Proof of concept – automated qualitative scoring of movement patterns using joint center positions collected using motion capture

BACKGROUND: Qualitative scoring of movement patterns is
a common element in gross motor test batteries for children and adults. The typical procedure in these tests is to
have a subject perform a given movement while an examiner observes and subsequently scores the movement by evaluating subject-posture at or between key events (e.g., in forward jumping: +1 point if “Arms are extended in
front of the body and above the head at toe-off”). Markerless motion capture makes it possible to easily capture and accurately quantify human locomotion without disturbing the natural movements of the subject being evaluated. Furthermore, most systems can describe subject posture as 3D joint-center positions on a frame-by-frame basis. This makes it possible to create software algorithms that automatically identify key events and evaluate
postures for most movements evaluated in popular movement screening batteries. The present work proves the concept of automating the qualitative scoring of standing broad jumps. The work shows how the identification of
key events and the subsequent postural evaluation of subjects performing standing broad jumps can be automated using joint center positions and simple algorithms.
DISCUSSION: The proposed method makes it possible to quickly and reliably evaluate large quantities of motion-capture data. However, the technique needs to be fine-tuned and validated before applying it in cohort studies or clinical practice. Furthermore, the accuracy of the automated qualitative scoring algorithm will be limited by the quality of the motion capture data it is applied to.
PERSPECTIVES: Primary proposed research: The Motor Skills in PreSchools (MiPS) cohort contains motion capture data of +600 children performing standing broad jumps at the ages of 3, 4, 5, and 6 years. If the proposed
method proves valid, future research can use it to describe the early locomotor development of jumps in typically developing children. Other potential research: The proposed method of using joint-center positions to break a movement down to key events and evaluate posture can be applied to a wide range of activities. Therefore, automated scoring of popular screening tests, such as the
Functional Movement Screen, are obvious targets for future development of the method.