Evaluation of Traumatic Brain Injury Patients Using a Shape-Constrained Deformable Model

Traumatic brain injury (TBI) is often associated with life long neurobehavioral effects in survivors. Imaging has historically supported the detection and acute management of life-threatening complications. However, in order to predict these long term consequences in the increasing number of individuals surviving TBI, there is an emerging need for structural neuroimaging biomarkers that would facilitate detection of milder injuries, allow recovery trajectory monitoring, and identify those at risk for poor functional outcome and disability. This paper presents a methodology capable of identifying such structural biomarkers in MR images of the brain. Results are presented demonstrating the quantitative accuracy of the approach with respect to (i) highly accurate annotations from expert tracers, (ii) an alternative segmentation method in FSL, and (iii) the ability to reproduce statistically significant differences in the volumes of specific structures between well-defined clinical cohorts (TBI vs age-matched healthy controls) in a retrospective analysis study.