Search Results (751)

Efficient routing optimization remains a pivotal challenge in Electronic Design Automation (EDA), as it profoundly influences circuit performance, power consumption, and manufacturing cost. The Rectilinear Steiner Minimum Tree (RSMT) problem plays a crucial role in this process by minimizing the routing length through the introduction of Steiner points. This paper proposes a reinforcement learning-driven RSMT construction model that incorporates a novel Selective Kernel Transformer Network (SKTNet) encoder to enhance feature representation. SKTNet integrates a Selective Kernel Convolution (SKConv) and an improved Macaron Transformer to improve multi-scale feature extraction and global topology modeling. Additionally, Self-Critical Sequence Training (SCST) is employed to optimize the policy by leveraging a greedy-decoded baseline sequence for the advantage computation. Experimental results demonstrate superior performance over state-of-the-art methods in wirelength optimization. Ablation studies further validate the contribution of this model, highlighting its effectiveness and scalability for routing. Full article

Background: The widespread use of social media among children and adolescents has raised increasing concern about its potential impact on mental health. Given the unique neurodevelopmental vulnerabilities during adolescence, understanding how digital platforms influence psychiatric outcomes is critical. Objectives: This narrative review aims to synthesize current evidence on the relationship between social media exposure and key psychiatric symptoms in youth, including depression, anxiety, body image disturbances, suicidality, and emotional dysregulation. Methods: We conducted a comprehensive narrative review of the literature, drawing from longitudinal, cross-sectional, and neuroimaging studies published in peer-reviewed journals. Specific attention was given to moderators (e.g., age, gender, and personality traits) and mediators (e.g., sleep, emotion regulation, and family context) influencing the relationship between social media use and mental health outcomes. Results: Evidence indicates that certain patterns of social media use, especially passive or compulsive engagement, are associated with increased risk of depression, anxiety, body dissatisfaction, and suicidal ideation. Adolescent girls, younger users, and those with low self-esteem or poor emotional regulation are particularly vulnerable. Neuroimaging studies show that social media activates reward-related brain regions, which may reinforce problematic use. Family support and digital literacy appear to mitigate negative effects. Conclusions: Social media use is not uniformly harmful; its psychological impact depends on how, why, and by whom it is used. Multilevel prevention strategies, including media education, parental involvement, and responsible platform design, are essential to support healthy adolescent development in the digital age. Full article

Timber structures and structural members have undergone rapid development in recent decades and are now fully competitive with traditional structures made of reinforced concrete or structural steel in many areas. Low self-weight, high durability, rapid construction assembly, and a favourable environmental footprint predispose timber structures for wider future use. A persisting drawback is the often-complicated joining of individual elements, especially when moment resistance is required. For CLT panels, this issue is more urgent due to their relatively small thickness and cross-laminated lay-up. This paper presents experimental research investigating parameters related to the actual behaviour of a moment-resisting embedded joint of CLT panels. The test programme consisted of four series (12 specimens) loaded in four-point bending to failure. The proposed and tested joint consists of high-strength steel rods glued into the two connected parts of the CLT panel. In addition to a detailed investigation of the resistance and stiffness of the joint, this research evaluates the effect of composite action with a reinforced-concrete slab on the performance of this type of joint. The experimental results and their detailed analysis are also extended to propose a framework concept for creating a theoretical (mechanical) model based on the component method. Full article

Recently, cognitive training programs using digital content with visuoperceptual stimulation have been developed and commercialized. In particular, digital intervention content for children with attention deficit hyperactivity disorder (ADHD) has been developed as games, enhancing motivation and accessibility for the target population. Active stimulation is required to elicit positive effects on self-regulation training, including attention control and impulse inhibition, through task-based content. Common forms of stimulation include emotional stimuli, such as praise and encouragement, and economic stimuli based on a self-directed token economy system. Economic stimulation can serve as active reinforcement because the child directly engages as the primary agent within the task content. This study applied and validated a token economy intervention using digital therapeutic content in children with ADHD. Behavioral assessments were conducted using the Comprehensive Attention Test (CAT) and the Korean version of the Child Behavior Checklist (K-CBCL). The developed digital intervention content implemented a user-centered token economy based on points within the program. In the CAT Flanker Task, the experimental group (0.84 ± 0.40) showed significantly higher sensitivity factor scores than the control group (0.72 ± 0.59) after 4 weeks, with a large effect size (F = 4.76, p = 0.038, partial η² = 0.150). Additionally, the rate of change in externalizing behavior scores on the K-CBCL showed a significant difference between the two groups (t = 2.35, p = 0.026, Cohen’s d = 0.860), demonstrating greater improvement in externalizing symptoms in the experimental group than in the control group. Therefore, this study suggests that the participant-centered implementation model using token economy mechanisms in digital intervention content may serve as a novel and effective therapeutic approach for children with ADHD. Full article

This study explores a novel approach of using large language models (LLMs) in the real-time Proportional–Integral–Derivative (PID) control of a physical system, the Quanser QUBE-Servo 2. We investigated whether LLMs, used with an Artificial Intelligence (AI) agent workflow platform, can participate in the live tuning of PID parameters through natural language instructions. Two AI agents were developed: a control agent that monitors the system performance and decides if tuning is necessary, and an Optimizer Agent that updates PID gains using either a guided system prompt or a self-directed free approach within a safe parameter range. The LLM integration was implemented through Python programming and Flask-based communication between the AI agents and the hardware system. Experimental results show that LLM-based tuning approaches can effectively reduce standard error metrics, such as IAE, ISE, MSE, and RMSE. This study presents one of the first implementations of real-time PID tuning powered by LLMs, and it has the potential to become a novel alternative to classical control, as well as machine learning or reinforcement learning-based approaches. The results are promising for using agentic AI in heuristic-based tuning and the control of complex physical systems, marking the shift toward more human-centered, explainable, and adaptive control engineering. Full article

This article explores how gender biases in STEM education persist despite formal commitments to equality. Based on data from the Erasmus+ project STEMGenderIN, we analyze responses from lower-secondary school teachers (ISCED 2; ages 11–15), of STEM subjects, in Portugal, Italy, Belgium, and Romania using the TPGESE scale, which assesses three dimensions: perceived gender equality in education (PGEE), the awareness of the effects of gender segregation (AEGSE), and the naturalization of gender stereotypes (GSNGI). Findings show a consistent gap between teachers declared support for gender equality and their limited awareness of structural and cultural barriers faced by girls in STEM. While most teachers affirm equality in principle, many attribute girls’ underrepresentation to personal choice or aptitude, overlooking the influence of stereotypes, social expectations, and systemic inequalities. The results point to a paradox: formal recognition of gender equality coexists with low engagement in reflexive practice or institutional change. Differences between countries suggest varying degrees of critical awareness, with some contexts showing greater openness to questioning dominant narratives. This study highlights the urgent need for teacher training that goes beyond rhetoric, promoting deep pedagogical transformation and equipping educators to create more inclusive STEM learning environments. We argue that addressing the perception–practice gap is essential to closing the gender gap in STEM. To situate these findings, we also note how national cultural–political debates—such as Portugal’s public controversy around so-called “gender ideology” in Citizenship and Development—may shape teachers’ perceptions and self-reports, reinforcing the need for context-aware training. Full article

Increasing environmental threats and accelerating effects of climate change serve to reinforce the perception of environmental safety not only as an ecological concern but also as a social and economic one. The extant research suggests that environmental factors have a significant impact on health and quality of life. However, the literature still lacks comprehensive analysis integrating objective environmental indicators with subjective measures of quality of life in a comparative and dynamic framework, particularly in European Union (EU) countries. The primary objective of this paper is to evaluate the environmental safety within European Union countries and its impact on self-perceived quality of life and health. The analysis integrates the multidimensional environmental indicators with subjective assessments of quality of life. To this end, the TOPSIS method is employed to construct a synthetic index for environmental safety (ES_Score). Moreover, pooled cross-sectional time-series regressions are utilised for formal analyses. The study encompasses data from 27 EU countries from 2018 to 2023. The findings of the study suggest that environmental safety exhibits considerable variation among EU countries yet remains relatively stable over time. This underscores the enduring nature of environmental advantages and deficits. Countries with superior environmental safety are also those which have been shown to exhibit a higher quality of life and better health. Proactive environmental investments and activities aimed at sustainable growth have the capacity to improve the quality of life of the population. However, some factors, such as high air emission intensity or excessive water abstraction may be challenging. The findings of this study demonstrate a significant relationship between environmental protection initiatives and social prosperity within European nations, thus offering valuable insights that can inform the development of public policy. Full article

The rapid expansion of communication networks and increasingly complex service demands have presented significant challenges to the intelligent management of network resources. To address these challenges, we have proposed a network self-optimization framework integrating the predictive capabilities of the Large Language Model (LLM) with the decision-making capabilities of multi-agent Reinforcement Learning (RL). Specifically, historical network traffic data are converted into structured inputs to forecast future traffic patterns using a GPT-2-based prediction module. Concurrently, a Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm leverages real-time sensor data—including link delay and packet loss rates collected by embedded network sensors—to dynamically optimize bandwidth allocation. This sensor-driven mechanism enables the system to perform real-time optimization of bandwidth allocation, ensuring accurate monitoring and proactive resource scheduling. We evaluate our framework in a heterogeneous network simulated using Mininet under diverse traffic scenarios. Experimental results show that the proposed method significantly reduces network latency and packet loss, as well as improves robustness and resource utilization, highlighting the effectiveness of integrating sensor-driven RL optimization with predictive insights from LLMs. Full article

Pharmacists are increasingly expected to deliver medication therapy management (MTM) services, yet many pharmacy students report insufficient confidence and preparedness in executing these complex tasks. This study evaluated a scaffolded MTM instructional series integrated into a second-year pharmacy skills laboratory, aiming to enhance student competence through progressive, structured learning. A mixed-methods design assessed changes in self-reported confidence, performance-based outcomes, and reflective insights among 154 students across three educational tracks. The 14-week intervention included sequential activities such as medication history interviews, drug-related problem (DRP) identification, care plan development, and comprehensive MTM simulations. Pre- and post-intervention surveys revealed statistically significant improvements in all 18 confidence domains, with the greatest gains in therapeutic recommendations and prescriber communication. Effect sizes ranged from small to very large (Cohen’s d 0.33–1.05), indicating gains that were both statistically reliable and educationally meaningful. Performance assessments showed consistent proficiency across MTM components, with average scores ranging from 90% to 96%. Qualitative reflections reinforced these findings, highlighting growth in communication, individualized patient care, and professional identity formation. The scaffolded approach aligns with accreditation standards and instructional design theory, offering a model for pharmacy curricula. Despite limitations such as lack of a comparator group and potential response bias, the study demonstrates that scaffolded MTM instruction effectively supports skill acquisition and confidence, preparing students for real-world clinical practice. Full article

Introduction: Stroke remains a major cause of disability worldwide, and precise identification of stroke lesions is essential for prognosis and rehabilitation planning. Machine learning has emerged as a powerful tool for automating stroke lesion segmentation and outcome prediction; however, these tasks are often studied in isolation. The two strategies are inherently interdependent since segmentation provides lesion-based features that directly inform prediction models. Methods: This narrative review synthesises studies published between 2010 and 2024 on the application of machine learning in stroke lesion segmentation and recovery forecasting. A total of 23 relevant studies were reviewed, including 10 focused on lesion segmentation and 13 on recovery prediction. Results: Convolutional Neural Networks (CNNs), including architectures such as U-Net, have improved segmentation accuracy on the Anatomical Tracings of Lesions After Stroke (ATLAS) V2 dataset; however, dataset bias and inconsistent evaluation metrics limit comparability. Integrating imaging-derived lesion characteristics with clinical features improves predictive accuracy at a higher level. Furthermore, semi-supervised and self-supervised methods enhanced performance where annotated datasets are scarce. Discussion: The review highlights the interdependence between segmentation and outcome prediction. Reliable segmentation provides biologically meaningful features that underpin recovery forecasting, while prediction tasks validate the clinical relevance of segmentation outputs. This bidirectional relationship underlines the need for unified pipelines integrating lesion segmentation with outcome prediction. Future research can improve generalisability and foster clinically robust models by advancing semi-supervised and self-supervised learning, bridging the gap between automated image analysis and patient-centred prognosis. Conclusion: Accurate lesion segmentation and outcome prediction should be viewed not as separate goals but as mutually reinforcing components of a single pipeline. Progress in segmentation strengthens recovery forecasting, while predictive modelling emphasises the clinical importance of segmentation outputs. This interdependence provides a pathway for developing more effective, generalisable, and relevant AI-driven stroke care tools. Full article

As infrastructure construction expands, the massive consumption of traditional concrete materials has led to resource shortages and environmental pollution. Utilizing industrial wastes such as coal gangue and fly ash to produce high-performance concrete is an important pathway toward a greener construction industry. However, concrete incorporating high volumes of fly ash and coal gangue (i.e., high-volume fly-ash coal-gangue self-compacting concrete, CGSC) suffers from low strength and high brittleness due to the inherent deficiencies of its constituents. This study proposes using glass fiber-reinforced polymer (GFRP) tubes for external confinement to improve the axial compressive capacity and deformability of CGSC. A total of 27 concrete cylinders were prepared and tested under axial compression, with real-time acoustic emission (AE) monitoring. The variables examined include the coarse aggregate type (coal-gangue and natural gravel), GFRP tube thickness (5 mm and 8 mm), and fly-ash content (80%, 85%, 90%). The stress–strain response of each specimen and the failure evolution of internal cracks were recorded throughout the loading process. The results show that GFRP tube confinement markedly increases the axial strength and ductility of CGSC. AE features exhibited staged behavior that closely mirrored the stress–strain curves. This correspondence reveals the progression of internal cracks under confinement and indicates that AE is an effective tool for damage monitoring in such composites. The findings provide a new technical approach for the efficient reuse of solid waste in concrete and offer a theoretical and practical basis for applying FRP composite structures in underground support engineering. Full article

Conventional strengthening measures for existing structures are usually not effective for the self-weight, which accounts for around 70% of the total load in reinforced concrete structures. Therefore, their effect on the overall load-bearing capacity is low. A self-weight-effective alternative for flexural strengthening is the thermal prestressing of additional reinforcement installed on the structure. In this method, reinforcing bars are slotted into the tensile zone, embedded in filler material, and tempered from the outside. They are thermally stretched, and once cooling starts, the bond with the hardened filler prevents re-deformation. The induced prestressing force counteracts dead loads and relieves the tensile zone, making the additional bars effective for the self-weight. In this paper, the effectiveness of the strengthening method is experimentally investigated in the serviceability and the ultimate limit states. Experiments involve strengthening a reinforced concrete beam under load by a thermally prestressed additional bar. Moreover, two reference tests are made to evaluate the method. An unstrengthened beam characterizes the lower capacity limit. Another beam with the same reinforcement amount as the strengthened one, but completely installed at casting, serves as the upper benchmark. All beams are loaded until bending failure. The strengthening method is assessed by means of the load-bearing behavior, deflection, crack development, and the strains in the initial as well as the added reinforcement. The results demonstrate the effectiveness of the strengthening method. The thermally prestressed bar achieves an effective pre-strain of approximately. 0.4‰ by heating at about 70 °C. The induced prestressing force and associated compression reduce tensile cracks by approx. 45% and increase stiffness. The strengthened beam reaches the maximum load of the upper benchmark, but with about 33% less deflection. The filler, which also expands thermally, generates an additional prestressing force that is effective up to about 20% of the load capacity. Beyond this, the filler begins to crack and its effect decreases, but the pre-strain in the reinforcing bar remains until maximum load. Full article

Adolescence is a crucial stage of development in which numerous habits that will shape future health are established. Participation in physical and sport activity is recognized as a key factor not only for improving physical condition but also for psychological and social well-being. However, its practice tends to decline during this vital stage. In this line, Self-Determination Theory (SDT) emerges as a useful approach to understand and promote quality motivation in sports practice. The aim of this study is to examine, through a scoping review, the effects of interventions based on SDT principles on variables related to motivation and well-being in adolescents. This scoping review was based on the PRISMA quality criteria, using the 14 databases included in the Web of Science platform. A total of 10 open access articles published in English and Spanish between 2021–2025 met inclusion criteria, with diverse designs and applied in school, family, and clinical contexts. The results reveal that interventions supporting autonomy, reinforcing competence, and fostering interpersonal relationships produce positive effects on self-determined motivation, active engagement, perceived well-being, and the intention to remain physically active. These findings support the importance of designing programs that are sensitive to the motivational context of adolescents, aimed at holistic development and the consolidation of active habits that become sustainable over time. Full article

Porous composites utilize their unique pore structures to effectively store and release lubricants, providing a fundamental mechanism for continuous lubrication in self-lubricating bearing cages. This study investigates carbon nanotube (CNT)-reinforced PEEK porous composites fabricated by fused deposition modeling (FDM) and subsequently subjected to heat treatment to improve tribological properties. Results show that incorporating 3 wt% CNTs significantly enlarges average pore size from 0.08 μm to 11.62 μm and increases porosity, resulting in an oil retention rate exceeding 80%. The composites achieve a 26.4–63.4% reduction in friction coefficient under dry sliding conditions at room temperature. After heat treatment, the material maintains a stable friction coefficient below 0.30 during high-temperature dry friction, demonstrating excellent lubricant slow-release capability and thermal stability. Full article

Cracks in concrete are a persistent issue that compromises structural durability, increases maintenance costs, and poses environmental challenges. Self-healing concrete has emerged as a promising innovation to address these concerns by autonomously sealing cracks and restoring integrity. This review focuses on two primary healing mechanisms: autogenous healing and microbial-induced calcite precipitation (MICP), the latter involving the biomineralization activity of bacteria, such as Bacillus subtilis and Sporosarcina pasteurii (formerly known as B. pasteurii). This review explores the selection, survivability, and activity of these microbes within the alkaline concrete environment. Additionally, the review highlights the role of fiber-reinforced cementitious composites (FRCCs), including high-performance fiber-reinforced cement composites (HPFRCCs) and engineered cement composites (ECCs), in enhancing crack control and enabling more effective microbial healing. The hybridization of natural and synthetic fibers contributes to both improved mechanical properties and crack width regulation, key factors in facilitating bacterial calcite precipitation. This review synthesizes current findings on self-healing efficiency, fiber compatibility, and the scalability of bacterial healing in concrete. It also evaluates critical parameters, such as healing agent integration, long-term performance, and testing methodologies, including both destructive and non-destructive techniques. By identifying existing knowledge gaps and performance barriers, this review offers insights for advancing sustainable, fiber-assisted microbial self-healing concrete for resilient infrastructure applications. Full article