Search Results (13,546)

As deep neural networks continue to scale and enable emerging applications such as agentic AI systems, training increasingly relies on distributed paradigms across heterogeneous edge devices. However, this shift introduces significant security challenges, particularly model poisoning attacks, which are largely underexplored in model-parallel settings. To address these challenges, we propose a trusted and attack-resilient mechanism for distributed DNN training that supports both data and model parallelism. The mechanism leverages a blockchain-enabled infrastructure to ensure the tamper-resistant and auditable execution of security-critical operations. It introduces a Loss-aware Credit Evaluation mechanism to assess agent reliability based on group-level training dynamics and a Shuffling-based Isolation Mechanism to progressively cluster and isolate malicious agents across training epochs. In addition, Byzantine-tolerant aggregation (BTA) is employed to further mitigate adversarial influence during model aggregation. Extensive experiments demonstrate that the proposed mechanism achieves superior robustness and efficiency compared with state-of-the-art methods under diverse poisoning attack scenarios. Full article

This study examines the determinants of non-performing financing (NPF) using country-level Islamic banking system aggregates for the six Gulf Cooperation Council (GCC) countries, with particular attention to the role of global economic policy uncertainty and U.S. monetary policy uncertainty. Using a panel dataset covering the period 2014Q4–2024Q3, this study applies fixed-effects estimation and panel quantile regression to capture both average effects and distributional heterogeneity in financing risk. The findings reveal that the determinants of NPF vary significantly across the conditional NPF distribution. Profitability and capital adequacy are positively associated with NPF, whereas GDP is negatively associated with NPF. Liquidity has a negative and statistically significant association mainly in the middle and upper quantiles, indicating a stronger stabilizing role under elevated risk conditions. Global economic policy uncertainty is significant only in the upper quantiles and has a negative coefficient, while U.S. monetary policy uncertainty shows limited statistical relevance. These results indicate that mean-based models may conceal important differences across financing risk regimes. The findings are interpreted as statistical associations rather than causal effects, given the country-level aggregation, limited cross-sectional dimension, and potential measurement and model specification constraints. This study contributes distribution-sensitive evidence on GCC Islamic banking systems and offers cautious implications for risk monitoring, liquidity management, and macroprudential supervision. Full article

Background: Lower limb exoprostheses often lead to stump dermatological pathologies. The mechanisms by which mechanical microtraumas progress to non-healing ulcerative defects due to dysbiosis remain poorly understood. The objective of this study was to analyze mechanical, inflammatory, and infectious stump skin complications and justify the role of bacterial and mycological dysbiosis in blocking tissue regeneration. Methods: A critical narrative review guided by SANRA principles was conducted (PubMed/Scopus, 1980–2026). Data were extracted with a structured query focusing on amputation stumps, prosthetic interfaces, and skin/microbiological complications (dysbiosis, biofilms, and inflammatory markers). Evidence was graded using predefined clinical matrices and integrated through structured evidence collations to synthesize stump–prosthesis pathogenesis. The PRISMA method was not applied due to study heterogeneity. Results: Skin damage dynamics were categorized into three stages: adaptation (up to 12 months), chronic reactive changes (12–24 months), and late proliferative-infectious destruction (>24 months). The sealed liner space creates 100% humidity and alkalization (pH > 6.5). This causes a mycological shift, where resident Malassezia spp. lose dominance to invasive Candida albicans and non-dermatophyte molds (Aspergillus spp., Fusarium spp.). These pathogens form polymicrobial biofilms with Staphylococcus aureus. At the molecular level, delayed regeneration is driven by “frustrated phagocytosis”: macrophages, unable to engulf large fungal hyphae, continuously release reactive oxygen species and enzymes, trapping the wound in the inflammatory phase. Excessive matrix degradation and suppressed angiogenic factors further block epithelialization. Conclusions: The skin under a prosthesis socket forms a unique pathological biotope. Successful regeneration requires preventive mycobiota correction and targeted management of biophysical parameters (pH, humidity) within the “skin–liner” interface. Full article

Two-dimensional transition-metal carbides and nitrides (MXenes) are increasingly adopted in advanced electronic devices, where their metallic conductivity, optical tunability, and chemically addressable surfaces support next-generation multifunctional optoelectronics. Their practical performance, however, depends not only on their intrinsic properties but also on the heterogeneous interfaces where charges, photons, and ions interact. Unlike earlier reviews organized around synthesis routes or separate device categories, this review takes interfacial chemistry as a single organizing principle and follows it from surface terminations through to integrated systems. The structural and surface-chemical characteristics of MXenes are described first, showing how dynamic terminations and interfacial dipoles regulate work functions and energy-level alignment. We then discuss molecular functionalization, defect passivation, and heterojunction formation as strategies for reducing Schottky barriers and improving charge-transfer kinetics. Optoelectronic platforms built on these engineered interfaces, including high-efficiency photovoltaics, broadband photodetectors, and stretchable wearable systems, are subsequently detailed, together with emerging architectures that merge self-powered sensing with neuromorphic visual functions, a scope seldom treated alongside conventional devices in previous surveys. By connecting surface chemistry with device integration, this review outlines a materials-to-systems pathway toward more reliable and scalable MXene-based optoelectronic technologies. Full article

The Artificial Intelligence Pilot Zone Policy is an important strategic initiative for building artificial intelligence (AI) innovation hubs. It provides new opportunities to enhance manufacturing firm resilience and promote the sustainable development of the manufacturing sector. The creation of the National New-Generation Artificial Intelligence Innovation and Development Pilot Zones (AI Pilot Zones) is viewed in this study as a quasi-natural experiment. Using data from Chinese A-share-listed manufacturing firms from 2015 to 2023, we employ a staggered DID model to evaluate the impact of the policy on manufacturing firm resilience. We find that the AI Pilot Zone policy increases manufacturing firm resilience by an average of 0.0282 units. The analysis of potential mechanisms shows that the policy significantly promotes digital talent agglomeration, stimulates urban innovation vitality, and improves firm-level supply chain efficiency. These findings are consistent with the theoretical expectations and provide supportive evidence that these factors may constitute potential mechanisms associated with the policy’s effect on manufacturing firm resilience. The heterogeneity analysis reveals a pronounced “weakness-compensating” effect. At the regional level, the resilience-enhancing effect is stronger for manufacturing firms located in areas with relatively weak digital infrastructure. At the industry level, the effect is more pronounced among firms in low-technology manufacturing industries. At the firm level, the effect is stronger for firms with lower levels of human capital, weaker innovation capacity, and lagging digital transformation. Overall, this study provides micro-level evidence on the resilience effects of the AI Pilot Zone policy and offers policy implications for integrating AI more effectively with the real economy. Full article

Power modules (PMs) from waste electrical and electronic equipment (WEEE) represent an underexploited source of strategic secondary raw materials. However, due to their high level of integration and heterogeneity, PMs remain difficult to recycle, resulting in low recovery rates for several materials. This study analyzes the material composition of different PM types to identify key challenges and opportunities related to their end-of-life management. A step-by-step comprehensive parametric inventory model of PM recycling is developed from data collection, literature review and a corresponding dataset from the Ecoinvent database. Parametric inventory models are used to carry out environmental impact assessment of PM recycling according to material selection and different recycling process options. The models are made simple to use for PM designers such that it can be useful to guide and support design decision-making to maximize material recovery rates and minimize recycling-related environmental impacts. Implemented during the design phase, the models support the development of more sustainable PMs. Models are also made simple for recycling practitioners to access important data regarding PM material compositions. Full article

Automatic modulation recognition (AMR) identifies the modulation scheme of received radio frequency (RF) signals under unknown channel conditions and underpins spectrum monitoring and signal demodulation in wireless systems. Under low signal-to-noise ratio (SNR), multipath fading, and limited observation length, however, the discriminative features of modulated signals are severely attenuated, degrading recognition robustness. We propose MSTFFNet, a multi-scale time-frequency fusion network that addresses these challenges with two designs. First, it fuses the raw in-phase/quadrature (I/Q) signal with its short-time Fourier transform (STFT) time-frequency map at the token level through dual-stream heterogeneous encoding, capturing complementary temporal and spectral features. Second, rather than relying on external SNR ground truth, the network self-estimates an SNR-bin probability from the I/Q features and generates a channel-quality embedding that conditions the classifier, requiring no SNR label at inference. On the RadioML2016.10a and 10b benchmark datasets, MSTFFNet achieves overall accuracies of 67.33% and 70.87%, outperforming state-of-the-art methods by 3.53% and 5.33%, with improvements of 5.91% and 9.52% in the low-SNR regime. These results demonstrate improved recognition performance across the SNR conditions represented in the two synthetic RadioML2016 benchmarks, particularly at low SNR. Full article

Automatic colon polyp segmentation is important for reliable endoscopic visual sensing. However, segmentation performance can be degraded by ambiguous lesion boundaries, heterogeneous appearance, and image quality variations during acquisition or transmission. This study proposes a stage-aware Swin-enhanced nnU-Net v2 framework, where Swin Transformer blocks are inserted into selected encoder stages while preserving the original nnU-Net v2 pipeline. Different insertion strategies were evaluated on an independent Kvasir-SEG test set, and robustness was further assessed under six synthetic corruption types with three severity levels. The results show that the insertion stage strongly influences the effectiveness of Swin enhancement. Among the evaluated variants, Stage5-Swin achieved the best overall trade-off between segmentation performance, robustness, and computational cost. It achieved the highest Dice scores across all corruption–severity combinations while maintaining comparable external-domain performance on CVC-ClinicDB. Additional FedAvg experiments demonstrated the compatibility of the proposed architecture with collaborative training workflows. Resource analysis further quantified the computational and communication overhead. The findings indicate that middle-to-deep encoder insertion provides a favorable balance between contextual modeling, spatial representation, and efficiency for robust endoscopic segmentation. However, improvements were metric- and corruption-dependent, and moderate overexposure revealed a Precision–Recall trade-off; practical deployment also remains to be validated. Full article

MXenes have attracted considerable attention as anode materials for sodium-ion batteries (SIBs) because of their metallic conductivity, hydrophilic surfaces, tunable surface terminations, and layered structures. However, pristine MXenes are limited by nanosheet restacking, oxidation instability, heterogeneous surface chemistry, low initial Coulombic efficiency, and insufficient electrode-level ion accessibility. These issues indicate that MXenes should be regarded not simply as standalone active materials but as multifunctional building blocks for composite electrode design. This review discusses recent progress in MXene-based composite anodes for SIBs, focusing on MXene/carbon composites, MXene/metal compound composites, polymer-assisted composites, and three-dimensional structured MXene composites for improving structural stability, interfacial chemistry, and sodium-storage kinetics. We emphasize that composite engineering can reshape sodium storage from diffusion-limited intercalation toward hybrid mechanisms involving interfacial adsorption, pseudocapacitive storage, heterointerface-driven redox reactions, ion desolvation regulation, and solid-electrolyte interphase stabilization. Key practical challenges, including oxidation control, initial Coulombic efficiency, high-mass-loading electrode design, gravimetric–volumetric performance trade-offs, scalable synthesis, and full-cell validation, are also discussed. Finally, we propose future design principles based on integrated materials chemistry, interfacial science, multiscale architecture engineering, and realistic cell-level evaluation for advancing MXene composites toward practical SIB anodes. Full article

Environmental noise is a major environmental and public health concern in rapidly urbanizing cities, yet integrated studies combining field measurements, GIS-based spatial analysis, and predictive modelling remain limited in Central Asia. This study assessed the spatial distribution of urban noise pollution in Samarkand (Uzbekistan) and explored its statistical association with selected public health indicators, forecasting future trends. Measurements were conducted at 50 georeferenced sites covering more than 300 streets. The measured data were processed and mapped using ArcGIS 10.5 (Esri, Redlands, CA, USA) to produce the spatial distribution of environmental noise across the study area. Official data on registered vehicles, industrial enterprises, and disease incidence (2014–2024) were analysed using Pearson correlation, Autoregressive Integrated Moving Average (ARIMA), and its extension incorporating exogenous variables (ARIMAX) models. Results revealed pronounced spatial heterogeneity in noise levels, highest along transport corridors and industrial zones. Industrial enterprises showed the strongest correlations with disease incidence; vehicle registrations were excluded from final models owing to collinearity with industrial activity. ARIMA projected continued industrial growth through 2030, while ARIMAX models identified significant associations between industrial activity and diseases of the ear and mastoid process and of the nervous system (MAPE 28.55% and 19.78%). As an ecological, exploratory study using infrastructural proxies rather than measured noise exposure, findings should be interpreted as associations rather than causation. The framework offers a transferable approach for environmental risk assessment and sustainable urban planning. Full article

This study employs panel data from 11 provinces and municipalities along the Yangtze River Economic Belt spanning 2010–2021. It measures green transport efficiency (GTE) using principal component analysis (PCA) and the undesirable Super-SBM model while constructing an economic high-quality development index (HQEDI) through an entropy-weighted CRITIC approach. The study combines coupling coordination degree modeling with spatial autocorrelation analysis (Global Moran’s I, LISA, hotspot/coldspot detection) to empirically investigate their synergistic evolution mechanism. The findings indicate the following: (1) Multidimensional policy combinations exhibit a nonlinear threshold effect on enhancing green transport efficiency, with efficiency significantly rebounding post-2015 as low-carbon policies deepened. (2) High-quality economic development displays a dual-stage ‘convergence-divergence’ pattern, where downstream regions lead in HQEDI but mid- and upstream regions show faster growth in coordination and green dimensions. (3) The coupling coordination degree exhibits pronounced spatial spillover effects, with the global Moran’s I mean reaching 0.485. High-value clusters form in downstream regions, while upstream areas predominantly exhibit low-value clusters, revealing an ‘east-high, west-low’ regional differentiation pattern. (4) The gradient divergence mechanism stems from heterogeneity in infrastructure density, industrial structure, and policy responsiveness elasticity. Accordingly, it is recommended to establish a multi-level governance mechanism to dismantle administrative barriers and to construct a tripartite policy package integrating ‘digital transport, ecological compensation, and industrial radiation’ to advance coordinated basin development. Full article

Aquaculture effluent commonly contains suspended solids, inorganic nitrogen, reactive phosphate and algal biomass, creating a need for low-input ecological treatment approaches. This study assessed endpoint water-quality and bacterial-community responses to Bellamya stocking density in aquaculture effluent. A no-snail control (CON) and four stocking-density treatments (LD, MD, MHD and HD) were established with three independent tank replicates per treatment. Suspended solids (SS), NH₄⁺-N, NO₂⁻-N, NO₃⁻-N, PO₄³⁻-P and chlorophyll a (Chl-a) were quantified, and bacterial communities were characterized by 16S rRNA gene sequencing of the V3–V4 region. At the 60-day endpoint, all Bellamya-stocked treatments had lower NH₄⁺-N and NO₃⁻-N concentrations than the control, and MHD and HD also had lower PO₄³⁻-P and Chl-a. SS showed a nonsignificant downward tendency, whereas NO₂⁻-N showed a nonsignificant upward tendency. Bray–Curtis NMDS visualized treatment-associated separation, and PERMANOVA detected significant differences among treatments (R² = 0.62, p = 0.001; stress = 0.0822). However, PERMDISP was also significant (F₄,₁₀ = 4.99, p = 0.001), indicating heterogeneous within-treatment dispersion and requiring cautious interpretation of the PERMANOVA result. Representative genera showed distinct treatment-associated abundance patterns, and 46 of 90 genus–environment associations remained significant after Benjamini–Hochberg correction. Predicted KEGG Level 3 pathways varied numerically among treatments, but none remained significant after false-discovery-rate correction. These findings indicate that Bellamya stocking may provide a low-input ecological component of aquaculture-effluent management, while the microbiome results should be interpreted as community-level associations and predicted functional trends rather than evidence of microbial causality or pathway activation. Full article

Tax incentives are central to environmental policy, yet the innovation effects of liquidity-based fiscal instruments remain underexplored. This study asks whether China’s value-added tax (VAT) credit refund promotes corporate green innovation, and how firm governance and the policy environment condition that effect. Exploiting the 2018 industry-targeted pilot as a quasi-natural experiment, we estimate two-way fixed-effects difference-in-differences models on 25,259 firm-year observations for 4044 Chinese A-share listed firms over 2014–2021, measuring green innovation by invention-patent applications and defining treatment by industry eligibility. The refund raises green invention-patent applications by 17.0 percent, significant at the 1 percent level. A sequential decomposition is consistent with transmission through eased financing constraints and higher R&D investment. The effect is stronger where internal control quality is higher and pre-treatment financial slack is greater, and is absent where climate policy uncertainty (CPU) is high. Results survive removing control firms drawn in by the 2019 policy expansion, sector-by-year and province-by-year fixed effects, heterogeneity-robust estimators, and a triple difference on the innovation-quality margin. Fiscal liquidity complements governance quality and regulatory predictability rather than substituting for them. Full article

Environmental, Social, and Governance practices have emerged as a critical mechanism for mitigating extreme climate risks and achieving Sustainable Development Goals (SDGs). Against this backdrop, this study focuses on Chinese listed manufacturing enterprises from 2015 to 2023. By incorporating the city-level climate risk expressions of public views index, we examine the impact of public concern and negative sentiment regarding climate risk (CR-PCNS) on corporate ESG performance. Our baseline findings indicate that elevated CR-PCNS significantly enhances corporate ESG performance, particularly within the environmental and social pillars, while exerting no significant effect on the governance dimension. Heterogeneity analysis reveals that this promotional effect is more pronounced among non-state-owned enterprises and firms located in the eastern region, with the most noticeable improvements manifested in their environmental performance. Furthermore, the moderation analysis demonstrates that higher executive educational attainment significantly amplifies the positive impact of CR-PCNS on corporate ESG performance, whereas local protectionism severely attenuates this promotional effect. These findings offer crucial policy implications for how to effectively harness public involvement to incentivize greater corporate engagement in ESG initiatives. Full article

Metastatic colorectal cancer remains associated with poor prognosis despite major therapeutic advances, highlighting the need for robust biomarkers to refine treatment selection and monitor disease dynamics. This review summarizes emerging predictive and prognostic biomarkers in metastatic colorectal cancer across molecular and cellular layers, encompassing both tissue and circulating biomarkers. At the tissue level, we discuss genomic alterations and mutational signatures, transcriptomic classification systems and immune-related gene expression tools, protein-level immune checkpoint markers, and cellular determinants including immune infiltrates, cancer-associated fibroblasts and microbiome features. At the circulating level, we review biomarkers derived from liquid biopsy and peripheral blood, including circulating tumor DNA kinetics, T-cell receptor repertoire diversity, soluble cytokines and proteins, immune cell phenotyping, and circulating tumor cells. We highlight major challenges limiting clinical translation, including tumor heterogeneity, methodological variability, and the absence of standardized analytical pipelines and thresholds. Finally, we discuss future perspectives, emphasizing the integration of multi-omics biomarkers and artificial intelligence-driven strategies to improve biomarker validation and enable more precise management of metastatic colorectal cancer, particularly for patients with microsatellite-stable tumors who derive limited benefit from immune checkpoint inhibition. Full article