Browse Articles
Discover research articles across all indexed journals
Short-term predictability of data-driven SMILE 4.0 nomograms for myopia: cycloplegic versus manifest refraction
Visibility-driven compensation model for driver pupillary response in road tunnels: a GWO-XGBoost and SHAP-based interpretable framework
Nested contextual change and the temporal compression of episodic memory
Abstract How does the structure of events influence the when and the where of life experiences in comparison to the what? We developed a novel virtual reality (VR) environment to understand how the quantity of information within nested structures influence participants’ memory for events. Participants moved through a series of virtual rooms (events) where images (items) appeared in randomised locations on a 3 by 3 grid located on a wall. Participants were asked to remember the what (old/new), when (timeline location), and where (grid location), of the images they experienced. Two types of nested events were tested (6 rooms, each containing 4 images; 3 rooms, each containing 8 images), all lasting an equivalent number of seconds. We found a strong temporal compression effect at nested levels in which participants remembered early items and events happening later, and later items and events happening earlier, than the original experience. Crucially, presenting four-item events resulted in a greater compression rate than eight-item events. We also found greater temporal distances between pairs of items occurring within eight-item events than pairs of items which occurred on either side of a boundary. Memory for when depends on the nested compression of information within events.
Lymphatic endothelial cells regulate neutrophil phenotypes and function in a microphysiological model of infection
A dual-domain spatio-temporal and frequency framework for robust deepfake detection
Artificial intelligence approach for predicting suicide-related behaviour in emergency departments
Abstract Suicide remains a major public health challenge, with emergency departments (EDs) representing a critical point of contact for individuals at risk. However, conventional suicide screening approaches are often time-intensive, disclosure-dependent, and difficult to implement consistently in high-acuity settings. This study developed and evaluated an artificial intelligence (AI)-based model for predicting suicide-related behaviour (SRB) within 30 days of ED presentation using routinely collected structured triage data available at admission. Five supervised learning models, logistic regression, random forest, extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and a multilayer perceptron (MLP), were developed and evaluated using patient-level train-validation-test splits. Performance was assessed using a multidimensional framework including area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), Recall, F2-score, Matthews Correlation Coefficient (MCC), Balanced Accuracy, threshold optimisation, decision-curve analysis, bootstrap confidence intervals, and SHapley Additive exPlanations (SHAP). Among the evaluated models, LightGBM showed the most favourable balance between overall discrimination, minority-class discrimination, sensitivity, and decision stability, achieving an AUROC of 0.88, an AUPRC of 0.25, a Recall of 0.79, and an F2-score of 0.41 on the independent test set. Threshold optimisation and decision-curve analysis further supported the model’s clinical utility, while SHAP analyses identified prior psychiatric history, age, and physiological variables as important contributors to risk prediction. These findings demonstrate that short-term SRB risk can be predicted using structured ED triage information available at presentation. Rather than replacing clinicians, the proposed model aims to provide transparent, scalable decision support to support prioritisation, follow-up planning, and early intervention. Although external validation and prospective studies remain necessary, the study provides a deployment-oriented framework for integrating AI-assisted suicide risk prediction into emergency care workflows.
Supplementation of hydrogen-rich water to goats modifies the ACE inhibitory activity and bioactive peptide profile detected by LC–MS–MS of colostrum and mature milk
Functional trait-mediated limiting similarity shapes weed biomass acquisition, fecundity, and reproductive phenology
Renovation dimensions and residents’ subjective well-being in old residential neighbourhoods: a structural equation modeling study in Hangzhou, China
Abstract Aligned with the global agenda for sustainable urban regeneration, this study examines how different dimensions of old residential community renovation relate to residents’ subjective well-being (SWB) in urban China. SWB is defined as residents’ situated evaluation of community life, operationalised through happiness, neighbourhood relations, and perceived safety. A dual-path framework is developed to link four renovation dimensions, physical environment, infrastructure, community services, and property management, to SWB directly and indirectly through resident participation, captured through consultation, information disclosure, feedback mechanism, and post-renovation evaluation. Based on 2,236 valid household responses from 36 renovated communities in Hangzhou, China, this study employs structural equation modelling to simultaneously estimate the direct paths from each renovation dimension to SWB and the indirect paths mediated through resident participation. The results show that physical environment and infrastructure exhibit significant positive direct paths to SWB, whereas community services and property management do not. Resident participation is positively associated with SWB and selectively mediates the relationships of physical environment and community services with SWB, while the indirect paths through infrastructure and property management remain non-significant. Heterogeneity analyses further indicate that SWB among lower-income residents is mainly related to improvements in the physical environment and infrastructure, whereas higher-income residents are additionally responsive to community services and participation. Among older residents, community services and participation, rather than infrastructure, emerge as salient correlates of SWB. These findings suggest that built-environment upgrading and participatory governance function as complementary but group-sensitive mechanisms of well-being. Policy efforts should therefore move beyond a narrow technical approach toward a staged and group-differentiated strategy that prioritises essential environmental improvements while embedding responsive services and institutionalised participatory channels to meet diverse needs.
Generalizable prediction of biological contaminants across aerobic and anaerobic membrane bioreactors
Demographic drivers of the far eastern leopard population recovery
Longitudinal characterization of medial gastrocnemius muscle growth deficits throughout childhood in spastic cerebral palsy
Balancing privacy and accuracy: federated learning for multi-class student performance prediction with adaptive noise differential privacy
Deepfakebuster: a confidence-calibrated adaptive ensemble framework for robust Deepfake image detection
A novel method for analysis of oxidant (OX = O3 + NO2) and its local and regional contributions using satellite remote sensing data
An improved Portia spider optimized SpikeApneaNet for ECG-based sleep apnea detection with FPGA implementation
Potential mediating role of triglycerides in the nonlinear association between body roundness index and MASLD in nondiabetic adults
Acid resistance and self-healing crack-sealing performance of Shewanella oneidensis-modified low-calcium fly ash geopolymer concrete
Abstract Durability of cementitious materials in sewer infrastructure is greatly affected by sulphuric acid attack and cracking. A low-calcium fly ash-based geopolymer concrete (GPC) enhanced with Shewanella oneidensis, a dissimilatory metal-reducing bacterium (DMRB), was tested for its resistance to acid in simulated sewer conditions and its self-healing crack-sealing abilities. Specimens were submerged in 3% H₂SO₄ (pH 0.66) for 60 days; crack-healing was also observed over 32 days through bacterial surface coating. The properties were examined using mass-loss, electrical resistivity, chloride migration, capillary water absorption, and microscopic characterisation using SEM-EDS, XRD, FTIR, and XRF. Bio-modified specimens showed greater strength development (42.41 vs. 39.22 MPa at 60 days), better post-acid strength retention (26.20 vs. 18.92 MPa; 38.5% advantage), lower percentage mass loss (0.6% vs. 1.9%), lower chloride migration coefficient (1.05 × 10⁻ 1 ⁰ vs. 1.09 × 10⁻ 1 ⁰ m 2 /s), and higher residual electrical resistivity (3.6 vs. 3.0 kΩ·cm). During acid attack, the surface showed dissolution and formation of gypsum-carbonate reaction layers, typical of diffusion-driven deterioration in low-calcium systems; ICP-OES of the acid leachate confirmed all monitored elements were below detection limits, consistent with precipitation-dominated leaching. Microstructural analysis confirmed the retention of Fe and Ca phases and the formation of carbonate crystals in bio-treated specimens. Crack repair analysis indicated calcium carbonate crystals forming at crack interfaces in both bio-treated and coated samples, with crack-closing ratios reaching up to 0.08 in surface-coated specimens, where capillary absorption was also reduced compared to controls. Bio-modification enhances acid resistance under simulated sewer conditions and enables autonomous crack sealing under standard water-immersion conditions, with potential for sustainable sewer infrastructure applications.