< Back to previous page

Publication

Focal Traumatic Brain Injury Biomechanics: An Experimental and Computational Framework towards the Evaluation of Protective Headgear

Book - Dissertation

Traumatic brain injury (TBI) is a global health concern and a major cause of mortality and lifelong disability, affecting millions of people of all ages yearly. Victims of bicycle accidents are particularly prone to focal and diffuse TBI, with cerebral contusion being one of the most frequently observed focal injuries. Since reversing the outcome of TBI is, as of yet, therapeutically impossible, research on prevention and protection strategies is crucial. Biomechanical investigation of TBI using experimental and computational techniques, often in synergy, provides the scientific basis to identify injury criteria to develop safer protective headgear such as bicycle helmets. However, there is still an open discussion in the TBI biomechanics community over the mechanical response of brain tissue to focal TBI. Moreover, the injury criteria used in the current bicycle helmet evaluation standards are suboptimal, as they neglect the critical role of brain tissue deformation, often linked to the rotational kinematics of the head, in the onset of brain injury. The focus of the dissertation is to bring insight into the response mechanism of brain tissue to focal TBI, aiming at providing tools to improve the assessment of bicycle helmet efficacy. In silico models, in casu finite element (FE) models, are widely used to study the mechanical response of the human head to TBI-like loading scenarios and determine injury criteria. The reliability of the predictions depends on the accuracy of the brain tissue material properties assigned to the FE model, which, however, are often calibrated either with animal experiments or with tests performed on human brain tissue under loading conditions not representative for TBI. To fill this research gap, we carried out localized force-relaxation tests on fresh human cadaveric brain samples under loading conditions relevant to focal TBI (35% strain and 10/s strain rate) using a microindentation apparatus. The samples were obtained from twelve brain regions, including the cerebrum, cerebellum, and brainstem. The experimental data were used to calibrate the material parameters of a neo-Hookean quasi-linear viscoelastic constitutive model. We observed marked regional differences, highlighting the need to update FE models with region-specific material properties. In addition, the regional human brain tissue material parameters were compared to the material parameters of pig, rat, and mouse brain tissue obtained through identical testing conditions. The results suggest that the pig brain tissue is a suitable surrogate in the absence of human brain tissue, supporting the relevance of animal experiments for biomechanical investigations of TBI. In previous work within our group, controlled cortical impact (CCI) tests were performed on pig brains in vivo to investigate the impact conditions leading to cerebral contusion since the onset mechanism of this type of focal TBI is still a matter for debate. We developed a FE model of the porcine brain to simulate the CCI tests and estimate the brain tissue strains and strain rates, with the aim to derive cerebral contusion tissue-level injury metrics. We also developed a machine learning surrogate of the FE porcine brain model to quantify, in a computationally efficient way, the uncertainty and sensitivity of these metrics to the experimental and modeling parameters. Our combined in vivo - in silico data analysis indicates a better performance of brain maximum principal strain and maximum shear strain as tissue-level injury metric for cerebral contusion, providing helpful information for the assessment of protective headgear. Our findings also highlight the potential of using machine learning for computationally efficient TBI biomechanics investigations. The last work of this dissertation describes pendulum impact tests performed on three different bicycle helmet designs with (MIPS, WaveCel) and without (conventional EPS-only liner) head rotation-damping technologies. Compared to the commonly used drop tests, the pendulum device produced amplified head rotational kinematics over linear kinematics. The impacts were performed at two impact velocities (4.1 m/s and 5.4 m/s) and four impact locations (frontal, temporal, frontotemporal, occipital) relevant to TBI-inducing bicycle accidents scenarios. The helmet's performance was assessed by considering head linear and rotational kinematics metrics as well as brain tissue deformation metrics, the latter obtained by simulating the impacts with the state-of-the-art KTH FE head model. The results indicate different performances depending on the impact location, injury metric, and brain region considered. This step beyond the current standards, which only consider the head's peak linear acceleration, stresses the need to perform more thorough helmet performance evaluations, including global head kinematics and tissue-level injury criteria, and multi-directional impacts. The experimental and computational research conducted during the doctorate contributed to the understanding of the mechanical response of brain tissue to focal TBI. Future protective headgear performance evaluation protocols should include brain tissue deformation metrics derived from FE head models, whose reliability could be improved with the insights brought by the research presented in this dissertation.
Publication year:2022
Accessibility:Open