Structured networks observed in resting fMRI noise

We observe structure similar to functional connectivity networks within the data we typically remove from fMRI time series during noise correction. Physiological and head motion regressors were used to extract noise variance from resting-state BOLD fMRI data in a general linear model. Independent component analysis identified several key networks, including the default mode network, in the noise datasets. These structures are also observed when using random noise regressors uncoupled to physiology or motion, indicating that this phenomenon is inherent to the removal of any regressors of similar frequencies.

This abstract and the presentation materials are available to members only; a login is required.

Join Here