Development of a finite element model of the human shoulder

Stapp Car Crash J. 2000 Nov:44:281-97. doi: 10.4271/2000-01-SC19.

Abstract

Previous studies have hypothesized that the shoulder may be used to absorb some impact energy and reduce chest injury due to side impacts. Before this hypothesis can be tested, a good understanding of the injury mechanisms and the kinematics of the shoulder is critical for occupant protection in side impact. However, existing crash dummies and numerical models are not designed to reproduce the kinematics and kinetics of the human shoulder. The purpose of this study was to develop a finite element model of the human shoulder in order to achieve a deeper understanding of the injury mechanisms and the kinematics of the shoulder in side impact. Basic anthropometric data of the human shoulder used to develop the skeletal and muscular portions of this model were taken from commercial data packages. The shoulder model included three bones (the humerus, scapula and clavicle) and major ligaments and muscles around the shoulder. This model was then integrated into a human thorax model developed at Wayne State University (WSU) along with pre-existing models of other body parts such as the pelvis and the lower extremities. Material properties used for the model were taken from the literature. The model was first used to simulate lateral shoulder impact study by the Association Peugeot- Renault (APR) followed by simulations of several of the 17 rigid and padded cadaveric impacts conducted on a side impact sled at WSU. Contact forces measured at the levels of shoulder, thorax, abdomen and pelvis were used as response variables to validate the model. Additionally, a cadaveric test involving the deployment of a generic side airbag was also used to check the validity of the model. Model prediction of accelerations of the shoulder matched well against those measured experimentally. The role of the shoulder in side impact protection and the reduction of injury to the ribcage are discussed, based on model results.