A structural enriched functional network: An application to predict brain cognitive performance

Abstract

The structure-function coupling in brain networks has emerged as an important research topic in modern neuroscience. The structural network could provide the backbone of the functional network. The integration of the functional network with structural information can help us better understand functional communication in the brain. This paper proposed a method to accurately estimate the brain functional network enriched by the structural network from diffusion magnetic resonance imaging. First, we adopted a simplex regression model with graph-constrained Elastic Net to construct the functional networks enriched by the structural network. Then, we compared the constructed network characteristics of this approach with several state-of-the-art competing functional network models. Furthermore, we evaluated whether the structural enriched functional network model improves the performance for predicting the cognitive-behavioral outcomes. The experiments have been performed on 218 participants from the Human Connectome Project database. The results demonstrated that our network model improves network consistency and its predictive performance compared with several state-of-the-art competing functional network models.

Keywords: Functional network; Graph-constrained elastic net; Simplex regression; Structure-function coupling.

Figure 1.. Visualizing the average network pattern across subjects.

Sub-figures (a) and (b) visualized the average network pattern before thresholding and after thresholding, respectively. The brain regions were ordered according to the Cole-Anticevic Brain-side Network (Ji et al., 2019).

Figure 2.. Boxplots of density values for seven different brain network approaches.

Boxplots of density values for seven different brain network approaches.

Figure 3.. The comparison of proportion of common connection (POC) for seven different approaches for building the networks.

The POC was computed after thresholding the networks. Each subfigure has two rows, where the first row visualized the heatmap of the POC, and the second row visualized the histogram of the POC. The histogram distribution following the power law denoted lower inter-subject variability.

Figure 4.. The activation pattern maps and the related cognitive topics for predicting WM-2bk-acc score.

Top row: the mean standardized regression coefficients from the prediction model and their decoding results using SFN_simplex. Bottom row: the mean standardized regression coefficients from the prediction model and their decoding results using FN_simplex. For each sub-figure, left column: mean standardized regression coefficients map. Center column: selection probability map. Right column: word clouds plot related to cognitive function in the Neurosynth database.

Figure 5.. The activation pattern maps and the related cognitive topics for predicting the gF score.

Top row: the mean standardized regression coefficients from the prediction model and their decoding results using SFN_simplex. Bottom row: the mean standardized regression coefficients from the prediction model and their decoding results using FN_simplex. For each sub-figure, left column: mean standardized regression coefficients map. Center column: selection probability map. Right column: word clouds plot related to cognitive function in the Neurosynth database.