Functional magnetic resonance imaging (fMRI) has great potential to evaluate how networks respond and compensate for network dysfunction caused by traumatic brain injury (TBI). In this study, sparse dictionary learning (sDL) and independent component analysis (ICA) were applied to resting-state fMRI (rs-fMRI) data, collected from a group of piglets 1-day (D1) and 7-days (D7) after TBI. Activation maps were generated using group ICA and group sDL, both with dual regression. Voxel-wise permutation tests were then applied to identify changes to six resting-state networks (RSNs). Consistency was observed through the two methods, indicating functional network activity changes after injury.