Performance analysis of Classification methods for Parkinson’s Disease with PPMI Dataset

Abstract

The motor and non-motor complications in older people leads to neurodegenerative disorder known as Parkinson’s Disease. Movement ailments are instigated by the dopamine deficiency in neurons. In this proposed work the dataset from PPMI applied to classify and identify symptoms using the classification techniques. The method opted to calculate the severity and progression of Parkinson’s Disease in patients is the Unified Parkinson’s Disease Rating Scale given as UPDRS. The machine learning classification methods such as nearest neighbors, extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), random forest, SVM, decision tree, Naive Bayes and logistic regression are used to categorize early PD from healthy person. The study brings out the approaches are highly accurate and have a large accuracy to distinguishing early PD from healthy normal.