Paper Contents
Abstract
Parkinsons Disease (PD) is a chronic and progressive neurodegenerative disorder that primarily affects motor functions due to the loss of dopamineproducing neurons in the brain. Early detection of PD is vital for mitigating its effects and enhancing patient outcomes. This study explores the use of machine learning (ML) techniques for diagnosing PD through voice-based biomarkers. Voice features, such as pitch, jitter, shimmer, and harmonic-to-noise ratio, are extracted from patient speech recordings, which often reveal subtle changes indicative of PD.Four ML models were evaluated for their effectiveness in classifying PD: Random Forest, Support Vector Machine (SVM), Logistic Regression, and K-Nearest Neighbors (KNN). Among these, the Random Forest classifier demonstrated superior performance with an accuracy of 91.83% and a sensitivity of 0.95, highlighting its potential for reliable early detection.To facilitate practical implementation, a user-friendly web application was developed using the Streamlit framework. This application enables remote testing by allowing users to upload voice recordings and receive diagnostic predictions in real time. The system offers a cost-effective, scalable, and non-invasive diagnostic solution, which is especially advantageous for telemedicine applications in neurology. By integrating advanced ML techniques with an accessible digital platform, this project aims to revolutionize early PD detection, providing a significant step forward in personalized healthcare and remote diagnostics
Copyright
Copyright © 2025 Mrs.HEMAVATHI R. This is an open access article distributed under the Creative Commons Attribution License.
