Accurately modeling the human brain functional correlations with hypergraph Laplacian

The relationship between the structural connectivity (SC) of human brain networks and their functional connectivity (FC) is of immense importance in understanding brain cognition and disorders and has gained significant attention in the neuroscience over the past decade. However, the underlying mechanism of how SC gives rise to the whole pattern of FC especially the negative correlations still remains poorly understood. In this paper, we propose a model that can accurately simulate the resting-state human brain functional correlations based on hypergraph Laplacian, including the negative correlations. We firstly assume that, for each brain region, there are some links showing positive correlations and some other links showing negative correlations. Then we derive a hypergraph model with the two matrices indicating positive and negative correlations respectively using the SC of human brain network. We apply the model to two empirical datasets and the results show that the simulated FC can achieve higher Pearson correlations with the empirical FC, outperforming the state-of-art graph diffusion (GD) model. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

The study explores the relationship between the structural connectivity (SC) and functional connectivity (FC) in human brain networks, proposing a model based on hypergraph Laplacian to simulate brain functional correlations, including negative correlations. By utilizing the SC of human brain networks, the model can accurately simulate empirical data and outperform the state-of-art graph diffusion model.