Early diagnosis of Alzheimer's disease based on VGG cascade model

Alzheimer's disease (AD) is one of the common degenerative diseases of the central nervous system, and early diagnosis of AD is the key to treatment. In order to improve the accuracy of classification, this study uses Visual Geometry Group (VGG) as the backbone network and Support Vector Machine (SVM) as the cascade learning model of the decider. The model can cascade analysis of the input structured magnetic resonance imaging (sMRI) to achieve early diagnosis of AD, and the final binary classification result of AD vs. NC is 90.1%, and the binary classification result of EMCI vs. NC is 75.0%. The results show that the proposed VGG cascade model has good performance in the early diagnosis of AD.