Multiple studies have demonstrated sex-related differences in types and risk of lower extremity injuries in motor vehicle crashes (e.g., Carter et al. 2014; Rupp and Flannagan 2011). These differences are thought to be in part due to sex-related differences in the sizes and shapes of lower extremity bones that affect the interaction between the body and the vehicle seat and restraint system (Besnault et al. 1998; Riggs et al. 2004; Wang et al. 2004). As a result, for crash test dummies or anthropomorphic test devices (ATDs) to reproduce sex-specific differences in lower extremity injury type and risk, they must appropriately capture the sex-specific differences in skeletal and soft tissue geometry. In the pelvis, it is critical for the size and shape of the iliac wings to be humanlike to have realistic interaction with seat belts and vehicle structures. The ischial tuberosities also must be in the correct location for reasonable interaction with vehicle seats. Despite differences between the sexes in lower extremity bone shape, the shapes of female ATD skeletal components have typically been established by scaling male geometry, usually based on a characteristic length (Schneider et al. 1983; Rhule and Backaitis 1998; Humanetics Innovative Solutions, Plymouth, MI). As a result, the bones of small female ATDs may have the appropriate size (or an appropriate correct dimension), but not a representative shape. In the pelvis, sex-specific differences in shape exist that may affect the interaction of the pelvis with vehicle belts and side structures (Wang et al. 2004). One way to account for this would be to use a single female pelvis as the ATD design target; however, such an approach does not necessarily result in a pelvis that has the typical size and shape. A better approach is to use a pelvis described by averaging skeletal surface landmark locations for a particular range of sizes, as was done by Reynolds et al. (1981). In this study, landmark locations were averaged from all pelvises from women under the 25th percentile in height in a sample of post-mortem pelvises; however, this approach involves some averaging and is somewhat limited by the idiosyncrasies of the sample. A still better approach is to use a statistical shape model that is based on a large number of pelvises to predict the geometry associated with a particular set of occupant characteristics. An early attempt to develop such a parametric model of the pelvis by Besnault et al. (1998) did not consider occupant characteristics, such as age and BMI, which affect injury risk. The first known definition of an ATD skeletal component via statistical shape analysis was the pediatric pelvis developed by Reed et al. (2009) and tested by Klinich et al. (2010). Later work by Klein (2015) resulted in statistical geometry models for the male and female pelvises that are parameterized by age, BMI, and bispinous breadth. In this report, the surface geometry for the small female pelvis is predicted using the Klein (2015) female pelvis model. The resulting geometry is compared to the Hybrid III small female pelvis geometry and an estimate of female pelvis geometry obtained by length scaling the midsize male pelvis based on bispinous breadth. (Author/publisher)
Samenvatting