Modular Reorganization of Brain Resting State Networks and Its Independent Validation in Alzheimer’s Disease Patients

• Main Authors: Guangyu eChen, Hong-Ying eZhang, Chunming eXie, Gang eChen, Zhi-Jun eZhang, Gao-Jun eTeng, Shi-Jiang eLi
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2013-08-01
• Series: Frontiers in Human Neuroscience
• Subjects: Alzheimer’s disease; Validation; graph theory; MCI; brain network; resting state functional connectivity
• Online Access: http://journal.frontiersin.org/Journal/10.3389/fnhum.2013.00456/full

Previous studies have demonstrated disruption in structural and functional connectivity occurring in the Alzheimer’s Disease (AD). However, it is not known how these disruptions alter brain network reorganization. With the modular analysis method of graph theory, and datasets acquired by the resting-state functional connectivity MRI (R-fMRI) method, we investigated and compared the brain organization patterns between the AD group and the cognitively normal control (CN) group. Our main finding is that the largest homotopic module (defined as the insula module) in the CN group was broken down to the pieces in the AD group. Specifically, it was discovered that the eight pairs of the bilateral regions (the opercular part of inferior frontal gyrus, area triangularis, insula, putamen, globus pallidus, transverse temporal gyri, superior temporal gyrus, and superior temporal pole) of the insula module had lost symmetric functional connection properties, and the corresponding gray matter concentration (GMC) was significant lower in AD group. We further quantified the functional connectivity changes with an index (index A) and structural changes with the GMC index in the insula module to demonstrate their great potential as AD biomarkers. We further validated these results with six additional independent datasets (271 subjects in six groups). Our results demonstrated specific underlying structural and functional reorganization from young to old, and for diseased subjects. Further, it is suggested that by combining the structural GMC analysis and functional modular analysis in the insula module, a new biomarker can be developed at the single-subject level.