Browse Articles
Discover research articles across all indexed journals
An explainable RoBERTa approach to analyzing panic and anxiety sentiment in oral health education YouTube comments
An explainable multi-task deep learning framework for crash severity prediction using multi-source data
Abstract Traffic accidents pose significant global challenges, causing substantial injuries, fatalities, and economic losses. Current research predominantly focuses on single-prediction objectives (e.g., fatality prediction) while neglecting property damage assessments and critical interactions between prediction tasks. Although neural networks demonstrate superior predictive capabilities, their application in traffic safety analysis remains constrained by inherent limitations in causal interpretability, coupled with challenges posed by data imbalance, heterogeneity, and complexity in crash datasets. This study proposes an interpretable multi-task learning framework (Adv MT-DNN) that synergistically integrates an enhanced deep neural network with post-hoc explanation methods for comprehensive crash severity prediction. Our dual-focused approach addresses multiple prediction targets (including fatalities, severe injuries, and property damage). It provides granular insights into contributing factors through SHAP-based feature importance rankings and interaction analysis. Validated using four-year (2018–2021) multi-source traffic data from China, the framework demonstrates significant improvements in prediction accuracy compared to baselines. Nonparametric estimation of the top-8 critical factors (e.g., blood alcohol content, collision type, and accident occurrence period) confirms statistically significant associations with crash severity. The explicit interpretation mechanism bridges the critical gap between predictive performance and model interpretability in traffic safety analytics, providing engineering-relevant insights. This research establishes a robust methodological foundation for developing data-driven road safety policies and intelligent transportation systems, particularly in developing countries with complex traffic ecosystems.
Innovative adaptation of coagulation-sedimentation-filtration process in lightly polluted urban rivers with seasonal high turbidity
3D collagen nanofiber scaffold with adipose derived stem cells for functional adipose tissue regeneration
An algorithm for road target detection of autonomous vehicles based on improved YOLOv8
AI-based routing algorithms improve energy efficiency, latency, and data reliability in wireless sensor networks
Integrated TE optical isolator based on magneto-optical perturbation in coupled waveguides
Abstract Photonic integrated circuits (PICs) increasingly require more advanced integrated functionalities and devices to meet the key application challenges. This evolution with the serial integration of optical functions in the same photonic circuit often results in internal reflections and optical feedback, which can specifically destabilize lasers. One solution to overcome this issue is to integrate an optical isolator at the output of the lasers. Among various proposed designs, magneto-optical-based isolators have significantly improved over the past decades in terms of compactness, insertion loss, isolation ratio, and spectral isolation bandwidth. Despite these improvements, the TE optical isolator still lacks performance. This paper introduces a new operational principle for a TE optical isolator based on modal beating in a transverse magneto-optical Kerr effect (TMOKE) coupled-waveguide system. This approach combines evanescent-coupled silicon waveguides and the magneto-optical effect, resulting in a nonreciprocal propagation that can be optimized for optical isolation. This new concept shows promise in achieving a high-performing TE optical isolator, as it does not depend on resonance and is free from constraints associated with interferometers. Based on data from magneto-optical garnet materials, the simulated device has a length of 500 $$\upmu \textrm{m}$$ and its 20 dB-isolation bandwidth is as high as 35 nm. With broadband and high isolation, this simulated device opens possibilities for miniaturizing complex photonic circuits used in optical communication, data communication, and optical sensing.
Research on the simulation methods and shock resistance performance of energy absorbing columns in rockburst roadway supports
Setdb1 ablation in macrophages attenuates fibrosis in heart allografts
Tissue fibrosis is commonly associated with organ malfunction and is strongly associated with the development of chronic rejection, cardiovascular diseases, and other chronic diseases. Fibrosis also contributes to immune exclusion in tumor tissues. Targeting fibrosis might be a strategy for prolonging allograft survival while suppressing cancer development. Here, single-cell transcriptomes of human and mouse heart allografts showed that macrophages accumulated in grafts with fibrosis were reprogrammed via histone methylation regulated by Setdb1, an H3K9 methyltransferase. Myeloid-specific deletion of Setdb1 prolonged heart allograft survival but reversed immune exclusion in tumor tissues. Interestingly, myeloid-specific Setdb1-knockout led to lower fibrosis in heart allografts and tumor tissues in mice. Our single-cell sequencing data showed that Setdb1 ablation impaired Fn1 + and SPP1 + profibrogenic macrophage reprogramming. Mechanistically, Fn1, which was induced by the CCR2-Creb/Setdb1 axis, upregulated the expression of genes related to fibrosis in fibroblasts and macrophages via ITGA5 and PIRA receptors. Blocking the interaction between FN1 and these receptors inhibited fibrosis in allograft and tumor tissues. Our results reveal a target, histone methylation in macrophages, for the treatment of fibrosis-related disease.
A rationally designed 2C inhibitor prevents enterovirus D68-infected mice from developing paralysis
Microsecond-lived quantum states in a carbon-based circuit driven by cavity photons
Identification of immunogenic and cross-reactive chikungunya virus epitopes for CD4+ T cells in chronic chikungunya disease
Synergistic Effect of Boron and Oxygen Coordination on Ruthenium Clusters for Industrial Water Splitting in Alkaline Medium
Abstract The alteration in the coordination environment of metal atoms can manipulate their electronic structure and regulate the electrocatalytic hydrogen evolution activity. In this work, synchrotron radiation tests prove that the boron (B) and oxygen (O) elements co‐coordinate with ruthenium clusters (Ru C ) on the surface of B–O modified reduced graphene oxide. The electrochemical tests demonstrate that this unique structure electrocatalyst presents an overpotential of 12 mV in 1 M KOH condition and for over 120 h at the current of −1 A cm −2 , indicating potential practical applications. The quasi in‐situ X‐ray photoelectron spectroscopy and in‐situ infrared spectroscopy confirmed that the B–O diatomic coordination can modulate the synergy between the substrate and the Ru C catalytic site, enhancing the intrinsic catalytic activity and ion migration efficiency. The first principles calculation further proves that B–O diatomic coordination will reduce the desorption barrier of H* and construct a complete hydrogen migration path. This study discloses the significance of the synergistic effect of two anions to enhance the catalytic activity of the catalyst by altering the coordination environment of ruthenium clusters.
The burden of age-related macular degeneration and its socioeconomic associates in the Eastern Mediterranean Region from 1990 to 2021
Compound climate events and their role in the decay of Greece’s cultural heritage
Synergistic effect of Agrococcus and Rossellomorea Marisflavi species assisted probiotic functional feed on Vibrio affected Nile tilapia fish
Characterizing full-shift worker exposures to VOCs in small-sized auto repair shops in the Tucson, Arizona, USA metropolitan area
Survival benefits of primary tumor resection in metastatic differentiated thyroid cancer: an analysis of SEER data
Multi-objective artificial-intelligence-based parameter tuning of antennas using variable-fidelity machine learning
Abstract Multi-objective optimization (MO) is an important topic in contemporary antenna design. Due to the reliance on computationally-expensive electromagnetic (EM) simulations, the use of conventional algorithms is prohibitive. These costs can be reduced by appropriate algorithmic tools involving surrogate modeling and soft computing methods. This study introduces an innovative artificial intelligence (AI)-based approach to antenna MO. Our algorithm is a machine learning (ML) procedure employing artificial neural network models. In each iteration, multiple infill vectors are produced, using Pareto ranking of the candidate solution set produced by a multi-objective evolutionary algorithm. The full-wave simulation results acquired for all infill points are incorporated into the dataset to refine the metamodel. Termination of the procedure is based on a comparison of non-dominated solutions obtained in subsequent iterations. Additional reduction of the expenses is enabled through the use of multi-resolution electromagnetic simulations. The presented methodology has been extensively demonstrated with the help of four planar devices, including broadband monopoles and a quasi-Yagi antenna. As shown, the average cost of MO is equivalent to approximately two hundred high-fidelity EM analyses. In absolute terms, 40% of relative speedup is achieved due to variable-fidelity modeling, and almost 90% savings over the one-shot approach. Comparative experiments indicate that the improved computational efficiency of the presented framework is not detrimental to reliability. Consequently, the introduced algorithm can be regarded a feasible alternative to the current MO methodologies for antennas, especially when computational budget is a critical constraint.