Search Results (134,332)

This study examines motivational and capability barriers to internet non-adoption in Thailand’s near-saturated digital economy. Using the 2025 Q4 ICT Household Survey conducted by Thailand’s National Statistical Office, the analysis focuses on 20,633 adult non-adopters who report either motivational or capability-related barriers. The dependent variable distinguishes capability non-adoption, defined as lack of skill or awareness, from motivational non-adoption, defined as lack of perceived need or privacy/security concerns. Weighted logistic regression with normalised population weights, PSU-clustered robust standard errors, and average marginal effects is used to estimate associations between household ICT access, age, education, employment, smartphone access, and barrier type. Motivational barriers account for 56.2% of the two-category non-adopter population and capability barriers for 43.8%. Although motivational reasons are the more common, household ICT access is positively—if modestly—associated with capability rather than motivational barriers (average marginal effect +1.7 percentage points): capability-constrained non-adopters are concentrated in connected households, the compositional signature predicted by the second-level digital divide. Age does not significantly moderate this association. Among older non-adopters, education, employment, and smartphone access are negatively associated with capability barriers, while household ICT access is not. The findings suggest that in post-access digital economies, household connectivity is insufficient for digital market inclusion; individual-level skills and device access become central to expanding effective digital market reach. Full article

Detecting and identifying faults in industrial chemical plants is essential for safe and stable operation, and modern monitoring systems increasingly rely on deep learning to classify faults from multivariate sensor data. A practical obstacle to adoption is trust: most deep-learning diagnosers reach their decisions through a classification layer that operators cannot inspect, making it hard to see how the model maps process signals to a particular fault. This study targets fault diagnosis on the Tennessee Eastman (TE) process, a standard benchmark of simulated chemical-plant sensor data, and asks whether this final decision stage can be made directly inspectable without sacrificing accuracy. We propose CNN–Transformer–KAN (CTKAN), a hybrid model that learns local temporal patterns with a one-dimensional convolutional encoder, captures global inter-time-step dependencies with a Transformer encoder, and classifies faults with a Kolmogorov–Arnold Network (KAN) head whose learnable B-spline activations can be plotted and examined individually, in place of a conventional multi-layer perceptron (MLP). On the TE benchmark, CTKAN attains a Macro-F1 of 91.38 ± 0.26% over ten independent runs, comparable to a CNN + Transformer + MLP ablation (91.21 ± 0.32%) and a capacity-matched MLP-head variant (91.43 ± 0.37%) within seed-to-seed variability. The main finding is therefore not a higher score: at matched capacity the KAN and MLP heads are statistically indistinguishable in accuracy, so the KAN head’s value is to add a directly inspectable view of the classification stage at no measurable accuracy cost, helping process engineers sanity-check how the diagnoser separates faults in safety-critical settings. Full article

Despite the growing attention paid to sustainability education, limited quasi-experimental research has examined how metacognitive guidance can integrate cognitive, affective, and agency-oriented learning in food-related climate education. This classroom-based quasi-experimental study, complemented by student interviews, investigated a six-lesson metacognitive guidance-based unit designed to strengthen ninth-grade students’ sustainable food literacy (SFL), climate-change perceptions and attitudes, and constructive hope. Participants were 59 students from two intact classes in northern Israel; one class received the intervention, and the other received traditional instruction on the same content. Quantitative data were collected through a sustainable food and climate change knowledge test and a climate change literacy questionnaire and were analyzed using mixed-design repeated-measures ANOVA, t-tests, and multiple regression. Qualitative data were obtained from individual semi-structured interviews with students in the experimental group. Results indicated significant intervention-related gains in SFL knowledge, climate-change perceptions, climate-change attitudes, and constructive hope, with moderate-to-large time × group effects across the main outcomes (partial η² = 0.16–0.33). Climate-change perceptions emerged as the strongest post-intervention predictor of constructive hope (β = 0.92, p < 0.001). Interviews illustrated how reflective prompts, self-monitoring, discussion, and learning artifacts supported conceptual understanding, moral responsibility, perceived agency, and self-reported short-term intentions for sustainable food choices. The findings suggest that metacognitive guidance can support integrative, hope-oriented sustainability learning among adolescents. These findings should be interpreted cautiously given the small non-random sample, the use of two intact classes, the short six-lesson intervention, and the reliance on short-term self-reported outcomes. The study’s novelty lies in integrating sustainable food literacy, climate-change perceptions and attitudes, and constructive hope within a metacognitively guided food–climate unit in a culturally underrepresented Druze school context. Full article

Offshore wind farms are increasingly and rapidly expanding due to their ability to harness strong and consistent wind energy resources. Large offshore wind farms are connected to mainland grids through High-Voltage Direct Current (HVDC) technology. However, offshore wind farms can often experience disturbances related to sudden wind changes, voltage drops/dips, faults related to converter switching, and unbalanced grid conditions which affect both the HVDC operation and wind turbine output. As a result, there is a growing need for more advanced and reliable modeling and monitoring tools. Moreover, traditional proportional-integral (PI) controllers are widely applied in wind turbines and HVDC systems due to their simple structure, easy implementation, and reliability. However, PI controllers perform poorly under non-linear and abnormal/fast-changing conditions, especially during sudden drops in wind power and grid faults. With this background, this paper first develops a digital twin model of an offshore wind farm that enables remote operation and monitoring of individual wind turbines. Also, an artificial intelligence (AI)-based controller, namely a fuzzy logic controller (FLC), is proposed to maintain transient stability of a full digital twin-based offshore wind farm connected to the HVDC grid under fault conditions. The effectiveness of the proposed FLC is demonstrated by considering a digital twin-enabled 700 MW offshore wind farm. The performance of the proposed FLC has been compared with that of the PI controller. Simulations performed by the MATLAB/Simulink software show that during the moderate voltage dip at 15 s, the PI controller experienced a 29.8% power reduction with a recovery time of approximately 9 s, whereas the FLC reduced the power drop to 23.1% and recovered within 6 s. During the severe converter disturbance at 15 s, the PI controller recorded a 36.9% power reduction compared to 23.4% for the FLC. Similarly, during the short-duration turbulence at 15 s, the PI controller exhibited a 36.73% power drop and recovered in approximately 7 s, while the FLC limited the power reduction to 19.17% and recovered within 5s. Overall, the FLC provided improved voltage stability, faster recovery, reduced oscillations, and superior fault ride-through capability compared with the conventional PI controller, demonstrating its effectiveness for digital twin-enabled offshore wind farm application. Full article

While individual determinants of students’ well-being are well established, less is known about the association with the institutional context. This study evaluates institutional-level factors associated with students’ mental health while controlling for individual characteristics. The cross-sectional analysis used data from 12 German institutions (n = 13,715) collected in the StudiBiFra survey on study conditions and student mental health. Individual-level variables included gender, age, study subject group, and four mental health variables (general well-being, depressiveness, cognitive stress, and exhaustion). Institution-level variables comprised institution type, excellence status, multi-campus structure, size, and satisfaction with the quality of health promotion services. Multilevel binary logistic regression models were applied to examine associations between institutional characteristics and mental health outcomes, adjusting for individual factors. Students enrolled at universities of applied sciences showed a lower likelihood of reporting depressiveness and exhaustion. Higher levels of depressiveness and cognitive stress were observed among students at medium-sized institutions compared to small ones. Students not enrolled at institutions with excellence status had lower risks of depressiveness, stress, and exhaustion. Additionally, higher satisfaction with institutional health promotion services was associated with reduced odds of depressiveness. Institutional factors are related to students’ mental health beyond individual characteristics, highlighting the need for a holistic, setting-based approach. Full article

Systemic arterial hypertension (SAH) is a multifaceted condition marked by sustained elevations in arterial blood pressure. Its occurrence is closely related to alterations in target organs, such as the liver. Non-pharmacological treatments have been proposed for these effects. Thus, the aim of this study was to investigate the effects of açaí supplementation and resistance training, applied individually or in combination, on blood pressure and liver structural parameters. An experimental, quantitative, and longitudinal study was conducted using young Wistar rats (~60 days old) and spontaneously hypertensive rat (SHR) strains. Fifty rats were divided into five experimental groups: Wistar Control (C), Hypertensive Control (H), Hypertensive Trained (HT), Hypertensive Açaí-Supplemented (HA), and Hypertensive Trained plus Açaí Supplementation (HAT). Each group consisted of ten animals. Subsequently, analyses were performed for the antioxidant capacity and proximate composition of the açaí pulp, systolic blood pressure assessment, and histological evaluation of the liver. The açaí used exhibited high antioxidant capacity. At the end of the experimental period, the trained groups increased their maximal load carried, along with a reduction in systolic blood pressure in all treated groups. Açaí supplementation resulted in lower relative liver mass compared with the H group. The hypertensive condition promoted extracellular matrix expansion and a reduction in hepatocyte proportion. Both interventions attenuated these effects, and the combined treatment (HAT) produced the greatest improvement, indicating an additive response. Hypertension also elevated hepatic glycogen concentration, and the treatments reduced this alteration. It is concluded that açaí supplementation and resistance training could promote positive adaptations in the liver of hypertensive animals. Full article

Deep learning models for power quality (PQ) disturbance classification remain critically vulnerable to adversarial perturbations, with classification performance degrading severely under white-box attacks. Existing defenses address individual models in isolation and provide no mechanism for operators to assess whether the system is under attack or which classifier remains trustworthy. This paper proposes a two-stage framework that combines adversarial training with large language model (LLM) reasoning to improve both robustness and interpretability. In the first stage, four architecturally diverse classifiers, including a proposed Multi-Scale Temporal Attention Network (MSTAN), are evaluated under four adversarial attacks (FGSM, PGD, C&W, and UAP), and their failure patterns are recorded as structured vulnerability fingerprints. The ensemble is then retrained via adversarial training on mixed clean and perturbed signals. In the second stage, an LLM analyzes the ensemble predictions alongside the fingerprint knowledge base to perform attack detection, fingerprint-guided meta-classification, and operator-facing threat report generation. On a 17-class, 255,000-signal synthetic benchmark, adversarial training recovers FGSM and PGD accuracy from below 25% to the 53–78% range, with MSTAN achieving the highest post-training robustness (78.26% under FGSM, 65.41% under PGD). The LLM reasoning layer provides an additional 3.5–6.2 percentage point improvement over majority voting by selecting the most reliable ensemble member based on the inferred attack condition, and detects adversarial attacks with 87.6% overall accuracy. To our knowledge, this is the first integration of LLM-based ensemble reasoning into the PQ adversarial robustness pipeline and the first application of the C&W optimization attack to power quality signals. Full article

Background/Objectives: Impacted maxillary canines are frequently associated with variations in maxillary arch morphology; however, the relationship between impaction position and three-dimensional arch characteristics remains unclear. This study aimed to evaluate the association between buccally and palatally impacted maxillary canines and maxillary arch morphology using CBCT and three-dimensional digital model analysis. Methods: This retrospective cross-sectional study included CBCT images and three-dimensional dental models of 86 individuals with unilateral impacted maxillary canines (mean age: 16.1 ± 0.72 years). Impacted canines were classified as buccal or palatal according to CBCT findings. Maxillary arch morphology was assessed using digital model analysis. Statistical comparisons between groups were performed using independent-samples t-tests (p < 0.05). Results: The buccally impacted group demonstrated significantly greater arch length, higher arch length-to-arch width ratios, greater mesiodistal width of the four maxillary incisors and increased tooth–arch discrepancy (p < 0.05). In contrast, intermolar width and available arch space were significantly greater in the palatally impacted group (p < 0.05). No significant differences were identified in arch width or palatal depth measurements between groups (p > 0.05). Intra-examiner reliability demonstrated excellent agreement (ICC > 0.90). Conclusions: Maxillary dental arch morphology differed according to the position of impacted maxillary canines. Buccal impaction was associated with sagittal arch elongation and increased tooth–arch discrepancy. In contrast, palatal impaction was not consistently associated with reduced transverse dental arch dimensions within the measurements evaluated in this study. These findings contribute to a better understanding of the association between impacted canine position and maxillary dental arch morphology and may assist clinicians in the morphological assessment of patients with impacted maxillary canines. Full article

Background/Objectives: This study aimed to evaluate the gingival crevicular fluid (GCF) levels of fractalkine/CX3CL1 and CX3CR1 in patients with gingivitis and periodontitis before and after non-surgical periodontal therapy. Methods: A total of 90 individuals comprising 30 with stage 3 periodontitis, 30 with gingivitis, and 30 periodontally healthy, were enrolled in the study. Gingivitis and periodontitis patients underwent non-surgical periodontal treatment. GCF samples were collected at baseline and at 1 and 3 months after treatment. CX3CL1 and CX3CR1 were measured by an ELISA analysis. Results: GCF CX3CL1 and CX3CR1 were significantly elevated in patients with periodontitis and gingivitis compared to healthy controls (p < 0.001). The periodontitis patients also showed higher GCF levels of CX3CL1 and CX3CR1 than those with gingivitis (p < 0.001). Significant decreases in GCF CX3CL1 and CX3CR1 were detected at 1 month after periodontal treatment compared to baseline values in both the gingivitis and periodontitis patients (p < 0.001). Moreover, the periodontitis patients exhibited significant decreases in both CX3CL1 and CX3CR1 levels at 3 months post-treatment compared to 1 month (p < 0.001), whereas no significant changes were observed between the two time points in the gingivitis patients (p > 0.05). Conclusions: Our findings suggest that the CX3CL1–CX3CR1 axis might contribute to the inflammatory processes of periodontal diseases and may represent a treatment-responsive component of the local host response. Full article

Background/Objectives: Accurate post-mortem interval (PMI) estimation remains challenging in forensic pathology. Although potassium (K⁺) is the most well-validated single biomarker in vitreous humor (VH), multivariate approaches may enhance precision by capturing the complex cascade of post-mortem biochemical changes. This study aimed to develop and validate a multivariate PMI estimation model incorporating three biochemical markers—potassium, ammonium (NH₄⁺), and albumin (ALB)—in vitreous humor using automated clinical chemistry platforms for practical forensic application. Methods: Vitreous humor samples from 38 autopsy cases with documented PMIs (39.5–285 h; mean, 105.5 h) were analyzed for K⁺ (Cobas C8000), NH₄⁺ (Cobas C8000), and ALB (Immage 800 nephelometry). Univariate and multivariate regression analyses were performed, with the residual standard error (RSE) as the primary measure of accuracy. Model validation was conducted by back-calculating PMI in four samples completely distinct from the training cohort. Results: All three analytes demonstrated strong individual correlations with PMI (R²: K⁺ = 0.88, ALB = 0.78, NH₄⁺ = 0.69; all p < 0.001). The multivariate regression model [PMI = 40.25[Alb] + 0.01573[NH₄⁺] + 5.339[K⁺] − 53.032] yielded an RMSE of ±15.5 h (MSE = 240.25 h²), outperforming potassium-only models (RMSE = ±22.6 h). Although NH₄⁺ showed limited statistical significance in the multivariate model (p = 0.128), its inclusion improved overall predictive accuracy. External validation in an independent cohort of four subjects (distinct from the 38 subjects in the training set) demonstrated a mean absolute error (MAE) of 20.4 h. Conclusions: The multivariate approach combining K⁺, NH₄⁺, and ALB in VH improves PMI estimation accuracy compared with single-marker methods. The use of automated clinical chemistry platforms enhances reproducibility and facilitates practical implementation in forensic laboratories. Full article

Disordered eating has been identified as a concern among individuals with type 1 diabetes mellitus (T1DM) and is associated with poorer glycaemic control, which may result in vascular complications and increased morbidity and mortality. Despite this, there is a paucity of research examining disordered eating behaviour (DEB) among individuals with T1DM in Ireland. This study aimed to examine the prevalence of DEB among adults with T1DM attending an Irish hospital outpatient clinic. This cross-sectional study recruited 95 adults with T1DM attending an outpatient diabetes clinic at Sligo University Hospital. Screening for DEB was conducted using self-reported measures, including the Diabetes Eating Problem Survey–Revised (DEPS-R) and the Eating Disorder Examination Questionnaire (EDE-Q 6.0). DEB was identified in 16.3% (n = 15) of the participants using the DEPS-R and in 3.2% (n = 3) using the EDE-Q 6.0. Higher levels of DEB were observed among the female participants; however, a notable prevalence was also identified among the male participants. DEB demonstrated a weak positive correlation with body mass index (BMI) and a negative correlation with age and height. No association was found between DEB and glycated haemoglobin (HbA1c). The substantial prevalence of DEB among adults with T1DM, with nearly one in six participants affected, highlights the presence of clinically relevant disordered eating symptoms in this population. The findings suggest potential differences in case identification between diabetes-specific and generic screening tools. These results support further evaluation of diabetes-specific screening instruments such as the DEPS-R in adult T1DM populations within larger studies before routine clinical implementation. Full article

This study examines how mothers raising children with disabilities in American Samoa experience the processes of seeking diagnosis, navigating special education, and advocating for services within an insular rural context. American Samoa, an unincorporated U.S. territory located 2600 miles from Hawaiʻi with a population under 50,000, represents a case of what we term insular rurality—a condition in which the structural disadvantages of rurality are intensified by oceanic isolation, territorial governance, and colonial history. Data were collected through three focus groups with fifteen mothers whose children hold a range of disability diagnoses, with a card sort activity at the outset of each session serving as an idiographic anchor to protect individual voice within the group format. Analysis followed Interpretative Phenomenological Analysis adapted for focus groups (IPA-FG), proceeding from line-by-line exploratory noting through Personal Experiential Themes and Group Experiential Themes within each focus group case to cross-case convergence and divergence analysis, interpreted through the Fonofale model of Pacific wellness. Findings reveal two overarching themes: systemic invalidation, in which mothers encountered deficit-based assumptions, stagnant educational goals, and institutional disengagement; and parent peer support as the primary infrastructure, in which mothers became de facto experts, built community-driven solutions, and envisioned more inclusive futures. Technology emerged as a contradictory force—valuable for parent learning but largely ineffective for children’s remote therapy. These findings suggest how workforce shortages and geographic isolation create conditions in which maternal advocacy becomes a systems-level necessity rather than a personal choice. Implications for rural education policy, IDEA implementation in U.S. territories, and culturally grounded family support are discussed. Full article

Background: Metabolic control in phenylketonuria (PKU) is known to deteriorate with age, but national-level data describing blood phenylalanine (Phe) control across the United Kingdom (UK) are limited. Objective: To characterise blood Phe control in individuals with PKU attending UK metabolic centres. Methods: Sixteen UK centres (nine paediatric, six adult, one mixed) retrospectively extracted blood Phe results collected between 2012 and 2018. Demographic, phenotypic and monitoring-related variables were analysed. Written consent for data collection was obtained from all patients or their caregivers. Results: Data were available for 871 individuals (55% female), of whom 744 (85%) were classified as follows: classical PKU, 75%, mild PKU, 22% and hyperphenylalaninaemia, 3%. Mean blood Phe concentrations were significantly higher in adults than children (491 ± 308 vs. 303 ± 199 µmol/L; p < 0.001), and the proportion of samples within target range declined steadily with age, from 78% in children <2 years to 36% in adults ≥41 years. Individuals with classical PKU had higher mean Phe concentrations and lower target attainment than those with HPA (386 vs. 300 µmol/L; 61% vs. 78%; p < 0.001), while mild PKU and HPA showed comparable control. Females generally demonstrated better metabolic control than males. More frequent dried blood spot sampling for blood Phe was strongly associated with improved metabolic control: weekly (254 ± 175 µmol/L; 82% within target), fortnightly (319 ± 207 µmol/L; 70%), monthly (397 ± 231 µmol/L; 61%), and less than monthly (624 ± 349 µmol/L; 44%). Nearly half of the blood Phe samples (47%) with recorded timing were taken in a non-fasting state. Conclusions: Achieving lifelong metabolic stability on a Phe-restricted diet alone remains challenging. These national data highlight the need for broader therapeutic options to support individuals with PKU across the lifespan. Full article

Melatonin occupies a paradoxical position in contemporary sleep medicine: despite its physiological role as a regulator of circadian timing, it is frequently used and perceived as a nonspecific “natural” hypnotic. Although melatonin demonstrates modest benefits for sleep initiation and clearer efficacy in circadian rhythm sleep–wake disorders, its clinical use is often undermined by diagnostic imprecision, inappropriate dosing, mistimed administration, inconsistent formulations, and inadequate patient counseling. Circadian disorders can be misclassified as primary insomnia, leading to symptomatic treatment approaches that fail to address the underlying phase misalignment. At the same time, supraphysiological doses and reflexive bedtime administration have become normalized despite evidence that melatonin acts primarily as a chronobiotic whose effects depend more on timing than dose. Regulatory inconsistencies and substantial variability in over-the-counter preparations further complicate safe and reproducible use. These factors contribute to avoidable treatment failure, inaccurate labeling of nonresponse, and persistent misconceptions regarding melatonin’s mechanism of action. Therefore, melatonin should be approached as a pharmacological intervention requiring the same diagnostic rigor, individualized dosing, and longitudinal assessment expected of other sleep therapeutics, particularly when integrated with behavioral and circadian interventions. Full article

Traditional Mediterranean grapevine landraces represent irreplaceable reservoirs of adaptive diversity, yet many regional germplasm pools remain poorly characterized, limiting conservation strategies and climate-resilient breeding. This study presents the first comparative genetic assessment of 154 local Vitis accessions from three historically interconnected but genomically underrepresented Mediterranean regions: Greece, Morocco, and Slovenia. Using 12 highly polymorphic nuclear SSR markers, we detected substantial genetic diversity (168 alleles; mean heterozygosity He = 0.881) with distinct regional signatures. Moroccan accessions exhibited the highest allelic richness and 11 private alleles, reflecting diverse agroecological adaptation. Slovenian germplasm formed a cohesive, genetically stable cluster with high effective allele numbers. Greek accessions exhibited the highest observed heterozygosity and 14 private alleles, consistent with the Aegean’s role as a major diversification hotspot. Despite >90% of variance occurring within individuals, AMOVA and pairwise FST (0.050–0.061) revealed low to moderate but significant geographic differentiation. Multivariate analyses (PCA, UPGMA) and Bayesian clustering (sNMF, K = 3) consistently resolved three regional genetic groups with varying admixture levels, consistent with a mosaic domestication model, as previously proposed for the Mediterranean basin, shaped by recurrent introductions, wild introgression, and region-specific selection. Our results show that peripheral Mediterranean germplasm harbors meaningful, regionally distinctive, substantial, non-redundant diversity not fully represented in surveys focused on climate adaptation, disease resistance breeding, and long-term genetic resource conservation. These findings challenge simplistic diffusion models and emphasize the strategic importance of geographically comprehensive sampling in grapevine conservation programs. Full article