In this review a notable contribution is made to medical diagnosis to classify Parkinson which is the second highest occurrence neurodegenerative disease .The objective of this study is to discuss the various machine learning algorithms that are used for classification of PD patients from Healthy patient list. We demonstrate the work to compare the efficiency of ML models to identify the Parkinson disease. In this study various dataset from different sources have been considered for the analysis .Now a days there exist many classification methods have been proposed to detect the Parkinson’s Disease from prodromal MCI (mild cognitive impairment ) phases and Healthy Controls(HC).A study of 48 papers have been explained with different machine learning models with results comparison and found the limitation that most of the paper have used the MRI images as input without any image enhancement techniques.