In patients with traumatic brain injury (TBI), diffuse axonal injury (DAI) accounts for a significant amount of parenchymal injury. Diffusion weighted magnetic resonance imaging (DWI) is known to be sensitive for detecting visible DAI lesions. We focused on detection of non-visible, quantifiable diffusion changes in specific normal-appearing brain regions, using apparent diffusion coefficient (ADC) maps. Thirty-seven adults with TBI were compared to 35 age-matched control patients. DWI was performed and ADC maps were generated. Thirty-one regions of interest (ROI) were manually drawn on ADC maps and ADC values extracted. Brain ROIs were categorized into five zones: peripheral gray matter, peripheral white matter, deep gray matter, deep white matter, and posterior fossa. ADC results were compared with the severity of injury based on the admission Glasgow Coma Scale (GCS 3-8; severe; GSC 9-15 mild/moderate) and with long-term outcome (6-12 months after injury) using the Glasgow Outcome Scale (GOS 1-3, unfavorable; GOS: 4-5, favorable) score. Mean ADC values in all five brain zones were significantly different between TBI subjects and controls (p<or=0.05). Patients with more severe injury (i.e., GCS 3-8) had significantly different mean ADC values than control patients (p<or=0.05) and patients with unfavorable outcomes had significantly higher mean deep gray and white matter ADC values compared to those with favorable outcomes and to controls (p<or=0.05). Thus, ADC maps can be used to detect non-visible DAI lesions. In addition, injuries in the deep gray and white matter may be useful to predict outcome in adult TBI patients.