中国科学院行为科学重点实验室知识库

Resting-state functional MRI (R-fMRI) has emerged as a promising neuroimaging technique used to identify global hubs of the human brain functional connectome. However, most R-fMRI studies on functional hubs mainly utilize traditional R-fMRI data with relatively low sampling rates (e.g., repetition time [TR] = 2 s). R-fMRI data scanned with higher sampling rates are important for the characterization of reliable functional connectomes because they can provide temporally complementary information about functional integration among brain regions and simultaneously reduce the effects of high frequency physiological noise. Here, we employed a publicly available multiband R-fMRI dataset with a sub-second sampling rate (TR = 645 ms) to identify global hubs in the human voxel-wise functional networks, and further examined their test-retest (TRT) reliability over scanning time. We showed that the functional hubs of human brain networks were mainly located at the default-mode regions (e.g., medial prefrontal and parietal cortex as well as the lateral parietal and temporal cortex) and the sensorimotor and visual cortex. These hub regions were highly anatomically distance-dependent, where short-range and long-range hubs were primarily located at the primary cortex and the multimodal association cortex, respectively. We found that most functional hubs exhibited fair to good TRT reliability using intraclass correlation coefficients. Interestingly, our analysis suggested that a 6-minute scan duration was able to reliably detect these functional hubs. Further comparison analysis revealed that these results were approximately consistent with those obtained using traditional R-fMRI scans of the same subjects with TR = 2500 ms, but several regions (e.g., lateral frontal cortex, paracentral lobule and anterior temporal lobe) exhibited different TRT reliability. Finally, we showed that several regions (including the medial/lateral prefrontal cortex and lateral temporal cortex) were identified as brain hubs in a high frequency band (02-03 Hz), which is beyond the frequency scope of traditional R-fMRI scans. Our results demonstrated the validity of multiband R4MRI data to reliably detect functional hubs in the voxel-wise whole-brain networks, which motivated the acquisition of high temporal resolution R-fMRI data for the studies of human brain functional connectomes in healthy and diseased conditions. (C) 2013 Elsevier Inc. All rights reserved.

We thank Dr. Jin-Hui Wang for valuable discussion on TRT reliability. This work was supported by the National Natural Science Foundation of China (Grant Nos. 11205041, 81030028, 81020108022, 81271652, 81201122 and 31221003), the National Science Fund for Distinguished Young Scholars (Grant No. 81225012, YH), the Beijing Natural Science Foundation (Grant No. Z111107067311036, YH), the Beijing Funding for Training Talents (YH), Major Project of National Social Science Foundation (Grant No. 11&ZD186), the Major Joint Fund for International Cooperation and Exchange of the National Natural Science Foundation (Grant No. 81220108014, XNZ).