Machine Learning Techniques for Parkinson's Disease Detection

A neurological disease is Parkinson's disease. It causes trembling in the hands, trouble walking, losing balance, and coordination. In the high-level stage, there is no access to medical care. Blood test reports, CT scan results, and X-ray reports are not accessible early enough. Early Parkinson’s disease detection is crucial to implement effective treatment. The purpose of the proposed effort was to identify Parkinson’s disease in early prediction using clinical imaging and machine learning technologies. Despite the fact that there are numerous methods for detecting Parkinson’s disease, using MRI scan images still it is a big challenge. In this study, an Adaboost classifier is used with a hybrid PSO algorithm to propose a novel technique for detecting Parkinson’s disease. Adaboost acted as the best classifier among other classifiers. Initially, MRI image best features are extracted and identified by the curvelet transform and principal component analysis. This Ad boost classifier receives optimal features as input. Finally, Adaboost classifieds the MRI images and gave excellent classification accuracy. To evaluate the proposed method three methods metrics namely accuracy, specificity, and sensitivity are used. Based on the results the proposed methods yield greater accuracy than the existing systems.