Early Alzheimer’s Disease Diagnosis with Paired Comparative Deep Learning The code was written by Hezhe Qiao, Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, 400714 Chongqing, China, University of Chinese Academy of Sciences, 100049 BeiJing, China.
Introduction we propose a paired comparative deep learning method that measures the differences of group category (G-CAT) and subject mini-mental state examination (S-MMSE), respectively, to enhance the sMRI features of groups and individuals. This proposed model has been evaluated on the ADNI-1, ADNI-2, and MIRIAD datasets.
Prerequisites
Linux python 3.7
Pytorch version 1.2.0
NVIDIA GPU + CUDA CuDNN (CPU mode, untested) Cuda version 10.0.61
The detailed information and pretriand model will be added after the paper is published.