Review on Parkinson’s Disease Detection using Machine Learning

Parkinson’s Disease (PD) is a disorder that affects the brain as it causes brain cells to deteriorate. Parkinson’s Disease is difficult to diagnose since it has vague symptoms. Early diagnosis is critical for the effective management of PD. This paper proposes a new methodology based on machine learning techniques that allows predicting the onset of Parkinson's Disease using clinical and speech datasets. In this work, the stages of data preprocessing, de-noising, normalization, feature extraction, and modeling are considered. Specific voice parameters including jitter, shimmer, RPDE, and PPE are used as key features for modeling the pathology. Multiple machine learning algorithms, such as SVM, random forest, KNN, and XGboost, are tested in the work. Several performance measures, including accuracy, precision, recall, F1 score, and ROC-AUC score are used to evaluate the results of models’ performance. As shown in experimental part, machine learning techniques can successfully recognize Parkinson's Disease and achieve high levels of reliability and accuracy.