Evaluation of Adaptive Neuro-Fuzzy Inference System with Artificial Neural Network and Fuzzy Logic in Diagnosis of Alzheimer Disease

In recent past, artificial intelligence techniques have been applied in various domains including medical one for enhancing the performance of the tasks to be done in these domains. Expert systems to be used in medical analysis are being developed using various soft computing techniques (like artificial neural network, Fuzzy Logic, Genetic algorithm, etc.) to assist medical practitioner in diagnosing the diseases effectively. But these techniques have limitations in dealing with uncertain and multidimensional data. To deal with such issues, hybrid systems like Neuro-fuzzy systems (NFS) are developed and used to build up systems with better performance as compared to the individual ones. In this paper, a comparative study has been done to evaluate the performance of hybrid system such as ANFIS with artificial neural network (ANN) and Fuzzy Logic system (FLS) in diagnosis of Alzheimer disease. To assess the performance of each technique, three measures that are root mean square error (RMSE), accuracy and F-score are implemented.