Parkinson's Disease Prediction Using Efficient Analysis of Machine Learning Algorithms

Parkinson's disease (PD) is a severe form of neurodegeneration that primarily impacts the CNS and results in bradykinesia. Dopamine levels drastically decrease as a result of the injury to the nerve cells. Genetic and environmental influences are the key ones that are now understood. In order to choose the optimal model for early disease prediction, we have examined numerous machine learning techniques in this study. Results were acquired after running the model's training data set, which contained records of Parkinson's patients, and a comparative analysis was conducted. KNN yielded a result of 89.74%, SVM yielded a result of 87.17%, XGBoost yielded a result of 94.87%, and Random forest yielded a result of 97.43% respectively. The goal of this study is to make a significant contribution to the improvement of healthcare and the well-being of PDpatients, so that novel therapies and treatments can be developed.