Alternative modelling procedures for pelvic marker occlusion during motion analysis

BackgroundMotion analysis of participants with different body shapes under diverse conditions can be problematic when vital markers are occluded. The markers located over the anterior superior iliac spines are commonly occluded in older patients and during analysis of activities with trunk and hip flexion which can prevent accurate calculation of lower limb joint kinematics. Options to modify standard body models exist but have not been described in detail, and the effects on the lower limb kinematics are not known.MethodsThree-dimensional motion analysis data were collected from 10 participants during level walking. A single trial from each participant was processed using the standard PlugIn Gait model and with four alternative modelling procedures where either one or both anterior pelvis markers were not labelled for all or part of the trial. Similarity of these alternative procedures to PlugIn Gait was assessed by comparison of peak kinematic characteristics and Root Mean Square (RMS) across the gait cycle.FindingsThe peak lower limb kinematics of all four alternative modelling procedures were similar to PlugIn Gait to within 4.57°. The alternative procedure most similar to PlugIn Gait was less than 1.24° different. The largest RMS was 2.88° and the smallest was 0.92°.InterpretationThis study has presented several options for researchers and clinicians to modify the standard body models of motion analysis so that lower limb kinematics may be calculated without reliance on continuous visualisation of anterior pelvic markers. Although the alternative modelling processes are subject to different sources of error which need to be considered, the error is minimal.