TY - GEN
T1 - Exploring Brain Function-Structure Connectome Skeleton via Self-supervised Graph-Transformer Approach
AU - Kang, Yanqing
AU - Wang, Ruoyang
AU - Shi, Enze
AU - Wu, Jinru
AU - Yu, Sigang
AU - Zhang, Shu
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.
PY - 2023
Y1 - 2023
N2 - Understanding the relationship between brain functional connectivity and structural connectivity is important in the field of brain imaging, and it can help us better comprehend the working mechanisms of the brain. Much effort has been made on this issue, but it is still far from satisfactory. The brain transmits information through a network architecture, which means that the regions and connections of the brain are significant. The main difficulties with this issue are currently at least two aspects. On the one hand, the importance of different brain regions in structural and functional integration has not been fully addressed; on the other hand, the connectome skeleton of the brain, plays the role in common and key connections in the brain network, has not been clearly studied. To alleviate the above problems, this paper proposes a transformer-based self-supervised graph reconstruction framework (TSGR). The framework uses the graph neural network (GNN) to fuse functional and structural information of the brain, reconstructs the brain graph through a self-supervised model and identifies the regions that are important to the reconstruction task. These regions are considered as key connectome regions which play an essential role in the communication connectivity of the brain network. Based on key brain regions, the connectome skeleton can be obtained. Experimental results demonstrate the effectiveness of the proposed method, which obtains key regions and connectome skeleton in the brain network. This provides a new angle of view to explore the relationship between brain function and structure. Our code is available at https://github.com/kang105/TSGR.
AB - Understanding the relationship between brain functional connectivity and structural connectivity is important in the field of brain imaging, and it can help us better comprehend the working mechanisms of the brain. Much effort has been made on this issue, but it is still far from satisfactory. The brain transmits information through a network architecture, which means that the regions and connections of the brain are significant. The main difficulties with this issue are currently at least two aspects. On the one hand, the importance of different brain regions in structural and functional integration has not been fully addressed; on the other hand, the connectome skeleton of the brain, plays the role in common and key connections in the brain network, has not been clearly studied. To alleviate the above problems, this paper proposes a transformer-based self-supervised graph reconstruction framework (TSGR). The framework uses the graph neural network (GNN) to fuse functional and structural information of the brain, reconstructs the brain graph through a self-supervised model and identifies the regions that are important to the reconstruction task. These regions are considered as key connectome regions which play an essential role in the communication connectivity of the brain network. Based on key brain regions, the connectome skeleton can be obtained. Experimental results demonstrate the effectiveness of the proposed method, which obtains key regions and connectome skeleton in the brain network. This provides a new angle of view to explore the relationship between brain function and structure. Our code is available at https://github.com/kang105/TSGR.
KW - Brain Function
KW - Brain Structure
KW - Graph Neural Network
KW - Self-supervised
KW - Transformer
UR - http://www.scopus.com/inward/record.url?scp=85174700129&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-43993-3_30
DO - 10.1007/978-3-031-43993-3_30
M3 - 会议稿件
AN - SCOPUS:85174700129
SN - 9783031439926
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 308
EP - 317
BT - Medical Image Computing and Computer Assisted Intervention – MICCAI 2023 - 26th International Conference, Proceedings
A2 - Greenspan, Hayit
A2 - Greenspan, Hayit
A2 - Madabhushi, Anant
A2 - Mousavi, Parvin
A2 - Salcudean, Septimiu
A2 - Duncan, James
A2 - Syeda-Mahmood, Tanveer
A2 - Taylor, Russell
PB - Springer Science and Business Media Deutschland GmbH
T2 - 26th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2023
Y2 - 8 October 2023 through 12 October 2023
ER -