Peiyu Chen

New paper in Imaging Neuroscience by Peiyu Chen, Zaixu Cui, et al:

Group-common and individual-specific effects of structure–function coupling in human brain networks with graph neural networks

doi.org/10.1162/imag...

Great thanks to my supervisor @zaixucui.bsky.social and all the collaborators!

The individual effects showed a positive association with the sensorimotor-association cortical axis ranks across all regions. The individual effects were the highest in association cortices and were lowest in sensorimotor cortices.

The group effects showed a negative association with the ranks of sensorimotor-association cortical axis across all regions. Particularly, the group effects were highest in sensorimotor cortices and were lowest in association cortices.

Finally, we found that the regional group and individual effects of the structure-function coupling were both hierarchically organized across the cortex along a sensorimotor-association cortical axis.

We found that the structure-function coupling was dominated by group-common effects, with subtle yet significant individual-specific effects. This finding was consistent using GNNs or Pearson correlation to evaluate coupling.

We proposed a graph neural network framework to predict unseen individuals’ functional connectivity with structural connectivity. We defined individuals’ structure-function coupling as the correlation between the actual and predicted functional connectivity.

Group-common and individual-specific effects of structure-function coupling in human brain networks with graph neural networks Abstract. The human cerebral cortex is organized into functionally segregated but synchronized regions bridged by the structural connectivity of white matter pathways. While structure-function couplin...

Our article "Group-common and individual-specific effects of structure-function coupling in human brain networks with graph neural networks" is now published online at Imaging Neuroscience!
direct.mit.edu/imag/article...