An Early Prediction of Parkinson’s Disease using Machine Learning Techniques

Abstract

Machine Learning has transformed Healthcare Domainproviding more efficient, faster, smarter ways to detect and cure various diseases. Machine learning approaches are widely used for Parkinson’s Disease (PD) prediction. The prediction of Parkinson’s disease challenging for doctors and researchers as the symptoms of the disease are examined in middle and later middle ages where condition has progressed over time. In this paper, we will build a predictive model which can be used for early and accurate detection of the presence of Parkinson’s disease in one’s body. We focused on XGBoost, a new Machine Learning algorithm,based on decision trees, designed with speed and performance in mind, to improve the accuracy of PD prediction. This approach using the XGBoost algorithm obtained higher accuracythan other machine learning techniques such as Naïve Bayes algorithm, binary logistic regression, random forest and support vector machine.

Keywords:Machine Learning, Parkinson’s disease, XGBoost, Decision Tree Based