A new discriminant model for Parkinson’s disease based on logistic regression

Abstract

Although Parkinson’s Disease has significantly affected patients’ normal family and life, it’s difficult to find the correlations of many kinds of indicators and an effective way of diagnosing timely. In this paper, an efficient Discriminant Analysis model based on Logistic Regression is provided to give the reduction of Parkinson’s patients’ indicators and diagnose whether or not you have the disease. Firstly, Principal Component Analysis and Clustering Analysis are implemented to reduce the data dimension of Parkinson’s indicators and give the reductive clusters to complete the feature extraction. The 22 physical indicators data of 195 patients is efficiently 3 main components and 4 clusters. Secondly, according to the extracted features and discriminant analysis based on Logistic Regression, the new discriminant function is established. At last, Python programming is designed to input the characteristic values and output whether or not have the disease to finish the efficient computer implementation of this model. Its result that the accuracy of 195 Parkinson cases of discrimination is 100% has verified the efficiency of this model.