Search Results (485)

Generative artificial intelligence (GenAI) is advancing rapidly and is increasingly integrated into educational settings. How to effectively leverage GenAI to support instructional design has thus become a critical issue in teacher education. While existing studies have validated the technical potential and functional value of GenAI in instructional design, there remains a notable gap in qualitative investigations into pre-service teachers’ subjective willingness to adopt GenAI and its underlying influencing factors. To address this gap, this present study employed grounded theory to explore the factors that shape pre-service teachers’ intention to use GenAI for instructional design. Semi-structured interviews were conducted with 23 pre-service teachers from Shaanxi Normal University, and the data were analyzed through open coding, axial coding, and selective coding. A theoretical model comprising four major dimensions was developed as follows: (1) technical factors (relative advantage and ease of use), (2) environmental factors (social impact, opinion leader, and facilitating conditions), (3) usage characteristics (purpose of use and method of use), and (4) psychological factors (trust, perceived risk, and a professional self-concept). The findings reveal that pre-service teachers’ intention to use GenAI is not shaped by a single factor but is instead the result of dynamic and interrelated interactions among the four dimensions. This study extends current technology acceptance theories and offers practical insights for the effective integration and promotion of GenAI in instructional design. Full article

In Industry 4.0 environments, edge intelligence plays a critical role in enabling real-time analytics and autonomous decision-making by integrating artificial intelligence (AI) with edge computing. However, deploying deep neural networks (DNNs) on resource-constrained edge devices remains challenging due to limited computational capacity and strict latency requirements. While conventional methods primarily focus on structural model compression, we propose an adaptive input-centric approach that reduces computational overhead by pruning redundant features prior to inference. A Bayesian network is employed to quantify the influence of each input feature on the model output, enabling efficient input reduction without modifying the model architecture. A bidirectional chain structure facilitates robust feature ranking, and an automated algorithm optimizes input selection to meet predefined constraints on model accuracy and size. Experimental results demonstrate that the proposed method significantly reduces memory usage and computation cost while maintaining competitive performance, making it highly suitable for real-time edge intelligence in industrial settings. Full article

E-commerce indicators are very complex and have a wide range of levels of complexity and applications. The digital economies that we are oriented towards also have complex features in terms of consumers and businesses. The research objectives are focused on determining the impact of e-commerce on the digital economy within countries with different stages of economic development, digitalization techniques, and e-commerce usage. This study evaluates how AI-based clustering reveals patterns in the e-commerce indicators influencing the digital economy. The research methods used are focused on AI techniques in order to evaluate and assess the usage of e-commerce in the digital economy. In this sense, the methods used in this research are clustering techniques in order to determine the stage of implementation of the digital economy. The research implications have a worldwide impact and soundness in establishing the evolution of the e-economy in different types of countries with different stages and levels of digitalization and different e-commerce development paths. The empirical results show there are significant differences between countries due to cultural, economic, social, and judicial differences. The conclusions of this study highlight that using AI techniques can be a solution for enhancing future digital economy development and labor market consolidation, especially by strengthening e-commerce indicator usage and application. Full article

Growing interest in organic food, along with European regulations limiting chemical usage, and the declining effectiveness of herbicides due to weed resistance, are all contributing to the growing trend towards mechanical weeding. For mechanical weeding to be effective, tools must pass near the crops in both the inter- and intra-row areas. The use of AI-based computer vision can assist in detecting crop lines and accurately guiding weeding tools. Additionally, AI-driven image analysis can be used for selective intra-row weeding with mechanized blades, distinguishing crops from weeds. However, until now, there have been two separate systems for these tasks. To enable simultaneous in-row weeding and row alignment, YOLOv8n and YOLO11n were trained and compared in a lettuce field (Lactuca sativa L.). The models were evaluated based on different metrics and inference time for three different image sizes. Crop lines were generated through linear regression on the bounding box centers of detected plants and compared against manually drawn ground truth lines, generated during the annotation process, using different deviation metrics. As more than one line appeared per image, the proposed methodology for classifying points in their corresponding crop line was tested for three different approaches with different empirical factor values. The best-performing approach achieved a mean horizontal error of 45 pixels, demonstrating the feasibility of a dual-functioning system using a single vision model. Full article

Frequent disruptions to global supply chains, driven by factors such as trade restrictions and geopolitical conflicts, brought supply chain resilience to the forefront of both academic research and industry practice. Concurrently, the rapid advancement of artificial intelligence (AI) technologies in supply chain management in recent years offers new perspectives for researching resilience. Based on the Organizational Information Processing Theory (OIPT), this study explores the direct and indirect mechanisms through which AI usage impacts supply chain resilience from an information processing perspective. Within the OIPT framework, we develop a theoretical model incorporating AI usage, supply chain resilience, supply chain efficiency, supply chain collaboration, and digital information technology capability. We empirically test the model using survey data collected from 231 Chinese manufacturing senior executives and supply chain managers, employing partial least squares structural equation modeling (PLS-SEM). The findings reveal that AI usage has a significant direct positive effect on supply chain resilience. Additionally, supply chain efficiency and collaboration act as mediators in this relationship. Furthermore, we examined the moderating role of a firm’s digital information technology capability and found that it positively moderates the impact of AI usage on supply chain resilience. Full article

Background/Objectives: Artificial intelligence (AI) tools for fracture detection in radiographs are increasingly approved for clinical use but remain underutilized. Understanding physician attitudes before implementation is essential for successful integration into emergency care workflows. This study investigates the acceptance of an AI-based fracture detection tool among physicians in emergency care settings, using the Unified Theory of Acceptance and Use of Technology (UTAUT) model. Methods: A cross-sectional, pre-implementation survey was conducted among 92 physicians across three hospitals participating in the SMART Fracture Trial (ClinicalTrials.gov: NCT06754137). The questionnaire assessed the four core UTAUT constructs—performance expectancy (PE), effort expectancy (EE), social influence (SI), facilitating conditions (FC)—and additional constructs such as attitude toward technology (AT), diagnostic confidence (DC), and workflow efficiency (WE). Responses were collected on a five-point Likert scale. Structural equation modeling (SEM) and confirmatory factor analysis (CFA) were performed to assess predictors of behavioral intention (BI). Results: PE was the strongest predictor of BI (β = 0.5882, p < 0.001), followed by SI (β = 0.391, p < 0.001), FC (β = 0.263, p < 0.001), and EE (β = 0.202, p = 0.001). These constructs explained a substantial proportion of variance in BI. WE received the lowest ratings, while internal consistency for SI and BI was weak. Moderator analyses showed prior AI experience improved EE, whereas more experienced physicians were more skeptical regarding WE and DC. However, none of the moderators significantly influenced BI. Conclusions: Physicians’ intention to use AI fracture detection is primarily influenced by perceived usefulness and ease of use. Implementation strategies should focus on intuitive design, targeted training, and clear communication of clinical benefits. Further research should evaluate post-implementation usage and user satisfaction. Full article

This study investigates the potential of fine-tuned large language models (LLMs) to enhance geospatial intelligence by translating natural language queries into executable Python code. Traditional GIS workflows, while effective, often lack usability and scalability for non-technical users. LLMs offer a new approach by enabling conversational interaction with spatial data. We evaluate OpenAI’s GPT-4o-mini model in two forms: an “As-Is” baseline and a fine-tuned version trained on 600+ prompt–response pairs related to geospatial Python scripting in Virginia. Using U.S. Census shapefiles and hospital data, we tested both models across six types of spatial queries. The fine-tuned model achieved 89.7%, a 49.2 percentage point improvement over the baseline’s 40.5%. It also demonstrated substantial reductions in execution errors and token usage. Key innovations include the integration of spatial reasoning, modular external function calls, and fuzzy geographic input correction. These findings suggest that fine-tuned LLMs can improve the accuracy, efficiency, and usability of geospatial dashboards when they are powered by LLMs. Our results further imply a scalable and replicable approach for future domain-specific AI applications in geospatial science and smart cities studies. Full article

The accelerated development of e-commerce has given rise to sophisticated systems defined by significant user interaction, a variety of product offerings, and considerable quantities of structured and unstructured data. Upholding trust and operational security is becoming ever more essential. E-commerce platforms are susceptible to deceptive practices, including counterfeit reviews, dubious transactions, and anomalous usage behaviors. This research introduces a framework for anomaly detection powered by artificial intelligence, integrating deep learning and natural language processing (NLP) with the isolation forest algorithm tree to enhance the identification of unusual activities on e-commerce platforms. We leveraged customer feedback, transaction logs, and user interaction data obtained from Kaggle. Textual reviews were interpreted using natural language processing (NLP), while deep learning was utilized to discern behavioral patterns. The isolation forest algorithm tree was employed to detect statistical anomalies in multidimensional data. The hybrid model surpassed conventional techniques in terms of detection accuracy, recall, and interpretability. It successfully detects suspicious actions and clarifies anomalies in their relevant context. The application of AI techniques, particularly natural language processing, deep learning, and isolation forest algorithm trees, establishes a solid foundation for anomaly detection in the realm of e-commerce. This approach fosters a more secure and trustworthy experience for online consumers. Full article

Background/Objectives: Artificial intelligence (AI) has been utilised in urological conditions such as urolithiasis, urogynaecology and uro-oncology. The aim of this study is to examine the attitudes and beliefs about AI technology amongst urology healthcare providers. Methods: A structured online questionnaire, created from a modified Delphi method with a panel of urologists and urology surgical trainees, was delivered through the Urological Asia Association’s annual congress. The questionnaire, with 25 items of mixed type responses (five-point Likert scale, nominal-polytomous and open-ended), acquired data regarding demographics, perception and attitudes towards general usage of AI in urological care. Results: A total of 464 respondents from 47 different countries were collected. The results showed that 83.4% of participants believed AI will improve efficiency and 18.8% believed they are knowledgeable in AI technology, with ordinal logistic regression showing both urology specialists and trainees are more likely to agree to these responses. Overall, 51.5% believed AI adoption will not replace clinical practice, and regression analysis found those with previous AI training are more likely to agree to this response. We found AI is commonly used in research, patient education and administrative tasks and identified key enablers as regulatory approval, AI clinical effectiveness and access to AI training. Conclusions: Overall attitudes and beliefs towards the use of AI in urology is positive and encouraging. AI training and education and regulatory reform needs to be addressed to allow integration of AI into clinical practice. A limitation of the study lies in its generalisability to global settings due to the demographics of the respondents. Full article

As cities densify, deep underground infrastructure construction such as mass rapid transit (MRT) systems increasingly demand smarter, digitalized, and more sustainable approaches. RC diaphragm walls, essential to these systems, present challenges due to complex rebar configurations, spatial constraints, and high material usage and waste, factors that contribute significantly to carbon emissions. This study presents an AI-assisted rebar optimization framework to improve constructability and reduce waste in MRT-related diaphragm wall construction. The framework integrates the BIM concept with a custom greedy hybrid Python-based metaheuristic algorithm based on the WOA, enabling optimization through special-length rebar allocation and strategic coupler placement. Unlike conventional approaches reliant on stock-length rebars and lap splicing, this approach incorporates constructability constraints and reinforcement continuity into the optimization process. Applied to a high-density MRT project in Singapore, it demonstrated reductions of 19.76% in rebar usage, 84.57% in cutting waste, 17.4% in carbon emissions, and 14.57% in construction cost. By aligning digital intelligence with practical construction requirements, the proposed framework supports smart city goals through resource-efficient practices, construction innovation, and urban infrastructure decarbonization. Full article

In the context of increasing Artificial Intelligence integration in higher education, understanding the factors influencing university teachers’ adoption of AI tools is critical for effective implementation. This study adopts a perception–intention–behavior framework to explores the roles of perceived usefulness, perceived ease of use, perceived trust, perceived substitution crisis, and perceived risk in shaping teachers’ behavioral intention and actual usage of AI tools. It also investigates the moderating effects of peer influence and organizational support on these relationships. Using a comprehensive survey instrument, data was collected from 487 university teachers across four major regions in China. The results reveal that perceived usefulness and perceived ease of use are strong predictors of behavioral intention, with perceived ease of use also significantly influencing perceived usefulness. Perceived trust serves as a key mediator, enhancing the relationship between perceived usefulness, perceived ease of use, and behavioral intention. While perceived substitution crisis negatively influenced perceived trust, it showed no significant direct effect on behavioral intention, suggesting a complex relationship between job displacement concerns and AI adoption. In contrast, perceived risk was found to negatively impact behavioral intention, though it was mitigated by perceived ease of use. Peer influence significantly moderated the relationship between perceived trust and behavioral intention, highlighting the importance of peer influence in AI adoption, while organizational support amplified the effect of perceived ease of use on behavioral intention. These findings inform practical strategies such as co-developing user-centered AI tools, enhancing institutional trust through transparent governance, leveraging peer support, providing structured training and technical assistance, and advancing policy-level initiatives to guide digital transformation in universities. Full article

Background/Objectives: Accurate diagnosis of brain tumors is one of the most important challenges in neuro-oncology since tumor classification and volumetric segmentation inform treatment planning. Two-dimensional classification and three-dimensional segmentation deep learning models can augment radiological workflows, particularly if paired with explainable AI techniques to improve model interpretability. The objective of this research was to develop a web-based brain tumor segmentation and classification diagnosis platform. Methods: A diagnosis system was developed combining 2D tumor classification and 3D volumetric segmentation. Classification employed a fine-tuned MobileNetV2 model trained on a glioma, meningioma, pituitary tumor, and normal control dataset. Segmentation employed a SegResNet model trained on BraTS multi-channel MRI with synthetic no-tumor data. A meta-classifier MLP was used for binary tumor detection from volumetric features. Explainability was offered using XRAI maps for 2D predictions and Gaussian overlays for 3D visualizations. The platform was incorporated into a web interface for clinical use. Results: MobileNetV2 2D model recorded 98.09% classification accuracy for tumor classification. 3D SegResNet obtained Dice coefficients around 68–70% for tumor segmentations. The MLP-based tumor detection module recorded 100% detection accuracy. Explainability modules could identify the area of the tumor, and saliency and overlay maps were consistent with real pathological features in both 2D and 3D. Conclusions: Deep learning diagnosis system possesses improved brain tumor classification and segmentation with interpretable outcomes by utilizing XAI techniques. Deployment as a web tool and a user-friendly interface made it suitable for clinical usage in radiology workflows. Full article

Background: The implementation of intravenous thrombolysis (IVT) and mechanical thrombectomy (MT) in acute ischemic stroke (AIS) has proven effective, offering significant benefits for patient outcomes. We therefore investigated trends in the implementation of IVT and MT in Germany between 2015 and 2023. Methods: We conducted a retrospective analysis using German Diagnosis-Related Group (DRG) statistics from 2015 to 2023. Treatment numbers were analyzed annually based on OPS codes. We examined the age and sex distribution of patients undergoing these treatments. Additionally, we analyzed all hospital cases coded with ICD-10 for acute ischemic stroke (AIS). Results: Between 2015 and 2023, the number of AIS cases in Germany slightly declined from 250,802 to 248,107 (−1.1%), with the largest annual decrease (−4.3%) occurring during the COVID-19 pandemic (2019–2020). Despite this, the use of IVT increased from 40,766 cases (16.25%) in 2015 to 48,378 (19.50%) in 2023. MT usage rose even more sharply, from 7840 cases (3.13%) to 22,445 (9.05%). Among MT recipients, the proportion of patients aged ≥80 years rose significantly, from 27.2% to 42.1%. In this age group, women consistently comprised the majority of MT patients—65.4% in 2015 and 65.5% in 2023. Conclusions: Despite a stable stroke incidence, the use of IVT—and particularly MT—continued to increase in Germany from 2015 to 2023, even during the COVID-19 pandemic. MT usage nearly tripled, especially among patients aged ≥80 years. These trends highlight a resilient stroke care system and underscore the need for future planning to meet the rising demand for endovascular treatment in an aging population. Full article

As artificial intelligence (AI) technology becomes increasingly widespread in organizations, its impact on individual employees’ psychology and behavior has garnered growing attention. Existing research primarily focuses on AI’s effects on organizational performance and job design, with limited exploration of its mechanisms influencing individual employees, particularly in the critical area of risk-taking behavior, which is essential to organizational innovation. This research develops a moderated mediation model grounded in social cognitive theory (SCT) to explore how AI usage affects the willingness to take risks. A three-wave longitudinal study collected and statistically analyzed data from 442 participants. The findings reveal that (1) AI usage significantly enhances employees’ willingness to take risks; (2) self-efficacy serves as a partial mediator in the connection between AI usage and the willingness to take risks; and (3) learning goal orientation moderates both the relationship between AI usage and self-efficacy, as well as the mediating effect. This research enhances our understanding of AI’s impact on organizational behavior and provides valuable insights for human resource management in the AI era. Full article

In the digital era, understanding the intention to use generative AI is critical, as it enhances productivity, transforms workflows, and enables humans to focus on higher-value tasks. Drawing upon the unified theory of acceptance and use of technology (UTAUT) and the technology threat avoidance theory (TTAT), this research integrates perceived AI literacy into the AI acceptance–avoidance framework as a central variable. This study gathered 583 valid survey responses from China and validated its model using a dual-phase, combined method that integrates structural equation modeling and artificial neural networks. Research findings indicate that the model explains 51.6% of the variance in generative AI usage intention. Except for social influence, all variables within the extended framework significantly impact the usage intention, with perceived AI literacy being the strongest predictor (β = 0.33, p < 0.001). Additionally, perceived AI literacy mitigates the adverse effect of perceived threats on the intention to use AI. Practical implications suggest that enterprises adopt a tiered strategy, as follows: maximize perceived benefits by integrating AI skills into reward systems and providing task-automation training; minimize perceived costs through dedicated technical support and transparent risk mitigation plans; and cultivate AI literacy via progressive learning paths, advancing from data analysis to innovation. Full article