Early Diagnosis of Parkinson's Disease Based on Spiral and Wave Drawings Using Convolutional Neural Networks and Machine Learning Classifier

Parkinson's disease (PD) is a neurodegenerative condition caused by dopamine-producing nerve cell loss. However, due to a lack of indicators, early detection of Parkinson's disease is difficult. The goal of this study is to create a system for the early detection of Parkinson's disease (PD) based on hand drawings, using pre-trained CNN models as a feature extractor and Machine Learning (ML) classifiers to differentiate PD from healthy persons. Several pre-trained CNN models, including VGG16, VGG19, ResNet-50, IncepetionV3, Xception, and Mobile Net V2, are used as feature extractors. The retrieved characteristics are fed into the various machine learning classifiers as input. The proposed system VGG16 as a feature extractor and Random Forest (RF) as ML classifier performed much better than existing state-of-the-art pretrained models, with a classification accuracy of 97%. The results of the experiment indicate that the proposed strategy works better on publicly available hand-drawn datasets for the early identification of Parkinson's disease.