Scholarlink

Parkinson’s Disease (PD) is a neurological disorder due to the loss of neurons that produce dopamine and it affects motor functions. Monitoring disease progression and detecting disease early are the challenges that patients often face as traditional diagnostic methods for PD are expensive and rely majorly on subjective assessments. This study leverages machine learning specially ensemble learning techniques to predict PD progression using voice analysis. The Parkinson’s Telemonitoring Voice Dataset is used to train and test various ensemble models like CART, Random Forest Regressor, AdaBoost Regressor and Random Forest-AdaBoost Ensemble technique to predict both motor UPDRS and total UPDRS scores. Among the models tested, Random Forest-AdaBoost Ensemble technique achieved the highest R2 values of 0.966 and 0.961for Motor and Total UPDRS, respectively.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Treatment Monitoring Using Ensemble Learning: A Case Study on Parkinson’s Disease Progression

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Meghna Manoj,
  • Geetika Diwan,
  • Anchal Jaiswal,
  • K. R. Seeja

摘要

Parkinson’s Disease (PD) is a neurological disorder due to the loss of neurons that produce dopamine and it affects motor functions. Monitoring disease progression and detecting disease early are the challenges that patients often face as traditional diagnostic methods for PD are expensive and rely majorly on subjective assessments. This study leverages machine learning specially ensemble learning techniques to predict PD progression using voice analysis. The Parkinson’s Telemonitoring Voice Dataset is used to train and test various ensemble models like CART, Random Forest Regressor, AdaBoost Regressor and Random Forest-AdaBoost Ensemble technique to predict both motor UPDRS and total UPDRS scores. Among the models tested, Random Forest-AdaBoost Ensemble technique achieved the highest R2 values of 0.966 and 0.961for Motor and Total UPDRS, respectively.