A Multiscale Computational Approach to Estimating Axonal Damage under Inertial Loading of the Head

Abstract

A computational modeling framework is developed to estimate the location and degree of diffuse axonal injury (DAI) under inertial loading of the head. DAI is one of the most common pathological features of traumatic brain injury and is characterized by damage to the neural axons in the white matter regions of the brain. We incorporate the microstructure of the white matter (i.e., the fiber orientations and fiber dispersion) through the use of diffusion tensor imaging (DTI), and model the white matter with an anisotropic, hyper-viscoelastic constitutive model. The extent of DAI is estimated using an axonal strain injury criterion. A novel injury analysis method is developed to quantify the degree of axonal damage in the fiber tracts of the brain and identify the tracts that are at the greatest risk for functional failure. Our modeling framework is applied to analyze DAI in a real-life ice hockey incident that resulted in concussive injury. To simulate the impact, two-dimensional finite element (FE) models of the head were constructed from detailed MRI and DTI data and validated using available human head experimental data. Acceleration loading curves from accident reconstruction data were then applied to the FE models. The rotational (rather than translational) accelerations were shown to dominate the injury response, which is consistent with previous studies. Through this accident reconstruction, we demonstrate a conceptual framework to estimate the degree of axonal injury in the fiber tracts of the human brain, enabling the future development of relationships between computational simulation and neurocognitive impairment.