nahid saberipour

A significant proportion of population in each community suffer from autism disorder. One of the challenges in diagnosing autism is the uncertainty in determining the severity of the disease. To this end, fuzzy systems based methods have been adopted in this study. The presented methods are based on 112 data driven from children and adolescents between the ages of 3 to 14 years. These data were collected from various rehabilitation centers in Tehran. The average performance accuracy of the proposed methods Using Genetic Algorithm with area under curve ROC compared to other methods (adaptive fuzzy neural inference system algorithm) proved to be 97/4% more reliable and efficient. The system designed in this article can be used as a medical diagnostics help tool for physicians.

A significant proportion of population in each community suffer from autism disorder. One of the challenges in diagnosing autism is the uncertainty in determining the severity of the disease. To this end, fuzzy systems based methods have been adopted in this study. The presented methods are based on 112 data driven from children and adolescents between the ages of 3 to 14 years. These data were collected from various rehabilitation centers in Tehran. The average performance accuracy of the proposed methods Using Genetic Algorithm with area under curve ROC compared to other methods (adaptive fuzzy neural inference system algorithm) proved to be 97/4% more reliable and efficient. The system designed in this article can be used as a medical diagnostics help tool for physicians.

A significant proportion of population in each community suffer from autism disorder. One of the challenges in diagnosing autism is the uncertainty in determining the severity of the disease. To this end, fuzzy systems based methods have been adopted in this study. The presented methods are based on 112 data driven from children and adolescents between the ages of 3 to 14 years. These data were collected from various rehabilitation centers in Tehran. The average performance accuracy of the proposed methods Using Genetic Algorithm with area under curve ROC compared to other methods (adaptive fuzzy neural inference system algorithm) proved to be 97/4% more reliable and efficient. The system designed in this article can be used as a medical diagnostics help tool for physicians.