P4‐158: Nonstationarity in the ‘resting brain's’ modular architecture

Task-free fMRI (TF-fMRI) has great potential for advancing the understanding and treatment of neurologic illness. However, as with all measures of neural activity, variability is a hallmark feature of intrinsic connectivity networks (ICNs) identified by TF-fMRI. This variability has hampered efforts to define a robust metric of connectivity to serve as a biomarker for neurologic illness. We hypothesized that some of this variability is related to the non-stationary nature of the connectivity within ICNs. We performed a large (N=892) high-dimensional independent component analysis decomposition of TF-fMRI data on cognitively normal (CN) subjects drawn from the Mayo Clinic Study of Aging. This was used to atlas the brain into 68 regions, categorized based on network of origin, anatomical locations, and a functional meta-analysis. These regions were used to construct dynamic graphical representations of the brain within a sliding time window. We calculated dwell time in particular network configurations in 892 CN and 28 AD subjects. The 80,280 graphs displayed a highly modular architecture relative to the null model (Q = 0.55 +/-0.06 vs. null Q = 0.16 +/- 0.01, p<0.001). The number of modules varied between 2 and 5, with 23.63% in a 2 module configuration, 72.38% in a 3 modular configuration, 3.98% in a 4 modular configuration, and less than 0.01% in a 5 module configuration. This distribution was similar for the dwell time from single subjects' scanning sessions. The regions of the brain assigned to the same module resembled the most commonly identified ICNs (Figure). We found that Alzheimer's subjects form brain states with a lower proportion of strong pDMN contributions and a higher proportion of strong aDMN contributions (Table). Clustering of Modular Assignment for the 68 Regions of Interest, from the 892 Cognitively Normal Subjects. The dendrogram for the clustering of the 68 ROIs is displayed above. The colored regions in the dendrogram correspond to the similarly colored overlaid collection of ROIs below.