Accuracy of a markerless motion capture system for balance related quantities

Abstract

Background Balance studies usually focus on quantities describing the global body motion, such as the position of the whole-body centre of mass (CoM), its associated extrapolated centre of mass (XCoM) and the whole-body angular momentum (WBAM). Assessing such quantities using classical marker-based approach can be tedious and modify the participant's behaviour. The recent development of markerless motion capture methods could bypass the issues related to the use of markers. Research question Can we use markerless motion capture systems to study quantities that are relevant for balance studies? Methods Sixteen young healthy participants performed four different motor tasks: walking at self-selected speed, balance loss, walking on a narrow beam and countermovement jumps. Their movements were recorded simultaneously by marker-based and markerless motion capture systems. Videos were processed using a commercial markerless pose estimation software, Theia3D. The position of their CoM was computed, and the associated XCoM and WBAM were derived. Bland-Altman analysis was performed and root mean square error and coefficient of determination were computed to compare the results obtained with marker-based and markerless methods across all participants and tasks. Results Bias remained of the magnitude of a few mm for CoM and XCoM position, and RMSE of CoM and XCoM was around 1 cm. Confidence interval for CoM and XCoM was under 2 cm except for one task in one direction. RMSE of the WBAM was less than 8% of the total amplitude in any direction, and bias was less than 1%. Significance Results suggest that the markerless motion capture system can be used in balance studies as the measured errors are in the range of the differences found between different models or populations in the literature. Nevertheless, one should be careful when assessing dynamic movements such as jumping, as they displayed the biggest errors.