An Early Prediction of Parkinson's Disease Using Facial Emotional Recognition

Abstract

Parkinson's disease (PD) is a neurodegenerative disorder which challenges the population due to its uncertainty in prediction of the disease. It is a progressive disorder of the nervous system marked by tremor, muscular rigidity and slow, imprecise movement, chiefly affecting middle-aged and elderly people. In our proposed work, we utilize the facial emotions of PD patients and normal person to identify their facial emotions like sad, happy, anger, and depression. For this predictive analysis, the datasets are acquired from Parkinson's Progression Markers Initiative (PPMI) which consists of 188 PD patients and 50 normal people for testing and training process. By utilize this dataset, we applying the CNN architecture of Alex Net, and Vgg 16 to achieve their performance in terms of accuracy, sensitivity, specificity, F1 score and area under curve. Finally, it proven that Vgg 16 gives 10% more accurate results than Alex Net. This research outcome will be very useful in diagnosis of early-stage Parkinson's disease in healthcare.