Development of a Detailed Finite Element Model of the BIPED MK2 and Verification of Fidelity in Two Cases of Blunt Impact.

Abstract

Physical surrogates of the human head are commonly used to model cranial impacts, assess helmet efficacy and assess likelihood of head injuries. The Brain Injury Protection Evaluation Device (BIPED mk2) is a head form that contains a brain simulant, cerebrospinal fluid layer (CSF), connective membranes, a skull and a skin layer, and can be configured to measure kinematics, pressures and strains. In design efforts to increase the biofidelity of surrogates, finite element models play a significant role in assessing design iterations that better mimic the biological response of the head during impact. This study aims to create a digital model of the BIPED mk2 and provide a robust comparison to experimental pressure and strain data, measured from specific impact scenarios. Kinematics from two separate frontal impact experiment campaigns were used to drive the BIPED mk2 finite element model. In the first experiments, brain pressure was extracted from in situ transducers. In the second, brain strain was extracted from post hoc imagery analysis. These pressure and strain data are the basis on which we verify the pressures and strains reported from the finite element model. Pressure and displacement time series responses were compared with experimental data using a CORrelation Analysis (CORA). The average CORA rating for pressure measurements taken at the front brain sensor was 0.701 using the kinematic model inputs and 0.851 for the force model inputs. For the rear brain sensor, the signals were deemed poor fits as the average CORA scores were 0.442 for the kinematic input and 0.255 for the force input. CORA ratings for the comparison of displacement data in the x (anterior-posterior) and z (superior-inferior) directions of the 18 nodes tested resulted in a range of values from 0.012 to 0.936. The results matched best in the interior but were poor along the perimeter of the brain depending on the location of the point in relation to the brain surface. We speculate the mixed findings are due in large part to the simplified CSF model, a potential focus for future model refinement.