Genetic algorithm based feature selection on diagnosis of Parkinson disease via vocal analysis

Abstract

Parkinson's disease is a neurological disorder that affects the quality of life of a patient adversely, especially in elder people. Parkinson's disease first manifests itself as slowness, imbalance and trembling in motor movements. There are many methods such as gait analysis and voice analysis for diagnosis of the disease. In this study, genetic algorithm based feature selection method is proposed for the voice analysis that one of the methods used in the diagnosis of Parkinson's disease. Classification success rates of selected attributes are calculated by SVM classification algorithm and validated by LOOCV method. The classification success rate of the proposed method was compared with previous studies using the same classification and validation method. As a result of the comparison, more successful results were obtained than those using SVM classification and LOOCV validation methods.