Browse Articles

Fault detection of high-speed train wheelset bearings based on improved auxiliary classifier generative adversarial networks and VAE

Fault detection in high-speed train wheelset bearings is paramount for ensuring operational safety. However, the scarcity of fault samples limits the accuracy of traditional detection methods. To address this challenge, this paper proposes a supervised generative model that integrates an Improved Auxiliary Classifier Generative Adversarial Network (IACGAN) with a Variational Auto-Encoder (VAE). Firstly, the method employs the VAE as the generator, introducing latent variables with prior information to optimize the encoding and generation process; Secondly, an independent classifier network is integrated into the ACGAN framework to enhance compatibility between classification and discriminative capabilities. Concurrently, a loss function incorporating Wasserstein distance and gradient penalty terms is designed to prevent gradient vanishing during training while satisfying Lipschitz constraints, thereby improving model stability. Experiments conducted on the XJTU bearing dataset validate that samples generated by the proposed method demonstrate superior quality assessment at both the data and feature levels compared to several GAN variants. Furthermore, the constructed VAE-IACGAN-CNN detection model achieves an average classification accuracy of 88.04%, representing a maximum improvement of 15.17% over comparative methods. This significantly mitigates accuracy degradation caused by sample imbalance, demonstrating the proposed approach’s efficacy in resolving low fault detection accuracy stemming from imbalanced high-speed train wheelset bearing samples.

A density-functional theory study of vapor, liquid, and mesophases in a symmetric, hard-sphere mixture with cross attraction at non-equimolar concentrations

We investigate the phase diagram of a model hard-sphere mixture consisting of two species of equal diameter, featuring a square-well cross attraction. The study is carried out using density-functional theory (DFT) in the mean-field approximation and extends to arbitrary species concentrations a former DFT study at equimolar composition [Prestipino et al., J. Chem. Phys. 159, 204902 (2023)]. In addition to the stripe phases found in the equimolar case, cylinder and cluster phases are also observed. While at high temperatures, the inhomogeneous domains can be accessed only from the high-density liquid; at low temperatures, these phases coexist with the low-density vapor, resulting in a notably rich phase diagram. The predictions of an analytic implementation of the theory—based on the Landau expansion of the free energy in powers of the density modulation amplitude—are compared with numerical DFT minimization. The Landau approach shows qualitative agreement, although it overestimates the extent of the stripe region at high density, where local concentration values may exceed their physical bounds. Differences and similarities between DFT and simulation results for narrow attractive wells are briefly discussed.

Study of ethnoveterinary medicinal plants in Metema and Quara districts, northwestern Ethiopia

Validation of the Mentalization Scale (MentS) in francophone control and clinical samples

Aims The Mentalization Scale (MentS) is a self-report measure for the assessment of mentalizing capacities, consisting of 28 items, yielding a three-factor structure: self-mentalizing, mentalizing others and motivation to mentalize. Its anglophone version has been validated for usage in both research and clinical contexts. The present study explores the psychometric properties of the francophone translation of the MentS, in both control and clinical samples. Method A total of 711 participants were enrolled in this study. The MentS was administered to a community sample (N = 302, 161 females, Mage = 37.1, SDage = 12.3), and to a clinical sample composed of participants diagnosed with borderline personality disorder (BPD), attention deficit hyperactive disorder (ADHD) and co-occurring BPD and ADHD (N = 409, 266 females, Mage = 32.5, SDage = 11.9). Confirmatory factory analysis was used to analyze the fit of the model in our data, followed by reliability and validity analyses. Results Results from confirmatory factor analysis (CFA) revealed a 27-item model to best fit the data for both control and clinical samples. In the control sample, good internal consistency was found for the total scale (α = 0.856, ω = 0.867) as well as for the three subscales MentS-Motivation (α = 0.789, ω = 0.801), MentS-Other (α = 0.792, ω = 0.798) and MentS-Self (α = 0.824, ω = 0.828). Similarly, good internal consistency was found in the clinical sample for the total scale (α = 0.871, ω = 0.879) and subscales (Motivation α = 0.770, ω = 0.783) (Others α = 0.842, ω = 0.847) (Self α = 0.808, ω = 0.815). The MentS demonstrated good temporal stability over a one-year interval, with excellent average-measures ICC for the total scale (ICC = .877, 95% CI [.843,.904], p  < .001), and strong reliability for the Motivation (.837), Others (.806), and Self (.837) subscales (all ps  < .001). The validity of the scale was confirmed using additional measures showing a coherent pattern of associations in relation to components underlying the construct of mentalization: reflective functioning, childhood trauma, cognitive emotion regulation, overall psychopathology distress and borderline symptomatology. Conclusion & Clinical Implications The MentS can be used for research and clinical purposes in francophone samples. Our results suggest that amongst French speaking samples, a 27-item solution may be most optimal in both clinical and control populations. Evidence shows the scale could be employed across diagnostic entities, and that participants scoring low in self-mentalizing (MentS-Self subscale) may be more likely to report increased manifestations of psychopathology.

Efficient evaluation of spin–orbit couplings in single-molecule magnets using DMRG within the DMET framework

The accurate characterization of complex electronic structures in single-molecule magnets (SMMs) is crucial for guiding their rational design. However, conventional multireference wavefunction theory faces significant computational bottlenecks for these systems, stemming from their intricate ligand environments and the necessity for large active spaces to capture strong electron correlation. To overcome these limitations, we introduce an efficient quantum embedding framework that synergistically combines density matrix embedding theory with ab initio density matrix renormalization group Relativistic spin–orbit coupling (SOC) effect is incorporated through a state-interaction procedure using the eXact two-component atomic mean-field spin–orbit (X2CAMF-SO) operator. Benchmarking on representative single-ion magnets confirms that our DMET-embedded DMRG-SI approach delivers results highly consistent with all-electron calculations, while offering a clear path for systematic improvement via active space expansion. Furthermore, by investigating SOC effects in a chromium dimer, we demonstrate the method’s potential for high-accuracy and scalable ab initio studies of larger-sized polynuclear SMMs.

Flow rate selection depends on the size of allergen particles to achieve optimal nasal irrigation

Correction: Domain-specific life satisfaction among older adults with and without children: The role of intergenerational contact

Imaging transient molecular configurations in UV-excited diiodomethane

Femtosecond structural dynamics of diiodomethane (CH2I2) triggered by ultraviolet (UV) photoabsorption at 290 and 330 nm are studied using time-resolved coincident Coulomb explosion imaging driven by a near-infrared probe pulse. We map the dominant single-photon process, the cleavage of the carbon–iodine bond producing rotationally excited CH2I radical, identify the contributions of the three-body (CH2 + I + I) dissociation and molecular iodine formation channels, which are primarily driven by the absorption of more than one UV photon, and demonstrate the existence of a weak reaction pathway involving the formation of short-lived transient species resembling iso-CH2I2-like geometries with a slightly shorter I–I separation compared to the ground-state CH2I2. These transient molecular configurations, which can be separated from the other channels by applying a set of conditions on the correlated momenta of three ionic fragments, are formed within ∼100 fs after the initial photoexcitation and decay within the next 100 fs.

Research on the pore size distribution and durability improvement mechanism of magnetic slurry

Minimum recommended micronutrient intake status and associated factors among pastoralist children aged 6–23 months in Aysaita district, Afar Region, Ethiopia, 2024: A community based cross-sectional study

Background Micronutrient deficiencies are among the most prevalent public health problems in developing countries like Ethiopia. Despite government efforts to address these deficiencies, progress remains slow, leaving many children to endure serious health consequences. There is a scarcity of studies examining the minimum recommended micronutrient intake among children aged 6–23 months in this study area. Objective This study aimed to assess minimum recommended micronutrient intake status and associated factors among pastoralist children aged 6–23 months in Aysaita District, Afar Region, Ethiopia, 2024. Methods A community-based cross-sectional study design was conducted among 614 children aged 6–23 months with mothers from August 1–30/2024. Multi-stage sampling technique was used to select study participants. Data were collected by KoboToolbox using an interview-based structured questionnaire. The collected data were exported to Excel and then into STATA version 17 software packages. Descriptive statistics such as proportion, mean, median, cross-tabulation, and frequencies were calculated and presented in tables. Bivariable and multivariable logistic regression analyses at a 95% confidence interval (CI) and a P-value of less than 0.05 in the multivariable logistic regression were considered significant predictors. Results All study participants were included for data collection by giving a response rate of 100%. In this study, only 35.67% of the children aged 6–23 months had received at least one minimum recommended micronutrient intake status (35.67%; 95% CI: 31.96, 39.54%). Being a male child (AOR = 2.04; 95% CI: 1.42, 2.92), having media exposure (AOR = 2.50; 95% CI: 1.34, 4.64), having exclusive breast feeding (AOR: 5.09, 95% CI: 2.72, 9.52), being in the income category 7.07–35.34 USD (AOR: 2.90, 95% CI: 1.49, 5.65) and spontaneous vaginal delivery (AOR: 2.71, 95% CI: 1.20, 6.10) were factors associated with the minimum recommended micronutrient intake status. Conclusions and recommendations This study revealed low adequate micronutrient intake among children aged 6–23 months, with only slightly over one-third (35.67%) meeting the minimum recommended status. Being a male child, having media exposure, having exclusive breast feeding, being in the income category 7.07–35.34 USD and spontaneous vaginal delivery were factors associated with the minimum recommended micronutrient intake status. To effectively address micronutrient intake, interventions must be more targeted and intensified. This involves strengthening programs that educate mothers on optimal infant and young child feeding, from exclusive breastfeeding to preparing diverse, nutrient-rich foods. Public messaging via community campaigns and media should be expanded to promote affordable, local food options and must explicitly target caregivers of girls to eliminate preferential feeding practices. Finally, to overcome economic barriers, initiatives like women’s economic empowerment, social safety nets, and homestead food production (e.g., kitchen gardens, poultry) should be supported to ensure consistent household access to nutritious food.

Multi-scale study of cobalt adsorption on TiO2 anatase (101): From DFT to force-field parameterization

Understanding metal–support interactions (MSI) is critical for designing stable and efficient catalysts, such as cobalt catalysts on TiO2, commonly employed in Fischer–Tropsch synthesis. This study investigates cobalt adsorption on the TiO2 anatase (101) surface through a computational approach, combining density functional theory (DFT), ab initio molecular dynamics (AIMD) simulation, and genetic algorithm-based force field parameterization. It is shown that Co clusters get substantially oxidized due to interaction with oxygen atoms of the support. The parameterization of Morse potential force field is achieved using an automated and combined genetic algorithm (GA) - molecular dynamics (MD) simulation approach, leveraging AIMD simulations to capture the dynamic nature of MSI effectively in the training dataset. This method incorporates simulations with the Large-scale Atomic/Molecular Massively Parallel Simulator into the GA framework, thereby streamlining the optimization process for the force field, facilitating an efficient exploration of the parameter space. This new force field, validated against DFT results, offers an efficient tool for modeling MSI at larger length scales and longer timescales. Our findings highlight how MSI influences cobalt cluster stability, electron transfer, and surface restructuring, directly impacting catalytic performance and resistance to sintering. The methodology presented in this study offers a versatile framework that can be adapted to other metal–support systems with system-specific reparameterization and validation, enabling comprehensive investigation of mesoscale MSI studies on explicit oxide surfaces, facilitating further advances in heterogeneous catalysis research and applications. The findings highlight the influence of MSI on cobalt cluster stability, electron transfer, and surface reconstruction, all of which are crucial to catalytic activity and sintering resistance.

Villin like suppresses nasopharyngeal carcinoma proliferation by downregulating MYC

ACCYolo: Transmission equipment inspection image detection method based on multi-scale and occluded targets

With the rising global demand for electricity, transmission infrastructure is becoming increasingly important as a key support for ensuring stable and reliable power supply.In recent years, UAVs have been widely used in the inspection and maintenance of transmission equipment due to their advantages of high efficiency, flexibility and intelligence, which have greatly improved the operation and maintenance efficiency and safety level.However, the transmission equipment itself is exposed to harsh natural environments during prolonged use, such as high temperatures, humidity changes, wind and sand erosion, as well as electromagnetic interference, coupled with complex topographical features, such as mountainous, hilly, and forested areas, which result in the transmission equipment inspection process being challenged by occlusion and large differences in dimensions.To cope with these problems, this paper proposes ACCYolo. a model based on the YOLOv10n architecture with the goal of improving image detection of transmission equipment under multi-scale and occluded targets in UAV-based scenes.On the one hand, the ACCYolo model, to solve the occlusion problem, incorporates the Acmix model, which incorporates the self-attention mechanism to achieve dynamic feature extraction, effectively improving the detection performance of the model in overlapping scenes.On the other hand, in order to cope with the size difference problem in multi-scale detection, the GELAN structure combines a lightweight design with the Programmable Gradient Information (PGI) mechanism to improve the accuracy of multi-scale target detection, while the ASFF module is designed to improve the accuracy of multi-scale target detection through adaptive spatial feature fusion.The experimental results show that. The proposed method shows significant advantages in transmission equipment monitoring tasks, Overall mAP@50 raise to 0.950, and provides an effective program to ensure the reliability of power supply.

QM/MM study of cytochrome P450 TxtE catalysis: Substrate reorganization enables selective aromatic nitration

Cytochrome P450 enzymes are heme-dependent catalysts found in all forms of life (plants, bacteria, and mammals), playing pivotal roles in the transformation of both endogenous and exogenous molecules. Recently, a new subfamily of P450 enzymes has been identified that catalyzes the direct and selective nitration of aromatic amino acids, such as L-tryptophan, in the presence of oxygen and nitric oxide. To investigate the catalytic mechanism, we applied quantum mechanical (QM)/Molecular Mechanics (MM) computational methods to explore potential nitration pathways. Our study confirms that the initial substrate binding conformation from the crystal structures is inactive for the following nitration reactions. Instead, the substrate undergoes a conformational change, leading to the alternative conformations that are essential for the following nitration reaction at the C4 position. Moreover, comparative QM/MM scanning and QM/MM-based free energy calculations showed that the conformational reorganization of the intermediate indole/·NO2, which is coupled to the reorganization of enzyme pocket, greatly enhances the subsequent nitration reaction and H-atom transfer reaction. This study underscores the key roles of enzyme-mediated conformational changes of the substrate and intermediates in facilitating the selective biotransformation. Our study shows that aromatic nitration at C4 site is energetically favored, in agreement with experimental findings.

Correction: Histone deacetylase inhibitors inhibit lung adenocarcinoma metastasis via HDAC2/YY1 mediated downregulation of Cdh1

Correction: Diagnostic performance of microvascular flow imaging for noninvasive assessment of liver fibrosis in chronic liver disease

MuAPBEK: An improved analytical kinetic energy density functional for quantum chemistry

Orbital-free density functional theory (OFDFT) enables full quantum-mechanical simulations based solely on electron densities but is limited by the lack of accurate kinetic-energy functionals. We improve upon the APBEK functional by tuning its μ parameter for a given system during density initialization and adding two non-empirical corrections based on Kato’s cusp condition and the virial theorem. The resulting functional, MuAPBEK, assessed on atoms and molecules, shows significantly lower energy errors than standard APBEK and Thomas–Fermi–von-Weizsäcker functionals, even when evaluated on converged Kohn–Sham density functional theory (KSDFT) densities. MuAPBEK also produces accurate densities that slightly deviate from those of KSDFT. Its density optimization step is over ten times faster than a single KSDFT SCF cycle and scales as O(N1.96), indicating that accurate, large-scale OFDFT simulations are feasible beyond practical KSDFT limits.

Correction: Effects of high-definition transcranial direct current stimulation for the treatment of chronic insomnia: a randomized, double-blind, controlled trial

The combination of zalfermin and semaglutide has additive therapeutic effects in a diet-induced obese and biopsy-confirmed mouse model of MASH

Fibroblast growth factor 21 (FGF21) analogs have significant therapeutic potential in metabolic dysfunction–associated steatohepatitis (MASH) but limited body weight effects in patients with MASH. This study investigated the effect of combined treatment with the FGF21 analog zalfermin and the glucagon-like peptide-1 receptor agonist semaglutide on body weight and plasma and liver biochemistry and histology in a mouse model of MASH. Amylin liver nonalcoholic steatohepatitis diet-induced obese-MASH mice with biopsy-confirmed MASH and fibrosis were administered (subcutaneous [SC], daily [QD]) vehicle, zalfermin (0.05 or 0.2 mg/kg), semaglutide (3 or 120 µg/kg), or zalfermin 0.05 mg/kg + semaglutide 3 µg/kg for 8 weeks (n = 11–12 per group). Vehicle-dosed (SC, QD) chow-fed mice served as normal controls (n = 10). Pre- to post-liver biopsy histology was compared for within-subject evaluation of changes in non-alcoholic fatty liver disease Activity Score (NAS), fibrosis stage, and quantitative histology. Additional endpoints included plasma/liver biochemistry and liver RNA sequencing. Combined low-dose zalfermin and semaglutide treatment resulted in super-additive body weight loss (−18%) vs. individual low-dose monotherapies (zalfermin, −6%; semaglutide, −4%) and was equally effective as high-dose zalfermin monotherapy (−16%) and semaglutide (−15%). Low-dose combination therapy promoted greater benefits on transaminases, total cholesterol and triglycerides, NAS, steatosis, and inflammation vs. individual low-dose monotherapies and high-dose semaglutide, and high-dose zalfermin was as effective as the low-dose combination therapy on most endpoints. Combination treatment reduced gene expression markers of fibrosis to a greater degree than monotherapies. In conclusion, combined low-dose zalfermin and semaglutide, as well as high-dose zalfermin, resulted in beneficial effects on body weight and biochemical and histological endpoints, supporting the clinical development of zalfermin as therapy for patients with MASH.

A kinetic model to simulate charge flow through an electro-chemical half-cell

A kinetic model of the electron transfer at the electrode/electrolyte interface is developed, implemented in a Monte Carlo framework, and applied to simulate this process in idealized systems consisting of the primitive model of electrolyte limited by an impenetrable conducting surface. In the present implementation, a charged, spherical interface surrounding an equally spherical sample of electrolyte is introduced to model a single-electrode system, providing the computational analog to the conceptual half-cell picture that is widely used in electrochemistry. The electron transfer itself is described as a simple surface hopping process underlying a first order reaction corresponding to one of the coupled M/M+ and X−/X half reactions. Then, the electron transfer at the interface is combined with the self-diffusion of ions in the electrolyte, whose role is to supply reagents and disperse products, allowing the system to settle in a stationary non-equilibrium state. Simulations for the primitive model of an electrolyte in contact with a charged impenetrable surface show that, after a brief transient, the samples sustain a steady current through the half-cell. The results quantify the dependence of the current on the overall charge of the electrode, the ionic strength of the electrolyte, its viscosity, and the kinetic parameter ke, which represents the rate of the electron transfer for each ion in contact with the electrode. Since the simulated interface is very idealized, strategies to overcome the limitations of the present model are outlined and briefly discussed.