An Artificial Neural Network Model to Diagnosis of Type II Diabetes
Introduction: Diabetes is a disease caused by an increase in blood glucose levels due to insulin secretion deficiency (type 1 diabetes) or impaired insulin activity (type 2 diabetes). More than 90% of people with this condition are diagnosed with type 2 diabetes. Due to sharply prevalence of type 2 diabetes in recent years, the prognosis and early diagnosis of the disease have become even more important. In this study, a model for diagnosis of type 2 diabetes was developed using Artificial Neural Network (ANN) method. Objectives: Minimizing the diagnosis faults of diabetes disease, using a hybrid ANN and the Genetic optimization algorithm.
Method: In this study, a hybrid ANN-Genetic Algorithm model was developed for classification of diabetic patients. Therefore, the number of optimal neurons as well as hidden layers was determined to design the architecture of the ANN model. To reduce the mean square error of the MSE network and optimize the accuracy of the diagnostic system a Genetic Algorithm (GA) was combined with the proposed ANN model. For experiment process, the model was considered on a dataset included 768 samples to diagnose the patients with type II diabetes from other cases.
Findings: The results showed a precision of 85% for diagnosing of type-2 diabetic patients. The proposed structure based on the lower mean square error of the MSE, indicated the best performance of the ANN with the MSE rate of 0.155.
Conclusion: The developed intelligent model showed an effective performance in comparison with existing methods with a minimum error and maximum confidence in the diagnosis process of diabetic disease.