Search Results (7,860)

As one of the most widely used high-performance kernels, General Matrix Multiplication, or GEMM, plays a pivotal role in diverse application fields. With the growing prevalence of training for Convolutional Neural Networks (CNNs) and Large Language Models (LLMs), the design and implementation of high-efficiency, low-precision GEMM on modern Neural Processing Unit (NPU) platforms are of great significance. In this work, HGEMM for Ascend NPU is presented, which enables collaborative processing of different computation types by Cube units and Vector units. The major contributions of this work are the following: (i) dual-stream pipeline scheduling is implemented, which synchronizes padding operations, matrix–matrix multiplications, and element-wise instructions across hierarchical buffers and compute units; (ii) a suite of tiling strategies and a corresponding strategy selection mechanism are developed, comprehensively accounting for the impacts from M, N, and K directions; and (iii) SplitK as well as ShuffleK methods are raised to address the challenges of memory access efficiency and AI Core utilization. Extensive evaluations demonstrate that our proposed HGEMM achieves an average 3.56× speedup over the CATLASS template-based implementation under identical Ascend NPU configurations, and an average 2.10× speedup relative to the cuBLAS implementation on Nvidia A800 GPUs under general random workloads. It also achieves a maximum computational utilization exceeding 90% under benchmark workloads. Moreover, the proposed HGEMM not only significantly outperforms the CATLASS template-based implementation but also delivers efficiency comparable to the cuBLAS implementation in OPT-based bandwidth-limited LLM inference workloads. Full article

Background: Streptococcus pneumoniae is a leading cause of child mortality in Nepal despite the introduction of the 10-valent pneumococcal conjugate vaccine (PCV10). Vaccine effectiveness is threatened by the emergence of non-vaccine serotypes (NVTs) and the multiple serotypes carriage which often fail to be detected by traditional methods. We aimed to study changes in serotype distribution before and after PCV10 immunization among infants, including serotype dominance in Nepalese infants in the post-vaccine era. Methods: We enrolled infants in a longitudinal cohort study (2020–2022) conducted in Bhaktapur, Nepal. Nasopharyngeal swabs were collected before PCV10 dose 1 (6 weeks) and at 9 and 12 months post-immunization. We used a sensitive nanofluidic qPCR platform to detect multiple serotypes and establish their hierarchy by quantifying the bacterial load of each strain. Inverse Probability Weighting (IPW) adjusted risk factor analysis was used to account for loss to follow-up. Results: PCV10 successfully reduced vaccine-type (VT) carriage, declining sharply from 32.8% at 6 weeks to 4.8% at 12 months. VTs were pushed from being the dominant strain to occupying subdominant roles in co-colonization. Conversely, NVTs rapidly filled the vacated niche, showing a significant increase in their dominant status (p < 0.001). The most common replacing NVTs that rose to dominance were 35B, 19A, 6C/6D, and 15B/15C. Significant risk factors for carriage included older infancy (aOR 3.4, 95%CI: 2.6–4.5 at 9 months), a household kitchen in the living area (aOR 1.4, 95%CI: 1.0–1.9), and winter (aOR 1.7, 95%CI: 1.5–2.7) and pre-monsoon seasons (aOR 2.0, 95%CI: 1.5–2.8). Conclusions: While PCV10 reduced overall VT circulation, the persistence of VTs in subdominant niches creates a continuous reservoir for potential re-emergence and antibiotic resistance. This clear hierarchical shift in dominance towards NVTs underscores the urgent need for a public health strategy that includes the adoption of a higher-valent PCV to provide broader protection, and interventions targeting environmental risk factors are essential to sustain long-term reductions in pneumococcal colonization. Full article

Antimicrobial resistance (AMR) represents a significant global concern, undermining the efficacy of treatments in both human and veterinary medicine. Livestock production plays a major role in the emergence and dissemination of AMR, primarily due to the extensive use of antibiotics for therapeutic, prophylactic, and metaphylactic purposes. Addressing this multifaceted issue necessitates a One Health approach. At the international level, regulatory frameworks are predominantly non-binding, relying on soft-law instruments developed by the World Health Organization (WHO), the Food and Agriculture Organization (FAO), and the World Organization for Animal Health (WOAH, formerly OIE), which advocate for harmonized guidelines and national action plans. In contrast, the European Union has implemented binding regulations, including Regulation (EU) 2019/6 and Regulation (EU) 2019/4, which restrict non-essential antimicrobial use (AMU) and reinforce veterinary accountability. Initiatives such as the Farm to Fork Strategy and platforms like ClassyFarm further advance antimicrobial stewardship by integrating animal welfare, sustainability, and access to EU funding. Achieving substantial reductions in AMR within livestock systems requires coordinated, cross-disciplinary, and multi-level governance efforts. The EU model illustrates how enforceable legal frameworks, combined with science-based monitoring and welfare incentives, can facilitate prudent antibiotic use and promote sustainable animal production. This review aims to provide an integrated overview of international and European strategies for regulating antibiotic use in food-producing animals, focusing on how scientific, veterinary and legal perspectives contribute to combating AMR and promoting animal welfare by emphasizing prevention, and a prudent and responsible AMU. Full article

The growing integration of immersive technologies in the tourism sector raises questions about their real impact on the visitor experience. This study investigates whether mixed reality effectively influences the tourism experience, seeking to understand the mechanisms through which tourism content and emerging technologies shape tourists’ perceptions. A quantitative approach was adopted through the application of a questionnaire. The data were analyzed using the PLS-SEM (Partial Least Squares Structural Equation Modeling) method, allowing us to test direct and indirect relationships between the constructs Tourism Content, Adoption of Mixed Reality, and Tourist Experience. The analysis revealed positive and statistically significant direct effects. Tourism content strongly influences the adoption of mixed reality (β = 0.725; p < 0.001). Moderate impacts of the adoption of mixed reality (β = 0.375; p < 0.001) and tourism content (β = 0.392; p = 0.001) on the tourist experience were found. The indirect effect mediated by the adoption of mixed reality proved to be significant (β = 0.272; p = 0.001), with a VAF (Variance Accounted For) of 41%. Mixed reality plays a complementary partial mediating role in the relationship between tourism content and visitor experience, confirming its relevance in the contemporary tourism experience. Full article

Background: Insulin resistance (IR) is a key metabolic abnormality underlying type 2 diabetes and cardiometabolic diseases. Although lifestyle and sociodemographic determinants are well described, the role of psychosocial constructs—such as purpose in life—remains insufficiently characterized. No prior study in large occupational samples has examined the associations between purpose in life and IR when evaluated through three complementary indices: the triglyceride–glucose index (TyG), the Single-Point Insulin Sensitivity Estimator for Insulin Resistance (SPISE-IR), and the metabolic score for insulin resistance (METS-IR). Objectives: To analyze the cross-sectional associations between purpose in life and IR indicators in a large working population and determine whether these associations persist after accounting for sociodemographic and lifestyle factors. Methods: A cross-sectional study was conducted among 93,077 Spanish workers aged 20–69 years undergoing routine occupational health examinations. IR was estimated using TyG, SPISE-IR, and METS-IR indices. Purpose in life was assessed using the 10-item Purpose in Life Test and categorized into three groups based on the empirical distribution of scores. Multinomial logistic regression models adjusted for age, sex, social class, smoking, Mediterranean diet adherence, physical activity, and BMI were used to examine associations. Results: Lower purpose in life was consistently associated with higher IR categories across all indices. Compared with individuals reporting high purpose, those with low purpose had higher odds of belonging to the high IR category (TyG ORa 1.59; 95% CI 1.45–1.74; SPISE-IR ORa 1.94; 95% CI 1.76–2.13; METS-IR ORa 2.21; 95% CI 1.98–2.47). Adding purpose in life to sociodemographic and lifestyle models modestly improved discrimination for identifying high IR categories. Conclusions: In this large occupational cohort, purpose in life was independently associated with insulin resistance as measured by three metabolic indices. These findings highlight the relevance of psychosocial factors in metabolic health. Longitudinal studies are needed to clarify temporal pathways and assess whether purpose-oriented approaches may contribute to improved metabolic profiles. Full article

This study evaluates the early impact of Central Bank Digital Currencies (CBDCs) on key financial indicators in The Bahamas, focusing on the introduction of the SandDollar—the world’s first fully implemented retail CBDC. Using the Synthetic Control Method (SCM), the analysis constructs counterfactual scenarios to assess the effects of CBDCs on three dependent variables: outstanding loans from commercial banks as a percentage of GDP, outstanding deposits as a percentage of GDP, and the number of deposit accounts per 1000 adults. Three separate SCM models were estimated for the period 2014–2024, incorporating a broad set of control variables reflecting financial infrastructure, economic performance, demographic characteristics, and digital readiness. The findings consistently show that the SandDollar’s implementation is associated with reductions in loan issuance, deposit levels, and deposit account ownership compared to their synthetic counterparts. These results support the hypothesis that direct CBDC models may amplify “deposit substitution” and increase liquidity risks by shifting financial activity away from commercial banks. Although the SCM provides a structured causal framework, the short post-treatment period and potential pandemic-related disruptions limit the scope of a long-term understanding. The study underscores the importance of careful CBDC design, particularly the role of intermediated models in mitigating unintended financial stability risks. Full article

A growing body of literature recognises media narratives’ influence in shaping public perceptions of leadership and governance. Studies suggest that women presidential aspirants are often framed within symbolic constraints, where they are perceived as capable leaders in supportive roles but not as legitimate rulers. This study systematically reviews news coverage of Malawi’s first female president, Joyce Banda, examining how the media differentiates women’s ability to “lead” and their perceived inability to “rule”. Specifically, the study seeks to answer two key questions: How does the media in Malawi frame women’s political leadership in terms of governance and executive power? And what recurring gendered narratives emerge in media portrayals of women seeking the presidency? This investigation employs a content analysis of Malawi24, utilising Framing Theory and Feminist Theory to examine the dominant themes in political reporting. Analysis showed that media coverage reinforces a symbolic barrier to power, portraying women as leaders within limits while positioning men as natural rulers. Various perspectives on women’s legitimacy in executive positions were expressed, with narratives frequently questioning their authority and decision-making capabilities. The findings of this study suggest that gendered media framing constrains women’s political ambitions by reinforcing patriarchal expectations of leadership. Addressing this bias requires greater media accountability and equitable portrayals of women in executive political roles. Full article

This paper explores the potential of patristic theology to offer therapeutic resources for evangelical Christians who experience a crisis of hope: affirming hope doctrinally while struggling to feel hopeful in daily life. Drawing on recent psychological research, we use a tripartite model of hope—cognitive, agentic, and affective—to describe how hopeful experience can be sustained or undermined. We suggest that some theological frameworks, shaped by individualistic and goal-oriented assumptions, can unintentionally constrict believers’ capacity to experience hope. In dialogue with this psychological model, we read Gregory of Nyssa as a resource for each dimension of hope: his account of epektasis reframes the content of hope; his expansive understanding of divine agency widens the horizon of hope; and his use of imagery supports the affective experience of hope. The paper illustrates how patristic ressourcement can enrich theological imagination and can play a role in renewing believers’ capacity to hope. Full article

This study provides a multi-faceted and detailed perspective on the relationships between logistics performance, environmental degradation, and economic growth in 38 OECD countries, using each as an individual target variable. In the Analysis section, the relationship between logistics and environment is examined within a broader context, taking economic indicators into account. This examination utilizes the machine learning algorithms Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM). For each algorithm, the dataset is split into training and testing sets using three different ratios: 90:10, 80:20, and 70:30. A comprehensive performance evaluation is conducted on each of these splits by applying 5-fold and 10-fold cross-validation (CV). Considering economic indicators, the analysis section examines how the logistics-environment relationship is shaped in a broader context using the machine learning algorithms RF, XGBoost, and LightGBM. MSE, MAE, RMSE, MAPE, and R² metrics are utilized to evaluate model performance, while MDA and SHAP are employed to assess feature importance. Furthermore, a bee swarm plot is leveraged for visualizing the results. The XGBoost algorithm can successfully predict carbon dioxide (CO₂) emissions from transport and economic growth with high accuracy. However, the logistics performance model achieves high performance only with the LightGBM algorithm using a 90% train, 10% test split, and 5-fold CV setup. Based on the variable importance levels of the best-performing algorithm for each of the three target variables separately, the prediction of logistics performance is largely dependent on the economic growth predictor, and secondly, on the trade openness predictor. In predicting CO₂ emissions from transport, economic growth is identified as the most effective predictor, while logistics performance and trade openness contribute the least to the prediction. The findings also reveal that transport-related emissions and environmental indicators are prominent in the prediction of economic growth, whereas logistics performance and trade openness play a supportive, yet secondary role. Full article

Bounding box regression (BBR) loss plays a critical role in object detection within computer vision. Existing BBR loss functions are typically based on the Intersection over Union (IoU) between predicted and ground truth boxes. However, these methods neither account for the effect of predicted box scale on regression nor effectively address the drift problem inherent in BBR. To overcome these limitations, this paper introduces a novel BBR loss function, termed Gaussian Adaptive Ochiai BBR loss (GAOC), which combines the Ochiai Coefficient (OC) with a Gaussian Adaptive (GA) distribution. The OC component normalizes by the square root of the product of bounding box dimensions, ensuring scale invariance. Meanwhile, the GA distribution models the distance between the top-left and bottom-right corners (TL/BR) coordinates of predicted and ground truth boxes, enabling a similarity measure that reduces sensitivity to positional deviations. This design enhances detection robustness and accuracy. GAOC was integrated into YOLOv5 and RT-DETR and evaluated on the PASCAL VOC and MS COCO 2017 benchmarks. Experimental results demonstrate that GAOC consistently outperforms existing BBR loss functions, offering a more effective solution. Full article

Objective: To explore the impact of physical exercise on depression among college students and the chain mediation effect of psychological resilience and sleep problems and to verify the effectiveness of exercise science in improving psychological state. Methods: A survey was conducted among 3589 college students nationwide, using the Physical Activity Rating Scale, the Simplified Version of the Psychological Resilience Scale, the Pittsburgh Sleep Quality Index, and the Simplified Chinese Version of the Self-Rating Depression-Anxiety-Stress Scale (depression dimension). Correlation analysis was performed using SPSS 27.0, and the chain mediation effect was tested using the PROCESS 4.3 plugin. Results: (1) Physical exercise can significantly negatively predict depression, and the direct predictive effect of physical exercise on college students’ depressive state is significant; (2) Physical exercise can significantly positively predict psychological resilience and negatively predict sleep problems; psychological resilience significantly negatively predicts sleep problems and depression; sleep problems can significantly positively predict depression; (3) Psychological resilience and sleep problems play a significant mediating role between physical exercise and college students’ depression. Among the three mediating paths, physical exercise → psychological resilience → depression (path 1), physical exercise → sleep problems → depression (path 2), and physical exercise → psychological resilience → sleep problems → depression (path 3) account for 48.67%, 14.09%, and 6.46% of the total effect, respectively. Conclusions: Physical exercise is significantly negatively correlated with college students’ depression. Physical exercise not only exerts a simple mediating effect on depression through psychological resilience and sleep problems but also influences college students’ depression through the chain mediation effect of psychological resilience and sleep problems. Full article

Weeds are a major contributor to crop yield losses, accounting for up to 30% in conventional systems and exceeding 50% in organic farming. This study evaluated the effectiveness of intercropping as an agroecological strategy for weed suppression in winter crops, using advanced monitoring tools including Unmanned Aerial Vehicles (UAVs). Results showed that the wheat–field pea intercrop achieved superior weed suppression compared to both the wheat–faba bean intercrop and the wheat monoculture. These findings suggest that crop species selection, intercropping strategies, and optimized sowing densities can play a critical role in reducing weed pressure and enhancing the sustainability of cropping systems. Full article

Malformations of cortical development (MCD) are a group of neurodevelopmental disorders caused by abnormalities in cerebral cortex development, leading to conditions such as intellectual disability and refractory epilepsy. The prenatal phenotypes of MCD are complex and non-specific, complicating accurate diagnosis and prognosis assessment. Genetic testing, particularly chromosomal microarray analysis (CMA) and whole-exome sequencing (WES), has become an important tool for prenatal diagnosis. This review synthesizes current research on prenatal MCD, focusing on the integration of imaging and genetic diagnostic strategies based on the biological foundation of cortical development and the classification system of MCD. Prenatal MCD phenotypes show significant developmental stage clustering, with proliferation-phase abnormalities (62.9%) being the most common and microcephaly as the core phenotype. Genetic studies have revealed a high degree of genetic heterogeneity in MCD, with etiologies encompassing chromosomal abnormalities and a wide range of single-gene mutations. These mutations are clustered by phenotype: microcephaly is associated with neuronal proliferation/DNA repair genes; macrocephaly is driven by genes in the PI3K-AKT-mTOR and RAS-MAPK signaling pathways; and gyral and sulcal abnormalities are closely linked to microtubule-associated genes and migration pathways. De novo mutations account for the majority of pathogenic genetic alterations identified in MCD (50.6%); up to 75.1% of pathogenic mutations cannot be detected by routine prenatal screening. Based on this, the review emphasizes that for fetuses with suspected MCD, NGS, with WES at its core, plays an increasingly important role in achieving early and accurate prenatal diagnosis. Future research should prioritize the advancement of integrated diagnostic methods and large-scale cohort studies to further elucidate genotype–phenotype associations. Full article

Background and Objectives: Acute pancreatitis (AP) has a wide range of clinical severity, and early prediction of disease progression is still challenging. Trypsinogen-activating peptide (TAP) and trypsin-2 serve as direct biomarkers for intrapancreatic proteolytic activation and may provide earlier pathophysiological information compared with traditional markers. Materials and Methods: In this retrospective cohort analysis involving 54 AP patients, we evaluated 24 h serum and urinary TAP and trypsin-2 concentrations by Bayesian correlation, mediation analysis, unsupervised K-means clustering, and supervised machine learning (Elastic Net and Random Forest). The analyses investigated the relationships of biomarkers with inflammation (CRP), enzymatic activities (amylase, lipase), and clinical factors, as well as inflammation severity (CRP levels). Results: Bayesian correlations indicated moderate evidence for a relationship between serum TAP and CRP (BF₁₀ = 8.42), as well as strong evidence linking age to serum TAP (BF₁₀ = 12.75). Serum trypsin-2 showed no correlation with CRP, while urinary trypsin-2 had a correlation with amylase (BF₁₀ = 6.89). Mediation analysis indicated that TAP and trypsin-2 accounted for 42–44% of the impact of CRP on pancreatic enzyme elevation. Clustering revealed three phenotypic subgroups (“Mild Activation”, “Moderate System”, and “Severe Pancreatic-Renal”), the latter showing the highest levels of CRP and biomarkers. Machine learning models highlighted urinary trypsin-2 and age as the most significant predictors of inflammation, with Random Forest achieving the highest performance (R² = 0.53). Conclusions: Early urinary trypsin-2 outperforms serum markers as a predictor of systemic inflammatory intensity, indicating total proteolytic impairment and renal clearance. This integrative analysis reveals unique biological phenotypes and highlights the potential of these biomarkers for early assessment of the inflammatory burden. Their role in predicting clinical disease progression requires prospective validation. Integrative biomarker analysis reveals unique biological phenotypes and improves assessment of inflammatory burden in PA. Larger cohorts are required for prospective validation to incorporate these biomarkers into precision-based diagnostic frameworks. Full article

Information cascades refer to a type of learning behavior in social networks where individuals make decisions by observing the actions of others, rather than relying solely on their own private information. Network centrality, which measures the relative importance or influence of a node within a network, plays a significant role in initiating and shaping information cascades across four key dimensions. First, nodes with high degree centrality often initiate information cascades due to their large number of direct connections to other nodes. Second, nodes with high betweenness centrality serve as bridges between different parts of the network, thereby controlling the flow of cascading information. Third, nodes with high closeness centrality can access and disseminate information more quickly, accelerating the spread of cascades throughout the network. Fourth, nodes with high eigenvector centrality augment the impact of information cascades through their visibility and connections to other influential nodes. Synthesizing research findings on executive and director networks from management, finance, and accounting, this entry provides insights into emerging trends in corporate governance by highlighting the interaction between network structure and information dissemination. Full article