OPTIMAL WAY TO FIND THE FRAME LENGTH OF THE SPEECH SIGNAL FOR DIAGNOSIS OF ALZHEIMER'S DISEASE WITH PSO

This paper proposes an efficient algorithm for Alzheimer's disease (AD) diagnosis based on automatic spontaneous speech analysis. Acoustic features were used for classification of control group and three stages of Alzheimer's disease. PSO algorithm was used for estimating the optimal segment length of speech signal to calculate the acoustic features. Then these features were given to Adaptive neuro-fuzzy inference system (ANFIS) for classification. The simulation results show the ability of PSO to solve the segment length in conjunction with Adaptive neuro-fuzzy inference system (ANFIS) as a classifier. The performance of the ANFIS model was evaluated in term of classification accuracies and the results confirmed that the proposed method has potential in Alzheimer's disease (AD) diagnosis.