Study of functional brain networks in Alzheimer's disease based on 11C-PiB PET images

40 Objectives Prior to this study, brain networks are constructed based on fMRI, FDG-PET, etc. This is the first study based on graph theory using 11C-PiB PET data to investigate the characteristics of whole-brain functional network in Alzheimer9s disease. Methods PiB-PET image data of 149 individuals, including 120 from ADNI database (https://ida.loni.usc.edu) and 29 from Huashan Hospital were analyzed. Among which, 34 were Alzheimer’s disease (AD), 43 Mild Cognitive Impairment (MCI) and 72 healthy control (HC). The imaging data was pre-processed using Statistical Parametric Mapping 8 (SPM8). The sparsity threshold method was used to determine the connection between the two brain regions. After thresholding, the correlation coefficient matrix was transformed into a binary matrix that was described as a network. To further investigate the detailed connectivity associated with the brain regions, seed ROI-based correlation analysis method was performed, using ORBinf.L as seed. Results At sparsity 24%, several brain regions were identified as functional hubs in three groups as shown in Figure 1. Among the hubs, significant changes were found in 11 brain regions in AD and MCI group compared with HC: AD>HC: ORBinf.L, PoCG.L, HES.R; (MCI>HC): PreCG.L, SFGdor.R, SMA.R, OLF.R, REC.R, ACG.L, PCG.R, ANG. L. as shown in Figure 2 and detailed in Table 1. Conclusions It is feasible to investigate functional network of AD using 11C-PiB PET imaging. Global efficiency was lower but local efficiency was higher in both MCI and AD compared with HC. The hub regions may play a crucial role in the pathogenesis of AD. Figure 1 hub nodes,HC(left, red), MCI(middle, black)and AD(right, blue) Figure 2 black spots: HC < MCI; blue spots: HC < AD $$graphic_5C6BC201-717C-468C-AA60-31DB9D8DC0F0$$ $$graphic_525AAF4D-72B0-4233-9A14-E5351E84461D$$ Table 1 Detailed information of altered hubs