Repositorio Universidad Europea del Atlántico
Not a member yet
2719 research outputs found
Sort by
End-to-end emergency response protocol for tunnel accidents augmentation with reinforcement learning
No full textAutonomous unmanned aerial vehicles (UAVs) offer cost-effective and flexible solutions for a wide range of real-world applications, particularly in hazardous and time-critical environments. Their ability to navigate autonomously, communicate rapidly, and avoid collisions makes UAVs well suited for emergency response scenarios. However, real-time path planning in dynamic and unpredictable environments remains a major challenge, especially in confined tunnel infrastructures where accidents may trigger fires, smoke propagation, debris, and rapid environmental changes. In such conditions, conventional preplanned or model-based navigation approaches often fail due to limited visibility, narrow passages, and the absence of reliable localization signals. To address these challenges, this work proposes an end-to-end emergency response framework for tunnel accidents based on Multi-Agent Reinforcement Learning (MARL). Each UAV operates as an independent learning agent using an Independent Q-Learning paradigm, enabling real-time decision-making under limited computational resources. To mitigate premature convergence and local optima during exploration, Grey Wolf Optimization (GWO) is integrated as a policy-guidance mechanism within the reinforcement learning (RL) framework. A customized reward function is designed to prioritize victim discovery, penalize unsafe behavior, and explicitly discourage redundant exploration among agents. The proposed approach is evaluated using a frontier-based exploration simulator under both single-agent and multi-agent settings with multiple goals. Extensive simulation results demonstrate that the proposed framework achieves faster goal discovery, improved map coverage, and reduced rescue time compared to state-of-the-art GWO-based exploration and random search algorithms. These results highlight the effectiveness of lightweight MARL-based coordination for autonomous UAV-assisted tunnel emergency response
Effects of lifestyle interventions in pregnancy on gestational diabetes: individual participant data and network meta-analysis
No full textObjectives To assess the effects of lifestyle interventions on gestational diabetes, determine whether the effects vary by maternal body mass index, age, parity, ethnicity, education level, or intervention, and rank interventions by effectiveness.
Design Individual participant data (IPD) and network meta-analysis.
Data sources Major electronic databases (January 1990 to April 2025).
Methods This meta-analysis included randomised trials on the effects of lifestyle interventions (physical activity based, diet based, or mixed) in pregnancy on gestational diabetes. Main outcomes were gestational diabetes defined by any criteria and by UK NICE (National Institute for Health and Care Excellence) criteria; other outcomes included IADPSG (International Association of Diabetes in Pregnancy Study Group) and modified IADPSG defined gestational diabetes. A two stage IPD meta-analysis estimated summary odds ratios and 95% confidence intervals and interactions (subgroup effects), along with absolute risk reduction estimates. Aggregate data from non-IPD trials were added to the meta-analysis when possible. Intervention effects were ranked using network meta-analysis.
Results 104 randomised trials (35 993 women) were included, with IPD for 68% of participants (24 391 women; 54 studies). Lifestyle interventions reduced gestational diabetes defined by any criteria by 10% in IPD trials (odds ratio 0.90, 95% confidence interval (CI) 0.80 to 1.02; absolute risk reduction 1.3%, 95% CI −0.3% to 2.6%), and by 20% when combining IPD and non-IPD trials (odds ratio 0.80, 95% CI 0.73 to 0.88; absolute risk reduction 2.6%, 95% CI 1.6% to 3.6%), and no reduction was observed using NICE criteria (odds ratio 0.98, 95% CI 0.84to 1.13). Lifestyle interventions reduced gestational diabetes defined using IADPSG criteria by 14% in IPD trials (odds ratio 0.86, 95% CI 0.75 to 0.97; absolute risk reduction 2.7%, 95% CI 0.6% to 5.0%) and by 18% when combining IPD and non-IPD trials (odds ratio 0.82, 95% CI 0.72 to 0.93; absolute risk reduction 3.5%, 95% CI 1.3% to 5.7%). Effects did not vary by maternal characteristics, except for education. Although women of all educational levels benefited from the intervention, the benefit was less in those with low education (low v middle interaction: odds ratio 0.68, 95% CI 0.51 to 0.90; low v high interaction: odds ratio 0.71, 95% CI 0.54 to 0.93). Benefits did not vary by intervention characteristics, except for greater effectiveness with group format (odds ratio 0.81, 95% CI 0.68 to 0.97; absolute risk reduction 2.5%, 95% CI 0.4% to 4.3%) and newly trained facilitators (odds ratio 0.82, 95% CI 0.69 to 0.96; absolute risk reduction 2.4%, 95% CI 0.5% to 4.2%). Physical activity based interventions ranked highest (mean rank 1.1, 95% CI 1 to 2) in preventing gestational diabetes.
Conclusions Lifestyle interventions in pregnancy are likely to prevent gestational diabetes, with effects varying according to diagnostic criteria. Implementation strategies should address inequalities by maternal education, and consider group formats, provider training, and physical activity based interventions to prevent gestational diabetes
Securing internet of things devices using a hybrid approach
Full text linkWith increased Internet of Things (IoT) devices, complexity and protection are more challenging. Lightweight cryptographic algorithms are secure and suitable for limited-resource environments; however, their hash functions provide encrypted data but not integrity. Strong security features are available, but setup is difficult and expensive. Network security mechanisms increase power consumption and latency. As IoT networks grow, managing cryptographic keys and securely authenticating large numbers of devices become complex tasks. Efficient key management strategies are required to ensure the scalability required. Existing state-of-the-art solutions lack standardization, scalability, complex and costly. Thus, this research proposes a secure solution for IoT resource-constrained devices, combining strong data integrity and lightweight encryption, and is thus named a hybrid. This hybrid approach integrates SHA-512 and the present cipher in our proposed approach and thus ensuring higher security than state-of-the-art models. This intelligent combination not only enhances the algorithm’s resistance against cryptographic attacks but also improves its processing speed. The proposed approach is used to reduce the processing time for encryption in the IoT platform and to preserve the trade-off between security and efficiency. In terms of memory use, execution time, and precision, the proposed approach is compared with recent state-of-the-art research. The experimental results indicate that our approach is efficient using the avalanche, authentication success rate, collision events, and execution time. The efficiency is 53% to 65%, and the avalanche effect indicates sensitivity to input variations, suggesting moderate-to-considerable reactivity to small data changes. The experimental tests conducted across 10,000 and 80,000 runs reveal no collisions and found that the proposed approach is resilient in managing unique IDs. Moreover, our approach performs consistently, with an average execution time of 0.088246 s, ranging from 0.075954 to 0.094583 s. Finally, our approach provides a practical and scalable solution for securing IoT devices in resource-constrained environments, addressing practical problems for IoT devices
Tecnologías para la creación de simuladores virtuales de aprendizaje basados en asistentes conversacionales
No full textLa línea de actividad de I+D que se propone se orienta a tecnologías educativas para
la creación de simuladores virtuales de aprendizaje basados en asistentes
conversacionales.
La iniciativa se encuadra totalmente en la digitalización de servicios y en concreto en
la digitalización de servicios educativos con aplicación inicial en el ámbito de la
sanidad.
Las tecnologías educativas engloban un espectro amplio de conocimientos y
aplicaciones que se centran en integrar de forma efectiva las herramientas
tecnológicas en el ámbito educativo. Este campo no sólo se orienta a la implantación
de soluciones mediante la digitalización sino que también promueve un cambio
paradigmático en la conceptualización de la enseñanza. Este cambio se manifiesta en
varias dimensiones clave del proceso educativo, resaltando la importancia de la
personalización del aprendizaje, el acceso universal al conocimiento y el desarrollo
de habilidades esenciales para el siglo XXI. Estos elementos cruciales evidencian la
relevancia de integrar tecnologías digitales en la educación.
La transición hacia la era digital está reformulado profundamente el panorama
educativo, especialmente en la creación y distribución de recursos didácticos cada
vez más ricos e interactivos. En este contexto, los sistemas virtuales simulados
permiten mejorar la accesibilidad y disponibilidad de los medios educativos, y así por
ejemplo en el campo de la salud, permiten reducir el riesgo de errores frente a los
pacientes, en las fases de aprendizaje. Estos sistemas han avanzado enormemente,
evolucionando desde simples sistemas que reproducen condiciones particulares y
previamente definidas hasta soluciones, que incorporando la inteligencia artificial, se
acercan más si cabe a poder reproducir situaciones adaptativas y realistas.
Así, la oportunidad tecnológica que se pretende aprovechar en el ámbito educativo
se refiere a las nuevas herramientas disponibles, en especial de los modelos de
aprendizaje automático y LLMs (Large Language Models) y las técnicas de
procesamiento de lenguaje natural (NLP o natural language processing).
Para llevar a cabo el objetivo general, se han definido los siguientes objetivos
específicos:
OE1. Analizar herramientas disponibles de inteligencia artificial aplicables a un
entorno de asistentes conversacionales.
OE2. Definir las funcionalidades de un sistema experimental piloto.
OE3. Establecer la arquitectura tecnológica y de servicios para optimizar la eficiencia
de la plataforma y facilitar su interoperabilidad.
OE4. Desarrollar un simulador digital prototipo.
OE5. Validar el funcionamiento del prototipo implementado
Learning English in the Digital Age: The Application of ChatGPT and MagicSchool at Istinye University
No full textIn the current context of globalization, proficiency in English has become an indispensable skill for academic and professional success. Aware of the challenges in acquiring a new language, Istinye University has adopted innovative approaches to improve English learning. This article focuses on the implementation of review sessions for English students, using artificial intelligence tools such as ChatGPT and MagicSchool during the 2023/2024 academic year. Adequate preparation for level exams, aligned with the Common European Framework of Reference for Languages, is essential to ensure the acquisition of linguistic competencies. AI tools allow for personalized learning, facilitating the design of materials tailored to levels A2, B1, and B2. This approach not only enhances the effectiveness of learning but also fosters student autonomy and engagement. The use of AI offers benefits to teachers, such as reducing administrative burdens and the ability to provide immediate feedback. However, the adoption of these technologies also presents challenges, such as the need for proper training for educators. The findings suggest that artificial intelligence can transform language teaching, improving the learning experience. Recommendations for future research are proposed, emphasizing the importance of exploring both the benefits and limitations of using AI in educational settings. This study opens new perspectives for English teaching, highlighting the relevance of combining traditional methods with advanced technological tools
Novel Transfer Learning Approach for Detecting Infected and Healthy Maize Crop Using Leaf Images
No full textMaize is a staple crop worldwide, essential for food security, livestock feed, and industrial uses. Its health directly impacts agricultural productivity and economic stability. Effective detection of maize crop health is crucial for preventing disease spread and ensuring high yields. This study presents VG-GNBNet, an innovative transfer learning model that accurately detects healthy and infected maize crops through a two-step feature extraction process. The proposed model begins by leveraging the visual geometry group (VGG-16) network to extract initial pixel-based spatial features from the crop images. These features are then further refined using the Gaussian Naive Bayes (GNB) model and feature decomposition-based matrix factorization mechanism, which generates more informative features for classification purposes. This study incorporates machine learning models to ensure a comprehensive evaluation. By comparing VG-GNBNet's performance against these models, we validate its robustness and accuracy. Integrating deep learning and machine learning techniques allows VG-GNBNet to capitalize on the strengths of both approaches, leading to superior performance. Extensive experiments demonstrate that the proposed VG-GNBNet+GNB model significantly outperforms other models, achieving an impressive accuracy score of 99.85%. This high accuracy highlights the model's potential for practical application in the agricultural sector, where the precise detection of crop health is crucial for effective disease management and yield optimization
Nut Consumption Is Associated with Cognitive Status in Southern Italian Adults
No full textBackground: Nut consumption has been considered a potential protective factor against cognitive decline. The aim of this study was to test whether higher total and specific nut intake was associated with better cognitive status in a sample of older Italian adults. Methods: A cross-sectional analysis on 883 older adults (>50 y) was conducted. A 110-item food frequency questionnaire was used to collect information on the consumption of various types of nuts. The Short Portable Mental Status Questionnaire was used to assess cognitive status. Multivariate logistic regression analyses were performed to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for the association between nut intake and cognitive status after adjusting for potential confounding factors. Results: The median intake of total nuts was 11.7 g/day and served as a cut-off to categorize low and high consumers (mean intake 4.3 g/day vs. 39.7 g/day, respectively). Higher total nut intake was significantly associated with a lower prevalence of impaired cognitive status among older individuals (OR = 0.35, CI 95%: 0.15, 0.84) after adjusting for potential confounding factors. Notably, this association remained significant after additional adjustment for adherence to the Mediterranean dietary pattern as an indicator of diet quality, (OR = 0.32, CI 95%: 0.13, 0.77). No significant associations were found between cognitive status and specific types of nuts. Conclusions: Habitual nut intake is associated with better cognitive status in older adults
Tensiomyography, functional movement screen and counter movement jump for the assessment of injury risk in sport: a systematic review of original studies of diagnostic tests
No full textBackground: Scientific research should be carried out to prevent sports injuries. For this purpose, new assessment technologies must be used to analyze and identify the risk factors for injury. The main objective of this systematic review was to compile, synthesize and integrate international research published in different scientific databases on Countermovement Jump (CMJ), Functional Movement Screen (FMS) and Tensiomyography (TMG) tests and technologies for the assessment of injury risk in sport. This way, this review determines the current state of the knowledge about this topic and allows a better understanding of the existing problems, making easier the development of future lines of research.
Methodology: A structured search was carried out following the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines and the PICOS model until November 30, 2024, in the MEDLINE/PubMed, Web of Science (WOS), ScienceDirect, Cochrane Library, SciELO, EMBASE, SPORTDiscus and Scopus databases. The risk of bias was assessed and the PEDro scale was used to analyze methodological quality.
Results: A total of 510 articles were obtained in the initial search. After inclusion and exclusion criteria, the final sample was 40 articles. These studies maintained a high standard of quality. This revealed the effects of the CMJ, FMS and TMG methods for sports injury assessment, indicating the sample population, sport modality, assessment methods, type of research design, study variables, main findings and intervention effects.
Conclusions: The CMJ vertical jump allows us to evaluate the power capacity of the lower extremities, both unilaterally and bilaterally, detect neuromuscular asymmetries and evaluate fatigue. Likewise, FMS could be used to assess an athlete's basic movement patterns, mobility and postural stability. Finally, TMG is a non-invasive method to assess the contractile properties of superficial muscles, monitor the effects of training, detect muscle asymmetries, symmetries, provide information on muscle tone and evaluate fatigue. Therefore, they should be considered as assessment tests and technologies to individualize training programs and identify injury risk factors
Análisis de la vida útil en tortillas
No full textEste proyecto ha sido desarrollado para determinar la vida útil de la tortilla de patatas fabricada por la empresa Regma mediante un riguroso análisis microbiológico. La metodología implementada comprende tres estudios independientes que analizan la evolución microbiológica del producto desde su elaboración hasta 8-10 horas después. El protocolo incluye el recuento total de aerobios mesófilos a intervalos regulares con muestras por duplicado, complementado con análisis de coliformes totales. Adicionalmente, se realiza un estudio específico de mohos y levaduras en los mismos puntos temporales. Esta caracterización microbiológica permite establecer parámetros objetivos sobre la seguridad alimentaria del producto y optimizar tanto sus condiciones de conservación como el tiempo máximo recomendado para su consumo, asegurando así la calidad de los consumidores
A flexible and lightweight signcryption scheme for underwater wireless sensor networks
Full text linkUnderwater wireless sensor networks (UWSNs) are a new research area gaining popularity. It has several key applications for instance; marine monitoring, surveillance, environmental sensing, etc. However, It has several challenges including security, node mobility, limited bandwidth, and high error rates. Thus, to solve these issues, herein we propose a new lightweight Signcryption scheme for UWSNs. The proposed scheme effectively balances computational complexity and enhances the security of UWSNS. In contrast to the other state-of-the-art cryptographic schemes, the proposed scheme consists of a single combined operation of encryption and signing processes, which significantly improves its computational and communicational performance to ensure confidence when transmitting data. We performed the experimentation, and the experimental results show that the proposed scheme performs well compared to the state-of-the-art model. In addition, the experimental results revealed that the proposed scheme had a 40% less computational cost, 30% less energy consumption, and 25% less communication overhead than the state-of-the-art methods. This makes the proposed scheme highly appropriate for resource-scarce UWSNs. The proposed scheme also showed good scalability, where the performance could be sustained from a small-scale network of 50 nodes to a bandwidth of 200 nodes. Further, the proposed model also kept the security and latency low for the mobile nodes in an environment with high node mobility over the underwater terrain. In addition, the proposed method ensures flexibility and scalability by offering compatibility with diverse network structures and seamless integration with various cryptographic approaches, making it adaptable for dynamic underwater environments and broader applications such as IoT and smart city networks