The goal of this study was to develop a motion-based injury criterion for brain injuries derived from the material response of the brain tissue, under the assumption that impact response of the brain tissue can be characterized by a standard linear solid. Focus was given to brain injuries that are deemed to correlate with the strain of the brain tissue, including subarachnoid hemorrhage, intracerebral hemorrhage and diffuse axonal injury. The criterion is based on rotational motion of the head because of incompressibility of the brain tissue that allows large strain primarily in rotation.
The stiffness and damping parameters of one-dimensional Kelvin model were determined for each axis of rotation of the head in such a way that scaled displacement time history matches strain time history of the brain tissue predicted by the Global Human Body Models Consortium (GHBMC) head-brain model. The convolution integral of the impulse response of the model was used to predict strain time history of the brain when an arbitrary rotational acceleration time history is applied to the head. The maximum value of the predicted strain was defined as a new brain injury criterion (Convolution of Impulse response for Brain Injury Criterion; CIBIC). Head rotational acceleration data were taken from a number of crash test data representing full frontal, oblique frontal and side impacts along with pedestrian impact simulation results to investigate correlation between the values of various brain injury criteria, including CIBIC, and the maximum principal strain from the head-brain model.
The injury criterion proposed by this study, CIBIC, resulted in a better correlation with the predicted maximum principal strain of the brain relative to those proposed by past studies in all of the four crash configurations (R 2 ranging from 0.624 to 0.864). It was also found that the coefficient of determination was smaller for the impact conditions resulting in multiple or long-duration loading than other impact configurations representing single short-duration loading.