Ensemble Learning for Parkinson's Detection: Optimizing Classification with Hybrid Boosting and Ridge Regression

Abstract

Parkinson's Disease (PD) is a neurodegenerative disorder characterized by motor and non-motor symptoms, where early diagnosis and accurate progression prediction are vital for effective management. This study explores the use of an Ensemble Ridge Classifier to predict PD onset and progression using clinical data, motor and non-motor symptoms, and imaging features. The Ensemble Ridge Classifier, which combines multiple ridge regression models, addresses challenges such as overfitting, underfitting, and model bias, improving prediction accuracy and robustness. By applying this model to publicly available PD datasets, the approach demonstrates enhanced performance over single classifiers, offering better generalizability and minimizing variance. The study also highlights the classifier's ability to identify key features contributing to disease progression, providing insights that could inform personalized treatment strategies. This research underscores the potential of ensemble learning for early PD diagnosis and monitoring, making it a promising tool for clinical decision support and disease management.