Material Characterization and Computational Modeling of Human Subcutaneous Adipose Tissue

Obesity is associated with higher fatality risk and altered distribution of occupant injuries in motor vehicle crashes (MVCs) partially because of the increased depth of abdominal soft tissue. This results in limited and/or delayed engagement of the lap belt with the pelvis and increases the risk of pelvis “submarining” under the lap belt, which exposes the occupant’s abdomen to belt loading. Highly automated vehicles (HAVs) may increase the range of seating positions chosen by vehicle occupants, potentially including an increased prevalence of reclined riding postures. These postures may lead to altered lap belt placement and increased posterior pelvis rotation, which could potentially increase the risk of submarining for occupants of all anthropometries, amplifying the challenge of restraining occupants and preventing submarining during MVCs.
Finite element human body models (FE-HBMs) are valuable tools to simulate occupant response and facilitate the understanding of belt-flesh-pelvis interaction during MVCs. However, current HBMs do not possess sufficient biofidelity, especially in the abdominal region, to adequately represent belt-flesh-pelvis interaction. This engineering problem has led to the identification of three research gaps, which motivated this dissertation. First, knowledge on mechanical behavior of human subcutaneous adipose tissue (SAT) under belt loading was lacking. Second, an appropriate formulation for computational implementation of adipose tissue constitutive models remains to be determined due to challenges of modelling ultra-soft impressible materials with Lagrangian elements. Finally, the sensitivity of these computational models of human adipose tissue on belt-flesh-pelvis interaction has not been evaluated at the whole-body level. These research gaps lead to the goal of this dissertation, which is to develop and evaluate constitutive models and advanced computational models for human SAT in conditions relevant to belt loading in MVCs.
To achieve the goal of this dissertation, the first comprehensive material level test series using human adipose tissue under impact loading was performed to inform constitutive model development. Computational models of human adipose tissue were then developed using both Lagrangian-based FE approach and Smoothed Particle Galerkin (SPG) meshfree approach, verified and validated against experimental data. Finally, these computational models were integrated into an FE-HBM to investigate the effect of material stiffness and computational formulation of adipose tissue on belt-flesh-pelvis interaction under belt pull loading. This study will advance the understanding on adipose tissue mechanics and further advance the field of injury biomechanics by providing biofidelic abdominal adipose tissue models as engineering tools.
The main contribution of this dissertation is the largest dataset on adipose tissue mechanical behavior generated in this study, as well as the developed constitutive and computational models for future applications. Incorporating this research into HBMs advanced the state-of-the-art in belt-flesh-pelvis interaction modeling. The findings and developed engineering tools will eventually guide the technical innovation of more effective safety countermeasures, thereby, reducing the societal burden of traffic injuries.