it to Ni4+, which we named as the “ripening” mechanism of Ovac. Ovac reduces the hydrogen proton desorption energy by regulating local charge redistribution, thus realizing the rapid transformation of Ni2+→Ni3+→Ni4+. Meanwhile, the hard Lewis acid Ni4+ has strong selectivity to OH−, avoiding the competitiveness and corrosiveness of chloride ions in seawater. This work provides an effective strategy for the simple and rapid construction of self‐rebuilding high‐valence Ni4+ electrocatalysts, and is expected to provide guidance for the development of seawater electrolysis.

Linking habitat preferences and fitness across scales for a relict bird species of the southern Andes

Abstract Animals select their habitats from available resources in a way that should maximize fitness, and thus habitat preferences are generally predicted to be adaptive. However, there may be a mismatch between habitat preferences and fitness due to factors such as limited availability or disturbance of nesting habitats. In this study, we examine whether preferred nesting habitat attributes are linked to fitness (nest survival and number of fledglings) of the White-throated Treerunner ( Pygarrhichas albogularis ), an obligate excavator and tree cavity nester, across four spatial scales: (1) cavity -for fitness influence only-, (2) nest-tree, (3) forest-stand, and (4) landscape. During eight breeding seasons (October to February), between 2010 and 2018, we found and monitored 65 Treerunner nests in Andean temperate forests, Chile. We obtained four main results. First, we found a multiscale response for both habitat preferences and fitness: variables at both nest-tree and landscape scales were the most influential for nesting habitat preferences, while variables at both cavity and nest-tree scales were the most influential for fitness. Second, the probability that a given habitat is used for nesting increased with larger trees, advanced tree decay classes, and forest cover. Third, nest survival was positively related with cavity entrance diameter, height, and distance from the forest edge. Fourth, the number of fledglings increased with south-oriented cavities and decay class, excepting for old dead trees where the breeding outcomes decreased. Combined, our results suggest a general match between habitat preferences and fitness, with a mismatch occurring with trees in advanced decay. The fact that the match occurs in areas with live unhealthy trees and recently dead trees, and a high forest cover, highlight the importance of (a) old-growth forests, as they comprise the best integration of multiscale habitat attributes for this species, and (b) maintaining the continuity of forest cover together with both live unhealthy and recently dead trees in managed and/or second-growth forests.

Effect of fluid intake on cognitive function in older individuals: A prospective study

Background Adequate fluid intake is essential for maintaining cognitive health in older adults. However, two key questions remain unanswered before recommending increased fluid consumption: (1) whether the relationship between fluid intake and cognitive improvement is linear and (2) the underlying mechanisms that mediate this association. Methods and Findings Thirty-three older adults residing in a geriatric health service facility and receiving nursing care were enrolled in this study. Fluid intake was recorded as part of routine clinical practice. Cognitive function was assessed twice during their stay using the Japanese version of the Mini-Mental State Examination (MMSE-J). Additionally, cerebral blood flow was evaluated bilaterally in the common carotid arteries using ultrasonography, with assessments conducted approximately 82.6 ± 14.9 days apart. Relationships among fluid intake per lean body mass (LBM), changes in MMSE-J scores, and ultrasonographic parameters were analysed using Spearman’s linear correlation analysis with non-parametric bootstrapping. Correlation analyses revealed a positive linear association between fluid intake and improvement in MMSE-J scores [P(FDR) = 0.012] when the intake was less than 42 mL/LBM (kg) per day. Furthermore, fluid intake was inversely correlated with the resistance index in the right common carotid artery [P(FDR) = 0.046], indicating altered cerebral blood dynamics. The main limitations of our study include (1) the inability to evaluate baseline hydration status or fluid intake prior to facility admission due to clinical constraints and (2) the observational design precluding causal inference between fluid intake, cognitive changes, and cerebral blood flow parameters. Conclusions Within moderate intake ranges, fluid consumption was linearly associated with cognitive improvement, an effect that appears to be mediated by changes in cerebral haemodynamic.

High capacitance behavior in Sr0.6Ba0.4Nb2O6-based tungsten bronze ferroelectrics

Dielectric capacitors with superior energy storage performance are attracting growing attention due to ultrahigh power density and rapid charge/discharge speed. In this work, perovskite Sr0.6Ba0.4TiO3 (SBT) is introduced into tungsten bronze Sr0.6Ba0.4Nb2O6 (SBN), maintaining an identical Sr/Ba ratio, to enhance the energy storage performance of Bi/Na/Pb-free (1 − x)SBN-xSBT ferroelectrics. The sample with increasing x ≤ 0.3 forms a solid solution corresponding to the unfilled–filled tungsten bronze structure transition, accompanied by dramatically decreased grain anisotropy and average grain size. However, the perovskite second phase gradually appears, and the average grain size further decreases slightly when x &amp;gt; 0.3. Such phase and microstructure evolution induces monotonously enhanced relaxor degree γ and breakdown strength Eb around x = 0.4. As a result, this optimal composition shows high recoverable energy storage density Wrec = 9.5 J/cm3, energy storage efficiency η = 82.1%, both of which vary less than 7% within 20–150 °C. Encouragingly, excellent charge–discharge performances, t0.9 = 23 ns and WD = 6.2 J/cm3, are obtained. This work provides insights into developing high energy storage performance tungsten bronze oxide through phase and microstructure design.

Establishment of a humanized mouse model of pulmonary fibrosis for advancing drug validation strategies

Abstract Idiopathic pulmonary fibrosis (IPF) is a progressive interstitial lung disease characterized by excessive extracellular matrix deposition and limited therapeutic options. Conventional preclinical models, such as the bleomycin (BLM)-induced lung fibrosis (BILF) model in immunocompetent mice, fail to replicate the complexity of the human immune environment and are thus unsuitable for evaluating human-specific antibody and cell therapies. To overcome this limitation, we developed two humanized mouse models with functional human immune components: one engrafted with peripheral blood mononuclear cells (PBMCs) and the other with hematopoietic stem cells (HSCs). Pulmonary fibrosis was induced using BLM in both humanized models. In the PBMC-humanized model, human T cells accounted for 79.9% ± 9.7% of CD45+ cells; however, this model was limited by high incidence of graft-versus-host disease (GvHD). In contrast, the HSC-humanized model demonstrated stable multilineage engraftment (T cells: 20.1% ± 28.3%; B cells: 56.8% ± 25.4%), absence of GvHD, and consistent fibrotic pathology. Both models exhibited increased inflammatory infiltration, fibroblast activation, and collagen deposition following BLM induction. Nintedanib treatment significantly ameliorated fibrotic changes in both models. These findings establish the HSC-humanized BILF model as a clinically relevant platform for investigating human immune responses in IPF and for preclinical evaluation of human-targeted therapies.

Community participation enhanced prenatal mental health through strengthening peers’ support and partners’ communication in Chinese mothers: A cross-sectional study

This study investigated the relationship between extensive community participation and prenatal mental health, focusing specifically on the mediating roles of peer support and partner communication. A cross-sectional survey was conducted in Jiangsu, China, involving 1,705 pregnant participants. Structural Equation Modeling was employed to examine how offline peer support, online peer support, and partner communication mediate the association between community participation and prenatal mental health. Among the 1,705 participants (Mean age = 29.57, SD age = 3.70, Max age = 43, Min age = 16), 1,000 (58.7%) were giving birth for the first time, and 975 (57.3%) were local residents. The mean score for community participation was 4.43, while that for prenatal mental health was 4.21. These findings indicate that extensive community participation does not directly affect prenatal mental health. Instead, peer support and partner communication serve as parallel mediators, while social media and partner communication function as sequential (chain) mediators. These results underscore the critical need to enhance both the quantity and quality of partner-centered maternal support networks by leveraging peer groups and digital platforms. Strengthening these relational pathways can facilitate more effective communication between partners and ultimately contribute to improved maternal mental health during pregnancy.

Monolithic integrated <b> <i>β</i> </b>-Ga2O3 inverters using charge trapping technique

In this work, high-performance monolithic integrated β-Ga2O3 inverters are proposed, in which the enhancement mode (E-mode) devices are implemented using a charge trapping layer (CTL) technique. Leveraging the threshold voltage tunability of the CTL-based E-mode devices, the inverters demonstrate robust operation across a wide range (5–18 V) of supply voltage (VDD). In addition, the inverter with a β value of 40 achieves an ultralow output low voltage (VOL = 0.01 V) at VDD of 18 V, as well as excellent output swing/VDD ratio (17.99 V/18 V), noise margin capability (NML/NMH = 7.65 V/7.55 V), maximum voltage gain (9.6 V/V), and maximum power consumption (3.7 × 10−4 W). Furthermore, dynamic testing was further conducted to investigate the relationship between the dimensional matching of E/D-mode devices and key parameters such as transmission delay and output characteristics. These findings provide insights into the fabrication of monolithic integrated β-Ga2O3 inverters.

Attenuation of natural killer cell cytotoxicity by interaction between NKp30 of NK cells and dipeptidase 1 of colon cancer cells

KGMP: Augmenting retrieval knowledge graph with multi-hop perceptron

The core challenge of Knowledge Base Question Answering (KBQA), as a bridge between natural language and structured knowledge, is to accurately map complex semantic queries into Graph Query Language (GQL). Compared with the traditional Text-to-SQL task, KBQA faces a dual challenge: the structural differences between GQL and SQL and the lack of high-order subgraph information in multi-hop inference of knowledge graphs. While existing approaches such as ChatKBQA have made progress, the limitation of subgraph scalability severely constrains multi-hop query performance. To this end, this study proposes Knowledge Graph Multi-hop Perceptron (KGMP) - a retrieval-generation framework fine-tuned based on open-source large language models, whose innovativeness is reflected in three aspects: 1. Dynamic Graph Traversal Mechanism: Through an iterative subgraph expansion strategy, KGMP effectively achieves dynamic traversal of problem oriented graphs with progressive reasoning. 2. Structured Interaction Protocol: Based on SparQL syntax, KGMP designs a lightweight interaction instruction set to build an efficient communication interface between LLM and knowledge graph. 3. Graph Structure Optimization Technique: Develop subgraph reordering algorithms and pruning strategies based on the reranker model to ensure that the subgraphs input to the LLM are both compact and semantically complete. By integrating KGMP as a retrieval module into the ChatKBQA framework and providing it with optimised multi-hop subgraph input, the experimental results show a performance improvement of 6.2% and 5.3% on the WebQSP and CWQ datasets, respectively. This study provides a new technical paradigm for deep collaboration between LLM and knowledge graph.

Chemical solution-immersion deposition of BiFeO3 films reduced to nanometer thickness with stabilized polarization

Bismuth ferrite (BiFeO3, BFO), a representative ferroelectric material exhibiting multiferroic characteristics at ambient temperature and demonstrating reversible polarization under external electric fields, has attracted significant attention in the development of prototype ferroelectric devices utilizing its films. Moreover, nanometer-thick, ultra-thin BFO films with stabilized polarization facilitate device miniaturization and low-power consumption. Herein, we report the fabrication of high-quality BFO films with enhanced crystallinity in a rhombohedral structure through the cost-effective chemical solution-immersion deposition method, which maintains compatibility with existing semiconductor technologies. Notably, piezoresponse force microscopy and Kelvin probe force microscopy characterizations demonstrate stabilized polarization with a low decay exponent of 0.014. The out-of-plane ferroelectric polarization can still be reversed at ambient temperature, even when the film thickness is reduced to approximately 3 unit cells. This research presents an effective approach for fabricating ultra-thin ferroelectric films, particularly suitable for future non-volatile memory devices.

Predicting macroelement content in legumes with machine learning

Study of collaborative filtering recommendation with user clustering incorporating implicit social relationships and trust relationships

With the rapid development of the internet, information overload has become a prevalent issue. In order to tackle information overload, recommendation systems serve as an effective tool that can offer personalized recommendation services to users. The efficiency of recommendation systems is, however, hampered by the prevalent problems with data sparsity and cold start issues in collaborative filtering recommendations. Researchers typically address these issues by utilizing user social information clustering methods. Nevertheless, in practice, previous studies have shown that inaccurate similarity calculations and poor clustering results have led to a decrease in prediction accuracy. This paper suggests a collaborative filtering recommendation algorithm that incorporates several relationships in order to overcome these difficulties. This method first calculates user similarity based on implicit social relationships and trust relationships. After clustering users using the spectral clustering technique, it makes use of user-based collaborative filtering recommendations within the cluster containing the target person. The collaborative filtering recommendation system that integrates many relationships effectively decreases prediction errors and improves recommendation accuracy, as shown by the results of simulated studies.

Dipole-molecular buffer interface engineering for mitigating hole injection barrier and interface defect in high-efficiency quasi-2D blue perovskite light-emitting diodes

The quasi-two-dimensional (quasi-2D) perovskites, characterized by strong quantum confinement and high exciton binding energy, hold significant promise for blue perovskite light-emitting diodes (PeLEDs). However, challenges including imbalanced charge injection and defect-related nonradiative recombination at the interface, particularly at the hole transport layer (HTL)/perovskite interface, hinder the development of efficient PeLEDs. This investigation systematically evaluates the effect of various dipolar molecules on the HTL/perovskite interface, aiming to enhance the efficiency of blue PeLEDs. The results indicate that the hole injection barrier of devices modified with diethyl difluoromethanephosphonate (DFMP) was significantly reduced due to the high polarity of DFMP, which maintains a balanced hole–electron mobility. Furthermore, the electron-rich P=O bond in DFMP effectively coordinates with the unsaturated Pb2+ at the buried interface to passivate defects that contribute to radiative recombination. As a result, the DFMP-modified PeLEDs achieve a peak external quantum efficiency of 9.8% at 480 nm and a prolonged operational lifetime. The study paves the way for interfacial molecular buffer layers to interface modulation in quasi-2D PeLEDs.

<i>N</i>‐Doped Nonalternant Nanoribbons with up to 29 Linearly‐Fused Rings and High Charge‐Carrier Mobilities

AbstractThe synthesis of nonalternant graphene nanoribbons (GNRs)—carbon‐based nanostructures featuring fused pentagonal or heptagonal rings that disrupt hexagonal symmetry—has remained a significant challenge despite their unique electronic properties, such as tunable bandgaps and topological states. Unlike conventional alternant hydrocarbons (e.g., benzene‐derived systems), nonalternant architectures exhibit distinct electron confinement effects, making them promising for nanoscale optoelectronics and quantum materials. However, achieving precise control over their length, solubility, and structural regularity has had limited progress. Herein, we report a versatile solution‐phase synthesis of a novel family of structurally precise N‐doped nonalternant nanoribbons (NRs: NR‐11 to NR‐29) through pentagon annulation of pyrazino[2,3‐g]quinoxaline and naphthalene subunits terminated with triptycene end‐caps. The longest one comprises 29 linearly fused rings (particularly including 10 pentagonal rings), setting a new length record of 7.18 nm among the soluble nonalternant NRs reported to date. Strategic incorporation of triptycene end‐caps and alkoxyphenyl substituents into their molecular backbones confers high solubility and a unique 3D configuration, enabling comprehensive solution‐phase structural characterization and systematic investigation of length‐dependent optoelectronic properties. Employing terahertz (THz) spectroscopy, we infer high intrinsic charge carrier mobilities in nonalternant nanoribbons, reaching up to ∼2000 cm2 V−1 s−1, highlighting this novel family of nanoribbons as promising candidates for future nanoelectronic applications.