Search Results (19,247)

Cybersecurity threats are becoming increasingly sophisticated, frequent, and diverse, posing a major risk to critical infrastructure, public trust, and digital economies. Traditional intrusion detection systems often struggle with detecting novel or rare attack types, particularly when data availability is limited or heterogeneous. The current study tries to address these challenges by proposing a unified, multimodal threat detection framework that leverages the combination of synthetic data generation through Generative Adversarial Networks (GANs), advanced ensemble learning, and transfer learning techniques. The research objective is to enhance detection accuracy and resilience against zero-day, botnet, and image-based malware attacks by integrating multiple data modalities, including structured network logs and malware binaries, within a scalable and flexible pipeline. The proposed system features a dual-branch architecture: one branch uses a CNN with transfer learning for image-based malware classification, and the other employs a soft-voting ensemble classifier for tabular intrusion detection, both trained on augmented datasets generated by GANs. Experimental results demonstrate significant improvements in detection performance and false positive reduction, especially when multimodal outputs are fused using the proposed confidence-weighted strategy. The findings highlight the framework’s adaptability and practical applicability in real-world intrusion detection and response systems. Full article

The increasing cultural and demographic diversity among higher education students highlights the challenges regarding accessibility and inclusion. The COVID-19 pandemic has accelerated the shift toward flexible, technology-based teaching practices. However, inclusive, and accessible pedagogical practices lack consistency, particularly when supporting students with disabilities or diverse learning needs. This study evaluates the effectiveness of the Learning Design for Flexible Education (FLeD) Tool—a web-based platform developed to support teachers in designing flexible and inclusive learning scenarios. The research adopts a qualitative approach, featuring semi-structured interviews with two Portuguese experts in accessibility and inclusion. The experts analyzed three learning scenarios designed using the FLeD Tool, through the lens of Universal Design for Learning standards. The collected dataset was analyzed using thematic analysis to identify common issues, strengths, and opportunities for improvement. The findings show a gap between institutional policies and their practical application, mainly due to inconsistent teacher training and technical limitations. While the FLeD Tool supports more flexible and inclusive pedagogical designs, experts have identified key shortcomings such as the lack of automated accessibility checks and limited support for specific disabilities. Despite the reduced number of participants (two experts) and dataset (three learning scenarios), which limits the study’s generalisability, the conclusions draw attention to the pivotal role of systematic teacher training, embedded accessibility features and solid institutional policies in bridging the gap between policy aspiration and effective inclusive practice. Full article

Autism Spectrum Disorder (ASD) remains significantly underdiagnosed in women, resulting in a persistent gender gap with important clinical, functional, and psychosocial implications. This narrative review explores the multifactorial barriers contributing to diagnostic disparities, including the male-oriented structure of current diagnostic criteria, the prevalence of co-occurring psychiatric conditions, and the phenomenon of social camouflaging shaped by culturally reinforced gender norms. These factors frequently lead to delayed identification, clinical misinterpretation, and suboptimal care. The review synthesizes evidence from clinical, psychological, and sociocultural research to demonstrate how the under-recognition of ASD in women impacts mental health outcomes, access to education, occupational stability, and overall quality of life. Special emphasis is placed on the consequences of missed or late diagnoses for healthcare delivery and the educational needs of clinicians involved in ASD assessment and care. This article concludes with actionable, evidence-based recommendations for enhancing diagnostic sensitivity, developing gender-responsive screening strategies, and integrating training on female autism presentation into medical and allied health education. Addressing these challenges is essential to reducing diagnostic inequities and ensuring timely, accurate, and person-centered care for autistic women throughout their lifespan. Full article

For the last few decades, the use of High-Speed Trains (HSTs) has been growing rapidly in various parts of the world. Despite rapid global expansion, many HST projects fail due to demand overestimation and cost overruns. This study analyzes factors influencing HST demand in Indonesia, aiming to identify impactful determinants from user perspectives. Employing a quantitative cross-sectional approach, this research utilized questionnaires distributed to users of different modes of transportation in the Jakarta–Bandung area, including trains, buses, travel services, and private cars. Structural Equation Modeling (SEM) via Lisrel software was used to analyze the data. The results indicate that Transit-Oriented Developments (TOD) and new urban areas significantly increase HST demand by facilitating urban growth and development. Additionally, supporting infrastructure and external factors such as road accessibility, parking availability, shuttle services, and environmental integration are pivotal in shaping commuter preferences. Although factors such as safety, comfort, and reliability are important, they alone may not be adequate to persuade consumers to use high-speed trains for their travel. Full article

Periodic visual inspections are currently conducted to maintain the condition of railway bridges. These inspections rely on direct visual assessments by human inspectors, often requiring specialized equipment such as aerial ladders. However, this method is not only time-consuming and costly but also involves significant safety risks. Therefore, there is a growing need for a more efficient and reliable alternative to traditional visual inspections of railway bridges. In this study, we evaluated and compared the performance of damage detection using U-Net-based deep learning models on images captured by unmanned aerial vehicles (UAVs). The target damage types include cracks, concrete spalling and delamination, water leakage, exposed reinforcement, and paint peeling. To enable multi-class segmentation, the U-Net model was trained using three different loss functions: Cross-Entropy Loss, Focal Loss, and Intersection over Union (IoU) Loss. We compared these methods to determine their ability to distinguish actual structural damage from environmental factors and surface contamination, particularly under real-world site conditions. The results showed that the U-Net model trained with IoU Loss outperformed the others in terms of detection accuracy. When applied to field inspection scenarios, this approach demonstrates strong potential for objective and precise damage detection. Furthermore, the use of UAVs in the inspection process is expected to significantly reduce both time and cost in railway infrastructure maintenance. Future research will focus on extending the detection capabilities to additional damage types such as efflorescence and corrosion, aiming to ultimately replace manual visual inspections of railway bridge surfaces with deep-learning-based methods. Full article

Background/Objectives: Mild Cognitive Impairment (MCI), a condition between normal aging and dementia, is characterized by cognitive changes that do not significantly affect instrumental activities of daily living. The Memory Support System (MSS), an evidence-based behavioral intervention developed by the Mayo Clinic, has been shown to aid those living with MCI and their support partners in coping with cognitive challenges. However, the MSS has not been offered clinically within the Canadian context. Therefore, we conducted a study assessing the feasibility of the MSS from the perspectives of individuals living with MCI and their support partners. Methods: Participants from an institutional registry of research participants, patients, and support partners at a memory clinic, as well as members of a local Dementia Society, were approached to complete an online or paper version of a survey assessing feasibility dimensions. Responses were compared between and within groups for differences in mean scores and associations between linked binary choice response questions. Results: A total of 77 responses were received; 39 surveys were completed by participants with MCI, and 38 by support partners. Respondents found the MSS to be acceptable and practical. On average, participants thought it would be more difficult to train in using the MSS than support partners. Both groups expressed interest in the intervention. On average, participants with MCI and support partners preferred virtual MSS training to in-person and indicated more interest in participating in training over six weeks as compared to two weeks. Conclusions: Flexibility in duration and format when offering the MSS are important considerations when offering the intervention as part of a clinical program. Future research should evaluate cost-effectiveness (e.g., financial, staff resources, etc.) of the MSS approach if it were to be institutionalized in the Ontario healthcare system. Full article

The rapid expansion of peptide libraries and the increasing functional diversity of peptides have highlighted the significance of predicting the multifunctional properties of peptides in bioinformatics research. Although supervised learning methods have made advancements, they typically necessitate substantial amounts of labeled data for yielding accurate prediction. This study presents MvAl-MFP, a multi-label active learning approach that incorporates multiple feature views of peptides. This method takes advantage of the natural properties of multi-view representation for amino acid sequences, meets the requirement of the query-by-committee (QBC) active learning paradigm, and further significantly diminishes the requirement for labeled samples while training high-performing models. First, MvAl-MFP generates nine distinct feature views for a few labeled peptide amino acid sequences by considering various peptide characteristics, including amino acid composition, physicochemical properties, evolutionary information, etc. Then, on each independent view, a multi-label classifier is trained based on the labeled samples. Next, a QBC strategy based on the average entropy of predictions across all trained classifiers is adopted to select a specific number of most valuable unlabeled samples to submit them to human experts for labeling by wet-lab experiments. Finally, the aforementioned procedure is iteratively conducted with a constantly expanding labeled set and updating classifiers until it meets the default stopping criterion. The experiments are conducted on a dataset of multifunctional therapeutic peptides annotated with eight functional labels, including anti-bacterial properties, anti-inflammatory properties, anti-cancer properties, etc. The results clearly demonstrate the superiority of the proposed MvAl-MFP method, as it can rapidly improve prediction performance while only labeling a small number of samples. It provides an effective tool for more precise multifunctional peptide prediction while lowering the cost of wet-lab experiments. Full article

Multimodal knowledge graphs have recently been successfully applied to tasks such as those relating to information retrieval, question and answer, and recommender systems. In this study, we propose a dual-path fine-tuning mechanism technique with a low-rank adapter and an embedded cueing layer, aiming to improve the generalization and accuracy of the model in analogical reasoning tasks. The low-rank fine-tuning (LoRA) technique with rank-stable scaling factor is used to fine-tune the MKGformer model, and a cue-embedding layer is innovatively added to the input layer, which enables the model to better grasp the scale of the relationship between entities according to the dynamic cue vectors during the fine-tuning process and ensures that the model achieves the best results during training. The experimental results show that the R-MKG model improves several evaluation indexes by more than 20%, which is significantly better than the traditional DoRA and FA-LoRA methods. This research provides technical support for multimodal knowledge graph analogical reasoning. We hope that our work will bring benefits and inspire future research. Full article

Smart safety helmets equipped with vision systems are emerging as powerful tools for industrial infrastructure inspection. This paper presents a comprehensive state-of-the-art review of such VSLAM-enabled (Visual Simultaneous Localization and Mapping) helmets. We surveyed the evolution from basic helmet cameras to intelligent, sensor-fused inspection platforms, highlighting how modern helmets leverage real-time visual SLAM algorithms to map environments and assist inspectors. A systematic literature search was conducted targeting high-impact journals, patents, and industry reports. We classify helmet-integrated camera systems into monocular, stereo, and omnidirectional types and compare their capabilities for infrastructure inspection. We examine core VSLAM algorithms (feature-based, direct, hybrid, and deep-learning-enhanced) and discuss their adaptation to wearable platforms. Multi-sensor fusion approaches integrating inertial, LiDAR, and GNSS data are reviewed, along with edge/cloud processing architectures enabling real-time performance. This paper compiles numerous industrial use cases, from bridges and tunnels to plants and power facilities, demonstrating significant improvements in inspection efficiency, data quality, and worker safety. Key challenges are analyzed, including technical hurdles (battery life, processing limits, and harsh environments), human factors (ergonomics, training, and cognitive load), and regulatory issues (safety certification and data privacy). We also identify emerging trends, such as semantic SLAM, AI-driven defect recognition, hardware miniaturization, and collaborative multi-helmet systems. This review finds that VSLAM-equipped smart helmets offer a transformative approach to infrastructure inspection, enabling real-time mapping, augmented awareness, and safer workflows. We conclude by highlighting current research gaps, notably in standardizing systems and integrating with asset management, and provide recommendations for industry adoption and future research directions. Full article

This study develops and validates an artificial intelligence (AI)-assisted internship learning platform for automotive electronics based on the Llama 3 large language model, aiming to enhance pedagogical effectiveness within vocational training contexts. Addressing critical issues such as the persistent theory–practice gap and limited innovation capability prevalent in existing curricula, we leverage the natural language processing (NLP) capabilities of Llama 3 through fine-tuning based on transfer learning to establish a specialized knowledge base encompassing fundamental circuit principles and fault diagnosis protocols. The implementation employs the Hugging Face Transformers library with optimized hyperparameters, including a learning rate of 5 × 10⁻⁵ across five training epochs. Post-training evaluations revealed an accuracy of 89.7% on validation tasks (representing a 12.4% improvement over the baseline model), a semantic comprehension precision of 92.3% in technical question-and-answer assessments, a mathematical computation accuracy of 78.4% (highlighting this as a current limitation), and a latency of 6.3 s under peak operational workloads (indicating a system bottleneck). Although direct trials involving students were deliberately avoided, the platform’s technical feasibility was validated through multidimensional benchmarking against established models (BERT-base and GPT-2), confirming superior domain adaptability (F1 = 0.87) and enhanced error tolerance (σ² = 1.2). Notable limitations emerged in numerical reasoning tasks (Cohen’s d = 1.15 compared to human experts) and in real-time responsiveness deterioration when exceeding 50 concurrent users. The study concludes that Llama 3 demonstrates considerable promise for automotive electronics skills development. Proposed future enhancements include integrating symbolic AI modules to improve computational reliability, implementing Kubernetes-based load balancing to ensure latency below 2 s at scale, and conducting longitudinal pedagogical validation studies with trainees. This research provides a robust technical foundation for AI-driven vocational education, especially suited to mechatronics fields that require close integration between theoretical knowledge and practical troubleshooting skills. Full article

The successful deployment of sustainable building construction (SBC) is connected to sound knowledge sharing. Concerning SBC, knowledge sharing has been identified to directly and indirectly increase innovation, environmental performance, cost saving, regulatory compliance awareness and so on. The necessity of enhancing SBC practice globally has been emphasised by earlier research. Consequently, this study aims to investigate knowledge-sharing elements to enhance SBC in South Africa (SA). Utilising a questionnaire survey, this study elicited data from 281 professionals in the built environment. Data analysis was performed with “descriptive statistics”, the “Kruskal–Wallis H-test”, and “principal component analysis” to determine the principal knowledge-sharing features (KSFs). This study found that “creating public awareness of sustainable practices”, the “content of SBC training, raising awareness of green building products”, “SBC integration in professional certifications”, an “information hub or repository for sustainable construction”, and “mentoring younger professionals in sustainable practices” are the most critical KSFs for SBC deployment. These formed a central cluster, the Green Education Initiative and Eco-Awareness Alliance. The results achieved a reliability test value of 0.956. It was concluded that to embrace the full adoption of SBC, corporate involvement is critical, and all stakeholders must embrace the sustainability paradigm. It is recommended that the principal knowledge-sharing features revealed in this study should be carefully considered to help construction stakeholders in fostering knowledge sharing for a sustainable built environment. Full article

This study examines the factors influencing the adoption of enterprise resource planning (ERP) systems within Romanian organizations. The objective is to develop a comprehensive framework for ERP adoption decisions, thereby advancing the field of knowledge and offering managerial insights. To accomplish this research goal, a questionnaire is envisioned, employing various research hypotheses, and distributed to a representative sample. Quantitative econometric regression analysis is employed, considering potential factors such as user training and education, competitive pressures, user involvement and participation, decentralized ERP features, top management support, data quality, the quality of the ERP system, cost and budget considerations, and business process reengineering. Of the 12 factors analyzed, 9 were found to be relevant in terms of influence on the decision to adopt ERP systems, in the context of the Romanian market. The other three factors were found to be irrelevant, thus obtaining results partially different from other areas of the world. By validating the hypotheses and answering the research questions, this work addresses a research gap regarding the lack of a comprehensive understanding of the influencing factors that shape the adoption process of ERP systems in Romania. Full article

Almost one in three firefighters develop mental health disorders at some point during their careers, a rate double that in the general population. Frequent exposures to potentially traumatic situations can contribute to symptoms of these disorders, two of the most common being depression and post-traumatic stress disorder (PTSD). While various psychological interventions have been implemented among this group, reports of their effectiveness include mixed results. To this end, the current study endeavored to test the effectiveness of a 12-week intervention combining occupationally-tailored high-intensity functional training (HIFT) and psychological resilience training (RES) in reducing depressive and post-traumatic stress symptoms (PTSSs), as well as increasing psychological resilience and mental wellbeing, in career firefighters. Thirty career firefighters completed four mental health measurements over 17 weeks while anthropometrics and physical performance (i.e., number of stations completed in 20 min during an eight-station simulated job-task circuit workout [T-CAC]) were measured pre- and post-intervention. Pre to post comparisons were made via repeated-measures t-tests. Significant mean differences were observed for T-CAC stations completed, PTSSs, and psychological resilience between pre- and post-intervention. In future interventions, researchers should actively engage firefighters, maximize integration with daily operations, and employ culturally-relevant practices to explore the links between physical and mental health. Full article

Monitoring the height of secondary forest regrowth is essential for assessing ecosystem recovery, but current methods rely on field surveys, airborne or UAV LiDAR, and 3D reconstruction from high-resolution UAV imagery, which are often costly or limited by logistical constraints. Here, we address the challenge of scaling up canopy height monitoring by evaluating a recent deep learning model, trained on data from the Amazon and Atlantic Forests, developed to extract canopy height from RGB-NIR Planet NICFI imagery. The research questions are as follows: (i) How are canopy height estimates from the model affected by slope and orientation in natural forests, based on a large and well-balanced experimental design? (ii) How effectively does the model capture the growth trajectories of Pinus and Eucalyptus plantations over an eight-year period following planting? We find that the model closely tracks Pinus growth at the parcel scale, with predictions generally within one standard deviation of UAV-derived heights. For Eucalyptus, while growth is detected, the model consistently underestimates height, by more than 10 m in some cases, until late in the cycle when the canopy becomes less dense. In stable natural forests, the model reveals seasonal artifacts driven by topographic variables (slope × aspect × day of year), for which we propose strategies to reduce their influence. These results highlight the model’s potential as a cost-effective and scalable alternative to field-based and LiDAR methods, enabling broad-scale monitoring of forest regrowth and contributing to innovation in remote sensing for forest dynamics assessment. Full article

Background: Since the outbreak of war in Ukraine, countless forcibly displaced individuals facing not only material loss, but also deep psychological distress, have sought refuge across Europe. For those traumatized by war, the absence of a shared language in therapy can hinder healing and exacerbate suffering. While cultural diversity in psychotherapy has gained recognition, the role of native-language communication—especially from a person-centered perspective—remains underexplored. Methods: This narrative review with a thematic analysis examines whether and how psychotherapy in the mother tongue facilitates access to therapy and enhances therapeutic efficacy. Four inter-related clusters emerged: (1) the psychosocial context of trauma and displacement; (2) language as a structural gatekeeper to care (RQ1); (3) native-language therapy as a mechanism of change (RQ2); (4) potential risks such as over-identification or therapeutic mismatch (RQ2). Results: The findings suggest that native-language therapy can support the symbolic integration of trauma and foster the core conditions for healing. The implications for multilingual therapy formats, training in interpreter-mediated settings, and future research designs—including longitudinal, transnational studies—are discussed. Conclusions: In light of the current crises, language is not just a tool for access to therapy, but a pathway to psychological healing. Full article