Search Results (3,943)

This study investigates the impact of nanofluids (TiO₂, Fe₃O₄, Al₂O₃, and graphene) on thermoelectric power generation within a rectangular heat exchanger equipped with internal winglets. The integration of internal winglets in heat exchangers, alongside the use of nanofluids, is a recent strategy aimed at enhancing convective heat transfer. This numerical research analyzes fluid dynamics and temperature variations on both the cold and hot sides of the thermoelectric generator (TEG). Three different heat exchanger models are evaluated: the first model features a pair of winglets in both ducts; the second model only has winglets in the hot duct; and the third model does not include any winglets. The performance of the nanofluids is systematically compared with that of distilled water. The results show that the Al₂O₃ nanofluid produces the highest power output at 7.8461 watts, which is 1.5% greater than that of TiO₂ and 1.22% higher than distilled water. Moreover, using Al₂O₃ in a heat exchanger with winglets in both ducts results in a 5% increase in power generation compared to a configuration without winglets and a 2% improvement over a model that has winglets only in the hot duct. This enhancement can be attributed to an increased heat transfer area and improved fluid mixing, which together facilitate more effective heat transfer to TEG. Full article

Forest tourism visitation in U.S. national forests has grown by approximately 8 percent over the past decade (from 2014 to 2022) from 147 million to 158.7 million visits per year, indicating a clear upward trajectory in demand for nature-based leisure experiences, yet the determinants of traveler satisfaction in this context remain insufficiently understood. Existing studies have primarily emphasized destination attributes, overlooking the interplay between psychological motivations, lifestyle orientations, and physical conditions. This omission is critical because it limits a holistic understanding of forest traveler’s experiences, which prevents us from fully capturing how internal dispositions, everyday life contexts, and well-being concerns interact with destination attributes to shape satisfaction. Therefore, the purpose of this study is to explore how motivation, lifestyle, and physical condition jointly shape satisfaction in forest tourism, drawing on Push–Pull Theory and environmental psychology. A dataset of 10,792 TripAdvisor reviews of U.S. national forests was analyzed using LIWC 2022 for psycholinguistic feature extraction and Ordered Logit Regression for hypothesis testing. Results show that positive emotional tone, leisure-oriented language, health references, and reward motivation significantly enhance satisfaction, while negative tone, illness, and work-related language reduce it. Curiosity and risk motivations were non-significant, and allure exerted only a marginal effect. These findings extend the Push–Pull framework by incorporating lifestyle and physical condition as moderating variables and validate emotional tone in user-generated content as a proxy for subjective evaluations. The study refines motivation theory by revealing context-specific effects of motivational dimensions. The results offer actionable insights for destination managers, service providers, marketers, and policymakers aiming to enhance forest travel experiences and promote sustainable tourism development. Full article

Texture, a significant visual attribute in images, plays an important role in many pattern recognition tasks. While Convolutional Neural Networks (CNNs) have been among the most effective methods for texture analysis, alternative architectures such as Vision Transformers (ViTs) have recently demonstrated superior performance on a range of visual recognition problems. However, the suitability of ViTs for texture recognition remains underexplored. In this work, we investigate the capabilities and limitations of ViTs for texture recognition by analyzing 25 different ViT variants as feature extractors and comparing them to CNN-based and hand-engineered approaches. Our evaluation encompasses both accuracy and efficiency, aiming to assess the trade-offs involved in applying ViTs to texture analysis. Our results indicate that ViTs generally outperform CNN-based and hand-engineered models, particularly when using strong pre-training and in-the-wild texture datasets. Notably, BeiTv2-B/16 achieves the highest average accuracy (85.7%), followed by ViT-B/16-DINO (84.1%) and Swin-B (80.8%), outperforming the ResNet50 baseline (75.5%) and the hand-engineered baseline (73.4%). As a lightweight alternative, EfficientFormer-L3 attains a competitive average accuracy of 78.9%. In terms of efficiency, although ViT-B and BeiT(v2) have a higher number of GFLOPs and parameters, they achieve significantly faster feature extraction on GPUs compared to ResNet50. These findings highlight the potential of ViTs as a powerful tool for texture analysis while also pointing to areas for future exploration, such as efficiency improvements and domain-specific adaptations. Full article

Extracellular matrix (ECM) bioscaffolds have demonstrated therapeutic potential across a variety of clinical and preclinical applications for tissue repair and regeneration. In parallel, these scaffolds and their components have shown the capacity to modulate the immune response. Unlike synthetic implants, which are often associated with chronic inflammation or fibrotic encapsulation, ECM bioscaffolds interact dynamically with host cells, promoting constructive tissue remodeling. This effect is largely attributed to the preservation of structural and biochemical cues—such as degradation products and matrix-bound nanovesicles (MBV). These cues influence immune cell behavior and support the transition from inflammation to resolution and functional tissue regeneration. However, the immunomodulatory properties of ECM bioscaffolds are dependent on the source tissue and, critically, on the methods used for decellularization. Inadequate removal of cellular components or the presence of residual chemicals can shift the host response towards a pro-inflammatory, non-constructive phenotype, ultimately compromising therapeutic outcomes. This review synthesizes current basic concepts on the innate immune response to ECM bioscaffolds, with particular attention to the inflammatory, proliferative, and remodeling phases following implantation. We explore how specific ECM features shape these responses and distinguish between pro-remodeling and pro-inflammatory outcomes. Additionally, we examine the impact of manufacturing practices and quality control on the preservation of ECM bioactivity. These insights challenge the conventional classification of ECM bioscaffolds as medical devices and support their recognition as biologically active materials with distinct immunoregulatory potential. A deeper understanding of these properties is critical for optimizing clinical applications and guiding the development of updated regulatory frameworks in regenerative medicine. Full article

Objectives: The persistent evolution of software vulnerabilities—spanning novel zero-day exploits to logic-level flaws—continues to challenge conventional cybersecurity mechanisms. Static rule-based scanners and opaque deep learning models often lack the precision and contextual understanding required for both accurate detection and analyst interpretability. This paper presents a hybrid framework for real-time vulnerability detection that improves both robustness and explainability. Methods: The framework integrates semantic encoding via Bidirectional Encoder Representations from Transformers (BERTs), structural analysis using Deep Graph Convolutional Neural Networks (DGCNNs), and lightweight prioritization through Kernel Extreme Learning Machines (KELMs). The architecture incorporates Minimum Intermediate Representation (MIR) learning to reduce false positives and fuses multi-modal data (source code, execution traces, textual metadata) for robust, scalable performance. Explainable Artificial Intelligence (XAI) visualizations—combining SHAP-based attributions and CVSS-aligned pair plots—serve as an analyst-facing interpretability layer. The framework is evaluated on benchmark datasets, including VulnDetect and the NIST Software Reference Library (NSRL, version 2024.12.1, used strictly as a benign baseline for false positive estimation). Results: Our evaluation reports that precision, recall, AUPRC, MCC, and calibration (ECE/Brier score) demonstrated improved robustness and reduced false positives compared to baselines. An internal interpretability validation was conducted to align SHAP/GNNExplainer outputs with known vulnerability features; formal usability testing with practitioners is left as future work. Conclusions: The framework, Designed with DevSecOps integration in mind, the system is packaged in containerized modules (Docker/Kubernetes) and outputs SIEM-compatible alerts, enabling potential compatibility with Splunk, GitLab CI/CD, and similar tools. While full enterprise deployment was not performed, these deployment-oriented design choices support scalability and practical adoption. Full article

As the global demand for energy-efficient solutions grows increasingly urgent, passive design strategies emerge not only as a means to support the reduction in energy consumption but also as a pathway to minimizing building operational costs while enhancing thermal comfort and architectural attractiveness. On the other hand, the recognition and significance of building environmental certification schemes are steadily increasing worldwide. Within this context, this research investigates the extent to which passive bioclimatic principles are understood, applied, and incentivized in contemporary sustainable building practices in Greece—focusing in particular on their representation within the LEED certification credit structure. Drawing on a questionnaire survey completed by 89 experienced Greek construction professionals, the findings indicate a significant gap between the theoretical value attributed to passive design and its practical implementation. The respondents attribute this gap to two key factors within the Greek context: the lack of adequate education and awareness among key project stakeholders, and the considerable complexity associated with the collaborative frameworks required from the early design stages. Additionally, LEED appears to offer limited incentives for integrating passive design strategies. Instead, it tends to favor technological solutions and follows a standardized structure with minimal scope for regional customization. Enhancing LEED’s region-specific features to reward passive strategies proven effective in local contexts would be particularly expedient in reinforcing its role as a robust and impactful tool for promoting sustainability. Full article

Carbonate fracture–cavity reservoirs are significant oil and gas reservoirs globally, and their efficient development is influenced by the connectivity between fracture–cavity units within the reservoir. These reservoirs primarily consist of large caves, dissolution holes, and natural fractures, which serve as the primary storage and flow spaces. The S91 unit of the Tarim Oilfield is a karstic fracture–cavity reservoir with shallow coverage. It exhibits significant heterogeneity in the fracture–cavity reservoirs and presents complex connectivity between the fracture–cavity bodies. The integration of static and dynamic data, including geology, well logging, seismic, and production dynamics, resulted in the development of a set of static and dynamic connectivity evaluation processes designed for highly heterogeneous fracture–cavity reservoirs. Methods include using structural gradient tensors and stratigraphic continuity attributes to delineate the boundaries of caves and holes; performing RGB fusion analysis of coherence, curvature, and variance attributes to characterize large-scale fault development features; applying ant-tracking algorithms and fracture simulation techniques to identify the distribution and density characteristics of fracture zones; utilizing 3D visualization technology to describe the spatial relationship between fracture–cavity units and large-scale faults and fracture development zones; and combining dynamic data to verify interwell connectivity. This process will provide a key geological basis for optimizing well network deployment, improving water and gas injection efficiency, predicting residual oil distribution, and formulating adjustment measures, thereby improving the development efficiency of such complex reservoirs. Full article

The development of generative AI Large Language Models (LLMs) raised the alarm regarding the identification of content produced by generative AI vs. humans. In one case, issues arise when students heavily rely on such tools in a manner that can affect the development of their writing or coding skills. Other issues of plagiarism also apply. This study aims to support efforts to detect and identify textual content generated using LLM tools. We hypothesize that LLM-generated text is detectable by machine learning (ML) and investigate ML models that can recognize and differentiate between texts generated by humans and multiple LLM tools. We used a dataset of student-written text in comparison with LLM-written text. We leveraged several ML and Deep Learning (DL) algorithms, such as Random Forest (RF) and Recurrent Neural Networks (RNNs) and utilized Explainable Artificial Intelligence (XAI) to understand the important features in attribution. Our method is divided into (1) binary classification to differentiate between human-written and AI-generated text and (2) multi-classification to differentiate between human-written text and text generated by five different LLM tools (ChatGPT, LLaMA, Google Bard, Claude, and Perplexity). Results show high accuracy in multi- and binary classification. Our model outperformed GPTZero (78.3%), with an accuracy of 98.5%. Notably, GPTZero was unable to recognize about 4.2% of the observations, but our model was able to recognize the complete test dataset. XAI results showed that understanding feature importance across different classes enables detailed author/source profiles, aiding in attribution and supporting plagiarism detection by highlighting unique stylistic and structural elements, thereby ensuring robust verification of content originality. Full article

Engineering phase-selective gel composites presents a promising route to enhance both CO₂ adsorption and conversion efficiency in photocatalytic systems. In this work, Cu₉S₅/TiO₂ gel composites were synthesized via a hydrazine-hydrate-assisted hydrothermal method, using TiO₂ derived from a microwave-assisted sol–gel process. The resulting materials exhibit a porous gel-derived morphology with highly dispersed Cu₉S₅ nanocrystals, as confirmed by XRD, TEM, and XPS analyses. These structural features promote abundant surface-active sites and interfacial contact, enabling efficient CO₂ adsorption. Among all samples, the optimized 0.36Cu₉S₅/TiO₂ composite achieved a methane production rate of 34 μmol·g⁻¹·h⁻¹, with 64.76% CH₄ selectivity and 88.02% electron-based selectivity, significantly outperforming Cu₉S₈/TiO₂ synthesized without hydrazine hydrate. This enhancement is attributed to the dual role of hydrazine: facilitating phase transformation from Cu₉S₈ to Cu₉S₅ and modulating the interfacial electronic environment to favor CO₂ capture and activation. DFT calculations reveal that Cu₉S₅/TiO₂ effectively lowers the energy barriers of critical intermediates (*COOH, *CO, and *CHO), enhancing both CO₂ adsorption strength and subsequent conversion to methane. This work demonstrates a gel-derived composite strategy that couples efficient CO₂ adsorption with selective photocatalytic reduction, offering new design principles for adsorption–conversion hybrid materials. Full article

Graph anomaly detection (GAD) is crucial for safeguarding the integrity and security of complex systems, such as social networks and financial transactions. Despite the advances made by Graph Neural Networks (GNNs) in the field of GAD, existing methods still exhibit limitations in capturing subtle structural anomaly patterns: they typically over-rely on reconstruction error, struggle to fully exploit structural similarity among nodes, and fail to effectively integrate attribute and structural information. To tackle these challenges, this paper proposes a structure-aware dual-path attention graph node anomaly detection method (DPGAD). DPGAD employs wavelet diffusion to extract network neighborhood features for each node while incorporating a dual attention mechanism to simultaneously capture attribute and structural similarities, thereby obtaining richer feature details. An adaptive gating mechanism is then introduced to dynamically adjust the fusion of attribute features and structural features. This allows the model to focus on the most relevant features for anomaly detection, enhancing its robustness and antinoise capability. Our experimental evaluation across multiple real-world datasets demonstrates that DPGAD consistently surpasses existing methods, achieving average improvements of 9.06% in AUC and 11% in F1-score. Especially in scenarios where structural similarity is crucial, DPGAD has a performance advantage of more than 20% compared with the most advanced methods. Full article

This investigation on novel antifungal agents featuring a thiol-reactive (dithioperoxo)thiolate chemical nucleus [-NC(S)S-SR] established that the optimal levels of fungal growth inhibition were achieved with thiomethyl-bound derivatives (R = Me). The most efficacious analogs had MIC₅₀/MIC₉₀ values of 2/2 µg/mL and an MIC range of 1 to 2 µg/mL for a ten-member panel of voriconazole-resistant A. fumigatus mutants. Pharmacodynamic studies revealed that the lead (dithioperoxo)thiolates impaired conidial germination and germling development more effectively than voriconazole for the triazole-resistant strain AR-1295. Moreover, glutathione and Cu²⁺ were shown to have antagonistic interactions, which was attributed to the thiol-reactive, pro-oxidant properties of the (dithioperoxo)thiolates and their metabolic conversion to chelating agents. Cytotoxicity studies further showed that the compounds were less toxic to human fetal kidney cells than squamous carcinoma cells. The collective findings of the investigation indicate that (dithioperoxo)thiolates are effective antifungal agents against A. fumigatus to merit additional research on their therapeutic potential. Full article

We theoretically investigate high-order harmonic generation from ethylbenzene (C₈H₁₀), toluene (C₇H₈), and benzene (C₆H₆) molecules driven by a circularly polarized laser field using time-dependent density functional theory. By comparing the harmonic spectra of these structurally related molecules, we find that ethylbenzene, which features a larger molecular size due to the ethyl group, exhibits a higher harmonic cutoff and stronger harmonic intensity than toluene and benzene. Time-resolved electron density distributions, together with the probability current density analysis, indicate that under long-wavelength conditions (e.g., 1200 nm), the ethyl group in ethylbenzene and the methyl group in toluene significantly enhance the probability of ionized electrons from neighboring nuclei colliding with nearby nuclei, thereby leading to stronger harmonic emission, with ethylbenzene > toluene > benzene. In contrast, under short-wavelength conditions (e.g., 200 nm), the harmonic intensities of the three molecules show little difference, and the effects of the ethyl and methyl groups on the harmonic yield can be neglected. The influence of laser intensity and wavelength on high-order harmonic generation is further analyzed, confirming the robustness of the structural enhancement effect. Additionally, we study the harmonic ellipticity of ethylbenzene under different carrier-envelope phases, and find that while circularly polarized harmonics can be obtained, their spectral continuity is insufficient for synthesizing isolated circularly polarized attosecond pulses. This limitation is attributed to the broken ring symmetry caused by the ethyl substitution. Our findings offer insight into the relationship between molecular structure and harmonic response in strong-field physics, and provide a pathway for designing efficient circularly polarized attosecond pulse sources. Full article

Background. Severe left ventricular hypertrophy (SLVH) is a high-risk structural cardiac abnormality associated with increased risk of heart failure. It is typically assessed using echocardiography or cardiac magnetic resonance imaging, but these modalities are limited by cost, accessibility, and workflow burden. We introduce a deep learning framework that classifies SLVH directly from chest radiographs, without intermediate anatomical estimation models or demographic inputs. A key contribution of this work lies in interpretability. We quantify how clinically relevant attributes are encoded within internal representations, enabling transparent model evaluation and integration into AI-assisted workflows. Methods. We construct class-balanced subsets from the CheXchoNet dataset with equal numbers of SLVH-positive and negative cases while preserving the original train, validation, and test proportions. ResNet-18 is fine-tuned from ImageNet weights, and a Vision Transformer (ViT) encoder is pretrained via masked autoencoding with a trainable classification head. No anatomical or demographic inputs are used during training. We apply Mutual Information Neural Estimation (MINE) to quantify dependence between learned features and five attributes: age, sex, interventricular septal diameter (IVSDd), posterior wall diameter (LVPWDd), and internal diameter (LVIDd). Results. ViT achieves an AUROC of 0.82 [95% CI: 0.78–0.85] and an AUPRC of 0.80 [95% CI: 0.76–0.85], indicating strong performance in SLVH detection from chest radiographs. MINE reveals clinically coherent attribute encoding in learned features: age > sex > IVSDd > LVPWDd > LVIDd. Conclusions. This study shows that SLVH can be accurately classified from chest radiographs alone. The framework combines diagnostic performance with quantitative interpretability, supporting reliable deployment in triage and decision support. Full article

The microstructure and magnetic properties of rare-earth-free, melt-spun Co₇₄Zr₁₆Mo₄Si₃B₃ alloys were investigated to enhance their hard magnetic response and elucidate their coercivity mechanism. The alloys exhibit a polycrystalline microstructure composed of randomly oriented, equiaxed grains, predominantly comprising the rhombohedral hard magnetic Co₁₁Zr₂ phase (92.4 wt.%). These materials display a favorable combination of magnetic properties, with coercive fields up to 581 kA/m, maximum magnetization reaching 0.30 T, and Curie temperatures as high as 751 K. An interpretation of the results, based on microstructural features, intrinsic magnetic parameters, and micromagnetic simulations, indicates that the coercivity mechanism of these melt-spun alloys can be attributed to the nucleation of reverse magnetic domains. Full article

Remote sensing objects often exhibit significant scale variations, high aspect ratios, and diverse orientations. The anisotropic spatial distribution of such objects’ features leads to the conflict between feature representation and boundary regression caused by the coupling of different attribute parameters: previous detection methods based on square-kernel convolution lack the overall perception of large-scale or slender objects due to the limited receptive field; if the receptive field is simply expanded, although more context information can be captured to help object perception, a large amount of background noise will be introduced, resulting in inaccurate feature extraction of remote sensing objects. Additionally, the extracted features face issues of feature conflict and discontinuous loss during parameter regression. Existing methods often neglect the holistic optimization of these aspects. To address these challenges, this paper proposes SODE-Net as a systematic solution. Specifically, we first design a multi-scale fusion and spatially orthogonal convolution (MSSO) module in the backbone network. Its multiple shapes of receptive fields can naturally capture the long-range dependence of the object without introducing too much background noise, thereby extracting more accurate target features. Secondly, we design a multi-level decoupled detection head, which decouples target classification, bounding-box position regression and bounding-box angle regression into three subtasks, effectively avoiding the coupling problem in parameter regression. At the same time, the phase-continuous encoding module is used in the angle regression branch, which converts the periodic angle value into a continuous cosine value, thus ensuring the stability of the loss value. Extensive experiments demonstrate that, compared to existing detection networks, our method achieves superior performance on four widely used remote sensing object datasets: DOTAv1.0, HRSC2016, UCAS-AOD, and DIOR-R. Full article