Voice analysis for detecting Parkinson's disease using genetic algorithm and KNN classification method

Parkinson's disease is a disorder of central nervous system. It is estimated that 90 percent of people with parkinson's disease suffer from speech and voice disorders. Vocal folds are usually undermined by this disease which would lead to creation of an improper voice in the patient's speech. In this paper, various features have been extracted from the voice signals of healthy people and people suffering from parkinson's disease. Afterwards, optimized features that influenced the process of data classification were detected using genetic algorithm and ultimately, based on various numbers of optimized features, the data classification was done using KNN classification method. It was shown that a classification accuracy percent of 93.7 per 4 optimized features, an accuracy percent of 94.8 per 7 optimized features and an accuracy percent of 98.2 per 9 optimized features could be achieved which is a notable result compared to other studies.