Author(s): SLEVIN, M., VALENTIN, S., UGBOLUE, U., EASTON, C., WHITE, A.D., FORREST, L.J., Institution: UNIVERSITY OF THE WEST OF SCOTLAND, Country: UNITED KINGDOM, Abstract-ID: 1962

INTRODUCTION:
Marker-based motion capture systems are deemed the gold standard method to collect and analyse human movement within sport. With technology constantly evolving, marker-less systems are now readily available to evaluate movement out with a laboratory environment. However, the validity of these systems within the applied sporting environment requires attention. This study aimed to validate the use of Vald Performance’s HumanTrak Movement Analysis System against the ‘gold standard’ Vicon Motion Capture System during single- and double-leg squatting exercises.
METHODS:
Twenty healthy, recreationally active adults (10 males; 10 females; age, 24.05 ± 3.62 years) completed two exercises: (1) single-leg squat (SLS) and (2) overhead squat (OHS) whilst kinematic data was simultaneously obtained from the marker-less system HumanTrak Movement Analysis System, and a gold standard 3D motion analysis system (Vicon Motion Capture System). Specifically, knee flexion angles were obtained during SLS and OHS, and knee varus/valgus was obtained for SLS only. Bland-Altman plots with 95% CI were used to depict levels of agreement (LoA) for knee angles and spearman’s correlation coefficients to report relationships between systems. Statistical significance was accepted at p < 0.05.
RESULTS:
No systematic bias was reported for knee flexion during both the SLS (-2.64; LoA between -38.15 and 32.88) and OHS (-2.07; LoA between -35.98 and 31.84), however wide limits of agreement report variability between the systems. A trend was illustrated during SLS knee varus/valgus (bias = 3.14; LoA between -28 and 34.27), the two systems systematically produced different results and wide limits of agreement reported variability between the two systems. There was a significant positive correlation between HumanTrak and Vicon during knee flexion of both the SLS (r = 0.683, p < 0.001) and OHS (r = 0.843, p < 0.001) but no significant relationship between HumanTrak and Vicon were identified for knee varus/valgus during the SLS (r = 0.104, p = 0.663).
CONCLUSION:
Knee flexion during both the single-leg and overhead squat exercises reported high accuracy between the two systems. Knee varus/valgus angles collected from HumanTrak were inaccurate as a result of very poor agreement when compared to Vicon during single-leg squat movements. HumanTrak Movement Analysis system is a valid marker-less system to assess knee flexion angles within an applied environment, however, should not be used to assess knee varus/valgus angles.