In an average year more than 1.7 million people will experience a traumatic brain injury (TBI) in the United States. It is known that atrophy occurs across a spectrum for TBI patients, ranging from mild to severe. Current conventional magnetic resonance imaging (MRI) methods are inconsistent in detecting this atrophy on the milder end of the spectrum. Also more contemporary imaging tools, although efficient, are too time consuming for clinical applicability. It is for these reasons that a quick and efficient measurement for detecting this atrophy is needed by clinicians. The measuring of third ventricle width had the potential to be this measurement, since it is known that ventricular dilation is an indirect measure of brain atrophy. This study used two different data sets acquired at multiple sites. A total of 152 TBI patients' MRI scans were analyzed with diagnosis ranging from mild to severe. They have been age matched with 97 orthopedic injury controls. All scans were analyzed using Freesurfer® auto-segmentation software to acquire cortical, subcortical, and ventricular volumes. These metrics were then used as a standard of efficacy which we tested the new third ventricle width protocol against. There was no statistically significant difference between the overall TBI group and OI group (Welch's F(1,238.435) = 1.091, p= .267). The complicated mild injury subgroup was significantly increased from the mild subgroup (p= .001, d= .87). The grand average third ventricle width measurement was the best prognosticator of all measures analyzed despite only predicting 35.1% of cases correctly. The findings suggest that the third ventricle width measurement is insensitive to atrophy between all groups as hypothesized.



College and Department

Life Sciences; Physiology and Developmental Biology



Date Submitted


Document Type





traumatic brain injury, traumatic axonal injury, third ventricle width