Abstract
Summary Statement
An alternative way to assess three dimensional skin motion artefacts of kinematic models is presented and applied to a novel kinematic foot model. Largest skin motion is measured in the tarsal region.
Introduction
Motion capture systems are being used in daily clinical practise for gait analysis. Last decade several kinematic foot models have been presented to gain more insight in joint movement in various foot pathologies. No method is known to directly measure bone movement in a clinical setting. Current golden standard is based on measurement of motion of skin markers and translation to joint kinematics. Rigid body assumptions and skin motion artefacts can seriously influence the outcome of this approach and rigorous validation is required before clinical application is feasible. Validation of kinematic models is currently done via comparison with bone pin studies. However, these studies can only assess major bones in a highly invasive way; another problem is the non-synchronous measurement of skin markers and bone pins. Recently the Glasgow Maastricht kinematic foot model, which comprises all 26 foot segments, has been presented. To validate the model we propose a novel non-invasive method for the assessment of skin motion artefact, involving loaded CT data.
Patients & Methods
25 subjects (healthy and pathological feet) have undertaken CT scans. These CT-scans have been obtained in 1 unloaded and 3 varying loading conditions. CT-slices are 3D reconstructed and segmented. The principal axes of the segmented bones were derived from the surface points of the bones. These principal axes are used to compute bone orientation. Subsequently, coordinate systems of bones in the different loading conditions were matched. Markers were translated and rotated to orientations of their corresponding bones. Maximal distance between markers is calculated per subject to asses the influence of skin motion.
Results
Preliminary results of 9 subjects show largest positional differences for markers associated with the cuneiform lateralis (5.7 ± 3.2 mm) and cuneiform intermedium (7.7 ± 3.7 mm). Smallest positional differences are found on the hallux proximalis (0.9 ± 0.34mm). Spatial resolution is too small to accurately calculate orientation of smaller bones, therefor distal phalanges 2–5 are not taken into account in the analysis.
Discussion/Conclusion
Skin motion is a major cause of inaccuracy in gait analysis. This is the first study presenting an automated non-invasive method to calculate the 3D orientation of skin markers with respect to the coordinate system of the corresponding bone(s). Largest skin motion is measured in the tarsal region. Future work will be in calculation of the effect of skin motion in the accuracy of joint angle calculation.
