Desarrollo de Algoritmos Basados en Modelos Matemàticos para Detectar Enfermedades
Fecha
2011-10
Autores
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad de Guayaquil. Facultad de Ciencias Matematicas y Fisicas. Carrera de Ingeniería en Sistemas Computacionales. Carrera de Ingeniería en Networking y Telecomunicaciones.
Resumen
La presente investigación ayudará a la comprensión de los procesos computacionales
que se dan en la medicina a partir de modelos matemáticos y servirá para abrir un
campo de investigación que en nuestro país es prácticamente inexplorado. El objetivo
es demostrar que sin necesidad de exámenes de laboratorio se puede detectar el
colesterol en base a la evaluación de los síntomas y estilo de vida de las personas que
han sido objeto de nuestro estudio. El algoritmo debe representar la realidad de la
población en estudio, es importante decir que el núcleo de la investigación son los
modelos matemáticos y que el mismo simula una red neuronal artificial tipo
perceptrón de capa simple. El presente estudio es un proyecto netamente investigativo
y de interés social. La encuesta fue dirigida a los padres de familia de la Escuela
Ruperto Arteta Montes, el tamaño total de esta población es de ochocientas personas
y se logró recoger una muestra de seiscientas personas. Dentro de la investigación
tenemos dos tipos de variables, la variable dependiente que estudia el hecho de tener o
no colesterol en base a las variables independientes que son referentes a los síntomas y
al estilo de vida. Nuestra propuesta abre un campo de investigación nuevo dado que
no es común tratar de detectar enfermedades basándonos en modelos matemáticos.
Con la información recopilada el modelo matemático podrá determinar el número de
personas de la muestra afectadas con el colesterol malo LDL (Lípidos de Baja
Densidad). Los beneficiarios son las personas que están dentro de la muestra, porque
este proyecto refleja un aspecto de su realidad, en este caso el colesterol.
This research will help to understand the computational processes given in medicine from mathematical models and serve to open a research field in our country is virtually unexplored. The goal is to show that without lab tests can detect cholesterol based on the evaluation of symptoms and lifestyle of people have been the subject of our study. The algorithm must represent the reality of the study population, it is important to say that the core of the research are mathematical models and that it simulates an artificial neural network single layer perceptron. The present study is a purely research project and social interest. The survey was aimed at parents of the School Arteta Ruperto Montes, the total size of this population is eight hundred people and it was possible to collect a sample of six hundred people. Within research we have two types of variables, the dependent variable studying or having no cholesterol based on the independent variables that are related to symptoms and lifestyle. Our approach opens a new field of research since is not common to try to detect diseases based on mathematical models. With the information gathered the mathematical model can determine the number of sample persons affected with the bad LDL (Low Lipids Density). The beneficiaries are the people who are within the sample, because This project reflects an aspect of reality, in this case cholesterol.
This research will help to understand the computational processes given in medicine from mathematical models and serve to open a research field in our country is virtually unexplored. The goal is to show that without lab tests can detect cholesterol based on the evaluation of symptoms and lifestyle of people have been the subject of our study. The algorithm must represent the reality of the study population, it is important to say that the core of the research are mathematical models and that it simulates an artificial neural network single layer perceptron. The present study is a purely research project and social interest. The survey was aimed at parents of the School Arteta Ruperto Montes, the total size of this population is eight hundred people and it was possible to collect a sample of six hundred people. Within research we have two types of variables, the dependent variable studying or having no cholesterol based on the independent variables that are related to symptoms and lifestyle. Our approach opens a new field of research since is not common to try to detect diseases based on mathematical models. With the information gathered the mathematical model can determine the number of sample persons affected with the bad LDL (Low Lipids Density). The beneficiaries are the people who are within the sample, because This project reflects an aspect of reality, in this case cholesterol.
Descripción
Adobe
Palabras clave
RED NEURONAL ARTIFICIAL, RECOLECCION DE DATOS