Universidad Israel

El desconocimiento del funcionamiento del negocio dentro de las organizaciones de desarrollo de software puede provocar interpretaciones erróneas de los requerimientos funcionales por parte de los desarrolladores, lo que genera retrasos en los proyectos y un aumento en los costos de desarrollo. En este contexto, las organizaciones buscan herramientas tecnológicas que faciliten el acceso al conocimiento organizacional y reduzcan la curva de aprendizaje del personal nuevo o en proceso de rotación. La presente investigación propone el desarrollo de un prototipo de chatbot inteligente entrenado con manuales técnicos y funcionales, capaz de proporcionar soporte técnico de primer nivel dentro de una organización de desarrollo de software. El objetivo del estudio es diseñar e implementar una herramienta basada en inteligencia artificial que permita mejorar el proceso de aprendizaje organizacional y optimizar la gestión del conocimiento. La metodología utilizada se fundamenta en el modelo de desarrollo en cascada, el cual permite estructurar el proceso en fases de definición de requisitos, diseño, construcción, pruebas e implementación. El sistema desarrollado emplea una arquitectura basada en recuperación aumentada por generación (RAG), integrando modelos de lenguaje de inteligencia artificial, bases de datos vectoriales y documentos corporativos como fuente de conocimiento. Los resultados evidencian que el chatbot permite consultar manuales técnicos, interpretar preguntas formuladas por los usuarios y generar respuestas contextualizadas basadas en documentos institucionales. Esta herramienta contribuye a mejorar la eficiencia del soporte técnico interno, reducir el tiempo de inducción del personal y fortalecer los procesos de transferencia de conocimiento dentro de la organización.

BackgroundHepatic cystic echinococcosis (HCE) remains a significant public health issue in endemic countries. Although recurrence is a recognized challenge, its independent impact on adverse clinical outcomes such as postoperative complications (POC), mortality, and length of hospital stay (LHS) remains poorly studied in Latin America. This study aimed to assess the risk of POC, mortality, and LHS in patients with recurrence of HCE.MethodsWe conducted a retrospective cohort study of patients who underwent surgery for HCE between 1993 and 2019 at two centers in southern Chile. Patients with recurrence (exposed group) were compared to those undergoing primary surgery (non-exposed group). The primary outcome was the presence of POC; secondary outcomes included mortality and LHS. Crude and adjusted relative risks (RR) with 95% confidence intervals were estimated using Poisson regression with robust errors. Linear regression models were applied to assess the effect of recurrence on LHS.ResultsA total of 154 patients with 271 cysts were included. Recurrence was identified in 43 patients (27.9%). POC occurred in 18.2% of the total cohort and were significantly more frequent in the recurrence group (41.9% vs. 9.0%, p < 0.001). Adjusted RR for POC in the presence of recurrence was 5.1 (95% CI 2.7–9.9). Mortality was higher in patients with recurrence (7.0% vs. 2.7%, RR: 2.6; 95% CI 0.5–12.3), though not statistically significant. LHS was 1 day longer in the recurrence group (7.3 ± 4.5 vs. 5.6 ± 3.4; p = 0.02), but this association lost significance in regression models.ConclusionsRecurrence of HCE increases the risk of POC. While trends toward higher mortality and prolonged LHS were observed, these did not reach statistical significance. These findings underscore the importance of long-term follow-up and the need to identify prognostic factors for recurrence to optimize outcomes in patients with HCE in endemic regions.Graphical abstract

This study presents a secure communication solution based on the Hub and Spoke model. The proposal uses GRE tunnels over IPsec and FortiGate firewalls in a simulation environment with GNS3, with the aim of evaluating its technical feasibility, efficiency in the use of resources and scalability. A virtual laboratory was designed that simulates a distributed business network. In this scenario, a central node manages encrypted traffic between branch offices, leveraging the encapsulation and security features offered by GRE and IPsec protocols. It was necessary to establish static and dynamic routes, security policies, authentication mechanisms and AES256 encryption, to demonstrate that this type of architecture guarantees the confidentiality and integrity of data, with optimal performance and easy management, becoming an optimal alternative to be implemented in small and mediumsized companies that have limited resources. The use of GNS3 made it possible to recreate real scenarios without the need to invest in expensive infrastructure, this makes it a secure solution, easily scalable for companies that are developing and that want to protect their communication and keep it fluid in all the offices they have.