AUTHOR=Hong Xin , Huang Kaifeng , Lin Jie , Ye Xiaoyan , Wu Guoxiang , Chen Longfei , Chen E. , Zhao Siyu 

TITLE=Combined Multi-Atlas and Multi-Layer Perception for Alzheimer's Disease Classification

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 14 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2022.891433

DOI=10.3389/fnagi.2022.891433

ISSN=1663-4365

ABSTRACT=Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease. To distinguish the stage of the disease, the AD classification technology challenge has been proposed in Pattern Recognition and Computer Vision 2021 (PRCV 2021), which provides the gray volume and average cortical thickness data extracted in multiple atlases from MRI.

Traditional methods either train with CNN-based method by MRI data to adapt the spatial features of images, or train with RNN-based method by temporal features to predict the next stage. However, morphological features from the challenge are neither spatial-based nor temporal-based. We present a multi-atlases multi-layer perceptron (MAMLP) approach to deal with the relationship between morphological features and the stage of the disease. Firstly, to preserve the diversity of brain features, the most representative atlases are chosen from groups of similar atlases, and one atlas is selected in each group; Secondly, each atlas is fed into one multi-layer perceptron to fetch the score of the classification; Thirdly, to obtain more stable results, scores from different atlases are combined to vote the result of the classification. 

Based on this approach, we rank 10th among 373 teams in the challenge. The results of the experiment indicate  below. 1) Group selection of atlas reduces the number of features required without reducing the accuracy of the model; 2) The MLP architecture achieves better performance than CNN and RNN networks in morphological features; 3) Compared with other networks, the combination of multiple MLP networks has faster convergence of about 40% and makes the classification more stable.