Browse Articles
Discover research articles across all indexed journals
Randomized trial of dexmedetomidine and remimazolam with remifentanil for sedation during hepatic radiofrequency ablation
Hybrid spectral learning framework for real-time material classification of end-of-life photovoltaic panels
Biodegradation potential of bacterial isolates Enterococcus faecalis RBRJ026 and Klebsiella pneumoniae RBRJ024 against benzyl butyl phthalate
Manual federated simulation for multiple sclerosis integrating XGBoost algorithm with SHAP explanation
Abstract Multiple sclerosis (MS) is a chronic autoimmune disorder of the central nervous system, underscoring the importance of early and accurate diagnosis. In this study investigates the predictive modelling of MS progression in patients with Clinically Isolated Syndrome (CIS), privacy-preserving for a federated and explainable Machine Learning (ML) framework. To address missing data while preserving inter-feature dependencies, Multivariate Imputation by Chained Equations (MICE) with iterative imputers was employed. Classification was performed using the Extreme Gradient Boosting (XGBoost) algorithm. Model interpretability was developed through Explainable Artificial Intelligence (XAI) techniques, specifically Shapley Additive Explanations (SHAP). To ensure data confidentiality and simulate decentralized clinical environments, an in silico federated learning framework was applied. Experimental results demonstrated strong predictive performance, achieving 96.7% accuracy and 99% ROC–AUC during training, 92.5% accuracy in validation, and 81.8% accuracy with an AUC of 88% on the test set. For the Federated Learning (FL) simulation, the model maintained competitive performance, yielding an accuracy of 76.3% and an AUC of 83.9%. The proposed approach supports early diagnosis, enhances clinical trust through interpretability, and promotes secure data collaboration, thereby contributing to more informed and transparent clinical decision-making and improved patient care.
AI-powered talent chain management with multi-agent systems for industry and innovation growth
ANN-powered design equations and graphical user interface for capacity prediction of short concrete-filled double skin steel tubular columns
Abstract Concrete-filled double steel tubular (CFDST) and concrete-filled double-skin steel tube (CFDSST) columns have emerged as an efficient steel–concrete composite system, offering improved axial capacity than conventional columns. However, traditional methods, such as experimental and numerical studies, for assessing their capacity can be both complex and time-consuming, highlighting the need for accurate and efficient data-driven predictive tools. Hence, this paper aims at developing an ANN-based model for predicting the axial capacity of short circular CFDSST columns, incorporating both normal and high-strength materials. A comprehensive database comprising 296 CFDSST specimens, consisting of 138 validated FEA results and 158 experimental data collected from the literature, was used for model development. The ANN model was trained using eight input parameters representing the geometric and material properties of the column. The developed ANN model achieved an overall correlation coefficient of 0.9978, demonstrating excellent predictive capability. Furthermore, the proposed model exhibited improved prediction accuracy when compared with existing design provisions and previously published theoretical models, using various performance indices, achieving RMSE of 207.74, MAPE of 4.62%, R 2 of 0.99, VAF of 99.38% and a20 index of 0.97, demonstrating improved prediction accuracy and reliability over existing design approaches. In addition, the Shapley additive interpretation (SHAP) technique is also adopted to examine the contribution of input design parameters for predicting axial capacity. To further strengthen the explainable AI framework, Partial Dependence Plots (PDPs) were also incorporated alongside SHAP to illustrate the nonlinear effect of the governing input variables on the predicted axial capacity. Lastly, an explicit ANN-based equation is proposed and based on these, a Graphical User Interface (GUI) is also developed to facilitate user-friendly approach for the prediction of the capacity of stub CFDSST columns. Overall, the proposed ANN framework, coupled with explainable AI and a user-friendly Excel-based design tool, proves an accurate, efficient and practical approach for predicting the axial capacity of the CFDSST columns, thereby supporting reliable design and optimization in engineering practice.
Characteristics and distribution of sweet spots in ultra-deep basement reservoirs in compressional tectonic settings: insights from the Northern Qaidam Basin
Single-cell pancancer analysis reveals epithelial–immune crosstalk in shaping the premetastatic niche of liver metastasis
An interpretable machine learning model for preoperative risk stratification of testicular viability in pediatric testicular torsion: A multicenter study
An intelligent decision framework for personalized sports training plan optimization using fermatean fuzzy CURLI based MCDM
Modelling farmer’s intention-behavior to participate in ecological governance in the Northwest China arid region
Chemical signatures, source characteristics, and transboundary transport of PM0.49 in Thailand–Myanmar border region
Abstract Air pollution, particularly biomass burning (BB)–derived particulate matter (PM), remains a major challenge in Southeast Asia, affecting atmospheric processes, climate, human health, and socio-economic systems. This study presents the chemical characterization and source contributions of PM 0.49 at a rural site along the Thailand–Myanmar border in 2023. PM 0.49 concentrations showed pronounced temporal variability, with smoke-haze (SH) concentrations (67.9 ± 46.0 µg/m 3 ) nearly six times higher than those during non-haze periods (non-SH; 10.9 ± 4.4 µg/m 3 ). Organic carbon (OC) dominated during SH, accounting for 46% of PM 0.49 mass, followed by secondary inorganic aerosols (SO 4 2− , NO 3 − , NH 4 + ; 13%) and BB tracers such as K + (1.4%) and levoglucosan (2.7%), indicating strong BB influence. K⁺ exhibited strong correlations with levoglucosan ( r = 0.95) and major anions ( r = 0.52–0.94), highlighting its dual role as both a neutralizing cation and a BB tracer in both fresh and aged BB aerosols. Elevated OC to elemental carbon ratios (OC/EC; 13.4 ± 4.6) and char-EC/soot-EC ratios (9.68 ± 4.82) indicate smoldering BB as the principal source. Positive matrix factorization attributed 57% of PM 0.49 to BB during SH (up to 69% during peak phase), whereas soil dust dominated during non-SH conditions (70%). Backward trajectories and fire hotspot analyses suggest substantial cross-border transport from Myanmar. Furthermore, the seasonal shift in the effective carbon ratio (ECR), from 0.63 (SH) to 1.13 (non-SH), indicates a transition from light-absorbing to light-scattering aerosol dominance, underscoring the role of PM 0.49 in aerosol transformation under BB influence. These findings suggest air quality and potential climate implications, supporting coordinated mitigation.
Whitefly detection in coconut tree leaves using deep learning
Correction: Optimization of enzyme-assisted extraction of polyphyllins from paris polyphylla var. yunnanensis rhizomes using response surface methodology
Highly efficient atmospheric water harvesting via a biomimetic condensation structure integrated with passive radiative cooling
Abstract Atmospheric water harvesting (AWH) via thermoelectric condensation offers a promising solution for localized water scarcity but is constrained by high energy consumption and low energy efficiency. In this study, we employ selective laser melting (SLM) technology to fabricate a biomimetic 3-D condensation structure. The synergistic effects of geometric curvature gradients and optimized surface wettability significantly enhance water collection efficiency compared to conventional fin designs. Electrochemical polishing is further applied to reduce surface pinning effects, thereby promoting droplet removal. Additionally, a superhydrophobic passive radiative cooling coating is introduced onto the structure’s surface. This coating reflects a substantial portion of incident solar radiation and continuously emits thermal radiation through the atmospheric window, effectively reducing cooling energy demand. Under an applied power of only 56.3 mW cm − 2 , the biomimetic structure achieves a water collection rate of 247.54 mg cm − 2 h − 1 , which is 4.7 times higher than that of a commercial fin condenser. Field testing demonstrates that incorporating the passive radiative cooling coating improves water harvesting efficiency by up to 39.8%. This study presents a viable approach to enhancing the performance of thermoelectric AWH systems, laying a foundation for mitigating the global water-energy conflict.
Higher order interactions in global language endangerment revealed by multilayer language country bipartite network analysis
Formal verification of integrity attacks on connected medical devices in battlefield hospitals
Abstract Field hospitals in battlefield settings increasingly rely on connected medical devices, including monitors, ventilators, infusion pumps, patient-identification services, and gateway-based communication. In such environments, integrity attacks may create patient-safety hazards even when the system appears operational, by tampering with data, replaying old messages, injecting false information, or swapping patient identities. This work presents a formal framework for analyzing these attacks as a stochastic clinical–cyber system that integrates patient-state evolution, intermittent communication, attacker actions, and gateway-centered defense policies. Safety is evaluated using bounded unsafe reachability, supported by formal verification and resource-aware strategy synthesis. A synthetic digital twin is used to generate benign and attacked traces without real patient data; therefore, the findings are interpreted as model-based and simulation-based evidence rather than clinical validation. Confirmatory Monte Carlo evaluation used 5000 trajectories per configuration under 20 fixed root seeds. Patient-ID swapping reached an empirical unsafe reachability of 0.297 (95% CI 0.284–0.310) at the largest evaluated per-step attack-capacity limit, while the combined adaptive attacker reached 0.384 (95% CI 0.371–0.397). Evaluation of the synthesized combined gateway policy reduced empirical model-estimated unsafe reachability from 0.392 (95% CI 0.378–0.406) to 0.119 (95% CI 0.110–0.128) in the evaluated synthetic scenario, corresponding to a 69.6% relative reduction in this formal safety proxy, but with increased workflow burden. A separate PRISM-games worst-case query returned 0.410 under the disclosed finite abstraction. Verification remained tractable at the evaluated scales, with median PRISM runtime increasing from 0.90 to 28.70 s as the abstract state space expanded. Overall, the study supports formal verification as an analytical framework for safety-aware cyber-defense evaluation in connected field-hospital medical systems under explicit modeling assumptions; it does not establish clinical risk reduction or deployment readiness.