Back

The Effect of Node Features on GCN-based Brain Network Classification: An Empirical Study

Wang, G.; Zhang, L.; Qiao, L.

2022-11-03 neuroscience
10.1101/2022.11.01.514695 bioRxiv
Show abstract

Brain functional network (BFN) analysis has become a popular technique for identifying neurological/mental diseases. Due to the fact that BFN is a graph, graph convolutional network (GCN) can be naturally used in the classification of BFN. Different from traditional methods that directly use the adjacency matrices of BFNs to train a classifier, GCN requires an additional input -- node features. To our best knowledge, however, there is no systematic study to analyze their influence on the performance of GCN-based brain disorder classification. Therefore, in this per, we conduct an empirical study on various node feature measures, including 1) original fMRI signals, 2) one-hot encoding, 3) node statistics, 4) node correlation, and 5) their combination. Experimental results on two benchmark databases show that different node feature inputs to GCN significantly affect the brain disease classification performance, and node correlation usually contributes higher accuracy compared to original signals and manually extracted statistical features.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

1
IEEE Journal of Biomedical and Health Informatics
● 37 papers in training set
Top 0.1%
11.8%
Similar papers in this journal
2
NeuroImage
○ 903 papers in training set
Top 2%
8.8%
Similar papers in this journal
3
Neuroinformatics
○ 46 papers in training set
Top 0.1%
7.8%
Similar papers in this journal
4
Biomedical Signal Processing and Control
○ 22 papers in training set
Top 0.1%
7.8%
Similar papers in this journal
5
Human Brain Mapping
○ 329 papers in training set
Top 1%
5.5%
Similar papers in this journal
6
PLOS ONE
● 5266 papers in training set
Top 31%
4.3%
7
Network Neuroscience
● 126 papers in training set
Top 0.5%
3.2%
8
Neurocomputing
○ 13 papers in training set
Top 0.1%
3.1%
50% of probability mass above
9
Scientific Reports
○ 3612 papers in training set
Top 39%
2.6%
10
Frontiers in Neuroscience
○ 256 papers in training set
Top 2%
2.4%
11
Brain Connectivity
○ 25 papers in training set
Top 0.2%
2.4%
12
Computers in Biology and Medicine
○ 128 papers in training set
Top 2%
2.1%
13
Medical Image Analysis
○ 35 papers in training set
Top 0.3%
2.1%
14
Magnetic Resonance Imaging
○ 23 papers in training set
Top 0.2%
2.1%
15
Journal of Neural Engineering
○ 221 papers in training set
Top 1%
1.9%
16
Cognitive Neurodynamics
○ 18 papers in training set
Top 0.1%
1.9%
17
Frontiers in Neuroinformatics
○ 41 papers in training set
Top 0.3%
1.7%
18
Frontiers in Psychiatry
○ 87 papers in training set
Top 1%
1.7%
19
Frontiers in Neuroimaging
○ 11 papers in training set
Top 0.1%
1.5%
20
Artificial Intelligence in Medicine
○ 17 papers in training set
Top 0.4%
1.5%
21
Frontiers in Computational Neuroscience
○ 60 papers in training set
Top 0.8%
1.4%
22
IEEE Access
● 35 papers in training set
Top 1.0%
1.1%
23
PLOS Computational Biology
● 1863 papers in training set
Top 17%
1.1%
24
Journal of Neuroscience Methods
○ 122 papers in training set
Top 1%
1.1%
25
Brain Structure and Function
○ 93 papers in training set
Top 1%
1.0%
26
IEEE Transactions on Biomedical Engineering
● 40 papers in training set
Top 0.9%
1.0%
27
Brain Informatics
○ 10 papers in training set
Top 0.2%
1.0%
28
IEEE Transactions on Neural Systems and Rehabilitation Engineering
● 49 papers in training set
Top 0.9%
1.0%
29
Frontiers in Human Neuroscience
○ 77 papers in training set
Top 2%
1.0%
30
IEEE Transactions on Medical Imaging
● 21 papers in training set
Top 0.4%
0.8%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.