Search Results (3,510)

Large-scale medical visual question answering (MedVQA) datasets are critical for training and deploying vision–language models (VLMs) in radiology. Ideally, such datasets should be automatically constructed from routine radiology reports and their corresponding images. However, no existing method directly links free-text findings to the most relevant 2D slices in volumetric computed tomography (CT) scans. To address this gap, a contrastive language–image pre-training (CLIP)-based key slice selection framework is proposed, which matches each sentence to its most informative CT slice via text–image similarity. This experiment demonstrates that models pre-trained in the medical domain already achieve competitive slice retrieval accuracy and that fine-tuning them on a small dual-supervised dataset that imparts both lesion- and organ-level awareness yields further gains. In particular, the best-performing model (fine-tuned BiomedCLIP) achieved a Top-1 accuracy of 51.7% for lesion-aware slice retrieval, representing a 20-point improvement over baseline CLIP, and was accepted by radiologists in 56.3% of cases. By automating the report-to-slice alignment, the proposed method facilitates scalable, clinically realistic construction of MedVQA resources. Full article

This study developed a rapid, non-destructive method for the quantitative detection of protein in cottonseed by integrating near-infrared (NIR) fiber spectroscopy with chemometric machine learning. The establishment of this method holds significant importance for the rational and efficient utilization of cottonseed resources, advancing research on the genetic improvement of cottonseed nutritional quality, and promoting the development of equipment for raw cottonseed protein detection. Fuzzy cottonseed samples from three varieties were collected, and their NIR fiber-optic spectra were acquired. Reference protein contents were measured using the Kjeldahl method. Spectra were denoised through preprocessing, after which informative wavelengths were selected by combining Uninformative Variable Elimination (UVE) with Competitive Adaptive Reweighted Sampling (CARS) and the Random Frog (RF) algorithm. Partial least squares regression (PLSR), least-squares support vector machine (LSSVM), and support vector regression (SVR) models were then constructed to predict protein content. Model performance was assessed using the coefficient of determination (R²), root-mean-square error (RMSE), residual predictive deviation (RPD), and range error ratio (RER). The results indicate that the standard normal variate (SNV) is the most effective preprocessing step. The best performance was achieved by the LSSVM model coupled with UVE + CARS, yielding R² = 0.8571, RMSE = 0.0033, RPD = 2.7078, and RER = 10.72, outperforming the PLSR and SVR counterparts. These findings provide technical support for the rapid detection of fuzzy cottonseed protein and lay the groundwork for the development of related detection equipment. Full article

While international cross-listing locations in host countries have been identified as integral to firm valuation gains, the influence of the home country information environment on firm financial market integration remains underexplored. This study examined how political stability, government effectiveness, and voice and accountability influence cross-listing destination choices amongst emerging-market firms seeking enhanced valuation gains. Using data on cross-listed firms from BRICS countries between 2000 and 2020, the study employed generalized linear models (GLMs), including probit and robit specifications, to analyze this relationship. The researchers found that stronger political stability; government effectiveness; and voice and accountability in home countries significantly increase the likelihood of BRICS firms cross-listing on advanced exchanges characterized by higher valuation gains. These results indicate that reduced political risk, effective government policy implementation and greater media freedom in BRICS emerging market countries facilitate cross-listing firms’ access to more efficient global capital markets by reducing asymmetric information, and help overcome traditional market segmentation barriers. Contrary to the conventional emphasis that home country proximity is significant for cross-listing valuation gains, these results highlight the signaling mechanism of home country governance quality as an appealing factor for firm cross-listing location in advanced exchange markets. Policymakers in emerging markets should consider governance reforms that enhance domestic firm competitiveness in global financial markets for higher valuation gains. Full article

Introduction: This study explores the impact of logistics information technology (LIT) on supply chain relationships and distribution performance in small and medium-sized enterprises (SMEs) using South Africa as a case study. Although digital supply chain solutions are increasingly important, there is limited evidence of SME efficiency in emerging markets using LIT. Methods: This study utilises a survey of 313 SMEs from four South African provinces. Bayesian structural equation modelling (Bayesian SEM) was then used to examine LIT’s effects on distribution performances in terms of timeliness, product availability, and condition. Results: The results show that the adoption of LIT strengthens buyer–seller networks (β = 0.524, CI = [0.434, 0.613]) and improves distribution by enhancing both timeliness performance (β = 0.237, CI = [0.098, 0.372]) and product condition performance (β = 0.175, CI = [0.042, 0.259], β = 0.222, p < 0.001). However, it does not directly enhance product availability performance (β = 0.085, CI = [−0.030, 0.199]), signifying that LIT adoption by itself fails to improve product availability. The results also demonstrate that SME network relationships mediate the connection between LIT adoption and distribution performance metrics. Discussion: This study’s findings contribute to the literature and offer valuable information and guidance to policymakers as they underscore the importance for SMEs to invest in LIT integration and compatibility, as well as inventory optimisation and improved supplier communication to minimise transit time variation. Policymakers should support SMEs’ digital transformation through interventions including funding and training for LIT adoption. This study confirms the essential role of LIT in SME supply chains and illustrates that technology-facilitated relationships enhance distribution performance, which enhances SME competitiveness. Full article

Performance management is critical for aligning human capital with organizational strategy, particularly in the increasingly competitive service sector. However, universally effective performance appraisal systems (PASs) exist, as effectiveness depends on contextual and organizational specificities. In Portugal, where services account for nearly three-quarters of gross value added, PAS implementation remains underdeveloped, highlighting a gap between strategic intent and practice. This study aims to address that gap by investigating how a performance appraisal model can be tailored to the service sector. A case study was conducted at PCI—Creative Science Park, S.A., a consulting firm, using a qualitative approach. The research design combined a literature review to identify theoretical dimensions of performance evaluation with an employee questionnaire to capture organizational perceptions and priorities. Integration of both strands of evidence informed the construction of the framework. The findings indicate that employees value objective-based evaluation as the most relevant dimension, complemented by customer feedback, adaptive performance, and organizational citizenship. Furthermore, the integration of 360° feedback mechanisms and regular review cycles emerged as key enablers of fairness and engagement. By combining theoretical insights with employee perspectives, this study contributes to a customized and flexible PAS that enhances strategic alignment in the service sector. The proposed model provides both scholarly value, by advancing the discussion on context-specific PAS design, and practical value, by offering a reference for organizations seeking to align human performance with mission-critical outcomes. Full article

Digital innovation is the core driving force to enhance the competitiveness of enterprises and promote sustainable development, and is a key enabler for achieving corporate ability goals. Executives with information technology (IT) backgrounds who have rich knowledge and skills in digital technology are the backbone of promoting the digital transformation of enterprises and optimizing the allocation of auditing resources. And they can lay the technological foundation for sustainable corporate development and play an important role in corporate audit decision-making. Based on the data of China’s A-share listed companies from 2015 to 2023, the impact of executives with IT backgrounds on auditor selection is empirically analyzed. The study shows that (1) the higher the proportion of executives with IT backgrounds in the executive team, the more the companies tend to choose high-quality auditors; (2) the degree of corporate digital innovation positively moderates the relationship between executives with an IT background and high-quality auditors; (3) the level of corporate internal control plays a mediating effect in the relationship between executives with an IT background and auditor selection; (4) for non-state-owned, large-scale, short executive tenures, and labor-intensive firms, executives with IT backgrounds exert a more significant influence on auditor selection. This study broadens previous research on corporate auditing behaviors from the perspective of executives with IT backgrounds, providing insights for companies to select suitable auditors, to make scientifically sound decisions regarding auditor selection in the context of digital innovation, further optimize internal management, enhance risk response capabilities, and thereby achieve sustainable corporate development. Full article

Background: Due to the wide host range of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), vaccination has been recommended for susceptible species in zoological collections, particularly to protect endangered species. The Zoetis^® Experimental Mink Coronavirus Vaccine (Subunit) was temporarily authorized in 2021–2024 for emergency use in North America for this purpose. However, there are limited data regarding its safety or efficacy in non-domestic mammals. The present study was conducted to assess the ability of this vaccine to elicit serum neutralizing titers against SARS-CoV-2 in selected animals from the Toronto Zoo (TZ) vaccinated during 2022. Methods: Serum samples were collected from 24 individuals across four families (Cervidae, Felidae, Ursidae, and Hyaenidae) and tested using a surrogate virus neutralization test (sVNT) and a plaque-reduction neutralization test (PRNT). Results: The results showed that all species developed some neutralizing titers after at least one vaccine dose, except for polar bears, which showed no seroconversion. Felids and hyenas had the highest neutralizing titers, which peaked at 3 and declined between 4 and 6 months after boost. These differences may stem from species-specific immune responses or lack of vaccination protocols tailored to individual species. Conclusions: While natural infection with SARS-CoV-2 could not be ruled out in the cohort of this study, insights from our results have the potential to inform future vaccine recommendations for non-domestic species. Furthermore, our study highlighted the value of competitive assays in assessing serological responses across a broad range of exotic species, for which reagents, such as anti-isotype antibodies, are often unavailable. Full article

Based on Agenda-Setting Theory, Media Agenda Synergy (MAS) can enhance the communication effectiveness of public issues (e.g., climate change, social justice, and public health) through the information resonance and agenda complementarity among cross-media platforms, thus reconstructing the public perception. In this paper, we focus on the dynamic impact of cross-media agenda synergy on public agenda intensity and innovatively propose a “HMM-Granger” hybrid modeling framework for Media Agenda Synergy: Firstly, we quantify the causal weights of agenda shifting based on the deconstruction of the nonlinear time-series dependence of multisource media data by using LSTM neural networks. Secondly, the state transfer probability matrix of the Hidden Markov Model reveals the dual paths of “explicit collaboration” (e.g., issue resonance) and “implicit competition” (e.g., agenda masking) in media agenda coordination. The results of this study show that the Agenda Synergy between mainstream media and social media during major events can generate an Agenda Multiplier Effect, resulting in a significant increase in the intensity of the public agenda. This study provides a computable theoretical paradigm for Inter-Media Agenda Network modeling and data-driven decision support for optimizing opinion guidance strategies. Full article

Using a panel of 3515 Chinese listed firms from 2011 to 2023, this study shows that the education level of the top management team (TMT) positively influences firm stock liquidity. The beneficial effect of TMT education on stock liquidity is stronger in settings with lower industry competition, higher information disclosure quality, and bull market periods. Mediation analysis indicates that analyst coverage provides a weak channel through which TMT education affects stock liquidity. Endogeneity concerns are alleviated by reverse causality tests, two-stage least squares regressions, propensity score matching, and generalized method of moments. The results are also robust to alternative liquidity measures and alternative definitions of TMT education. This study offers practical implications for investors, corporate executives, and policymakers seeking to promote market efficiency and liquidity. Full article

Objectives: To systematically review and analyze electronic systems in competitive motorcycles (2020–2025), examining their technical specifications, performance impacts, and technological evolution across MotoGP, World Superbike (WSBK), MotoE, British Superbike (BSB), and Spanish Championship (ESBK) categories. Eligibility criteria: Included studies reporting technical specifications or performance data of electronic systems in professional motorcycle racing, published between January 2020 and December 2025 in English, Spanish, Italian, or Japanese. Excluded: opinion pieces, amateur racing, and studies without quantitative data. Information sources: IEEE Xplore, SAE Technical Papers, Web of Science, Scopus, and specialized motorsport databases were searched through 15 December 2025. Risk of bias: Modified Cochrane Risk of Bias tool for experimental studies and Newcastle-Ottawa Scale for observational studies. Synthesis of results: Synthesis of results: Random-effects meta-analysis using DerSimonian-Laird method for homogeneous outcomes; narrative synthesis for heterogeneous data. Included studies: 87 studies met inclusion criteria (52 experimental, 38 simulation, 23 technical descriptions, 14 comparative analyses). Electronic systems were categorized into six domains: Engine Control Units (ECU, 28 studies, 22%), Vehicle Dynamics (23 studies, 18%), Traction Control (19 studies, 15%), Data Acquisition (21 studies, 17%), Braking Systems (18 studies, 14%), and Emerging Technologies (18 studies, 14%). Note that studies could address multiple domains. Limitations of evidence: Proprietary restrictions limited access to 31% of technical details; 43% lacked cross-category comparisons. Interpretation: Electronic systems are primary performance differentiators, with computational power following Moore’s Law. Future developments point toward distributed architectures and 5G telemetry. Full article

Artificial intelligence has become crucial in sports, leveraging its analytical capabilities to enhance the understanding and prediction of complex events. Machine learning algorithms in sports, especially basketball, are transforming performance analysis by identifying patterns and trends invisible to traditional methods. This technology provides in-depth insights into individual and team performance, enabling precise evaluation of strategies and tactics. Consequently, the detailed analysis of every aspect of a team’s routine can significantly elevate the level of competition in the sport. This study investigates a range of machine learning models, including Logistic Regression (LR), Ridge Regression Classifier (RR), Random Forest (RF), Naive Bayes (NB), K-Nearest Neighbors (KNNs), Support Vector Machine (SVM), Stacking Classifier (STACK), Bagging Classifier (BAG), Multi-Layer Perceptron (MLP), AdaBoost (AB), and XGBoost (XGB), as well as deep learning architectures such as Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNNs), to compare their effectiveness in predicting game outcomes in the NBA and WNBA leagues. The results show highly acceptable prediction accuracies of 65.50% for the NBA and 67.48% for the WNBA. This study allows us to understand the impact that artificial intelligence can have on the world of basketball and its current state in relation to previous studies. It can provide valuable insights for coaches, performance analysts, team managers, and sports strategists by using machine learning and deep learning models to predict NBA and WNBA outcomes, enabling informed decisions and enhancing competitive performance. Full article

Amid intensifying global competition, industrial product quality has become a critical determinant of competitive advantage. However, persistent quality-related faults in production environments threaten product integrity. To address this challenge, a Fusion Maximal Information Coefficient-based Quality-Related Kernel Component Analysis (FMIC-QRKCA) methodology is proposed in this paper by capitalizing on information fusion principles and statistical metric theory. Based on information fusion principles, a Fusion Maximal Information Coefficient (FMIC) strategy is first studied to quantify correlations between process variables and multivariate quality indicators. Subsequently, by integrating the proposed FMIC method with Kernel Principal Component Analysis (KPCA), a Quality-Related Kernel Component Analysis (QRKCA) method is proposed. In the proposed QRKCA strategy, the complete latent variable space is first obtained; on this basis, FMIC is further applied to quantify the correlation between each latent variable and quality variables, thereby completing the screening of quality-related latent variables. Additionally, the $T^2$ and squared prediction error monitoring statistics are used as the key indices to determine the occurrence of faults. This integration overcomes the limitation of conventional KPCA, which does not explicitly consider quality indicators during the principal component extraction, thereby enabling precise isolation of quality-related fault features. Validation through the numerical case and the industrial process case demonstrates that FMIC-QRKCA significantly outperforms established methods in detection accuracy for quality-related faults. Full article

Greenwashing threatens both consumer trust and the integrity of planetary health initiatives. Transparency in sustainability claims is therefore critical for promoting ecological wellbeing, strengthening food security, and fostering equitable development in the Anthropocene. This paper investigates greenwashing by adapting the Gompers Governance Index methodology to the context of sustainability claims. The focus of our greenwashing index in this case is the sustainability claims made by canned tuna brands in Australia. The index is created from a comprehensive set of criteria for environmental claims, based on the Australian Competition and Consumer Commission (ACCC)’s principles for trustworthy claims. We show that the canned tuna brands form two clusters: one at a very high level of achievement and a second group with notable opportunities to improve on their sustainability communication and transparency. The results also highlight several key issues, most notably a lack of information regarding future sustainability transition plans across most brands. A deeper analysis of the scoring scheme shows that the brands with third-party sustainability certification generally achieved a better alignment with the ACCC principles than other brands. Future iterations of this analysis could incorporate online transparency and third-party verification to provide a more comprehensive assessment. Overall, this study underscores the need for clearer sustainability messaging, greater regulatory enforcement, and improved accountability among brands to ensure consumers can make informed choices. Full article

Cannabidiol (CBD) is a chemical produced by the cannabis plant that acts as an allosteric modulator of cannabinoid receptors resulting in non-competitive receptor antagonism in the central nervous system. This mechanism of action leads to anti-convulsant, anti-anxiety, and analgesic properties with minimal psycho-activity, which has led to significant interest in the use of CBD as a medication. Legislation around cannabis has changed in recent years, with many states permitting the use of CBD-based products as “medication” without approval from the Federal Drug Administration. This has led to a proliferation of products with associated marketing claims that are often unsubstantiated. This review summarizes the evidence for cannabidiol as a medical treatment, focusing on epilepsy, mental health, behavioral and substance use disorders occurring in pediatric and adolescent populations for which information is available. CBD preparations have been approved by the FDA to treat epilepsy in childhood; no other indications currently exist, and the literature remains inconclusive. Few adverse effects related to CBD use have been reported. However, endogenous cannabinoids play an important role in guiding brain development, and the long-term impact of modulating the endocannabinoid system during periods of brain growth during childhood and adolescence is unknown. While there is excitement about the potential for the development of CBD medications, currently, there is very limited information about the long-term safety of CBD, especially in children and adolescents, and caution is recommended regarding the use of unregulated, unapproved CBD preparations that are currently available over the counter. Full article

The Capacitated Vehicle Routing Problem (CVRP) is a central challenge in combinatorial optimization, with critical applications in logistics and transportation. Traditional methods struggle with large-scale instances, due to the computational demands, while learned construction models often suffer from degraded solution quality and constraint violations. This work proposes SPORL, a Scalable Partitioning and Optimization via Reinforcement Learning framework for large-scale CVRPs. SPORL decomposes the problem using a learned partitioning strategy, followed by parallel subproblem solving, and employs a greedy decoding scheme at inference to ensure scalability for instances with up to 1000 customers. A key innovation is a context-based attention mechanism that incorporates sub-route embeddings, enabling more informed and constraint-aware partitioning decisions. Extensive experiments on benchmark datasets with up to 1000 customers demonstrated that SPORL consistently outperformed state-of-the-art learning-based baselines (e.g., AM, POMO) and achieved competitive performance relative to strong heuristics such as LKH3, while reducing inference time from hours to seconds. Ablation studies confirmed the critical role of the proposed context embedding and decoding strategy in achieving high solution quality. Full article