Fig. 3: Functional brain fingerprinting performance across methods for fMRI resting-state data. | Nature Communications

Fig. 3: Functional brain fingerprinting performance across methods for fMRI resting-state data.

From: Higher-order connectomics of human brain function reveals local topological signatures of task decoding, individual identification, and behavior

Fig. 3

a For each method, we report the differential identifiability scores obtained when considering the functional connections having at least one node in the functional network analyzed, namely, visual (VIS), somatomotor (SM), dorsal attention (DA), ventral attention (VA), limbic (L), frontoparietal (FP), default mode network (DMN), and subcortical (SC). For graphical clarity, the scaffold method is not reported (<9% for all functional networks). In almost all functional networks, the triangle method outperforms all other approaches. b To investigate the patterns of brain activation that are subject-specific, we report the coefficient of variation (cv) for the triangle nodal strength on the cortical brain surface when averaged over the 100 HCP subjects. Interestingly, the way somatosensory areas interact with higher-order networks (i.e., FP and DMN) is quite variable across subjects, as reflected in the high values of the coefficient of variation. Results are obtained by sampling 80 subjects from a total of 100 and repeated this process 100 times (n = 100). The box plots show the median and interquartile range (IQR), with whiskers extending to Q1-1.5IQR and Q3+1.5IQR. Individual data points, including outliers, are shown using scatter plots.

Back to article page