Search Results (92,873)

Heavy metals in water are a global concern due to their persistence, toxicity, and bioaccumulation potential, highlighting the need to assess their effects on aquatic organisms. The ecotoxicity of eight potentially toxic metal nitrates (Cd (0.002 mg/L), Co (0.02 mg/L), Cr (0.02 mg/L), Cu (0.001 mg/L), Mn (0.04 mg/L), Ni (0.005 mg/L), Pb (0.002 mg/L), Zn (0.1 mg/L) up to 1000 mg/L for all metals) was assessed on the aquatic macrophyte Lemna minor, in a multilevel study. Growth responses, measured as frond number and average specific growth rate, were used to generate dose–response curves, but also to calculate EC₅₀ (Median Effective Concentration) and E_rC₅₀ (50% Effect Concentration based on growth rate) values. Based on EC₅₀, toxicity followed the order Cd > Co > Ni > Zn > Cu > Cr > Pb > Mn, while E_rC₅₀-based ranking was Cd > Ni > Co > Zn > Cu > Cr > Pb > Mn. Metals were classified as highly toxic (Cd, Ni; Co based on EC₅₀), moderately toxic (Co based on E_rC₅₀, Zn based on EC₅₀), slightly toxic (Cu, Cr; Zn based on E_rC₅₀, Pb based on EC₅₀), or practically non-toxic (Pb based on E_rC₅₀, Mn). Biochemical responses regarding photosynthetic pigments, primary metabolites, and oxidative-stress-related enzymes were investigated, revealing physiological effects for most metals. Species sensitivity distribution-based comparison indicated that L. minor sensitivity varies across metals, ranging from typical to relatively low within the broader species sensitivity distribution. Although the toxicity of potentially toxic metals to L. minor has been extensively investigated, data on ErC50 values for several of the metals investigated in the present study remain limited, highlighting the need for further ecotoxicological research. Full article

Understanding how urban park landscape features are associated with service population and service radius can inform park planning and management. In Changsha, China, this study measured both indicators for 29 parks on one weekday and one weekend day in spring 2025 using mobile signaling data, quantified landscape features from multisource data, and examined their associations. Service population and service radius were not highly correlated. The adjusted R² values for the two indicators were 71.8% and 84.5%, respectively. Park area and the number of parking lots within the park were significantly associated with both indicators in the same direction. Percent vegetation cover and the number of stores within the park were significantly associated with the two indicators in opposite directions. Elevation, presence of large-scale flower landscapes, distance to the nearest subway station, number of surrounding bus stops, distance to the city center, slope, percent water body, edge density of vegetation patches, number of public restrooms, and number of surrounding residential quarters were significantly associated with only one indicator. These findings reveal synergistic, trade-off, and indicator-specific association patterns between landscape features and urban park recreational services, and may inform differentiated urban park planning and management. Full article

High-resolution remote sensing object detection is a fundamental task in Earth observation. However, the storage, transmission, and downstream processing of high-resolution images impose substantial bandwidth, memory, and computational burdens. Compressed sensing (CS) samples and compresses the signals simultaneously, thereby reducing data transmission and storage overhead. Unfortunately, most existing CS-based pipelines require explicit image reconstruction before downstream inference, leading to heavy computational overheads and poor scalability for high-resolution remote sensing images (RSIs). This work focuses on post-acquisition image compression and explores algorithmically, rather than from a physical hardware implementation perspective, whether explicit image reconstruction is an indispensable intermediate step prior to object detection. To this end, we propose HR²SIOD-CL, an end-to-end compressed learning (CL) framework that performs object detection directly on CS measurements of high-resolution RSIs without explicit image reconstruction. HR²SIOD-CL integrates an entropy-driven content-aware adaptive sampling strategy and a measurement-domain detection backbone for multi-scale feature extraction. Although jointly optimized during training, the adaptive sampling and detection modules can be decoupled for flexible deployment. For fair comparison, an extra lightweight reconstruction network equipped with a single-step data-consistency correction is constructed as the baseline. Extensive experiments on the NWPU VHR-10 and DIOR datasets show that across various sampling ratios, HR²SIOD-CL surpasses the reconstruction-based detection method when integrated into two-stage detectors, and achieves comparable or superior detection performance to the reconstruction-based counterparts when integrated into single-stage detectors. Meanwhile, its computational overhead and GPU memory consumption are merely 6.97% and 35.74% of those of the reconstruction-based counterpart, respectively. These results indicate that CS measurements can function as an effective intermediate representation for object detection, and explicit image reconstruction is not a prerequisite when detection is the primary objective. Full article

The slow evolution and inconspicuous nature of buffer-layer ablation in high-voltage cables pose a significant challenge for early fault diagnosis. To tackle this issue, we propose a hybrid diagnostic approach that integrates frequency-domain impedance measurement with a convolutional neural network (CNN). A cable simulation model is first established using transmission-line theory and a distributed-parameter framework. We examine the impedance and phase responses at the cable’s sending end, revealing a consistent decreasing trend with rising frequency alongside periodic resonant peaks. The simulator generates a diverse set of spectral signatures corresponding to various cable health states. The CNN then extracts discriminative features from these waveforms, and a probabilistic clustering preprocessing step further refines the input data. Experimental results on a test set of 78 samples—comprising 52 experimentally measured normal spectra and 26 experimentally calibrated simulated spectra for mild and severe ablation—demonstrate a classification accuracy of 0.95, confirming that the proposed methodology enables reliable, non-intrusive detection of buffer-layer ablation without cable disassembly. Full article

Background: Inherited metabolic disorders (IMDs) are a heterogeneous group of genetic diseases, some of which require lifelong disease-specific dietary management. Children with phenylketonuria (PKU), organic acidemias (OAs), and glycogen storage diseases (GSDs) are particularly susceptible to nutritional imbalances and metabolic disturbances that may adversely affect bone metabolism and skeletal health. However, data regarding bone turnover markers and bone health in these pediatric IMD populations remain limited. Objective: This study aimed to evaluate bone metabolism by assessing biochemical parameters, bone turnover markers, and bone mineral density in children with PKU, OA, and GSD receiving long-term dietary treatment. Methods: This cross-sectional study included 95 children, comprising 25 patients with PKU, 20 patients with organic acidemias, 20 patients with glycogen storage diseases, and 30 healthy age-matched controls. Demographic characteristics, anthropometric measurements, and dietary intake were recorded. Biochemical evaluation included serum calcium, phosphorus, alkaline phosphatase (ALP), parathyroid hormone (PTH), and 25-hydroxyvitamin D levels. Bone turnover was assessed using C-terminal telopeptide of type I collagen (CTX), procollagen type I N-terminal propeptide (P1NP), and osteocalcin (OC) levels. Bone mineral density was evaluated by dual-energy X-ray absorptiometry (DXA). Results: Age and sex distributions were similar among the groups. Height was lower in OA than in controls and PKU, and in GSD than in PKU; weight and BMI were lower in both OA and GSD than in controls and PKU (p < 0.05). Significant differences were observed in dietary energy and macronutrient intake according to disease-specific nutritional regimens. Serum calcium, phosphorus, and 25-hydroxyvitamin D levels were significantly lower in all IMD groups compared with healthy controls (p < 0.001). ALP and PTH levels were significantly higher in PKU, OA, and GSD patients than in controls (p < 0.001). Among bone turnover markers, P1NP and osteocalcin levels significantly differed among groups, with the highest values observed in PKU patients (p = 0.006 and p = 0.028, respectively). CTX levels did not significantly differ among the groups (p = 0.183). DXA Z-scores were significantly lower in all IMD groups than in controls, with the lowest mean Z-scores observed in GSD patients. Conclusions: Children with PKU, OA, and GSD receiving long-term dietary treatment exhibited significant alterations in bone metabolism, characterized by lower vitamin D, calcium, and phosphorus levels, higher ALP and PTH concentrations, differences in bone formation markers, and lower DXA Z-scores compared with controls. Significant differences were observed in the bone formation markers P1NP and osteocalcin, whereas CTX levels remained comparable among the study groups. These findings underscore the importance of regular assessment of bone health and optimization of nutritional management in children under dietary treatment, while acknowledging that disease-specific mechanisms of skeletal involvement may differ among these conditions. Full article

Gender bias in generative artificial intelligence (GenAI) is both a technical and a social phenomenon: it emerges from historically patterned data, model design, and interactions in institutional use, and it cannot be understood by engineering or by social critique alone. This critical integrative review develops a more differentiated account. It connects feminist epistemology, Science and Technology Studies, critical AI scholarship, natural language processing, and governance research to examine five levels: historical knowledge production, technical representation and generation, benchmark evaluation, institutional deployment, and accountability. The review explains tokenization, next-token prediction, transformers, and the transition from static embeddings to contemporary language models before assessing evidence from standard fairness tests—coreference tests (WinoBias), sentence-pair tests (CrowS-Pairs), and stereotype tests (StereoSet)—as well as open-ended generation, multilingual testing, and text-to-image systems. It shows that measured bias varies with task, prompt, language, model version, and metric. What a test records and what that record means are therefore distinct questions: measurements are situated and depend on the instrument, and their interpretation draws on theory rather than following from the numbers alone. Evidence from employment, education, healthcare, and translation further indicates that the relevant unit of analysis is the model-in-context—the model together with the institution and workflow in which its outputs are used. Technical mitigation can reduce specific harms but does not repair unequal criteria, incomplete evidence bases, or weak institutional accountability. The review proposes a multilevel governance approach combining technical evaluation, documentation, professional and community oversight, appeals, remedies, and public-interest knowledge infrastructure. Its distinctive contribution is to connect three observations usually kept apart—how bias is measured, how generative systems concentrate epistemic authority, and how statistical learning is oriented toward past data—and to show why democratic and feminist governance can keep alternative technological futures open. Full article

Against the backdrop of China’s 14th Five-Year Plan, digital economy strategy, and dual-carbon goals, this study draws on panel data from Chinese A-share-listed firms over 2008–2024 to construct a provincial digital intelligence (DI) index using the entropy-weighting method, measure corporate green total factor productivity (GTFP) using the SBM–GML model, and examine the effect of DI on GTFP and the mechanisms underlying this effect. The results show that regional DI is significantly and positively associated with GTFP, and this association remains robust across alternative specifications and endogeneity treatments. Mechanism tests indicate that lower financing constraints and lower ownership concentration may serve as potential channels linking DI to GTFP. Human capital strengthens this positive effect, whereas total asset turnover weakens it. Heterogeneity analysis further shows that the productivity gains from DI are more pronounced among large firms. Although subgroup estimates are larger for firms without executives who have overseas experience, the between-group difference is not statistically robust and is therefore interpreted as exploratory rather than causal. These findings highlight the importance of integrating DI with green transformation, improving firms’ access to finance and human capital, and adopting differentiated support policies. Full article

Quantitative EEG features such as frontal alpha asymmetry, spectral ratios and signal-complexity measures are often presented as interpretable biomarkers of emotion. Such claims require the markers to generalize across individuals, yet common evaluation protocols allow overlapping epochs and recordings from the same participants to appear in both training and test sets. We re-evaluated qEEG-based valence and arousal recognition on DEAP and DREAMER under trial-grouped, participant-independent, within-participant and cross-dataset protocols. Epoch-pooled evaluation on DEAP gave ROC-AUC values of 0.689 for valence and 0.711 for arousal, whereas participant-independent evaluation of the same features and model returned 0.493 and 0.447. Grouping epochs by trial accounted for about 0.06 of that difference and separating participants for a further 0.13 to 0.17. The same features identified participants with accuracy of 0.998 on DEAP and 0.891 on DREAMER, and a predictor that used no EEG, assigning each trial its participant’s training-set positive rate, accounted for 42 to 84 percent of the above-chance discrimination of the epoch-pooled model. Emotion-related effects were reproducible within participants on DEAP but close to zero on DREAMER, and their direction reversed for about 40 percent of features across participants. In a matched participant-level comparison using a single fixed estimator in both arms, training on a participant’s own data improved DEAP valence by 0.092 AUC (95% CI 0.029 to 0.157, Holm-adjusted $p=0.042$) and gave no reliable benefit for DEAP arousal or for either DREAMER target. Across the channels shared by the two datasets, per-feature arousal effect sizes correlated moderately, although no individual feature reached false-discovery-rate significance in both datasets. Pooled qEEG emotion-recognition scores can therefore reflect participant-specific recording structure rather than transferable affective information. Population-level claims require participant-independent evaluation, while personalization should be considered only where stable within-person effects are demonstrated. Full article

The increasing tourism pressure on the UNESCO World Heritage Altai Mountains calls for efficient environmental monitoring tools. This study presents a Google Earth Engine framework that couples the Remote Sensing Ecological Index (RSEI) with land use/land cover (LULC) mapping to assess ecological changes in the Lake Manzherok area between 2020 and 2025. RSEI was derived from Sentinel-2 and Landsat imagery by combining four indicators (NDVI, MNDWI, NDBSI, LST) through principal component analysis. LULC classification was carried out using Random Forest trained exclusively on Sentinel-2 spectral bands. The results confirm that RSEI effectively captures ecological gradients in complex mountainous terrain, with the first principal component explaining 57–62% of the total variance. While 92% of the study area remained stable, 5.9% showed a decline in ecological status, spatially coinciding with a near doubling of built-up and bare surfaces from 9.89 km² to 18.17 km². The largest negative RSEI changes were associated with transitions from forestland (ΔRSEI = −0.29) and grassland (ΔRSEI = −0.20) to built-up/bare land, whereas reverse transitions displayed positive ΔRSEI values. These spatial patterns are consistent with the visible development related to tourism. However, because the built-up/bare land class also includes naturally bare surfaces, and because interannual climate variability may affect the RSEI components, it is important to interpret the ΔRSEI values as relative changes rather than absolute measurements of tourism impact. The proposed framework provides a reproducible and transferable tool for monitoring ecological quality in data-scarce mountain regions, delivering spatially explicit evidence that can support conservation and land-use planning. Full article

Bottom-up ship emission inventories derived from Automatic Identification System (AIS) data are normally reported on kilometre-scale grids, which merge port waters that perform very different functions. Working with an AIS-based STEAM inventory for the Yangtze River Delta (YRD), we ask a question that gridded inventories rarely separate: are the cells where ships accumulate time the same cells where they emit? To answer it, we define the Static–Dynamic Ratio (SDR), the ratio of hotelling (auxiliary-engine) to propulsion (main-engine) emissions within a cell, and use it to classify YRD waters without recourse to external port charts. Hotelling-dominated cells occupy 17.7% of the sea area and accumulate 59.3% of all ship-hours, a ship-hour density seven times that of transit-dominated fairways, yet they carry only 22.4% of NO_x. A vessel at anchor emits about one-eighth as much NO_x per hour as one under way, and the two effects nearly cancel. The cancellation is species dependent: low-load correction factors are steeper for sulfur and particulate species than for NO_x, so hotelling zones reach relative SO₂ and PM_2.5 densities of 1.21 and 1.09 against 0.89 for NO_x. Coarsening the same activity field from 100 m to 1 km drops the share held by the busiest 1% of cells from 70% to 50%, showing how kilometre grids manufacture apparent continuity along shipping lanes. Activity hotspots are therefore not emission hotspots, and anchorage-targeted measures such as shore power are best justified by particulate and sulfur exposure near populated coasts rather than by their share of the regional NO_x burden. Full article

Background: Census data in Hungary indicate a substantial decline in declared religious affiliation, with Christian identification falling from 74.4% in 2001 to 42.5% in 2022, underscoring the need for culturally validated measures of religious faith. Methods: This study examined the psychometric properties of the Hungarian versions of the Santa Clara Strength of Religious Faith Questionnaire (SCSORF-10) and its 5-item short form (SCSORF-5) in a cross-sectional online sample of 430 participants (M_age = 22.28, SD = 9.23). Results: Confirmatory factor analysis supported a unidimensional structure for both scales, with acceptable support when considering the full pattern of indices, but with elevated RMSEA as a limitation. All models were estimated with the mean- and variance-adjusted weighted least squares estimator (WLSMV) on polychoric correlations, under which fit improved markedly for the SCSORF-10 (CFI = 0.992, TLI = 0.989, SRMR = 0.040, RMSEA = 0.080). Modification indices identified a single interpretable residual covariance between the two explicitly devotional items, which did not alter the solution and was not retained. Measurement invariance across gender was tested with the sequence appropriate to ordered categorical indicators, and configural, threshold, threshold-and-loading, and strict invariance were all supported. Internal consistency was high (SCSORF-10: α = 0.95, ω = 0.95; SCSORF-5: α = 0.89, ω = 0.89). Convergent validity was demonstrated by strong associations with spiritual well-being, while correlations with self-esteem were weak, providing limited discriminant evidence against a conceptually distant construct. Conclusion: The short form showed psychometric properties closely comparable to the full scale; because its items are a subset of the full version, this correspondence is partly expected. Both instruments appear to be reliable and valid for assessing religious faith in Hungary. Known-group evidence was strong: scores separated self-described religious respondents from atheists (d = 1.37) and from believing-but-not-religious respondents (d = 1.17), and the religiosity category accounted for 27.5% of score variance, whereas scores were unrelated to gender and age. Full article

Shoreline determination is fundamental to coastal research, spatial planning, and legal boundary delineation but remains challenging in low-relief coastal areas where small sea-level variations can produce substantial horizontal shoreline displacements. This study compares shoreline determination methods based on tide gauge (TG) observations, LiDAR data, Sentinel-1 synthetic aperture radar (SAR) and Sentinel-2 optical satellite imagery using the low-relief coast of Pärnu Bay, Estonia, as a case study. The comparison was based on shorelines derived from Sentinel-1 and Sentinel-2 imagery acquired on selected common acquisition dates within the 2015–2025 study period, rather than on a temporally continuous annual dataset, and compared with temporally matched LiDAR-derived shorelines extracted from a Digital Terrain Model (DTM) generated from a 2021 LiDAR survey. The LiDAR-derived shorelines were extracted using the mean sea level (MSL) observed at the Pärnu and Häädemeeste TGs at the satellite overpass time, while the satellite-derived shorelines were additionally validated against RTK GNSS measurements. The results demonstrate that the evaluated methods produce substantially different shoreline positions. Sentinel-2-derived shorelines generally corresponded more closely to the temporally matched LiDAR-derived shorelines than Sentinel-1-derived shorelines and most accurately represented the instantaneous land–water boundary during field validation. These findings demonstrate that different shoreline determination methods represent different shoreline definitions. Consequently, shoreline datasets should be interpreted according to their intended purpose rather than treated as directly interchangeable. Full article

Rock cohesion ($c$) and angle of internal friction ($\phi$) are core parameters for rock mass stability analysis and engineering design; however, traditional triaxial tests are costly and time-consuming, limiting their availability in preliminary engineering assessments. To address this limitation, this study develops a machine learning framework that predicts these parameters from easily measurable physical properties, enabling rapid and cost-effective estimation without the need for complex laboratory testing. Based on a total of 199 sets of measured data from four rock types (shale, limestone, quartzite, and quartz-mica schist) in the Himalayan region, this study uses P-wave velocity ($V_p$), density ($\rho$), uniaxial compressive strength (UCS), and tensile strength (TS) as input variables. It employs four models: Support Vector Regression (SVR), Random Forest (RF), Multi-Layer Perceptron (MLP), and extreme gradient boosting (XGBoost) to predict $c$ and $\phi$. Hyperparameters were tuned using grid search and Bayesian optimization. We compared unified modeling with rock-type-specific modeling, performed interpretability analysis using SHapley Additive exPlanations (SHAP), and tested robustness by introducing Gaussian noise. The results show that XGBoost produced the best predictions at$c$ (test set $R^2$ = 0.9901, RMSE = 0.512 MPa), while the Bayesian-optimized SVR model yielded the best results at $\phi$ ($R^2$ = 0.9776, RMSE = 0.744°). Rock-type-specific modeling improved the $R^2$ for limestone at $\phi$ by 0.3541; the SHAP contribution for UCS and TS exceeded 70%; Random Forest demonstrated the best noise resistance, with a decrease in $R^2$ of less than 0.04 under 10% noise. In summary, the strategy proposed in this paper allows for the selection of prediction schemes based on data quality and lithological differences, providing a feasible approach for rapidly obtaining rock strength parameters. Full article

Chemical hazards generated during food production and processing represent a major food safety concern because of their toxic and carcinogenic potential. This study investigated the occurrence and distribution of acrylamide reported through the EU Rapid Alert System for Food and Feed (RASFF) alongside research data to evaluate temporal trends, food categories at risk, and the implications of current regulatory and mitigation measures. A total of 98 notifications were identified between 2009 and 2025, with a marked increase following the implementation of EU regulatory measures in 2017 and 2019. However, consistently high numbers of notifications were observed only during the final three years of the study period. Acrylamide was predominantly associated with thermally processed foods, particularly cereals and bakery products, and prepared dishes and snacks. The geographical distribution of notifications partially reflected established production and trade patterns, with Bosnia and Herzegovina, Serbia, Ukraine, and the Netherlands representing the most frequently reported countries of origin. Classification of the reported products according to the NOVA system, combined with RASFF risk decisions and notification classifications, revealed a marked concentration of high-risk notifications in ultra-processed foods. These products accounted for 85.7% of notifications classified as posing a serious risk and 77.3% of alert notifications, underscoring their disproportionate contribution to acrylamide-related food safety concerns. Overall, the findings demonstrate that, despite the implementation of regulatory measures and mitigation strategies, acrylamide remains a persistent challenge throughout the food chain, highlighting the need for continuous monitoring, enhanced analytical surveillance, and more effective mitigation approaches. Full article

Objective: To examine the reliability, construct validity, and discriminative capacity of the Spanish version of the Economic Living Standards Index Short Form (ELSI_SF) in a sample of older adults in Spain. Methods: This analysis was conducted using baseline data from 133 participants (mean age = 66.2 years, SD = 5.6; 53.8% women) recruited from community health and social centers in the provinces of Cuenca, Albacete, and Toledo within the PEPE cohort. Participants completed the ELSI_SF (25 items), sociodemographic measures, the SF-12 Health Survey, and the International Fitness Scale (IFIS). Internal consistency was assessed using Cronbach’s alpha. Convergent validity was examined through correlations between ELSI_SF scores and education, occupation, physical and mental health, and perceived fitness. Discriminative ability for health-related variables was analyzed using ANOVA and partial η². The internal structure was explored using exploratory factor analysis (EFA) and subsequently examined against the original four-factor theoretical model using confirmatory factor analysis (CFA), based on polychoric correlations and estimation methods appropriate for ordinal data. Results: The mean ELSI_SF score was 23.0 ± 3.9, with most participants classified as having comfortable or good living standards. Internal consistency was acceptable for the total scale (α = 0.779), whereas the ownership restrictions (α = 0.441) and self-rating (α = 0.319) subscales showed low reliability. Significant associations were observed between the ELSI_SF total score and selected socioeconomic and health-related measures. Factor-analytic findings provided only partial support for the original four-domain structure. Although CFA showed favorable CFI, TLI, and RMSEA values, the elevated SRMR and instability of the self-rating factor warrant cautious interpretation of the factorial solution. Conclusions: The Spanish ELSI_SF provides preliminary psychometric evidence supporting the use of its total score among relatively socioeconomically advantaged older adults. However, limitations in subscale reliability, factorial stability, and the restricted socioeconomic variability of the sample preclude definitive validation across the full spectrum of material living standards. Further evaluation in larger, independent, and more socioeconomically heterogeneous samples, particularly including older adults experiencing material deprivation, is required. Full article