Static postures effect on scaling accuracy of shoulder complex

The aim of the current study was to compare two common static postures (anatomical and abducted arm postures). The scaling is an important step of musculoskeletal modeling which alters the anthropometry of a model so that it matches a particular subject as closely as possible. Scaling is typically performed by comparing experimental marker data to virtual markers placed on a model. In order to scale the generic model marker data of static posture was used. The common static posture for scaling and marker registration is anatomical posture, although many postures were used to create the scaled model in recent years. Sixteen subjects (۶۴.۳±۵.۵ kg, ۱۷۱±۸.۱ cm, ۲۷.۵±۶.۵ years) participated in this study. Ten Vicon motion captures (Vicon MX, Oxford, UK, ۲۰۰ Hz) were used to record the surface marker trajectories. Four musculoskeletal models in OpenSim were used: (M۱) a model with only three rotational DOFs between humerus and trunk (GHJ), (M۲) a model with three rotational DOFs for STJ, ACJ, and GHJ, (M۳) a model with coupled motions for scapula, clavicle, and humerus, (M۴) a model with an ellipsoid mobilizer for the SternoClavicular Joint. Significant differences between the two postures observed for all models except model M۴. According to results, abducted arm posture had lower marker error for all models. Therefore, the abducted arm posture is a better posture to scaling the shoulder musculoskeletal models.