phat 24(1):

Research Article

Prognoza: Parkinson’s Disease Prediction Using Classification Algorithms

Download63 downloads
  • @ARTICLE{10.4108/eetpht.9.3933,
        author={Mithun Shivakoti and Sai Charan Medaramatla and Deepthi Godavarthi and Narsaiah Shivakoti},
        title={Prognoza: Parkinson’s Disease Prediction Using Classification Algorithms},
        journal={EAI Endorsed Transactions on Pervasive Health and Technology},
        keywords={Parkinson's disease, Classification, Machine Learning, CATBoost, Random Oversampling},
  • Mithun Shivakoti
    Sai Charan Medaramatla
    Deepthi Godavarthi
    Narsaiah Shivakoti
    Year: 2023
    Prognoza: Parkinson’s Disease Prediction Using Classification Algorithms
    DOI: 10.4108/eetpht.9.3933
Mithun Shivakoti1, Sai Charan Medaramatla1, Deepthi Godavarthi1,*, Narsaiah Shivakoti2
  • 1: Vellore Institute of Technology University
  • 2: GDC Ibrahimpatnam Rangareddy Telangana
*Contact email:


Parkinson's Disease (PD) is a persistent neurological condition that has a global impact on a significant number of individuals. The timely detection of PD is imperative for the efficacious treatment and control of the condition. Machine learning (ML) methods have demonstrated significant potential in forecasting Parkinson's disease (PD) based on diverse data sources in recent times. The present research paper outlines a study that employs machine learning [ML]techniques to predict Parkinson's disease. A dataset comprising clinical and demographic characteristics of both patients diagnosed with PD and healthy individuals was taken from Kaggle. The aforementioned dataset was utilized to train and assess multiple machine learning models. The experimental findings indicate that the CatBoost model exhibited superior performance compared to the other models, achieving an accuracy rate of 95.1% and a root mean squared error of of 0.34.In summary, our research showcases the capabilities of machine learning methodologies in forecasting Parkinson's disease and offers valuable insights into the crucial predictors for PD prognosis. The results of our study could potentially contribute to the advancement of diagnostic methods for the timely identification of PD, with increased precision and efficacy.