ParkinNet: a Novel Approach to Classifying Alzheimer's and Parkinson's Diseases Using Brain Structural MRI

Abstract

Both Parkinson's disease (PD) and Alzheimer's disease (AD) are forms of neurodegeneration, which are linked to the same biochemical alterations in the brain. The mixed pathology of these diseases may cause diagnostic dilemmas, which may lead to misdiagnosis. Because of this, classification of AD and PD is essential to reduce extra healthcare costs and the patients' stress. However, the classification of AD and PD can be challenging because of the overlapping symptoms and risk factors. Therefore, the purpose of this study is to develop a model named ParkinNet to classify AD and PD. The current study used Global Average Pooling and Adam optimiser with a batch size of 64. For evaluation, seven deep learning algorithms are used, including MobileNetV2, EfficientNetB2, InceptionResNetv2, VGG16, VGG19, InceptionV3 and ResNet50, along with the proposed ParkinNet model. The proposed ParkinNet model outperforms the other existing models examined in this study and yields an accuracy of 98. 54 %. The precise classification of these diseases may contribute to the diagnosis process of AD and PD.