Access Type

Open Access Dissertation

Date of Award

January 2020

Degree Type


Degree Name



Biomedical Engineering

First Advisor

Liying Zhang


Traumatic brain injury (TBI) is caused by local tissue deformation at the time of trauma, leading to neurological dysfunction. In the United States alone, 2.87 million people sustain a TBI each year, of which one-fifth results in death. Traumatic axonal injury (TAI) is a well-recognized consequence of every fatal head injury and more than 85% of vehicular crash-related blunt head injuries. The most common and important pathologic feature of TBIs are multifocal changes to axons in the white matter produced by rapid head acceleration/deceleration during a traumatic event with consequent local shear/tension on neural tissue and axons contributing to secondary cellular injury. The Head Injury Criterion (HIC) is currently used by the US government for the evaluation and design of vehicular safety systems. HIC does not account for local tissue responses. Currently, the relationship between the local mechanical responses within the brain tissue and subsequent injury to this tissue are not well understood.

Finite element (FE) modeling of in vivo TBI is an effective approach to compute local mechanical response and correlate to the injury location and severity. To our best knowledge, none of the other FE rat models are capable of simulating in vivo closed head impact acceleration injury and predicting widespread TAI in the rat brain.

The goal of this research was to determine mechanical thresholds for white matter injury at the tissue level by correlating the local mechanical response (e.g. strain, stress) from an FE model with axonal pathology from in vivo experiments. An anatomically detailed FE rat head/brain model with a simplified body was developed, and FE simulations of in vivo experiments were performed.

The anatomically detailed rat head model consisted of over 724,000 elements, of which 301,000 were in the brain (200 x 200 x 200 micron). The white matter tissues with highly aligned axonal fibers were modeled with transversely isotropic materials to simulate impact direction-dependent injury in the brain. The rat head/body model was validated against in vivo rodent dynamic cortical deformation, brain-skull displacement, and head impact acceleration experimental data. A series of FE parametric studies were conducted to identify various biomechanical factors contributing to the variability of injury severity observed among experiments and across different labs.

The FE rat model simulated in vivo TAI in rat brains from closed-head impact acceleration experiments. The correlation of the local biomechanical parameter map with the severity and extent of axonal injury map at tissue levels was established. This improved our understanding of tissue-level injury mechanism for white matter injury. The tissue level thresholds for white matter injury was established by logistic regression analysis. The localized severe TAI in cerebral white matter (corpus callosum region) was best predicted by intracranial pressure (81 kPa) and maximum principal strain (0.26), while the white matter tracts in the brainstem (pyramidal tracts) were best predicted by localized maximum principal strain (0.18) response. The tissue level thresholds developed from this study can be directly translated to the FE human head model. This information will enhance the capability of the human head model in predicting brain injury.