The optimal deployment of Low-Power Wide-Area Networks (LPWANs) such as LoRaWAN in complex urban environments remains an NP-Hard Set Covering Problem. Traditional network planning often relies on 2D mathematical grids that ignore physical RF barriers, leading to topographic shadowing and single points of failure. This research proposes the Native 3D Memetic Spatially Aware Genetic Algorithm (3D-M-SAGA), an optimization framework that operates over a Morphological Digital Twin. By fusing OpenStreetMap (OSM) vector topologies with NASA SRTM elevation data and autonomous urban clutter classification, the framework evaluates physical constraints—including ITU-R knife-edge diffraction and dielectric absorption—directly within the evolutionary loop. To counteract the epistatic variance inherent to standard genetic algorithms, the 3D-M-SAGA integrates a vectorized memetic “Smart Repair” operator driven by heuristic attraction and repulsion forces. Formulated as a multi-objective optimization problem balancing Capital Expenditure (CAPEX) and topological Quality of Service (QoS) through K-coverage, the framework is evaluated using a 36-scenario parametric grid search and a 50-iteration Monte Carlo benchmark. Results show that the 3D-M-SAGA tightly bounds stochastic CAPEX variance ($\sigma =\pm 0.51$ gateways) while reducing single-point-of-failure network fragility ($K=1$) by up to 20%, guaranteeing fault tolerance ($K\ge 2$) without over-provisioning civic infrastructure.

Epilepsy, one of the most common neurological disorders worldwide, significantly affects the quality of life of millions of people due to its unpredictable seizure episodes. Early detection of seizures becomes a critical area to improve the diagnosis, treatment, and monitoring of the disease. This study conducts a systematic review and a detailed bibliometric analysis of recent research focused on the use of advanced machine learning (ML) techniques for the early detection of epileptic seizures. This study analyzes relevant scientific literature, obtained through an exhaustive search in the Scopus database, covering the period 2020-2024. The PRISMA methodology was applied to ensure transparency and rigor at each phase of the process. The results show a significant increase in scientific production on this topic, highlighting the predominant use of hybrid techniques that combine convolutional neural networks (CNNs), support vector machines (SVMs), and geometric deep learning methods, demonstrating high effectiveness in real clinical settings. In addition, a growing trend was identified toward the use of wearable devices and multimodal analysis, offering innovative and highly accurate solutions for real-time early alerts. The research also revealed persistent challenges in effective clinical implementation, such as interindividual variability, limited availability of labeled data, and the need for robust validation under real-world conditions.

La digitalización del sector fitness ha impulsado la adopción de tecnologías orientadas a la personalización de la experiencia del usuario, en Ecuador, la gestión de planes de entrenamiento y alimentación continúa siendo mayoritariamente manual, lo cual limita la escalabilidad del servicio, incrementa la dependencia del entrenador y dificulta la estandarización del seguimiento. Por esta razón, en este artículo se presenta el desarrollo e integración de un chatbot basado en inteligencia artificial dentro de un aplicativo web orientado a la administración de rutinas y planes alimenticios personalizados. El sistema fue implementado utilizando Laravel 12, Livewire 3, Blade, MVVM y la API Gemini 2.0 Flash como modelo de lenguaje natural y para la evaluación de los resultados se aplicó una encuesta a 25 usuarios, así como la ejecución de pruebas funcionales tipo caja negra. En consecuencia, los resultados permiten validar la factibilidad técnica y perceptual de la integración de IA generativa en un entorno fitness real, en condiciones controladas en eficiencia operativa, accesibilidad a la información y autonomía del usuario final, lo que demostró que la inteligencia artificial permite generar recomendaciones contextualizadas con respuestas en tiempo real. Con base a lo anterior, se concluye que el modelo de integración propuesto es viable, replicable y representa una referencia práctica para la transición de gimnasios tradicionales hacia servicios fitness inteligentes.