Resting-State Functional MRI of Healthy Adults: Temporal Dynamic Brain Coactivation Patterns

Background The aging brain is typically associated with aberrant interactions of large-scale intrinsic networks. However, the dynamic variation of these networks’ coactivation or deactivation across the adult lifespan remains unclear. Purpose To promote the interpretation of dynamic brain network variations underlying the complex aging process by quantifying activation levels and obtaining a clear definition of coactivation patterns (CAPs) with resting-state functional MRI (rsfMRI). Materials and Methods In a retrospective study (October 2010 to September 2013), rsfMRI data from healthy participants in the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository were used to generate CAPs by applying single-volume temporal clustering analysis. Spatial clustering analysis was then performed to capture dynamic coactivation and deactivation within or between primary sensory networks and high-order cognitive networks (including the default mode network [DMN], attentional network [AN], and frontoparietal network [FPN]). Linear relationships between dynamic metrics and age were revealed with Spearman partial correlations. Results A total of 614 participants (mean age, 54 years ± 18 [SD]; 311 women) ranging in age from 18 to 88 years were evaluated. There was a negative correlation of the CAPs (Spearman correlations: r = −0.98, P < .001) with loss of coactivation (partial correlations: r = −0.17, P < .001) and deactivation (partial correlations: r = 0.216, P < .001) with aging. The CAPs, characterized by negative correlation patterns between the DMN and AN, occurred (partial correlations: r = 0.14, P = .003) and dwelled (partial correlations: r = 0.10, P = .04) more with aging. Moreover, the AN and DMN CAP transitioned more to the AN and FPN CAP with aging (partial correlations: r = 0.17, P < .001). Conclusion The dynamics of the healthy aging brain are characterized mainly by more flexibility of the high-order cognitive networks while maintaining primary sensory functions (networks). Online supplemental material is available for this article. © RSNA, 2022 See also the editorial by Holodny in this issue.