Trabajo final de grado
Acceso abierto
GlucoForecast AI: asistente para gestión inteligente de Diabetes Tipo 1
Cargando...
Fecha
Autores
Sebastiao, Matias
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Siglo 21
Resumen
La gestión de la Diabetes Mellitus tipo 1 (DM1) presenta un desafío significativo debido al gran volumen de datos generados por los sistemas de Monitoreo Continuo de Glucosa (CGM), cuya interpretación requiere un considerable esfuerzo cognitivo por parte de los pacientes. Este proyecto abordó dicha problemática a través del desarrollo de GlucoForecast AI, un asistente inteligente diseñado para transformar datos glucémicos complejos en información comprensible. Se implementó un prototipo funcional que integra los datos de CGM del usuario, los cuales son analizados por un motor de Inteligencia Artificial Generativa basado en el modelo de lenguaje Google Gemini. El sistema genera reportes, resúmenes y explicaciones en lenguaje natural que son presentados al paciente a través de una interfaz web. Los resultados demostraron que el prototipo cumplió con el objetivo de crear una herramienta de apoyo que facilita la comprensión de los patrones glucémicos, promoviendo el empoderamiento del paciente y una autogestión más informada y efectiva de su condición.
The management of Type 1 Diabetes Mellitus (T1DM) presents a significant challenge due to the large volume of data generated by Continuous Glucose Monitoring (CGM) systems, the interpretation of which requires considerable cognitive effort from patients. This project addressed this issue through the development of GlucoForecast AI, an intelligent assistant designed to transform complex glycemic data into understandable information. A functional prototype was implemented that integrates the user's CGM data, which is then analyzed by a Generative Artificial Intelligence engine based on the Google Gemini language model. The system generates reports, summaries, and explanations in natural language that are presented to the patient through a web interface. The results demonstrated that the prototype successfully met the objective of creating a support tool that facilitates the understanding of glycemic patterns, promoting patient empowerment and more informed and effective self-management of their condition.
The management of Type 1 Diabetes Mellitus (T1DM) presents a significant challenge due to the large volume of data generated by Continuous Glucose Monitoring (CGM) systems, the interpretation of which requires considerable cognitive effort from patients. This project addressed this issue through the development of GlucoForecast AI, an intelligent assistant designed to transform complex glycemic data into understandable information. A functional prototype was implemented that integrates the user's CGM data, which is then analyzed by a Generative Artificial Intelligence engine based on the Google Gemini language model. The system generates reports, summaries, and explanations in natural language that are presented to the patient through a web interface. The results demonstrated that the prototype successfully met the objective of creating a support tool that facilitates the understanding of glycemic patterns, promoting patient empowerment and more informed and effective self-management of their condition.