Comparative Analysis of the Classification Performance of Machine Learning Classifiers and Deep Neural Network Classifier for Prediction of Parkinson Disease

Abstract

The accurate diagnosis of Parkinson disease specifically in its initial stages is extremely complex and time consuming. Thus the accurate and efficient diagnosis of Parkinson disease has been a significant challenge for medical experts and researchers. In order to tackle the accurate diagnosis of Parkinson disease issue we proposed machine learning and deep neural networks based non-invasive prediction system for accurately and on time diagnosis of Parkinson disease. In the development of the system machine learning predictive models such as support vector machine, logistic regression and deep neural network were used for people with Parkinson disease and healthy people classification. The data set was splits into 70% for training purpose and 30% for testing. Furthermore, performance evaluation metrics such as classification accuracy, sensitivity, specificity and Matthews's correlation coefficient were utilized for model performance evaluation. The Parkinson disease dataset of 23 attributes and 195 instances available on UCI machine learning repository was used for testing of the proposed system. Through our experimental results analysis shows that the proposed system classified the Parkinson disease and healthy people effectively. We also investigated that deep neural performance of classification was excellent as compared to traditional machines learning classifiers. These finding suggest that the proposed diagnosis system could be used to accurately predict Parkinson disease.