We review two complementary MRI imaging modalities to characterize structure and function of neuronal networks in the human brain, and their application to subjects with severe brain injury. The structural imaging modality, diffusion tensor imaging, is based on imagining the diffusion of water protons in the brain parenchyma. From the diffusion tensor, several quantities characterizing fiber structure in the brain can be derived. The principal direction of the diffusion tensor has been found to depend on the fiber direction of myelinated axons. It can be used for white matter fiber tracking. The anisotropy (or directional dependence) of diffusion has been shown to be sensitive to developmental as well as white matter changes during training and recovery from brain injury. The functional MRI imagining modality, resting state fMRI, Concerns the functional connectivity of neuronal networks rather than their anatomical structure. Subjects undergo a conventiaon fMRI imaging protocol without performing specific tasks. Various resting state network patterns can be computed by algorithms that reveal correlations in the fMRI signal. Often, thalamic sturcture are involved, suggesting that resting state fMRI could reflect global brain network functionality. Clinical applicaitons of resting state fMRI have been reported, in particular relating singnal abnormalities to neurodegenerative processes. To better understand to which degree resting state patterns reflect neuronal network function, we are comparing network patterns of normal subjects with those having severe brain lesions in a small pilot study.