The probability of traumatic brain injuries based on tissue-level reliability analysis

Motor vehicle crashes are one of the leading causes of traumatic brain injuries. Restraint systems of cars are evaluated by crash tests based on human tolerance data, however, the reliability of data currently used has been questioned several times in the literature due to the neglect of certain types of effects, injury types and uncertainties. Our main goal was to re-evaluate the currently applied risk curve by taking the previously neglected effects into account.

In this paper, the probability of traumatic brain injury was determined by reliability analysis where different types of uncertainties are taken into account. The tissue-level response of the human brain in the case of frontal crashes was calculated by finite element analyses and the injury probability is determined by Monte Carlo simulations. Sensitivity analysis was also performed to identify which effects have considerable contribution to the injury risk.

Our results indicate a significantly larger injury risk than it is predicted by current safety standards. Accordingly, a new risk curve was constructed which follows a lognormal distribution with the following parameters: μLN = 6.5445 and SLN = 1.1993. Sensitivity analysis confirmed that this difference primarily can be attributed to the rotational effects and tissue-level uncertainties.

Results of the tissue-level reliability analysis enhance the belief that rotational effects are the primary cause of brain injuries. Accordingly, the use of a solely translational acceleration based injury metric contains several uncertainties which can lead to relatively high injury probabilities even if relatively small translational effects occur.