Characterize cerebrovascular reactivity by measuring the functional topology-informed synchronization of fMRI signals

Motivation: The conventional approach to assess cerebrovascular reactivity involves employing the general linear model to analyze functional MRI (fMRI) signals, with end-tidal CO2 concentration as the regressor. However, spatial variations in delays of CO2-evoked responses due to complex vasculature of the brain may lead to inaccurate detection of cerebrovascular reactivity. Goal(s): Our study is to propose a new approach to characterize cerebrovascular reactivity without the linear assumption. Approach: We proposed a model-free approach to characterize cerebrovascular reactivity by measuring increases in functional topology-informed synchronization of fMRI signals. Results: Our proposed model-free approach could detect cerebrovascular reactivity in a more accurate manner. Impact: The model-free approach characterizes cerebrovascular reactivity by measuring changes in BOLD signal synchronization based on functional topology with a high spatial specificity and sensitivity, which is expected to serve as a standardized tool to measure cerebrovascular reactivity.