| Summary: | Alzheimer's disease (AD) is a common neurodegenerative disease in the elderly. Early diagnosis and prediction plays an important role in early intervention and delaying disease progression of AD. This paper focused on the principles and process of pattern classification method, and its application in the clinical study and auxiliary diagnosis of AD. The biomarkers, neuroimaging and cognitive ability scales are important features for pattern classification. Various classification algorithms including Bayesian networks, decision trees, support vector machines (SVM) and multilayer perception have been adopted to distinguish AD, mild cognitive impairment (MCI) and normal aging subjects. Besides, they can effectively trace and analyze MCI patients.
|
|---|