Speech Signal-Based Detection and Staging of Parkinson's Disease

This study uses voice signal analysis to create a machine learning method for Parkinson's disease stage classification. Based on voice-related biomarkers, the study uses a thorough technique that integrates many classification algorithms to forecast the course of the disease. In order to identify five different disease phases, the study employs a unique multi-stage classification approach that converts speech signal elements into a composite score. Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), Decision Tree, and Naive Bayes were the five machine learning models that were assessed. The study demonstrates the feasibility of using machine learning algorithms for objective, data-driven Parkinson's disease stage classification, providing a foundation for potential clinical diagnostic tools.