Predicting Metamorphic Changes In Parkinson's Disease Patients Using Machine Learning Algorithms

Parkinson's disease is a nervous disorder mainly it affects the motor activities of the human body. Manifestations start step by step; at later point it becomes the greatest obstacle to do our day to today activities. Individuals influenced with Parkinson's ailment should go through lifestyle changes and enthusiastic changes like dozing issues, disposition swings, stultification, and skin issues. The proposed methodology is to analyse the proportion of metamorphic changes of a person affected by Parkinson's disease using machine learning techniques. Principal Component Analysis (PCA), recurrent neural network and logistic regression algorithms are used for prediction. The accuracy, precision, recall and F I measure is used to assess the performance of the prediction algorithms. The dataset includes activities of daily living which from PPMI (Parkinson's Progression Markers Initiative) was taken for experimentation. Logistic regression can predict metamorphic changes with a higher accuracy of 92% for sleep dataset and 95% for Olfactory(smell) dataset when compared to other two algorithms.