Machine Learning-Based Early Detection of Parkinson's Disease Using Voice Features

This study presents a machine learning-based approach for the early detection of Parkinson’s disease (PD) using voice features. Parkinson’s disease is a progressive neurodegenerative disorder that affects motor control and vocal characteristics, making speech analysis a promising diagnostic tool. In this work, an optimized Support Vector Machine (SVM) model was trained and evaluated on a publicly available Parkinson’s voice dataset from Kaggle. Preprocessing techniques such as feature scaling and normalization were applied to reduce overfitting and improve robustness. The model achieved high classification accuracy (~90-95%), with balanced precision and recall, demonstrating reliable differentiation between Parkinson’s patients and healthy controls. Unlike many prior studies, this work emphasizes systematic preprocessing, cross-validation, and parameter tuning to improve generalizability and reduce bias. The results suggest that vocal biomarkers, when combined with machine learning, can serve as effective, non-invasive tools for clinical decision support in Parkinson’s disease screening. This research contributes to the growing field of AI in healthcare by highlighting the importance of explainability, reproducibility, and lightweight models for real-world applications. Keywords: Parkinson’s disease, Machine learning, Support Vector Machine, Voice biomarkers, Neurodegenerative disorders, Healthcare AI