Browse Articles
Discover research articles across all indexed journals
Dynamics analysis of a novel conservative chaotic system and its application in image encryption
A novel multi-criteria group decision-making technique in traffic flow and safety assessment using interval-valued complex spherical fuzzy soft set
Extreme temperature and infant deaths in rural Spain, 1920–1950: a case-crossover analysis
Simulation study on nonlinear permeability characteristics of coalbed methane mining driven by mixed N2/CO2 pressure
Federated TinyML and digital twin framework for secure and resilient IoMT-based ICU monitoring
Vehicle lane change prediction with explainable soft mask attention mechanism and heterogeneous information encoder
Abstract Lane-change intention prediction is critical for intelligent vehicles, and driver decisions depend on the perception and processing of driving context information. Despite advances in deep learning in this domain, further exploration of the driving context processing remains essential. This study proposes a soft mask attention mechanism to adaptively enhance or suppress input features from target and surrounding vehicles. After that, the masked features are extracted using a heterogeneous information encoder, thereby differentiating between target and surrounding vehicle information processing. The encoded features are then integrated using a multi-head attention mechanism, and the lane change probabilities are output through convolution operations and decoder layers. Experiments demonstrate: (1) The proposed SMILE-LC (Soft Mask Information with Lane-based Encoding for Lane Change) achieves optimal prediction performance across all perception ranges with strong efficiency and generalization. (2) The soft mask attention mechanism intuitively reveals the information processing patterns in prediction model. The target vehicle’s lateral acceleration is the most critical feature, while the position information of surrounding vehicles is more important than their velocity and acceleration. (3) The heterogeneous information encoder significantly improves lane change prediction performance, and the proposed lane-based encoding strategy outperforms other encoding architectures. These results can significantly advance the development of Advanced Driver Assistance Systems (ADAS) and enhance lane-change safety.
Prevalence and associated factors of depression, anxiety, and stress among women of childbearing age in Lorestan, Iran: a cross-sectional study
A simulation-based investigation of the production routes of the positron-emitting radionuclide 128Cs using GEANT4, TALYS, and EMPIRE codes for medical applications
Sexual experience of urostomy patients with bladder cancer: a qualitative explorative study
A facile and effective approach to prepare К-carrageenan / polyvinyl alcohol hydrogel as a potential wound dressing
Abstract Since К-carrageenan (КC) can be crosslinked by potassium cations while polyvinyl alcohol ( PVA) can be crosslinked by borated anions, the present work was undertaken to crosslink the КC/PVA blends in one step using potassium tetraborate tetrahydrate (PTB) as a crosslinker for forming КC/PVA/PTB hydrogels. Factors affecting crosslinking of that blends such as PTB concentration, steeping time in PTB aqueous solution, КC/PVA weight ratio, and crosslinker type were studied. The results indicated that the optimal conditions for preparing the КC/PVA/PTB gels are: КC/PVA weight ratio, 1; PTB concentration, 4%; and steeping time, 8 h. Among boric acid, potassium chloride, borax, and PTB as potential crosslinkers for the КC/PVA (50/50) matrix, the later, i.e. PTB, achieves the highest crosslinking magnitude and produces KC/PVA/PTB gel having a superior liquid-holding capacity (1020%) as well as high gel fraction (98.6%) properties. The chemical structure of the formed КC/PVA/PTB gel was confirmed using FTIR analysis. The potential application of the КC/PVA/PTB gel as wound dressings was investigated by treating non-woven cotton fabric samples with КC/PVA (50/50) matrices in absence and presence of curcumin or silver nano-particles (Ag-NPs) as bio-additives followed by crosslinking with PTB. The prepared dressings exhibits superior antibacterial activities as well as significant mechanical properties expressed in the good tensile strength, elongation at break, elastic modulus, stiffness, air permeability, and porosity properties. Besides, the treated fabric exhibits moderate cytotoxicity. The prepared Ag-NPs were evaluated via UV–Vis, TEM, and XRD analysis whereas the prepared dressings were characterized using SEM and EDX analysis.
Association of surrounding greenness with preterm birth and the potential mediating role of ambient air pollution
Novel index for assessing the intestinal environment in prebiotic-supplemented healthy adults via non-invasive gas analysis
When predators unexpectedly decrease instead of increase plant production
Directional propagation of interface modes in topological acoustic metamaterials via spin-momentum locking
Symbolic and domain-generalized machine learning for interpretable solubility modeling in supercritical CO₂
Abstract Accurate prediction of drug solubility in supercritical CO₂ remains challenging due to the limited generalizability of compound-specific correlations and the black-box nature of most machine learning models. This study proposes a domain-aware symbolic regression framework that discovers closed-form analytical expressions for ln( y ), enabling interpretable and transferable solubility modeling across chemically diverse pharmaceutical compounds. A leave-one-drug-out (LODO) validation strategy is employed to rigorously assess extrapolation performance on unseen drug domains across a curated dataset of 196 experimental data points spanning 9 antihypertensive compounds. Under this evaluation, the domain-adversarial neural network (DANN) component of the proposed framework achieved RMSE = 0.33 ± 0.08, MAE = 0.24 ± 0.06, and R² = 0.89 ± 0.05, outperforming standard machine learning baselines including Random Forest, XGBoost, and Multi-Layer Perceptron in cross-domain generalization. The symbolic regression component additionally recovered compact closed-form analytical expressions that accurately represent the full experimental dataset (in-sample expression fit: R² = 0.962, RMSE = 0.031, MAE = 0.024) while remaining physically interpretable and analytically tractable. The combined framework demonstrates that domain-aware learning and symbolic discovery can jointly address the dual challenge of predictive robustness under domain shift and model interpretability in supercritical pharmaceutical solubility modeling.
Enhancing the tensile performance of GFRE composites: a comparative study of micro- and nano-copper fillers
Impact of electrode drying time on capacitive performance of honey-derived graphene nanosheets for supercapacitors
Abstract Graphene nanosheets have a significant impact in the energy storage field, particularly in the realm of supercapacitors and capacitive deionization, due to their exceptional properties. This research aims to develop a facile method for preparing graphene nanosheets from biomass and examine the influence of the electrode drying time on the electrochemical properties of the prepared electrodes. Honey, as a carbon source, offers advantages over typical biomass because of its uniform composition, high carbon content, and ability for controlled low-temperature carbonization, which improves porosity and wettability. This chemical process was followed by KOH chemical activation with N 2 gas injection. The physical and chemical properties of graphene nanosheets were examined with X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM), energy-dispersive X-ray (EDX), Transmission Electron Microscope (TEM), low-temperature nitrogen adsorption-desorption for isothermal characterization, zeta potential, and particle size measurements. The honey-based graphene nanosheets have a specific surface area of 1427 m² g⁻¹ and a pore volume of 0.654 cm³g⁻¹. Consequently, the electrochemical performance of the prepared electrodes was assessed via galvanostatic charge-discharge (GCD), electrochemical impedance spectroscopy (EIS), and cyclic voltammetry (CV). However, the optimum drying time (24 h) had a significant effect on the specific capacitance of electrodes (maximum of 240 Fg⁻¹), which was achieved at a current density of 0.3 Ag⁻¹ when tested in a 0.5 M Na 2 SO 4 aqueous electrolyte. These results present promising rate capability and competitive performance for supercapacitor applications compared with other graphene-based supercapacitor electrodes. Moreover, this sustainable and cost-effective method may enhance the performance of supercapacitor electrodes.