Search Results (6,502)

The tumor microenvironment (TME) has a substantial impact on the progression of gastric cancer. Collagen, the most abundant protein in the extracellular matrix (ECM), forms a dense physical barrier that regulates anti-tumor immunity in the TME. It is a significant regulator of the signaling pathways of cancer cells, which are responsible for migration, proliferation, and metabolism. ECM proteins, particularly remodeling enzymes and collagens, can be modified to increase stiffness and alter the mechanical properties of the stroma. This, in turn, increases the invasive potential of tumor cells and resistance to immunotherapy. Given the dynamic nature of collagen, novel therapeutic strategies have emerged that target both collagen biosynthesis and degradation, processes that are essential for addressing ECM stiffening. This review delineates the upregulation of the expression and deposition of collagen, as well as the biological functions, assembly, and reorganization that contribute to the dissemination of this aggressive malignancy. Furthermore, the review emphasizes the importance of creating 3D in vitro models that incorporate innovative biomaterials that avoid the difficulties of traditional 2D culture in accurately simulating real-world conditions that effectively replicate the distinctive collagen microenvironment. Ultimately, it investigates the use of decellularized ECM-derived biomaterials as tumor models that are designed to precisely replicate the mechanisms associated with the progression of stomach cancer. Full article

The aim of this study was to address the issue of significant performance degradation in existing defogging algorithms under extreme fog conditions. Traditional Taylor series-based deformable convolutions are limited by local approximation errors, while the heavy-tailed characteristics of the Cauchy distribution can more successfully model outliers in fog images. The following improvements are made: (1) A displacement generator based on the inverse cumulative distribution function (ICDF) of the Cauchy distribution is designed to transform uniform noise into sampling points with a long-tailed distribution. A novel double-peak Cauchy ICDF is proposed to dynamically balance the heavy-tailed characteristics of the Cauchy ICDF, enhancing the modeling capability for sudden changes in fog concentration. (2) An innovative Cauchy–Gaussian fusion module is proposed to dynamically learn and generate hybrid coefficients, combining the complementary advantages of the two distributions to dynamically balance the representation of smooth regions and edge details. (3) Tree-based multi-path and cross-resolution feature aggregation is introduced, achieving local–global feature adaptive fusion through adjustable window sizes (3/5/7/11) for parallel paths. Experiments on the RESIDE dataset demonstrate that the proposed method achieves a 2.26 dB improvement in the peak signal-to-noise ratio compared to that obtained with the TaylorV2 expansion attention mechanism, with an improvement of 0.88 dB in heavily hazy regions (fog concentration > 0.8). Ablation studies validate the effectiveness of Cauchy distribution convolution in handling dense fog and conventional lighting conditions. This study provides a new theoretical perspective for modeling in computer vision tasks, introducing a novel attention mechanism and multi-path encoding approach. Full article

Fast, spark-free detection of hydrogen leaks is indispensable for large-scale hydrogen deployment, yet electronic sensors remain power-intensive and prone to cross-talk. Optical schemes based on surface plasmons enable remote read-out, but single-metal devices offer either weak H₂ affinity or poor plasmonic quality. Here we employ full-wave finite-difference time-domain (FDTD) simulations to map the hydrogen response of nanohole arrays (NAs) that can be mass-produced by colloidal lithography. Square lattices of 200 nm holes etched into 100 nm films of Pd, Mg, Ti, V, or Zr expose an intrinsic trade-off: Pd maintains sharp extraordinary optical transmission modes but shifts by only 28 nm upon hydriding, whereas Mg undergoes a large dielectric transition that extinguishes its resonance. Vertical pairing of a hydride-forming layer with a noble metal plasmonic cap overcomes this limitation. A Mg/Pd bilayer preserves all modes and red-shifts by 94 nm, while the predicted optimum Ag (60 nm)/Mg (40 nm) stack delivers a 163 nm shift with an 83 nm linewidth, yielding a figure of merit of 1.96—surpassing the best plasmonic hydrogen sensors reported to date. Continuous-film geometry suppresses mechanical degradation, and the design rules—noble-metal plasmon generator, buried hydride layer, and thickness tuning—are general. This study charts a scalable route to remote, sub-ppm, optical hydrogen sensors compatible with a carbon-neutral energy infrastructure. Full article

The utility of nanostructured TiO₂ in the degradation of organic compounds and the disinfection of pathogenic microorganisms represents an important endeavor in photocatalysis. However, the low photocatalytic efficiency of TiO₂ remains challenging. Herein, we report a robust photocatalytic route to benzene removal rendered by enhancing its adsorption capacity via rationally designed mesoporous SiO₂-coated TiO₂ colloids. Specifically, amorphous, mesoporous SiO₂-coated TiO₂ nanoparticles (denoted T@S NPs) are produced via a precipitation-gel-hydrothermal approach, possessing an increased specific surface area over pristine TiO₂ NPs for improved adsorption of benzene. Notably, under UV irradiation, the degradation rate of benzene by T@S NPs reaches 89% within 30 min, representing a 3.1-fold increase over that achieved by pristine TiO₂. Moreover, a 99.5% degradation rate within 60 min is achieved and maintains a stable photocatalytic activity over five cycles. Surface coating of TiO₂ with amorphous SiO₂ imparts the T@S composite NPs nearly neutral characteristic due to the formation of Ti-O-Si bonds, while manifesting enhanced light harvesting, excellent stability, adhesion, and photocatalytic bacteriostatic effects. Our study underscores the potential of T@S composites for practical applications in photocatalysis over pristine counterparts. Full article

Establishing a sustainable framework for remediating environmental degradation caused by historical mining operations in the Sierra Minera of Cartagena-La Unión, southeastern Spain, is a critical imperative. When the reclamation requirements of the post-mining district are considered in the context of its critical location, nested among conflicting land uses, the development of practical solutions to restore ecological and cultural functions emerge as a landscape planning challenge. The greenway approach emphasizes the primary ecological and functional corridors that sustain the vitality of the region; therefore, it is essential to preserve and enhance these critical lifelines. This study aimed to design a localized greenway network to support the conservation of key ecological, agricultural, and cultural resources within the area, while simultaneously promoting reclamation activities in degraded zones. The greenway corridor is built upon key elements: conservation areas, post-mining cultural resources, dry riverbeds, and agricultural zones. In the light of greenway approach, planners and land managers can make their decisions more judiciously by considering the priority zones. The protection, leveraging, and reclamation of significant resources can be provided through a multifunctional greenway approach as seen in the case of Cartagena-La Unión Post-Mining District. Full article

To address the issue of key component degradation in hybrid electric commercial vehicles under complex driving cycles negatively impacting system economy and durability, this paper proposes a model predictive control (MPC)-based energy management co-optimization strategy. Firstly, dynamic degradation models for the key components are established, enabling high-fidelity characterization of component health status. Secondly, a system-level model incorporating vehicle dynamics, power battery, and electric drive motor is developed, with the degradation feedback mechanism deeply integrated. Building on this foundation, an MPC-based energy management strategy for multi-objective optimization is designed. Its core functionality lies in the cooperative allocation of power sources within a rolling horizon framework: by integrating component degradation status as critical feedback into the control loop, the strategy proactively coordinates the optimization objectives between operational economy (minimization of equivalent energy consumption) and key component durability (degradation mitigation). Simulation results demonstrate that, compared to traditional energy management strategies, the proposed strategy significantly enhances system performance while ensuring vehicle drivability: equivalent energy efficiency improves by approximately 3.5%, component degradation is reduced by up to 87%, and superior state of charge (SOC) regulation capability for the battery is achieved. This strategy provides an effective control method for achieving intelligent, long-life, and high-efficiency operation of hybrid electric commercial vehicles. Full article

Machine learning (ML) has greatly improved intrusion detection in enterprise networks. However, ML models remain vulnerable to adversarial attacks, where small input changes cause misclassification. This study evaluates the robustness of a Random Forest (RF), a standard neural network (NN), and a Transformer-based Network Intrusion Detection System (NIDS). It also introduces ADV_NN, an adversarially trained neural network designed to improve resilience. Model performance is tested using the UNSW-NB15 dataset under both clean and adversarial conditions. The attack types include Projected Gradient Descent (PGD), Fast Gradient Sign Method (FGSM), and Black-Box transfer attacks. The proposed ADV_NN achieves 86.04% accuracy on clean data. It maintains over 80% accuracy under PGD and FGSM attacks, and exceeds 85% under Black-Box attacks at $ϵ=0.15$. In contrast, the RF, NN, and Transformer-based models suffer significant degradation under adversarial perturbations. These results highlight the need to incorporate adversarial defenses into ML-based NIDS for secure deployment in real-world environments. Full article

The integration of artificial intelligence (AI) in medical imaging has revolutionized diagnostic capabilities, yet the black-box nature of deep learning models poses significant challenges for clinical adoption. Current explainable AI (XAI) approaches, including SHAP, LIME, and Grad-CAM, predominantly focus on post hoc explanations that may inadvertently undermine clinical decision-making by providing misleading confidence in AI outputs. This paper presents a systematic review and meta-analysis of 67 studies (covering 23 radiology, 19 pathology, and 25 ophthalmology applications) evaluating XAI fidelity, stability, and performance trade-offs across medical imaging modalities. Our meta-analysis of 847 initially identified studies reveals that LIME achieves superior fidelity (0.81, 95% CI: 0.78–0.84) compared to SHAP (0.38, 95% CI: 0.35–0.41) and Grad-CAM (0.54, 95% CI: 0.51–0.57) across all modalities. Post hoc explanations demonstrated poor stability under noise perturbation, with SHAP showing 53% degradation in ophthalmology applications (ρ = 0.42 at 10% noise) compared to 11% in radiology (ρ = 0.89). We demonstrate a consistent 5–7% AUC performance penalty for interpretable models but identify modality-specific stability patterns suggesting that tailored XAI approaches are necessary. Based on these empirical findings, we propose a comprehensive three-pillar accountability framework that prioritizes transparency in model development, interpretability in architecture design, and a cautious deployment of post hoc explanations with explicit uncertainty quantification. This approach offers a pathway toward genuinely accountable AI systems that enhance rather than compromise clinical decision-making quality and patient safety. Full article

Emergency diesel is prone to degradation during long-term storage, and experimental evaluations are costly and slow. Three-dimensional computational fluid dynamics (CFD) simulations were employed to model the diesel conditioning process. A physical model based on the actual dimensions of the storage tank was constructed. The volume of fraction (VOF) model tracked the gas–liquid interface, and the species transport model handled mixture transport. A UDF then recorded inlet and outlet flow rates and velocities in each cycle. The study focused on the effects of different baffle structures and numbers on conditioning efficiency. Results showed that increasing the baffle flow area significantly delays the mixing time but reduces the cycle time. Openings at the bottom of baffles effectively mitigate the accumulation of high-concentration conditioning oil caused by density differences. Increasing the number of baffles decreases the effective volume of the tank and amplifies density differences across the baffles, which shortens the mixing time. However, excessive baffle numbers diminish these benefits. These findings provide essential theoretical guidance for optimizing baffle design in practical diesel tanks, facilitating rapid achievement of emergency diesel quality standards while reducing costs and improving efficiency. Full article

Cost-effective procedures usually cannot achieve complete removal of priority contaminants present in water at very low concentrations (as pesticides or pharmaceuticals). Advanced oxidation processes (AOPs) represent promising technologies for removing priority contaminants from water at trace concentrations, yet practical implementation remains limited due to technical and economic constraints. This study presents an innovative flow-through photodegradation device designed to overcome current limitations while achieving efficient contaminant removal at industrial scale. The device integrates a UVC 254 nm lamp-equipped flow chamber with automated dosing pumps for hydrogen peroxide and/or solid catalyst suspensions, coupled with a 30 nm porous membrane filtration system for catalyst recirculation. This configuration optimizes light–catalyst–pollutant contact while enabling combined catalytic processes. Performance evaluation using acesulfame (ACE) and iohexol (IHX) as model contaminants demonstrated rapid and effective removal. IHX degradation with UVC and 75 μM H₂O₂ achieved complete removal with t₉₅% = 7.23 ± 1.21 min (pseudo-order 0.25, t₁/₂ = 3.27 ± 0.39 min), while ACE photolysis (with UVC only) required t₉₅% = 14.88 ± 2.02 min (pseudo-order 1.27, t₁/₂ = 2.35 ± 0.84 min). The introduction of t₉₅% as a performance metric provides practical insights for near-complete contaminant removal requirements. Real-world efficacy was confirmed using tertiary wastewater treatment plant effluents containing 14 μg/L IHX, achieving complete removal within 8 min. However, carbamazepine degradation proved slower (t₉₅% > 74 h), highlighting the need for combined catalytic approaches for recalcitrant compounds. Spiking experiments (1000 μg/L) revealed concentration-dependent kinetics and synergistic effects between co-present contaminants. Analysis identified degradation byproducts consistent with previous studies, including tri-deiodinated iohexol (474.17 Da) intermediates. This scalable system, constructed from commercially available components, demonstrates potential for cost-effective industrial implementation. The modular design allows adaptation to various contaminants through adjustable AOP combinations (UV/H₂O₂, photocatalysts, ozone), representing a practical advancement toward addressing the gap between laboratory-scale photocatalytic research and full-scale water treatment applications. Full article

High-Temperature Proton-Exchange Membrane Fuel Cells (HT-PEMFCs) represent a promising clean energy technology and are valued for their fuel flexibility and simplified balance of plant. Their commercialization, however, is critically hindered by the prohibitive cost and resource scarcity of platinum-group metal (PGM) catalysts. The challenge is amplified in the phosphoric acid (PA) electrolyte of HT-PEMFCs, where the severe anion poisoning of PGM active sites necessitates impractically high catalyst loadings. This review addresses the urgent need for cost-effective alternatives by providing a comprehensive assessment of recent advancements in non-precious metal (NPM) catalysts for the oxygen reduction reaction (ORR) in HT-PEMFCs. It systematically explores synthesis strategies and structure–performance relationships for emerging catalyst classes, including transition metal compounds, metal–nitrogen–carbon (M-N-C) materials, and metal-free heteroatom-doped carbons. A significant focus is placed on M-N-C catalysts, particularly those with atomically dispersed Fe-Nx active sites, which have emerged as the most viable replacements for platinum due to their high intrinsic activity and notable tolerance to phosphate poisoning. This review critically analyzes key challenges that impede practical application, such as the trade-off between catalyst activity and stability, mass transport limitations in thick electrodes, and long-term degradation in the harsh PA environment. Finally, it outlines future research directions, emphasizing the need for a synergistic approach that integrates computational modeling with advanced operando characterization to guide the rational design of durable, high-performance catalysts and electrode architectures, thereby accelerating the path to commercial viability for HT-PEMFC technology. Full article

This paper presents a Ka-band cylindrical conformal transceiver active phased array (CCTAPA) with a full-azimuth scanning gain fluctuation of 0.8 dB and low power consumption. The array comprises 20 panels of 4 × 4 antenna elements, RF beam-control circuits, a Wilkinson power divider network, and frequency converters. The proposed three-subarray architecture enables ±9° beam scanning with minimal gain degradation. By dynamically switching subarrays and transceiver channels across azimuthal directions, the array achieves full 360° coverage with low gain fluctuation and power consumption. Fabrication and testing demonstrate a gain fluctuation of 0.8 dB, equivalent isotropically radiated power (EIRP) between 50.6 and 51.3 dBm, and a gain-to-noise-temperature ratio (G/T) ranging from −8 dB/K to −8.5 dB/K at 28.5 GHz. The RF power consumption remains below 8.73 W during full-azimuth scanning. This design is particularly suitable for airborne platforms requiring full-azimuth coverage with stringent power budgets. Full article

Autonomous driving is a complex and fast-evolving domain at the intersection of robotics, machine learning, and control systems. This paper provides a systematic review of recent developments in reinforcement learning (RL) and imitation learning (IL) approaches for autonomous vehicle control, with a dedicated focus on the CARLA simulator, an open-source, high-fidelity platform that has become a standard for learning-based autonomous vehicle (AV) research. We analyze RL-based and IL-based studies, extracting and comparing their formulations of state, action, and reward spaces. Special attention is given to the design of reward functions, control architectures, and integration pipelines. Comparative graphs and diagrams illustrate performance trade-offs. We further highlight gaps in generalization to real-world driving scenarios, robustness under dynamic environments, and scalability of agent architectures. Despite rapid progress, existing autonomous driving systems exhibit significant limitations. For instance, studies show that end-to-end reinforcement learning (RL) models can suffer from performance degradation of up to 35% when exposed to unseen weather or town conditions, and imitation learning (IL) agents trained solely on expert demonstrations exhibit up to 40% higher collision rates in novel environments. Furthermore, reward misspecification remains a critical issue—over 20% of reported failures in simulated environments stem from poorly calibrated reward signals. Generalization gaps, especially in RL, also manifest in task-specific overfitting, with agents failing up to 60% of the time when faced with dynamic obstacles not encountered during training. These persistent shortcomings underscore the need for more robust and sample-efficient learning strategies. Finally, we discuss hybrid paradigms that integrate IL and RL, such as Generative Adversarial IL, and propose future research directions. Full article

Improving the end-of-life performance of polylactic acid (PLA) for food packaging requires strategies that enhance biodegradability, solubility, and dispersibility without compromising essential material properties. PLA-based films were produced by melt extrusion using polyvinylpyrrolidone (PVP) as a hydrophilic modifier, aiming to enhance the water uptake and affinity of PLA, which may potentially lead to faster environmental degradation. Two PVPs with distinct molar masses at varying concentrations were used to investigate their effects on the structural, thermal, mechanical, optical, and barrier behavior of the films. Thermal analysis revealed a slight depression in glass transition temperature, more evident in blends with low-molecular-weight PVP10, indicating increased chain mobility and partial miscibility. A two-step degradation process with extended thermal decomposition profiles was observed upon the inclusion of PVP. SEM and ATR-FTIR analyses confirmed enhanced dispersion and non-covalent interactions in PVP10-based blends, in contrast to the pronounced phase separation and micro-voids observed in PVP40-based systems. Mechanically, films containing 5 and 10 wt.% of PVP10 retained tensile strength and stiffness, whereas PVP40 led to embrittlement. Optical properties were modified by increasing the PVP content, resulting in greater opacity and color differences, which potentially offer benefits for light-sensitive packaging. Altogether, PLA films containing 5 and 10 wt.% of PVP10 demonstrated the most favorable balance between water affinity-oriented design and packaging-relevant performance. Full article

To overcome poor catalyst recovery and inefficient mass transfer in photocatalytic water treatment, this study presents novel Triply Periodic Minimal Surface (TPMS) photocatalytic reactors (PCRs) fabricated via Stereolithography (SLA) 3D printing. Five TiO₂-loaded reactors (Fischer-Radin-Dunn (FRD), Neovius (N), Diamond (D), I-graph Wrapped Package (IWP), Gyroid (G)) with hierarchical porosity were designed. Using methylene blue (MB) as the target pollutant, the photocatalytic degradation performance of TPMS-PCRs is evaluated and Computational Fluid Dynamics (CFD) hydrodynamic simulations are conducted to analyze their flow characteristics under both horizontal and rotational flow fields. The catalytic efficiency of TPMS reactors is influenced not only by pore characteristics, specific surface area, and inter-pore connectivity, but also by the flow velocities on both the reactor surface and within its internal channels. The FRD-type TPMS-PCR loaded with 2.5 wt% TiO₂ exhibited optimal photocatalytic performance, achieving 95.36% degradation efficiency under rotational flow within 2.5 h, compared to 88.2% under horizontal flow. Remarkably, after five degradation cycles, its efficiency further improved to 96.7%, demonstrating its excellent stability. The rotational flow field enhanced the average flow velocity by approximately sixfold compared to horizontal flow, with the D-type reactor reaching a maximum surface velocity of 5.3 × 10⁻² m/s. Full article