Browse Articles
Discover research articles across all indexed journals
Cpeb4 regulates cardiomyocyte apoptosis in heart failure with association to Eif4a2 splicing modulation
Deep learning enables rabies virus detection and antibody quantification in cell culture
Abstract Rabies remains a major neglected disease, causing tens of thousands of deaths annually in endemic regions. Cell culture-based assays are central to rabies diagnostics and antibody quantification, yet interpretation relies on manual fluorescence image reading, which is labour-intensive, subjective, and difficult to scale. Here, we present, what is to our knowledge, the first AI-based workflow for detecting rabies virus infection in BHK-21 cell cultures and quantifying antibody titres. Three state-of-the-art architectures–ResNet, Swin Transformer V2, and MambaOut–were trained on a large dataset of 2344 fluorescence images (36,822 annotated patches) capturing biological and technical variability. Evaluation at patch, image, and well levels showed MambaOut consistently outperformed other models, achieving 0.996 accuracy, 0.993 F1-score, and 0.999 AUC-ROC at the well level. To assess applicability, models were tested on unseen images from the WHO/WOAH-recommended fluorescent antibody virus neutralisation assay. MambaOut demonstrated complete agreement with ground truth in predicting antibody concentrations, confirming workflow reliability for vaccine assessment and pet travel schemes. These results establish a foundation for AI-assisted rabies diagnostics and open opportunities for applying deep learning to other cell culture-based viral assays. AI-driven automation for high-throughput image analysis offers a scalable and time-efficient solution for diagnostic laboratories, particularly in endemic countries.
Prognostic significance of the endothelial activation and stress index in acute respiratory distress syndrome: a retrospective cohort study
An instance-aware segmentation and optical flow based DVI-SLAM for dynamic environments
Experimental and numerical investigation of stress spatiotemporal response of fault plane during underground coal seam advancement
Abstract Fault slip induced by coal seam mining is a major hazard that can trigger dynamic disasters such as coal bursts. Understanding the evolution of shear and normal stress on the fault plane is essential for revealing the mechanical mechanism of mining-induced fault activation. Based on physical experiment and numerical simulation of the 21,221 mining face in Qianqiu coal mine, this study investigates the stress spatiotemporal response during mining face advancement. Results show that shear stress fluctuates in repeated cycles of abrupt drop followed by rapid rise, with cycles becoming less frequent and intense farther from the coal seam, indicating a distance-dependent disturbance effect. Mining activity also induces migrating stress relief and concentration zones, the relief zone expands toward the near coal seam side while the concentration zone shrinks and shifts away. A rise in normal stress accompanied by a drop in shear stress serves as a key slip precursor of fault, whereas a simultaneous sharp decline in both shear and normal stress marks fault instability and energy release. Therefore, coupled monitoring of normal and shear stress evolution law provides valuable early warning of fault slip risk.
Born–Jordan frequency distribution based E-GMANN controller for vibration analysis in SSI based asymmetric floor buildings with ATMD
Magnetically insulated diode: existence of solutions and complex bifurcation. I
Pharmacological rescue of isolation-induced social and neurochemical deficits in Drosophila melanogaster
Abstract Social isolation (ISO) is a major risk factor for mood disorders, reflected in disrupted neurochemistry and social behavior. To test whether five days of ISO induce a depression-like state in male Drosophila melanogaster , we measured brain monoamines and social interaction networks (SINs), focusing on dopamine (DA) and octopamine (OA), key regulators of motivation and reward. ISO reduced DA, OA, and tyramine (TA) while increasing acetylcholine, and produced disorganized SINs marked by weaker, shorter interactions, reduced clustering and centrality, and hyperactive, uncoordinated locomotion. To test for rescue, male flies were fed L-DOPA (L-DA), OA, or both during ISO. All treatments improved monoamine levels, with the combined supplementation normalizing or exceeding DA and OA and reducing acetylcholine to control levels. A condition-specific SINs metrics enabled consistent, quantitative comparisons across groups. SINs analyses showed that L-DA or OA alone partially restored connectivity among individuals, whereas combined supplementation most effectively rescued network cohesion, clustering, and centrality, approaching or exceeding mixed sex controls. Null-model comparisons confirmed that the restored networks reflected genuine, non-random structure rather than chance co-movement. Our findings show that five days of ISO induce neurochemical and behavioral changes in male D. melanogaster that resemble depression, and that combined L-DA+OA supplementation robustly restores monoaminergic balance and complex social organization. These results suggest a mechanistic link between brain monoamines and group-level social behavior, highlighting the neurochemical basis of social deficits and their pharmacological rescue.
Biomass ash as supplementary cementitious material: microstructural mechanism underlying cement paste strength evolution
Non-invasive estimation of hemoglobin using machine learning algorithms from i-PPG and PPG signals
Design of a single inductor bidirectional DC converter for V2V energy transfer: On board charger architecture
Abstract This paper presents a Single-Inductor Bidirectional Converter (SIBC) for unified onboard Electric Vehicle (EV) charging, integrating grid, battery, and wireless ports within a single power stage. The topology enables native Grid-to-Vehicle (G2V), Wireless-to-Vehicle (W2V), and Vehicle-to-Wireless (V2W) operation without hardware reconfiguration, eliminating cascaded converter–inverter structures. Non-ideal steady-state and small-signal models are developed, revealing mode-dependent dynamics including a right-half-plane zero in V2W mode that constrains bandwidth. A two-loop Average Current-Mode Control (ACMC) is proposed to mitigate this limitation, achieving 40× bandwidth improvement over conventional voltage-mode control. Parametric sensitivity analysis of inductor equivalent series resistance establishes quantitative design boundaries for sustaining conversion efficiency above 90%. Scalability assessment to 3 kW operation demonstrates compatibility with silicon carbide devices and a bridgeless totem-pole PFC front-end achieving THD less than 4% and power factor higher than 0.99, satisfying IEC 61000-3-2 Class A with higher than 35 dB ripple rejection at the battery terminals. Experimental validation using a 136 W prototype achieves 93.4% transmitter and 95.07% receiver DC–DC efficiency, with an overall end-to-end efficiency of 75.03% including an 84.5% wireless link. A 1 kW interim hardware test confirms scalable operation, while simulation validates CC-CV battery charging compatibility. The results demonstrate that the SIBC architecture is not the dominant source of system losses and provides a compact, scalable foundation for advanced bidirectional EV charging systems.
Akkermansia muciniphila supplementation alters inflammatory profiles across diverse models of colitis
Childhood family attachment and physical abuse: the moderating role of gender
Abstract Childhood physical abuse has lasting implications for individual well-being, yet empirical evidence on protective family factors and their gendered effects during emerging adulthood remains limited. This study offers novel insight by examining gender as a moderating factor in the relationship between childhood family attachment and childhood physical abuse among emerging adults. The objective of this study is to assess whether stronger childhood family attachment reduces experiences of childhood physical abuse and to determine whether this protective effect differs by gender. Data were collected using a structured self-administered questionnaire distributed to university students through academic networks, with lecturers assisting in disseminating the survey to their students. A total of 416 responses were analysed using descriptive statistics and regression-based moderation analysis. The findings indicate that childhood family attachment is significantly and negatively associated with childhood physical abuse, confirming its protective role. Gender is found to be a significant predictor and moderator, with results showing that the buffering effect of childhood family attachment against physical abuse is stronger for males than for females. From a policy perspective, the results emphasise the need for attachment-based and gender-sensitive family interventions. Policies should prioritise strengthening parent–child relationships, promoting non-violent disciplinary practices, and incorporating gender-responsive approaches in child protection and parenting programmes to prevent childhood physical abuse effectively.