Classification of Alzheimer’s Disease Based on Core-Large Scale Brain Network Using Multilayer Extreme Learning Machine

Various studies suggest that the network deficit in default mode (DMN) is prevalent Alzheimer’s disease (AD) and mild cognitive impairment (MCI). Besides DMN, some reveal alteration occurs salience motor networks large scale network. In this study we performed classification of AD MCI from healthy control considering alterations DMN. Thus, constructed brain functional magnetic resonance (fMR) images. Pearson’s correlation-based connectivity was used to construct Graph features were converted feature vectors using Node2vec graph-embedding technique. Two classifiers, single layered extreme learning multilayered machine, for together with selection approaches. We test on different sizes including network, whole combined Experimental results showed least absolute shrinkage operator (LASSO) method generates better accuracy size, adaptive structure (FSAL) technique small size.