Assessing the Utility of a Video-Based Motion Capture Alternative in the Assessment of Lumbar Spine Planar Angular Joint Kinematics

Abstract

Markerless motion capture is a novel technique to measure human movement kinematics. The purpose of this research is to evaluate the markerless algorithm, DeepLabCut (DLC) against a 3D motion capture system (Vicon Motion Systems Ltd., Oxford, UK) in the analysis of planar spine and elbow flexion-extension movement. Data were acquired concurrently from DLC and Vicon for all movements. A novel DLC model was trained using data derived from a subset of participants (training group). Accuracy and precision were assessed from data derived from the training group as well as in a new set of participants (testing group). Two-way SPM ANOVAs were used to detect significant differences between the training vs. testing sets, capture methods (Vicon vs. DLC), as well as potential higher order interaction effect between these independent variables in the estimation of flexion extension angles and variability. No significant differences were observed in any planar angles, nor were any higher order interactions observed between each motion capture modality and the training vs. testing datasets. Bland Altman plots were also generated to depict the mean bias and level of agreement between DLC and Vicon for both training, and testing datasets. Supplemental analyses, suggest that these results are partially affected by the alignment of each participant’s body segments with respect to each planar reference frame. This research suggests that DLC-derived planar kinematics of both the elbow and lumbar spine are of acceptable accuracy and precision when compared to conventional laboratory gold-standards (Vicon).