Sistema inteligente para realizar diagnóstico previo de pacientes con diabetes mellitus tipo 2

• Autores: Paredes Vasquez, Juan Carlos, Celis Acosta, Jordi Valente
• Formato: tesis de maestría
• Fecha de Publicación: 2023
• Institución: Universidad Nacional De La Amazonía Peruana
• Repositorio: UNAPIquitos-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.unapiquitos.edu.pe:20.500.12737/9702
• Enlace del recurso: https://hdl.handle.net/20.500.12737/9702
• Nivel de acceso: acceso abierto
• Materia: Redes neuronales (informática); Inteligencia artificial; Diagnóstico por computador; Pacientes; Diabetes mellitus tipo 2; https://purl.org/pe-repo/ocde/ford#2.02.04

The objective of this research was to develop a computer tool to carry out previous diagnoses of type 2 diabetic patients. The design was of an applied type with a quantitative approach, using a population of 740 medical information data of patients from the Sylhet Diabetes Hospital in Sylhet, Bangladesh. . The sample consisted of 25 patients diagnosed by an endocrinologist, using electronic medical records and patient interviews as data collection instruments. Through the analysis of machine learning techniques, an artificial neural network model for the diagnosis of type 2 diabetes was built, demonstrating an accuracy of 88%, a sensitivity of 94% and a specificity of 75%. It was concluded that the diagnostic model based on artificial neural networks is an effective tool for the early identification of type 2 diabetes, recommending adjustments to improve the specificity in the identification of patients without diabetes. This research contributes to the development of medical diagnostic tools based on artificial intelligence, being useful in clinical practice for the early diagnosis of type 2 diabetes.