Application of Radial Basis Function Neural Network in Diagnosis of Ms Disease

Abstract

Artificial intelligence consist of research in computer science that it's targets is creating a computer that could reason such as human and help them make decision better. Decision support system is computer software that design for diagnosis system. Multiple Sclerosis is a kind of disease that starts with damage the myelin of brain and has some neurological symptoms like reducing the power of vision, spastic paralysis of limbs and impotency. In this case, the major problem is the lack of diagnosis. To improve diagnosis, Radial Basis Function Neural Network is used. In this study, we use learning power of Neural Network. K-fold cross validation has been used for optimizing the input/output. Implementation has been done in MATLAB. The Dataset consist of 600 patients that each one has 6 columns, 5 of them is input and one a of them is output that shows diagnosis. Proposed Method RBF compared with ANFIS and MLP. The result shows that RBF's accuracy is 96%. Its accuracy is a little more than ANFIS but it's training time is much more that the other. ANFIS's accuracy is almost 96%. By consideration of too much training time of RBF and ignore the slightly different accuracy, ANFIS give us the best result.