Search Results (11,295)

Artificial intelligence (AI) is deeply integrated into daily life. Emerging evidence suggests AI may help change the dynamics of social relationships by influencing social interactions, connectivity, and interpersonal relationships, and by providing new avenues for communication and contributing to improved social well-being. Therefore, this review aims to explore the potential of artificial intelligence (AI) technologies as a tool to enhance social health, focusing on current applications, advantages, challenges, and ethical considerations associated with their implementation, as well as opportunities for future development. The literature on the relationship between the connectedness of social health dimensions and AI as a tool to better understand how interactions with AI technologies may influence social well-being. In this current review, key terms such as “Artificial Intelligence”, “Social Health”, “social inequalities”, “AI algorithm”, “AI technology”, “social connection”, “digital communication”, “social participation”, “social support”, “social isolation”, “loneliness”, “mental wellbeing”, were used to search relevant literature on Google Scholar, PubMed, Scopus and Web of Sciences. In addition, relevant aspects of the multidimensional impacts of AI on social health dimensions are also discussed. The use of AI technologies by individuals within societies was found to hold profound potential to reshape social health through enhancing social relationships, bridging communication gaps in diverse populations, stimulating social dynamics, and understanding human emotions. It may contribute to reducing social inequalities, promoting equity, accommodating individual differences, and enhancing the effectiveness of many tasks in the social and health care systems through deep learning, natural language processing, and machine learning techniques. This reduces social exclusion and increases accessibility and quality of health and social services. However, AI has also posed distinguishable challenges to its adoption, specifically in terms of data quality, privacy and security, algorithmic bias, ethical issues, public trust and acceptance, and regulatory and policy gaps. Evidence suggests that building public trust in the future of AI in social health requires interdisciplinary collaboration among health providers and professionals, social scientists, community members, and policymakers. Such collaboration is crucial to ensure that AI platforms do not perpetuate social inequalities or biases by maintaining transparency, explainability, and demonstrated effectiveness. In conclusion, the integration of AI into social health dimensions holds promise for social health transformation. As we move forward, several key areas need to be addressed to develop a robust governance and regulatory framework, along with ethical guidelines to ensure privacy protection, respect for human rights, transparency, and the promotion of the common good. Full article

Cardiovascular disease remains the leading global health challenge, claiming approximately 19.8 million lives annually. The convergence of wearable technology and artificial intelligence represents a transformative shift in cardiovascular healthcare, enabling continuous real-time monitoring beyond conventional clinical settings. This narrative review synthesises current evidence on integrating consumer-grade and medical-grade wearable devices with AI algorithms for continuous cardiovascular monitoring applications, with particular attention to real-world translational applicability and global health equity. This review examined the technological landscape of wearable cardiovascular monitoring devices, including smartwatches with photoplethysmography and electrocardiogram capabilities, continuous cardiac monitoring patches, and emerging biosensor technologies. Also, the review explored AI methodologies, particularly machine learning and deep learning architectures, employed in processing complex physiological data streams from these devices. Clinical applications demonstrate impressive capabilities: arrhythmia detection with sensitivity rates exceeding 98%, continuous blood pressure monitoring through cuffless technologies, heart failure decompensation prediction, and cardiovascular risk stratification. However, substantial challenges persist, including data quality assurance, algorithm interpretability, regulatory compliance, and seamless clinical workflow integration. Privacy concerns, health disparities in algorithm performance, and the need for robust validation across diverse populations remain critical considerations. AI-enhanced wearable systems hold considerable potential for shifting cardiovascular care from reactive treatment paradigms towards predictive, preventive, and precision medicine approaches. Future directions include edge computing architectures, federated learning approaches, personalised AI models, enhanced interoperability with electronic health records, and expansion to resource-limited settings, ultimately improving patient outcomes whilst reducing healthcare costs. Full article

Climate change has increased the frequency of natural disasters, posing growing challenges to the high-quality development of the grain industry. Using China‘s provincial panel data from 2010 to 2023, this study examines the direct, moderating, and spatial effects of natural disasters on the high-quality development of the grain industry. A multidimensional index of high-quality development is constructed using the CRITIC weighting method. The results show that natural disasters significantly hinder the high-quality development of the grain industry, and this finding remains robust across a series of robustness tests. Further analysis indicates that agricultural irrigation and rural internet penetration mitigate the adverse effects of natural disasters, suggesting that both engineering resilience and digital resilience enhance the adaptive capacity of grain production systems. Heterogeneity analysis reveals that the negative impacts are more pronounced in major grain-producing regions and in the central, western, and northeastern regions of China. In addition, the spatial econometric results reveal significant negative spatial spillover effects, indicating that disaster shocks extend beyond directly affected regions through spatial interactions. This study contributes to the literature by extending disaster-impact research from grain output losses to the high-quality development of the grain industry, integrating engineering and digital resilience resources within a unified analytical framework, and providing new evidence on the spatial spillover effects of natural disasters. Full article

This paper examines the moderating effect of corporate digital transformation in the relationship between audit quality and aggressive tax avoidance in the sample of industrial companies listed on the ASE and operating during the 2020–2025 period. They were based on data of a balanced panel of 30 industrial companies listed on the ASE 180 observations. The primary estimator used was the feasible generalized least squares (FGLS) method that was employed after it was established that first-order autocorrelation, groupwise heteroskedasticity, and partial cross-sectional dependence existed. System-GMM estimator was used to confirm the robustness of the results, and to deal with the endogeneity that may arise due to reverse causality between auditor selection and result. There are three key findings of the study. First, there is a strong and consistent negative relationship between affiliation with one of the Big Four audit firms and aggressive tax avoidance in all the models studied, confirming that reputation-based audit quality is an effective institutional deterrent a finding of particular importance given that 73.3% of the Jordanian industrial firms in the sample rely on local auditors and therefore lack similar governance controls. Second, aggressive tax avoidance is positively related to higher audit fees, which are indicative of a more complex client base and an economic dependence on clients by the auditor, rather than a signal of greater monitoring rigour and, therefore, as a challenge to the fee-as-quality assumption common to the developed-market framework. Third, although digital transformation demonstrates a direct positive correlation with aggressive tax avoidance—indicating that firms can use digital capabilities to enhance tax planning and not compliance in the pre-JoFotara regulatory environment—its moderating effect on Big Four affiliation is not statistically significant. It is important to note that the relationship between the intensity of audit fees and digital transformation is positively significant, which is in line with the economic dependence argument. The implications of the findings are important to the Jordan Securities Commission, the tax authorities as well as regulatory bodies who are looking to enhance corporate tax compliance in a dynamic digital regulatory environment, and raise important questions of the portability of audit quality assumptions across institutional settings. Full article

Magnetic Flux Leakage (MFL) sensors are pivotal for the non-destructive inspection of oil and gas pipelines. However, the accuracy of defect quantification is severely compromised by pervasive noise in field-acquired MFL sensor signals, leading to substantial measurement uncertainty. To address this, we introduce a novel hybrid denoising framework that synergizes Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) and Wavelet Packet Decomposition (WPD). The key innovation is the employment of the Rime Optimization Algorithm (RIME) to automatically fine-tune the critical parameters of ICEEMDAN—the signal-to-noise ratio (SNR) and the number of noise additions—thereby customizing the decomposition for superior sensor signal enhancement. This optimization effectively suppresses mode aliasing and yields intrinsic mode functions that faithfully represent underlying defect features. The framework’s efficacy is rigorously validated through mathematical modeling, COMSOL Multiphysics 6.3-based finite element simulation, and real-field MFL sensor data. Results demonstrate remarkable improvements in sensor signal quality: a 53.69% increase in the SNR and reductions of 61.03% in MAE and 62.05% in RMSE over conventional methods. Crucially, the method achieved a Feature Preservation Rate (FPR) of 97.18% on simulated defects, underscoring its exceptional capability to retain critical metrological features for defect sizing. This work provides a robust signal-processing framework that significantly advances the measurement fidelity of MFL sensors, enabling more reliable pipeline integrity assessment. Full article

This study examines the relationship between renewable energy consumption (REC) and carbon dioxide (CO₂) emissions and investigates the moderating role of country-level governance in 39 Renewable Energy Country Attractiveness Index (RECAI) economies over the 1996–2020 period. Drawing on institutional and environmental governance theories, this study evaluates whether governance strengthens the environmental benefits of renewable-energy consumption. Unlike previous studies that primarily focus on aggregate governance measures, this study separately examines six governance dimensions: control of corruption, political stability, rule of law, government effectiveness, regulatory quality, and voice and accountability. Using panel data from the World Bank, the analysis employs fixed-effects estimation and validates the findings through robust regression, the generalized method of moments (GMM), and panel-corrected standard error (PCSE) approaches. The results indicate that renewable energy consumption significantly reduces both CO₂ emissions per capita and total CO₂ emissions in the long run. The findings further show that governance significantly moderates the renewable energy–CO₂ emissions relationship, with stronger governance enhancing the emission-reducing effect of renewable energy consumption. Governance mitigates the environmental impacts associated with urbanization and industrialization, highlighting the importance of institutional quality in managing the structural drivers of emissions. The robustness analyses confirm the consistency of the main results across the alternative estimation techniques. This study contributes to the literature by providing evidence of the distinct roles of individual governance dimensions in shaping environmental outcomes in economies actively engaged in renewable energy development. These findings suggest that renewable energy expansion and governance reforms should be pursued simultaneously to achieve more effective carbon emission reduction strategies. Full article

Agapé is a light–touch relationship enhancement smartphone app. This study used data from a longitudinal study of couples using the Agapé app to explore change within an array of behavioral processes to uncover patterns of interconnected change, thereby providing some of the first quantitative insights into how various relationship processes might be linked as relationships change over time. A sample of 405 couples in long-term relationships (810 partners, 50% women, 75% white, together M = 4.5 yrs, 50% living together, 33% currently dissatisfied) completed assessments across their first month of using Agapé. Men and women significantly improved on 15 of the 16 relationship processes assessed. As the study lacked a randomized control condition, it remains unclear if those improvements were due to using the Agapé app or to factors like expectancy effects, regression to the mean, self-selection, demand characteristics, or general participation effects. Network analyses explored correlational linkages among the self-reported pre–post changes observed. Results highlighted increases in three processes (quality time spent together, perceived partner responsiveness, and gratitude toward partner) as the processes most proximally linked to increases in relationship quality. The network findings also uncovered a number of patterns of interconnected change to be explored in future studies (e.g., increases in couples talking about their relationships to increases in mindful awareness within those relationships to increases in gratitude and quality time to increases in relationship quality). Thus, the results offer some of the first comprehensive multivariate (albeit correlational) insights toward understanding how relationship processes might work in concert with one another within a broader multivariate pattern of self-reported relationship change. Full article

As the application scale of battery energy storage gradually increases, the accurate prediction of the remaining service life of large-capacity energy storage batteries is crucial for high-quality development in this field. To address the issues of insufficient reliability and poor generalization in large-capacity energy storage battery life prediction, a deep learning framework based on a long short-term memory (LSTM) neural network is developed. Early aging data from the first 150 cycles is used for the model, with outliers removed and noise reduced through Savitzky–Golay (SG) filtering. Data normalization and a sliding window method are employed for training. The model is validated on two batches of large-capacity batteries under GB/T 36276-2023 conditions at 25 °C and 45 °C, achieving the root mean square errors (RMSEs) of 0.86% and 0.50%, respectively, over 1000 cycles. Additionally, the method is tested on small-capacity batteries from an MIT dataset, achieving an RMSE of 4.3%. A transfer learning module fine-tunes the model using cycles 151–300, reducing RMSEs to 0.18%, 0.10%, and 3.1% for the three battery sets. This enhances the model’s generalization and offers a practical solution for life prediction in battery inspection and evaluation. Full article

Controlled environment agriculture (CEA), including hydroponics, vertical farming (VF), and soilless cultivation, is expanding rapidly as food production shifts toward resource-efficient and climate-resilient systems. However, conventional substrates such as rockwool, peat, coco coir, and perlite present limitations related to nutrient buffering, structural stability, and environmental sustainability. Biochar has emerged as a promising alternative substrate component due to its porous structure, surface functionality, and ability to modify root-zone conditions. This review synthesizes current knowledge on the role of biochar in controlled cultivation systems, focusing on its physicochemical properties, substrate interactions, and plant physiological responses. Biochar incorporation influences water retention, aeration, nutrient availability, and microbial activity within confined root environments, thereby improving root architecture, photosynthetic performance, crop quality, and plant uniformity. Applications across hydroponic, VF, and soilless cultivation systems demonstrate improved moisture regulation, nutrient buffering, and substrate stability. Biochar interactions with conventional media such as coco peat, perlite, and peat moss further highlight its role in engineered growing substrates. Despite these advantages, challenges remain, including feedstock variability, pH and electrical conductivity effects, lack of standardized specifications, and limited long-term performance data in recirculating systems. Emerging research areas such as engineered biochar, nano-biochar, microbial integration, and precision cultivation technologies offer opportunities to optimize biochar performance in controlled environments. Overall, biochar represents a versatile and sustainable substrate component for CEA, with potential to enhance crop productivity, substrate durability, and resource efficiency. Future research should focus on material standardization, system-specific optimization, and large-scale validation to support commercial adoption. Full article

Ice–snow tourism in Northeast China is developing rapidly. Against this backdrop, revealing the spatial network structure of ice–snow tourism cities and assessing their disturbance resistance capacity is of great significance for achieving high-quality development of regional ice–snow tourism. This study takes 25 cities across the three northeastern provinces as network nodes, using data covering the period from January 2024 to March 2025. Integrating a complex network analysis framework, this paper comprehensively employs an accessibility model, tourism symbiotic linkage intensity model, and core–periphery model to distinguish core and peripheral cities within the network, analyze its structural characteristics and spatial patterns, and evaluate network vulnerability by simulating two scenarios: random attacks and deliberate attacks. The results indicate that: (1) Accessibility presents a concentric zonal pattern that attenuates gradually from the center to the periphery, accompanied by pronounced north–south disparities. Urban symbiosis intensity is strongly influenced by transportation distance, exhibiting a distinct proximity symbiosis pattern. (2) An ice–snow tourism symbiotic network has initially taken shape among northeastern cities. The network displays small-world properties; however, urban development is unbalanced, with marked hierarchical differentiation. Based on geographic location and resource endowments, the network can be divided into four cohesive subgroups. (3) The symbiotic network proves robust under random attacks, whereas connectivity declines sharply under deliberate attacks, embodying typical “robust-yet-vulnerable” structural characteristics. Both expanding the scale of core nodes and optimizing inter-node connection weights can significantly enhance network robustness. The static identification and dynamic dependency evaluation framework constructed in this study can effectively identify key nodes and vulnerable links within ice–snow city networks, and can serve as a reference for the coordinated development and structural optimization of ice-snow tourism in Northeast China. Full article

With the acceleration of urbanization, the production of household waste continues to increase, while traditional manual sorting methods suffer from low efficiency, high cost, and unstable accuracy. Deep-learning-based image classification technology provides an effective solution for automated waste classification. However, household waste images present challenges such as large variations in object scale, complex backgrounds, densely packed small objects, and easily confusable categories, making it difficult for existing models to meet practical classification requirements. To address these issues, this paper proposes BE-YOLOv8, an improved household waste image classification model based on YOLOv8, which integrates multiple strategies including data augmentation, attention mechanisms, and edge feature enhancement. First, to tackle the problems of limited training samples and class imbalance, an improved LMix data augmentation method is proposed. By introducing a label smoothing strategy, dynamically correcting mixed label weights, and adding a regularization penalty term to the loss function, the generalization ability of the model is effectively improved. Second, an Edge-Guided Multi-Scale Hybrid (EGMSH) attention mechanism is designed, which enhances the model′s perception of edge contours and multi-scale texture features through online edge computation, multi-scale depthwise separable convolutions, and adaptive gating fusion. Finally, a learnable BoundaryEdge feature enhancement module is proposed, which utilizes a trainable color projection layer and fixed-weight Sobel operators to generate high-quality edge features online and embeds them into the network via residual connections, significantly improving the discrimination of shape-similar and easily confusable categories. Experiments are conducted on a household waste image dataset containing 26,994 images across 20 categories. The results demonstrate that BE-YOLOv8 achieves a Top-1 accuracy of 83.5% on the test set, improving by 1.1% over the baseline YOLOv8. The hazardous waste category exhibits the most significant improvement, with a Top-1 accuracy of 92.5%. It is demonstrated that the model has excellent robustness in scenarios with complex backgrounds and easily confusable categories, providing a high-precision technical solution for practical waste classification applications. Full article

Atmospheric haze is a major source of image degradation in Earth observation systems, reducing visibility, distorting spectral information, and obscuring surface details in remote sensing imagery. Physics-based dehazing methods often hinge on simplified atmospheric assumptions, whereas purely data-driven networks struggle with ultra-high-resolution overhead imagery and the wide spatial variability of haze. To address these challenges in a way that respects the characteristics of very large remote sensing scenes, we introduce WaveDiff-R, a wavelet-guided diffusion framework with residual sub-band enhancement. Rather than running diffusion directly in the full spatial domain, WaveDiff-R performs a multi-level discrete wavelet transform (DWT) to separate low- and high-frequency components in a geometry-aware manner. The wavelet-guided diffusion module (WGDM) performs conditional diffusion only on the low-frequency approximation coefficients $A^K$ after a K-level DWT, reducing the denoising target by $4^K$ while restoring global luminance and chromaticity. In parallel, the residual sub-band enhancement module (RSEM), built with residual state space blocks (RSSBs), refines the high-frequency sub-bands, recovering sharp edges and textures by jointly modeling long-range dependencies and local details. This collaborative design couples global consistency with fine-grained fidelity while maintaining an efficiency suitable for real-world remote sensing pipelines. Extensive experiments on six benchmark datasets covering synthetic and real scenarios showed that WaveDiff-R achieved consistently strong results, surpassing state-of-the-art natural-image and remote sensing dehazing baselines in both quantitative metrics and visual quality. Full article

In recent years, data-driven remaining useful life (RUL) prediction for bearings has gained significant attention in industrial prognostics. However, real-world applications face challenges due to the scarcity and noise of run-to-failure data, as well as the high sensitivity of deep models to hyperparameters. Traditional methods often optimize data enhancement and prediction modules separately, which may limit end-to-end performance improvement and reduce generalization in real-world scenarios. To address this, this paper introduces ROA-DE-LSTM, a novel framework that jointly enhances prediction accuracy and generalization under limited and imperfect data conditions. The framework features an adaptive Remora Optimization Algorithm (ROA) that dynamically adjusts optimization strategies, thereby improving both data augmentation and RUL prediction. Specifically, an ROA-optimized Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) is employed to generate high-quality augmented samples, while an ROA-optimized LSTM network captures long-term temporal dependencies for stable RUL estimation. Unlike traditional approaches, ROA-DE-LSTM synchronously optimizes the data enhancement and prediction stages within a unified framework. Extensive experiments on the XJTU-SY and IEEE PHM2012 datasets demonstrate that ROA-DE-LSTM outperforms representative baselines and exhibits competitive advantages over other optimization-based strategies. Cross-dataset experiments verify the effectiveness of the proposed data enhancement and optimization strategies, showing that the framework maintains stable and robust performance under different operating conditions. Full article

The wheel hub is an important component of automotive components, and its quality directly affects the stability, safety, and comfort of the vehicle during driving. In the face of increasingly stringent industry standards, traditional runout measurement methods are no longer sufficient to fully characterize the quality attributes of wheels. In response to this demand, the processing algorithm for determining wheel hub quality through harmonic analysis is proposed in this paper, and the 12-point coordinate method (12-PCM) is applied to taking 12 pairs of measurement values at equal intervals in the wheel hub measurement data, and 12 pairs of measured values are substituted into the harmonic analysis formula to obtain the Fourier factor. According to the obtained Fourier factor, the amplitude and phase of various harmonics are calculated; furthermore, the various harmonic curve graphs of the wheel hub are drawn. The runout of various harmonics obtained is compared with the given judgment threshold to determine whether the hub quality is qualified. The threshold for determining the quality of qualified wheel hubs is set at 0.06 mm in this paper, and the allowable error threshold for multiple repeated measurements is set at 0.03 mm. The reliability and repeatability of the algorithm are verified by single loading single measurement and single loading repeated measurement. The experimental results show that the algorithm proposed has high stability and repeatability in this paper. Judging wheel hub quality through harmonic analysis, the intuitiveness of wheel hub quality judgment is enhanced, and the reliable basis for judging the quality of the wheel hub is provided. Full article

High-quality synthetic datasets are critical for environmental monitoring due to the high cost of primary geochemical data collection. However, selection criteria for generative frameworks under small-sample constraints remain poorly defined. This study evaluates two architectures—the deep-learning-based Tabular Variational Autoencoder (TVAE) and the parametric Gaussian copula—using an empirical geochemical dataset (N = 297) of 25 log-transformed elements. Synthesis quality was benchmarked via marginal distribution fidelity (KSComplement), correlation preservation (CorrelationSimilarity), privacy (LogisticDetection), and a composite QualityScore, supplemented by divergence metrics and a rare-earth element (REE+Th+U) subgroup analysis. In this dataset, the Gaussian copula demonstrated superior global correlation preservation (CorrelationSimilarity: 0.9633 vs. 0.9482), particularly for weak-to-moderate dependencies. Conversely, TVAE better replicated marginal distributions (KSComplement: 0.8742 vs. 0.8596), maintained localized correlations (MAD: 0.0728 vs. 0.1196), and showed enhanced privacy (LogisticDetection: 0.5757 vs. 0.3063). These complementary profiles suggest that, for this case study, the Gaussian copula may be preferable for dependency modeling, while TVAE appears better suited for secure open-data dissemination. Further validation on additional datasets is needed to assess the generalizability of these findings. Full article