Parkinson's Disease Diagnosis and Severity Prediction with Machine Learning Techniques

Abstract

Diagnosis and forecasting disease progression is critical for effective treatment of Parkinson's disease (PD). The motivation of this study is to use machine learning methods such as neural networks, logistic regression, and random forests to diagnose PD and to determine whether the severity of the disease increases based on clinical information from early onset. We used data-driven models including traditional machine learning models and deep neural networks to determine and predict the PD condition. The proposed methods were validated with the Parkinson's Progression Markers Initiative (PPMI) dataset, which is the most widely known and validated source of PD data. The Hoehn & Yahr (H&Y) scale was used to determine if there was a change in disease severity. The results show that the accuracy of applying neural network to diagnose PD is 94.26%, and the accuracy of applying the random forest model to predict changes in disease severity achieved 82.24%.