Comparison of peak joint angles from non-concurrent squat and upper-limb assessment using a markerless Kinotek system...

Three-dimensional motion capture systems are essential for assessing human movement patterns, optimizing performance, and monitoring recovery. Marker-based systems are considered the gold standard but are often impractical in clinical settings due to their complexity. Markerless systems, such as Kinotek, offer a portable and user-friendly alternative. This study compared peak joint angles measured from non-concurrent trials using the Kinotek markerless system with a single-camera setup and the Vicon marker-based system.

Twenty-one healthy adults performed shoulder flexion, shoulder abduction, squat, and trunk rotation movements. Movements were captured using the Kinotek markerless system and a Vicon marker-based system in separate trials in randomized immediate succession. Paired t-tests or Wilcoxon signed-rank tests compared differences in peak joint angles between the two systems, while Pearson's (r) or Kendall's Tau-b (τb) correlation coefficients assessed the relationship between systems. Intra-class correlation coefficients (ICC(A,5)) with 95% Confidence Intervals (CIs) evaluated within-system repeatability across five repetitions.

No significant differences between marker-based and markerless systems were observed for hip and shoulder flexion. Significant differences between systems were found for knee flexion, ankle dorsiflexion, trunk rotation, and shoulder abduction. Moderate correlations were found for left knee flexion (τb = 0.62) and left shoulder abduction (r = 0.58), fair correlations were found for right knee flexion (τb = 0.42), and weak or negligible correlations were found for the other assessed movements. Kinotek demonstrated lower repeatability (ICC range = -0.013-0.908) than the Vicon system (ICC range = 0.324-0.996), with good repeatability for Kinotek in ankle dorsiflexion and shoulder abduction.

The Kinotek system demonstrated limited agreement with the Vicon system for peak joint angle measurements, with moderate or fair correlations observed in 3 of the 12 investigated variables. Significant differences between systems were identified, suggesting that the two systems should not be used interchangeably for all kinematic assessments, and highlights the need for further optimization of portable single-camera markerless technology. While the Kinotek system offers practical advantages for movement evaluation in clinical and field-based environments where traditional marker-based may be impractical, further methodological refinement and validation against gold-standard reference methods are needed before broader application in biomechanical analysis. This study adds to the growing literature on markerless motion capture by identifying areas of limited reliability alongside its practical advantages, supporting its continued development as a more accessible approach to movement analysis.