Online-Journals.org (International Association of Online Engineering)
Not a member yet
9805 research outputs found
Sort by
Design of a Mobile AR-Based English Contextual Teaching System and Cognitive Analysis of Interactive Behavior
Get PDFWith the proliferation of mobile smart terminals and the maturation of augmented reality (AR) technology, the field of education has been undergoing a transformation toward immersive and contextualized learning. In contrast, conventional English instruction remains constrained by static textbooks and abstract explanations, which have proven inadequate for constructing authentic language application scenarios. To address this gap, mobile AR-based English contextual teaching has emerged as a promising solution by converting linguistic knowledge into perceptible virtual contexts through the integration of real and virtual techniques. Despite its potential, existing studies reveal significant limitations. A number of investigations have focused solely on resource development without a comprehensive system architecture, while others have incorporated interaction analysis but lacked a foundation in cognitive theory. Moreover, assessment methods have primarily relied on simplistic feedback mechanisms, with limited incorporation of dynamic data tracking. To overcome these challenges, a mobile AR-based English contextual teaching system was designed, accompanied by an in-depth cognitive analysis of group interactive behavior. A hybrid context generation framework was constructed, functional modules for a mobile application were developed, and the mechanisms through which group interaction influences language cognition were examined. This study provides a systematic approach for the deep integration of technology and English education and offers a theoretical basis for optimizing contextual teaching strategies
A Hybrid Chaotic Zebra Optimization Algorithm for Cost-Effective Healthcare Team Formation
Get PDFThis paper presents a hybrid approach to enhancing the zebra optimization algorithm (ZOA) by integrating the chaotic map for cost-effective healthcare team formation. Healthcare team formation is one of the complex optimization problems that is essential in resource allocation, cost efficiency, and skill diversity. Traditional methods struggle to find optimal solutions, which makes the metaheuristic algorithm a valuable approach to solving complex challenges. Metaheuristic algorithms are inspired by natural and evolutionary processes and have been widely implemented in optimization problems due to their ability to explore large solution spaces and bring optimal solutions. Among these, ZOA has shown the ability to solve optimization problems where it is inspired by zebra natural behaviors, which face some limitations on diversity, exploration, and resource allocation, particularly in finding the best team formation by random skill set. The standard ZOA’s randomization of data lacks strategic diversity, which leads to inefficient solutions and slower convergence. To overcome these limitations, the chaotic tent-map will be integrated with ZOA to improve the algorithm’s exploration and heterogeneity or solution capabilities. The enhanced ZOA performance will be compared with the original ZOA and other metaheuristic algorithms. The performance of the improved algorithm is endorsed using real data information from expert doctors in Malaysia, displaying improved outcomes in terms of both cost efficiency and team formation size
Usability and Performance Evaluation of a Mobile Application for Learning PID Controllers: A Case Study with Engineering Students
Get PDFIn the academic field, especially in contexts with limited technological resources, it is essential to provide tools that facilitate both the theoretical understanding and the practical application of PID controllers. This study evaluates the usability and performance of TempRes ControlSys Simulator, a mobile application developed to enhance practical learning of automatic control. Through a non-experimental case study, the User Experience Questionnaire (UEQ) was applied to assess the perception of usability, as well as BlazeMeter to analyze performance. The results reflect a moderate usability of the application, with 60% of the students highlighting its clarity, efficiency, and precision. In terms of performance, the results show that the mobile application responded optimally to the expected demand within the case study. It is concluded that the mobile application represents a viable and accessible alternative in educational environments with technological limitations, favoring inclusion by facilitating access to practical simulation tools. Future research should integrate external sensors for real-time signal acquisition, allowing its application in real scenarios such as the handling of robotic arms or the automatic orientation of solar panels, reinforcing the connection between simulation and experimentation in mechanical and electrical engineering students
STRIDE-Based Threat Modeling and Risk Assessment Framework for IoT-enabled Smart Healthcare Systems
Get PDFThe increase in the evolution of Internet of Things (IoT) architectures and their use in different domains, such as agriculture, smart cities/homes, industry, transport and logistics, and others, has triggered a proportional increase in vulnerabilities, threats, and security risks that violate security objectives. The criticality of the information circulating in the architecture and assets exposed to public networks such as the internet imperatively implies effective management beforehand of the factors that can put a system at risk. This article focuses mainly on the IoT in the healthcare sector and uses it as a pilot for an in-depth study of vulnerabilities using a threat modeling approach based on the spoofing, tampering, repudiation, information disclosure, denial of service, and elevation of privilege (STRIDE) method. This will be carried out by carefully following the steps of explaining the use case, drawing a data flow diagram (DFD) using the tool offered by Microsoft Threat Modeling (MTM) tool, and identifying the assets in question. This is followed by an identification of the threats linked to the DFD and identified assets, then an assessment of the risks caused, and finally proposals for security patches to be applied as far as possible to ensure the efficient and secure use of an architecture that offers many advantages in terms of services and ease of management of modern domains but which at the same time puts at risk all the assets that can cause significant damage and impact if they are exposed to malicious hands
Swin-BSSeg: A Novel Swin Transformer-Enhanced Architecture for Accurate Ischemic Stroke Lesion Segmentation in MRI Images
Get PDFIschemic stroke, caused by obstructed cerebral blood flow, remains a leading cause of mortality and disability, necessitating precise magnetic resonance imaging (MRI)-based lesion detection. This paper proposes Swin-BSSeg (brain symmetry segmentation) algorithm, an enhanced version of the brain symmetry segmentation network (BSSNet), for improved stroke lesion segmentation. Swin-BSSeg integrates Swin Transformers to capture global context and long-range dependencies through a hierarchical attention mechanism. The encoder replaces traditional convolutions with depth-wise separable convolution (DWSC) blocks comprising three cascaded depth-wise convolutional layers for efficient feature extraction and parameter reduction. Feature transfer to the decoder is accomplished via concatenation operations. The decoder also employs DWSC blocks to reduce computational demands and incorporates attention-guided connections (AGC) to refine the contextual diversity—defined as model’s ability to capture varied and spatially distributed lesion features. The model was evaluated on the public datasets (Anatomical Tracings of Lesions after Stroke) and ISLES (Ischemic Stroke Lesion Segmentation), which achieved a Dice Coefficient of 0.842 and an accuracy of 0.87 on the ATLAS dataset, while a Dice Coefficient of 0.8049 and an accuracy of 0.84 on ISLES, outperforming BSSNet in contextual understanding and boundary precision. The improvements on the ATLAS dataset were statistically significant (p < 0.05), confirming the reliability of the proposed enhancements
Applications and Challenges of Artificial Intelligence in Oncologic Surgical Education
Get PDFSurgical training faces significant challenges due to the technical complexity of procedures and the imperative to ensure patient safety during the learning process. In this context, artificial intelligence (AI), machine learning, and robotic platforms are transforming traditional models of surgical education. This paper presents a narrative overview of the impact of these tools on oncologic surgical training. The reviewed evidence indicates that artificial intelligence contributes to improved clinical decision-making, enhances surgical planning, and enables the implementation of automated evaluation systems with objective, real-time feedback. These solutions promote personalized learning pathways and strengthen competency standardization. Moreover, robotic surgery assisted by artificial intelligence provides advanced simulation environments that facilitate the acquisition of technical skills in a safe and controlled manner. Recent developments also emphasize the importance of ethical governance, multi-phase validation in clinical settings, and alignment with international standards such as ISO/IEC 42001:2023 to ensure equitable and effective adoption of these technologies. Nonetheless, significant limitations remain, such as limited validation in real clinical settings, methodological heterogeneity across existing studies, and the high costs associated with implementation. These barriers hinder equitable and sustainable adoption. Future efforts should focus on validating AI-based training systems in real clinical environments to ensure their effectiveness, safety, and relevance in surgical oncology education. This study was registered in the Open Science Framework under the code 10.17605/OSF.IO/QUTC4
Intelligent Decision Support System Based on Heart Failure
Get PDFIn the ever-changing realm of healthcare, the integration of advanced technologies and the wealth of medical data presents exciting opportunities to enhance diagnostic processes. This study presents the Intelligent Diagnostic Decision Support System (IDDSS), employing data analysis and AI to aid doctors in making informed diagnostic decisions. The IDDSS harnesses a wealth of patient data, encompassing medical records, test findings, and demographic details. By employing advanced data analysis techniques, it reveals valuable insights, identifies patterns, and establishes correlations within the data. Furthermore, to elevate diagnostic precision, the IDDSS integrates state-of-the-art AI algorithms and machine learning models. These models, trained on extensive datasets, excel in recognizing intricate patterns, categorizing illnesses, and predicting outcomes. Continuously adapting to incorporate the latest medical advancements, the IDDSS remains at the forefront of enhancing healthcare efficacy and patient care. This system has significant potential to advance healthcare diagnostics, ultimately serving as a valuable decision support tool enabling physicians to provide exceptional care and improve patient outcomes
Advancing Osteoporosis Diagnosis through State-of-the-Art CNNs and Vision Transformers with Ensemble Strategies
Get PDFOsteoporosis is a common bone disorder marked by reduced mineral density and microarchitectural deterioration, increasing fracture risk. Early, accurate detection is vital for clinical intervention and personalized care. This study applies deep learning to binary classification of osteoporosis (normal vs. osteoporotic) using medical imaging. A curated dataset of bone-related images was used to train and evaluate advanced models and ensemble strategies. Evaluated architectures include EfficientNetB2, InceptionV3, InceptionResNetV2, ResNet50V2, Xception, Vision Transformer (ViT_B32), and Faster R-CNN. Accuracy served as the main metric. ResNet50V2 outperformed all with 97.83% accuracy, ahead of EfficientNetB2 and ViT_B32 (95.65%), InceptionV3 and Xception (95.22%), InceptionResNetV2 (93.91%), and Faster R-CNN (76.96%). Ensembles—average, weighted, and hard voting—further improved accuracy to 96.96% and 96.09%. The results validate the benefit of ensemble learning in boosting model robustness. ResNet50V2 stands out as the top single model, and ensemble techniques show strong promise for reliable, automated osteoporosis detection. These findings support deploying deep learning in clinical radiology for early diagnosis and decision support
The Mediating Role of Resource Allocation Optimization in AI-Enhanced Corporate Performance: An Integrated Framework
Get PDFThe application of artificial intelligence (AI) is a key factor to promote the sustainable development of the global economy. Based on the panel data of A-share listed companies from 2001 to 2022, this study explores the promoting effect and influence mechanism of AI on the business performance of high-tech enterprises. The study found that the development of AI technology effectively improves the business performance of high-tech enterprises. Mechanism analysis shows that AI technology can improve the overall operating performance of enterprises by improving the mechanism of enterprise factor structure. Further analysis shows that there are significant regional differences in the business performance of enterprises, with the strongest positive impact on high-tech enterprises in the eastern region. This study enriches the research on the influencing factors of the business performance of high-tech enterprises and provides reference for improving the business performance of high-tech enterprises, strengthening the data-driven policy, and building a perfect AI system
Needs Analysis from Teachers’ Experiences in Online Academic English Writing Model for Chinese EFL College Students
Get PDFThe integration of technology in education has gained significant attention, particularly in China’s higher education sector. While much research examines the role of digital tools in education, there is limited focus on understanding the specific needs of online courses from the perspectives of educators. In the context of online academic English writing for Chinese English as a Foreign Language (EFL) learners, teachers encounter various challenges that have not been thoroughly explored. This study employs a qualitative approach (N = 6) to conduct a needs analysis focused on the experiences of instructors teaching online academic English. By examining the difficulties teachers face and the improvements they seek, this study highlights key issues in three areas: pedagogical strategies and resources, course scheduling, and learner engagement. The thematic analysis of data from six participants reveals a strong desire for more interactive and collaborative learning frameworks in online teaching model development. These findings offer valuable, context-specific insights into teachers’ needs in online writing instruction and EFL pedagogy within the Chinese higher education context, providing implications for enhancing online teaching practices tailored to this particular setting