TY - JOUR
T1 - DS-GCNs
T2 - Connectome Classification using Dynamic Spectral Graph Convolution Networks with Assistant Task Training
AU - Xing, Xiaodan
AU - Li, Qingfeng
AU - Yuan, Mengya
AU - Wei, Hao
AU - Xue, Zhong
AU - Wang, Tao
AU - Shi, Feng
AU - Shen, Dinggang
N1 - Publisher Copyright:
© 2020 The Author(s). Published by Oxford University Press. All rights reserved.
PY - 2021/2/1
Y1 - 2021/2/1
N2 - Functional connectivity (FC) matrices measure the regional interactions in the brain and have been widely used in neurological brain disease classification. A brain network, also named as connectome, could form a graph structure naturally, the nodes of which are brain regions and the edges are interregional connectivity. Thus, in this study, we proposed novel graph convolutional networks (GCNs) to extract efficient disease-related features from FC matrices. Considering the time-dependent nature of brain activity, we computed dynamic FC matrices with sliding windows and implemented a graph convolution-based LSTM (long short-term memory) layer to process dynamic graphs. Moreover, the demographics of patients were also used as additional outputs to guide the classification. In this paper, we proposed to utilize the demographic information as extra outputs and to share parameters among three networks predicting subject status, gender, and age, which serve as assistant tasks. We tested the performance of the proposed architecture in ADNI II dataset to classify Alzheimer's disease patients from normal controls. The classification accuracy, sensitivity, and specificity reach 90.0%, 91.7%, and 88.6%, respectively, on ADNI II dataset.
AB - Functional connectivity (FC) matrices measure the regional interactions in the brain and have been widely used in neurological brain disease classification. A brain network, also named as connectome, could form a graph structure naturally, the nodes of which are brain regions and the edges are interregional connectivity. Thus, in this study, we proposed novel graph convolutional networks (GCNs) to extract efficient disease-related features from FC matrices. Considering the time-dependent nature of brain activity, we computed dynamic FC matrices with sliding windows and implemented a graph convolution-based LSTM (long short-term memory) layer to process dynamic graphs. Moreover, the demographics of patients were also used as additional outputs to guide the classification. In this paper, we proposed to utilize the demographic information as extra outputs and to share parameters among three networks predicting subject status, gender, and age, which serve as assistant tasks. We tested the performance of the proposed architecture in ADNI II dataset to classify Alzheimer's disease patients from normal controls. The classification accuracy, sensitivity, and specificity reach 90.0%, 91.7%, and 88.6%, respectively, on ADNI II dataset.
KW - Connectome
KW - GCN
KW - fMRI
UR - http://www.scopus.com/inward/record.url?scp=85100125670&partnerID=8YFLogxK
U2 - 10.1093/cercor/bhaa292
DO - 10.1093/cercor/bhaa292
M3 - Article
C2 - 33078190
AN - SCOPUS:85100125670
SN - 1047-3211
VL - 31
SP - 1259
EP - 1269
JO - Cerebral Cortex
JF - Cerebral Cortex
IS - 2
ER -