Independency of Functional Connectivity States on Spatiotemporal Resolution of fMRI Data

Functional magnetic resonance imaging (fMRI) is widely used to explore the brain because of its high temporal and spatial resolution. Resting-state fMRI data were studied by many researchers who found the existence of dynamic functional connectivity (dFC). However, it is unclear whether estimation of functional connectivity (FC) states is dependent on temporal and spatial resolution of sampling in fMRI data. In this paper, we addressed this concern by comparing the FC states with varying spatiotemporal resolution of data where different number of regions of interest (ROIs) were randomly chosen to extract the timecourses. These timecourses were then down-sampling to different temporal resolution. Finally, a sliding-window approach was used to estimate the potential FC states in resting-state dFC. The results show that the detection of brain connectivity is insensitive to the spatial and temporal resolution of sampled data in fMRI data, which provides a dimension reduction perspective for research based on fMRI data.