Parkinson's disease classification from voice analysis using machine learning

Abstract

Parkinson's disease (PD) diagnosis is often challenging, leading to delayed treatment and reduced patient quality of life. This study investigates the use of machine learning, specifically Support Vector Machine (SVM), in early PD detection through dysphonia measurements via phonatory voice analysis. Two feature selection methods, L1-norm Support Vector Machine (L1-norm SVM) and Recursive Feature Elimination (RFE), are compared within the SVM framework with the data from the UCI Machine Learning Repository and the Chulalongkorn Centre of Excellence for Parkinson’s Disease & Related Disorders (ChulaPD). Results show that models incorporating feature selection consistently perform better than models that did not. L1-norm SVM proves more effective for PD diagnosis than RFE, demonstrating superior accuracy, positive recall score, and F1-macro score. 95.97% accuracy, 97.16% recall score, and 95.69% F1-macro score were achieved using L1-norm SVM with the combined UCI and ChulaPD datasets. These findings suggest the potential for more accurate PD diagnosis methods in phonatory voice analysis and provide insights into enhancing PD diagnosis through machine learning and voice analysis. Further research into feature selection methods, datasets, and diagnosis algorithms is recommended to enhance early PD diagnosis performance.