Search Results (13,101)

Extended Reality (XR) and Artificial Intelligence (AI) are increasingly converging within cyber–physical infrastructures, including digital twins, the Spatial Web, and smart-city systems. These environments require new frameworks for understanding how human performance emerges through sustained interaction with immersive interfaces and adaptive computational agents. This paper introduces the TAXI–XI-CAP framework, a two-layer model that links psychobiological mechanisms of XR–AI interaction to higher-level, experimentally testable capability constructs. The TAXI layer defines 42 mechanisms spanning perception, cognition, physiology, sensorimotor control, and social coordination, while XI-CAP organizes these into capability patterns such as remote dexterity, distributed cognition, and adaptive workload regulation. Derived through a theory-guided synthesis across XR, neuroscience, and human–automation interaction, the framework models performance as emerging from interacting mechanisms under real-world constraints. A validation-oriented research agenda is proposed, emphasizing mechanism-level measurement, capability-level evaluation, and longitudinal testing. The TAXI–XI-CAP framework provides a structured basis for hypothesis generation, comparative analysis, and empirical validation of XR–AI systems, supporting the development of reliable, scalable, and human-centered Extended Intelligence infrastructures. Full article

Adaptive reuse of industrial heritage from a tourism perspective presents a complex design challenge requiring a balance between heritage preservation, functional innovation, and diverse stakeholder expectations. However, current practices often face issues such as ambiguous demand interpretation and a disconnect between design generation and systematic evaluation. Addressing these limitations, this paper proposes and illustrates a human–machine collaborative design paradigm that integrates generative AI into a closed-loop process of “demand analysis–intelligent generation–comprehensive evaluation.” The method first employs Kansei Engineering and the KANO model to qualitatively extract and quantitatively prioritise heterogeneous user needs, translating subjective perceptions into structured design constraints and optimisation objectives. Next, these needs are encoded as text prompts to drive targeted spatial exploration by the generative AI tool Nano Banana AI. Finally, the TOPSIS method is applied for multi-criteria performance evaluation and solution selection. A case study of Shanghai Libo Brewery suggests that this paradigm can enhance design efficiency and show potential to outperform traditional methods across dimensions such as historical preservation, public accessibility, ecological integration, social inclusivity, and formal innovation. The research offers a quantifiable and systematically documented intelligent design methodology for industrial heritage renewal, while acknowledging the exploratory nature of the generative phase. Furthermore, it provides a visitor-demand-driven innovation pathway for developing industrial heritage tourism destinations, thereby potentially enhancing cultural experiences and tourism appeal at heritage sites. This research illustrates a move from an experience-driven paradigm toward a data- and value-driven approach, contributing theoretical methodologies to the intersection of cultural tourism and artificial intelligence. Full article

Renewable energy technologies are becoming more and more relevant in a variety of engineering fields as a result of the move toward low-carbon, sustainable energy systems. Although research has historically concentrated on power generation, it now covers a broad range of applications, including precision agriculture, smart grids, energy storage, healthcare devices, and sustainable buildings. However, existing review studies are often limited to single disciplines or specific technologies, lacking a unified cross-disciplinary perspective that captures the interconnected nature of modern renewable energy systems. This gap motivates the need for a comprehensive review that bridges multiple engineering domains. This review provides a comprehensive synthesis of literature on renewable energy applications in electrical and electronics, computer, environmental, biomedical, architectural, and agricultural engineering. In electrical and electronics engineering, the use of renewable energy sources is largely based on the efficient generation of electricity from natural resources such as solar, wind, and ocean energy. Computer engineering contributes through artificial intelligence (AI), Internet of Things (IoT) architectures, digital twins, and cybersecurity solutions, optimizing energy management. Environmental engineering emphasizes life cycle assessment, carbon footprint reduction, and circular economy strategies. In biomedical engineering, energy harvesting and self-powered devices illustrate micro-scale applications of renewable energy. Architectural engineering integrates renewable systems through building-integrated photovoltaics, net-zero energy designs, and smart building management, while agricultural engineering uses solar-powered irrigation, biomass utilization, agrivoltaic systems, and other sustainable practices. To support a low-carbon future with integrated and sustainable engineering solutions, this study not only highlights innovations within individual fields but also showcases how different disciplines can connect and work together. Overall, the review offers a novel cross-disciplinary framework that advances the understanding of renewable energy systems beyond isolated applications and provides direction for future integrative research. Full article

This study proposes a scalable, AI-driven, and Geographic Information System (GIS)-integrated framework for intelligent route-level classification in large-scale airport transportation networks to support airport operations, logistics planning, and network-level decision-making. The framework addresses the need for practical artificial intelligence applications that combine spatial network analysis with supervised machine learning to improve route assessment and resource allocation in complex air transport systems. A structured dataset was developed using operational and traffic-related attributes, including route distance, aircraft capacity, weekly frequency, annual passenger volume, demand variability, and route performance indicators, with additional normalized features to improve data representation. A Gradient Boosting ensemble classifier was trained to categorize routes into high-, medium-, and low-priority classes. The model achieved strong predictive performance, with a testing area under the ROC curve of 0.961, accuracy of 0.922, F1-score of 0.915, precision of 0.918, and a recall of 0.922. Feature importance analysis identified demand variability and route-density indicators as the main drivers of classification, enhancing interpretability and practical trust. The proposed framework demonstrates the real-world potential of AI for scalable, explainable, and efficient decision support in airport logistics and transportation network management. Full article

Generative artificial intelligence (AI) has demonstrated significant potential in education, yet empirical research on its application in “hardware-driven” interdisciplinary design courses remains scarce. This study employed a randomized crossover experimental design in an IoT Hardware and Design Innovation course at Hunan University. Twelve industrial design undergraduates with no prior IoT background alternated between AI-assisted (ChatGPT-4o) and traditional learning resource conditions across six short-cycle tasks. The crossover design enabled each participant to serve as both experimental and control subjects, yielding 72 observation-level data points. Grounded in Cognitive Load Theory, the study examined three dimensions: process efficacy, difficulty structure, and switching adaptation costs. Results indicated that AI significantly improved perceived task completion efficiency, self-reported goal attainment, and learning experience, yet self-assessed knowledge transfer did not differ significantly between conditions. AI reduced the total number of reported difficulties but altered the difficulty-type distribution: resource-retrieval difficulties decreased while information-verification difficulties increased—a phenomenon we term “difficulty restructuring”. Furthermore, switching from AI back to traditional resources incurred significantly higher adaptation costs than the reverse transition, revealing emerging dependency risks. These findings suggest that generative AI may function more as a “difficulty restructurer” than a “difficulty eliminator” in hardware-driven design education, providing exploratory empirical evidence for incorporating verification literacy into future course design and calling for calibrated scaffold fading that may help mitigate emerging dependency risks. Full article

Multiple myeloma (MM) and osteosarcoma (OS) are two biologically distinct osseous malignancies with similar molecular networks that present translational challenges for their computational modeling. This comparative research analyzes MM and OS biology relevant to in silico approaches, focusing on PI3K-AKT-mTOR signaling, the RANK-RANKL-OPG axis, angiogenic factors (VEGF, TGFs), and immune mediators in MM, alongside the transcription factors (SOX9, RUNX2), signaling pathways (PI3K-AKT-mTOR, NOTCH), immune cell state (TAM2), and interleukins in OS. Based on this pathophysiologic foundation, the review outlines five computational paradigms: (i) mechanistic models; (ii) data-driven/machine learning schemes; (iii) hybrid mechanistic approaches; (iv) digital twins/virtual cohorts, and (v) MIDD/PBPK models for real-world applications. A cross-cancer comparison section summarizes common and distinct biological axes and their computational translation as well as the overlapping features from the bone microenvironment. For both MM and OS, the research assesses strengths, limitations, and data needs of current models, outlining the strategic objectives for next-generation multiscale, AI-enabled models providing a roadmap for tissue engineers, oncology scientists, and translational researchers to design clinically relevant preclinical tests and accelerate safer, more effective strategies for tumor-affected bones. The differences between MM and OS impose distinct biological constraints, so their comparisons are rare. Combining all these features with artificial intelligence capabilities will underpin a promising transition in the development of in silico adaptive and learning models. Full article

Cloud-based telephony platforms face growing fraud risks including voice phishing (vishing), subscription abuse, and organizational impersonation, with detection being especially challenging in low-resource languages such as Arabic. We present an Artificial Intelligence (AI)-based security architecture for fraud detection in Arabic cloud call centers, combining onboarding verification, behavioral monitoring, domain-adapted Automatic Speech Recognition (ASR), semantic transcript search, and Large Language Model (LLM)-based entity verification. The domain-adapted Langa ASR model achieves a Word Error Rate (WER) of 41.0% and Character Error Rate (CER) of 18.2%, outperforming all evaluated commercial baselines. LLM-based entity extraction with multi-call consensus achieves 97.3% company-name accuracy (Generative Pre-trained Transformer 4, GPT-4) and 92.0% in the cost-effective deployed configuration (GPT-3.5 with log-probability filtering). Evaluated on production data from a Middle East and North Africa (MENA)-region provider spanning more than 1000 accounts, the pipeline flagged 47 accounts of which 41 were confirmed fraudulent (directly observed precision 87.2%, 95% confidence interval (CI): 74.3–95.2%; estimated recall 51–82% under conservative base-rate assumptions—not directly measured), providing evidence for the viability of a unified, threat-model-driven architecture for low-resource telephony fraud detection. Full article

Flexible deformation sensors inevitably suffer from sensitivity degradation and severe measurement errors during long-term cyclic stretching due to structural fatigue. Traditional material-level optimizations are costly and lack dynamic adaptability. Herein, we propose an artificial intelligence (AI)-driven predict-and-compensate framework for the online calibration of flexible sensors. To overcome training sample scarcity, a generative adversarial network (GAN) performs temporal data augmentation. Subsequently, a hybrid deep learning framework integrating long short-term memory (LSTM) networks and a Sequence Attention mechanism is employed. This architecture accurately captures both local signal fluctuations and multiscale long-term decay trends, enabling precise multi-step prediction and output compensation. Experimental evaluations validate that this strategy significantly suppresses sensor response drift. Under cyclic loading, an initially substantial relative measurement error of 48.63% plummets to 7.16% post-calibration, with typical errors consistently reduced to the ~1% level. Furthermore, when deployed in a smart glove gesture recognition system, this method successfully restores the recognition accuracy from a fatigue-induced low of 75.73% (after 200 stretch cycles) back to 97.70%. This generative and attention-based deep learning paradigm offers robust, real-time error calibration, providing a highly viable solution for extending the long-term reliability and stability of flexible sensor systems. Full article

In the last few years, universities have increasingly implemented online learning environments, allowing students to study at their own pace. These environments utilize technological tools and implement methods to support training, deliver content, and promote the acquisition of new knowledge and skills. As an example of these technologies, eye tracking has emerged as a powerful tool for studying visual attention, cognitive processes, and learning behaviors. The main aim of this study is to provide a scoping review of recent eye-tracking research across diverse learner populations, ranging from K-12 students to university-level learners and educators. The present study examined recent advances in eye-tracking technologies, focusing on their potential, especially when combined with artificial intelligence (AI) techniques such as machine learning. It analyzed 54 empirical studies in the last few years, highlighting their applicability, strengths, and limitations. The research findings highlight the promise of eye-tracking technology to transform educational practices by providing data-driven insights regarding student behavior and cognitive processes. Future research must address implementation and data-analysis challenges to maximize the educational benefits of eye tracking. Full article

Lung cancer remains the leading cause of cancer-related mortality worldwide, and although immunotherapy has transformed the treatment landscape of advanced non-small cell lung cancer (NSCLC), durable benefit is limited to a subset of patients. PD-L1 immunohistochemistry and tumor mutational burden, while clinically utilized, demonstrate imperfect predictive capacity, underscoring the need for more robust biomarkers. This review highlights emerging multimodal biomarkers—including circulating tumor DNA (ctDNA), the gut microbiome, and artificial intelligence (AI)-driven radiomics—as promising tools to enhance the prediction of immunotherapy response. Longitudinal ctDNA monitoring offers a minimally invasive method to assess tumor burden dynamics, detect early molecular response, distinguish pseudo-progression from true progression, and stratify risk, with ctDNA clearance correlating with improved survival outcomes. The gut microbiome has also been associated with ICI efficacy, as specific bacterial taxa and composite scoring systems correlate with treatment response, though methodological heterogeneity limits clinical translation. Radiomic analyses leveraging CT and PET imaging extract quantitative tumor features that, when integrated with clinical and molecular data, demonstrate improved predictive performance compared to single-modality approaches. Despite promising advances, challenges including assay standardization, external validation, data harmonization, interpretability of AI models, and infrastructure requirements remain barriers to widespread adoption. Multimodal integration of genomic, microbiome, and imaging biomarkers represents a critical step toward precision immuno-oncology, with prospective validation needed to translate these approaches into improved outcomes for patients with advanced NSCLC. Full article

Vector–matrix multiplication (VMM), implemented through multiply–accumulate (MAC) operations, represents the dominant computational primitive in many artificial intelligence (AI) workloads. When executed on conventional von Neumann architectures, VMM operations suffer from important energy consumption and latency due to the separation between memory and processing units. To overcome these limitations, crossbar arrays built from Resistive Random Access Memory (RRAM) cells have been proposed for accelerating VMM computations. In this work, we investigate the key optimization trade-offs associated with implementing RRAM-based neural networks for classification applications. A simple two-layer neural network is first defined and trained in software to generate the weight matrices and bias parameters. Next, three hardware implementation scenarios are evaluated depending on whether negative floating-point numbers are used: Positive Weights Only (PWO), Positive and Negative Weights Only (PNWO), and Positive and Negative Weights with Biases (PNWB). The different implementations are analyzed at the hardware level by examining classification accuracy, energy efficiency, latency, and area overhead. The study further incorporates important RRAM limitations, including restricted conductance range and device variability. Hardware results show that the PWO scenario offers the lowest energy consumption (189 fJ/MAC) and area overhead but results in the lowest accuracy. PNWO and PNWB significantly improve accuracy (+177% and +180%) but increase energy consumption (+63% and +87%) and area (×2 and ×2.1). Under variability effects, PWO achieves better accuracy (94.65%), followed by PNWO (93.11%) and PNWB (92.11%). Full article

The integration of Generative Artificial Intelligence (GAI) into software design tools has transformed the early stages of mobile application development, particularly prototype creation from natural-language prompts. This study evaluates the usability and effectiveness of GAI-assisted prototyping tools for generating high-fidelity mobile application prototypes. A controlled laboratory usability study was conducted in which undergraduate Information Technology Engineering students used and evaluated four widely adopted prototyping platforms: Figma, Uizard, Visily, and Stitch. Participants employed these tools to recreate mobile interfaces corresponding to the interaction model of the Duolingo application. The System Usability Scale (SUS) was used to assess perceived usability and effectiveness from the users’ perspective. The results indicate that all evaluated tools enabled rapid prototype generation; however, significant differences emerged in usability, structural fidelity, and perceived control. Figma and Stitch achieved the highest usability scores and demonstrated greater alignment with the reference prototype (82.86 and 80.36, respectively). Visily achieved a favorable usability score (78.57), while Uizard obtained a moderate score (67.14). Although Uizard and Visily exhibited strong automation capabilities and faster initial generation, their outputs required additional manual refinement to achieve higher fidelity and customization. Participant feedback emphasized the importance of output quality, responsiveness, and foundational design knowledge in achieving satisfactory results. Overall, the findings suggest that current GAI-based prototyping tools are effective and valuable in real-world software development contexts. However, their effectiveness appears closely related to the degree of user control, responsiveness, and the ability to iteratively refine AI-generated interface components. Full article

Artificial intelligence has been considered as a transformative element capable of reshaping STEM education into equitable, resource-efficient, and scalable learning environments. However, realizing this potential requires striking a careful balance between technological innovation, pedagogical considerations, and ethical concerns. This study sought to examine the implementation of artificial intelligence (AI) tools by STEM university faculty members in Saudi Arabia to promote Sustainable Development Goal 4 (quality education). While doing so, the study attempted to explore how Saudi STEM university faculty members integrated AI tools in their instructional practices and analyze their perceptions towards these tools. To achieve these goals, the study employed an explanatory sequential mixed-methods design. In the first phase of data collection, a close-ended questionnaire was applied to a random sample of (324) STEM university faculty members. The second phase involved gathering qualitative data using a semi-structured interview administered to 12 purposively selected experts. Key quantitative findings revealed an overall AI integration at a medium level with a mean of (2.71) and standard deviation of (0.36) across three instructional practices, namely planning, implementation, and assessment. The highest integration level was in assessment (M = 2.93, medium) while the lowest was in planning (M = 2.61, medium). The results also revealed that the participants’ perceptions towards integrating AI tools were highly positive (M = 4.00, high), albeit with some concerns regarding the effect of excessive and unguided use of AI tools on students’ higher-order thinking skills, particularly the risk of AI functioning merely as an information delivery mechanism rather than serving its more pedagogically valuable role as a brainstorming scaffold. Furthermore, the study unveiled a number of barriers to integrating AI tools, including the weakness of digital infrastructure, lack of professional development, the limited credibility of AI-generated content, and ethical concerns related to academic integrity and copyrights. The research suggests the establishment of a sustainable digital environment by improving the infrastructure, providing specific training in accordance with the principles of sustainability, and implementing policies that promote equitable, transparent, and responsible integration of AI. These strategies can coordinate the growth of technology with the larger needs of the quality of education, inclusion, and sustainability of STEM education in the long term. Full article

This study aims to identify and synthesize the major global trends that shaped 2025 by applying the DeflyCompass methodology to a curated corpus of strategic foresight reports. The study synthesizes insights from 23 strategic reports published by leading international organizations, including the World Economic Forum, Accenture, Euromonitor, and major technology firms. Methodologically, DeflyCompass operationalizes a structured hybrid human–AI pipeline comprising the deployment of multi-agent AI systems, automated knowledge graph construction, semantic clustering, and hybrid human–AI validation processes, reducing an initial set of 816 preliminary signals to a validated catalog of 50 high-priority trends across six PESTEL domains: Political, Economic, Social, Technological, Environmental, and Legal/Governance. Key findings indicate that artificial intelligence functions as a systemic enabling technology across all domains, climate and sustainability imperatives permeate multiple domains, geopolitical fragmentation introduces systemic tension, and trust deficits emerge as a critical vulnerability. The study contributes a replicable and scalable framework for global-level strategic foresight that operationalizes human–AI integration within a rigorous expert-driven validation process, complementing existing hybrid analytical approaches in the literature. Implications extend to decision-making in technology governance, sustainability strategy, social adaptation, and scenario planning, highlighting the necessity of integrating AI augmentation with human expertise for effective future-oriented planning. Full article

Integrating artificial intelligence (AI) into design education is no longer optional; it has become an essential tool for enhancing innovative design and preparing students for data-driven practice and rapid technological acceleration. However, ignoring AI risks professional irrelevance; it introduces a range of concerns about students’ cognitive skills and comes with many drawbacks in the education process, as it threatens the attainment of learning outcomes, renders a fair assessment process unachievable, and places academic integrity in a vulnerable position. Using a qualitative case study approach, this research employs semi-structured interviews with 27 senior-year students in the interior design department to gain in-depth academic insights into how AI influenced their design process in their term project and its impact on their cognitive development and decision -making. Instructors’ observations on students’ skills, their pace in the project, and their end-products were documented. This study demonstrates that integrating AI into design education cannot be avoided, making a new paradigm for addressing design education inevitable. Based on the analysis, the paper proposes a conceptual framework outlining key dimensions in teaching and assessing strategies in design education adopting AI, focusing on analysis, critical thinking, reasoning, and process rather than on the end-product and its presentation. Full article