Sistema inteligente de apoyo al diagnóstico de preeclampsia en mujeres embarazadas en zonas rurales de Santa Cruz, Cajamarca
Cargando...
Fecha
2026
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Católica Santo Toribio de Mogrovejo
Resumen
En el presente trabajo se evidencia el desarrollo de un sistema inteligente de apoyo al diagnóstico de preeclampsia en mujeres embarazadas que residen en zonas rurales en un centro de salud en Santa Cruz. Para ello, se realizaron encuestas dirigidas a especialistas y gestantes para conocer cómo tradicionalmente se detecta la preeclampsia y cómo la asistencia a los controles prenatales se vinculan a esta. Por otro lado, para cumplir con los objetivos se siguió la metodología del ciclo de vida de Machine Learning la cual permitió el análisis de algoritmos que se relacionen con el proyecto; tomando en cuenta características como, algoritmos de clasificación con tipo de aprendizaje supervisado; del mismo modo, la metodología del Ciclo de vida Modelo Mobile Sprint (MMS) que permitió la integración del modelo en una aplicación móvil y la implementación de la interfaz de usuario. Posteriormente, se validó el modelo seleccionado para la predicción de preeclampsia, para ello, el algoritmo seleccionado fue Random Forest con un porcentaje de precisión del 85%. Finalmente, una gestante y una obstetra utilizaron el sistema inteligente, para así poder evaluar su usabilidad y calidad del producto, esto mediante una encuesta basado en la ISO25010. Por tal motivo, la implementación de este sistema inteligente ha permitido el diagnóstico temprano de preeclampsia en gestantes que residen en zonas rurales, debido a que estas tienden a tener limitaciones de acceso a controles prenatales tradicionales, haciendo que la detección de enfermedades no se detecte a tiempo.
This paper demonstrates the development of an intelligent system to support the diagnosis of preeclampsia in pregnant women living in rural areas at a health center in Santa Cruz. To do this, surveys were conducted for specialists and pregnant women to find out how preeclampsia is traditionally detected and how attendance at prenatal check-ups is linked to it. On the other hand, to meet the objectives, the Machine Learning life cycle methodology was followed, which allowed the analysis of algorithms that are related to the project; taking into account characteristics such as classification algorithms with a supervised learning type; likewise, the Mobile Sprint Model Life Cycle (MMS) methodology that allowed the integration of the model in a mobile application and the implementation of the user interface. Subsequently, the selected model was validated for the prediction of preeclampsia; for this, the selected algorithm was Random Forest with an accuracy percentage of 85%. Finally, a pregnant woman and an obstetrician used the smart system to evaluate its usability and product quality, through a survey based on ISO25010. For this reason, the implementation of this smart system has allowed the early diagnosis of preeclampsia in pregnant women living in rural areas, since these tend to have limited access to traditional prenatal check-ups, causing the detection of diseases not to be detected in time.
This paper demonstrates the development of an intelligent system to support the diagnosis of preeclampsia in pregnant women living in rural areas at a health center in Santa Cruz. To do this, surveys were conducted for specialists and pregnant women to find out how preeclampsia is traditionally detected and how attendance at prenatal check-ups is linked to it. On the other hand, to meet the objectives, the Machine Learning life cycle methodology was followed, which allowed the analysis of algorithms that are related to the project; taking into account characteristics such as classification algorithms with a supervised learning type; likewise, the Mobile Sprint Model Life Cycle (MMS) methodology that allowed the integration of the model in a mobile application and the implementation of the user interface. Subsequently, the selected model was validated for the prediction of preeclampsia; for this, the selected algorithm was Random Forest with an accuracy percentage of 85%. Finally, a pregnant woman and an obstetrician used the smart system to evaluate its usability and product quality, through a survey based on ISO25010. For this reason, the implementation of this smart system has allowed the early diagnosis of preeclampsia in pregnant women living in rural areas, since these tend to have limited access to traditional prenatal check-ups, causing the detection of diseases not to be detected in time.
Descripción
Palabras clave
Preeclampsia, Aprendizaje automático, Diagnóstico temprano, Preeclampsia, Machine learning, Early diagnosis
Citación
E. Barturen, “Sistema inteligente de apoyo al diagnóstico de preeclampsia en mujeres embarazadas en zonas rurales de Santa Cruz, Cajamarca,” tesis de licenciatura, Fac. de Ingeniería, Univ. USAT, Chiclayo, Perú, 2026. [En línea]. Disponible en: https://hdl.handle.net/20.500.12423/10336
