Classification of Parkinson's Disease by Analyzing Multiple Vocal Features Sets

摘要

Parkinson's disease (PD) is a growing and chronic neurodegenerative disease with a great amount of motor and non-motor symptoms. In the initial stages, most of the PD patients face difficulties in regular movements. Vocal disorders are one of the common symptoms of them. Vocal disorder centric diagnosis systems are one of the leading areas in recent PD detection studies. In this paper, the dataset was taken from the UCI Machine Learning repository and a feature extraction technique was applied. The Analysis of Variance (ANOVA) is used for extracting the features as the dataset was full of features and the topmost 50 features are selected according to ANOVA F-score. Multiple machine learning classification methods were applied and compared with other related existing works. Experimental results show that the highest accuracy score of 0.91 was achieved with the Random Forest Classifier method by feeding the selected features. ANOVA as a feature extraction technique successfully extracted the significant features that differentiate PD patients from healthy individuals and also improve the classification accuracy.