Exploring the Influence of Cortical Microstructural Diffusion on Functional Brain Networks

Objective The cortex is the primary source of blood-oxygenation-level dependent (BOLD) signals, and it is often believed that the brain structural modules serve as the cornerstones of the functional networks. Although a great deal of work had been done in the past to map the white matter fiber connections among various cortical regions in order to clarify the functional connectivity, it is still unclear how the diffusion property of the inner cortex's microstructure relates to the functional brain networks. This study aims to investigate the connection between the canonical brain functional networks and the complexity of cortical microstructural diffusion. Methods Kurtosis diffusion (DK) and resting state functional MRI data from 30 healthy volunteers were collected. The group level networks of default mode network (DMN), executive control network (ECN), dorsal attention network (DAN), salience network (SN), sensorimotor network (SMN) and visual network (VN) were extracted, and network masks were made on the T1 gray matter images after segmentation. Then, the diffusion parameter maps with kurtosis and tensor properties were calculated from the DK data, and co-registrated with the cortical T1 images. These diffusion kurtosis parameters of AK, RK, MK, KFA, and tensor parameters of AD, RD, MD, FA values in each individual were extracted based on each of the function network mask. The diffusion parameter values of above networks were analyzed by ANOVA method of non-parametric test. Results Statistical analysis showed that the values of AK, RK, MK and KFA in low-order networks (SMN, VN) were significantly higher than those in high-order networks (DMN, ECN, DAN and SN), and no significant differences were observed within either the low-order or high-order networks respectively. The values of RD in SN were significantly lower than those in VN and SMN. The values of FA in SN and ECN were significantly lower than those in SMN. Conclusions Our knowledge of the foundations of brain networks has advanced as a result of the findings, which suggested that the kurtosis diffusion of microstructure within the human cerebral cortex is topologically distributed and corresponds to the hierarchy between low-order and high-order functional networks. The DKI-specific diffusion model is suitable for mapping the networks of structural inner cortices.