Study at a glance

In 25 female varsity soccer players tracked across a full competitive season, Theia3D-based markerless motion capture combined with a novel three-dimensional movement-variability analysis found that players with a history of concussion showed persistently weaker motor control regulation during a simple squat task than players without, a difference that a conventional two-dimensional version of the same analysis was not sensitive enough to detect.

Summary

Disclaimer: The following summary details independent academic research. Theia3D is intended for research, sports performance and educational use only. It is not cleared as a medical device and is not intended to be used for the diagnosis, treatment, mitigation or prevention of any disease, injury or medical condition.

Editor's note: The paper does not report a specific Theia3D software version number or camera brand/model for the markerless system. Here, Theia3D served as the motion-capture tool supporting a novel mathematical movement-analysis framework, rather than as a system being validated against marker-based motion capture within this particular study.

Study Overview

Design

A longitudinal, repeated-measures biomechanical study comparing a novel three-dimensional movement-variability analysis framework against its conventional two-dimensional counterpart for detecting persistent motor control differences associated with concussion history, assessed at three points across a competitive soccer season.

Participants

Twenty-five female varsity soccer players from a single university program (eleven with a self-reported, medically diagnosed history of concussion within the prior three years; fourteen without), all at least 18 years of age, cleared for athletic participation, and able to complete the study's squat task.

Data Collection

Testing was conducted at three time points across a competitive season: pre-season (week 1), mid-season (week 9), and post-season (week 15). At each session, participants performed 40 repetitions of a bilateral bodyweight squat synchronized to a 40-beats-per-minute metronome (1.5 seconds down, 1.5 seconds up per repetition), with the final 30 repetitions retained for analysis to ensure participants had reached a consistent movement pattern. Whole-body kinematics were captured using an eight-camera markerless motion capture system (60 Hz) processed through Theia3D, with center-of-mass trajectories subsequently extracted in Visual3D.

Analysis

The researchers applied an enhanced, three-dimensional Goal Equivalent Manifold (GEM) framework to the squat movement, which mathematically separates trial-to-trial movement variability into components that do not affect task performance ("goal-equivalent") and components that do ("goal-relevant"), using the ratio between them to calculate a Performance Index. Unlike the conventional two-dimensional version of this analysis, which considers only the center of mass's displacement and velocity, the three-dimensional version explicitly incorporated timing error relative to the metronome beat as a third execution variable. Performance Index values and temporal persistence metrics (how consistently a player's variability pattern repeated from one squat to the next) were compared between the concussion-history and non-concussion groups across the three season time points using a generalized linear model accounting for repeated measures.

Key Findings

  • Players with a history of concussion showed a significantly lower three-dimensional Performance Index across the season than players without a concussion history, reflecting weaker regulation of goal-relevant movement variability during the squat task.
  • This difference remained consistent across the pre-season, mid-season, and post-season testing points, with no significant change over time or interaction between season stage and concussion history, suggesting a stable, persistent effect rather than one that narrowed as the season progressed.
  • The conventional two-dimensional version of the same analysis, using only movement displacement and velocity, did not detect a statistically significant difference between the concussion-history and non-concussion groups at any point in the season.
  • Timing error, the new third dimension added in this study's enhanced framework, accounted for an average of about 12% of the total movement variability captured across the season, directly demonstrating that incorporating timing precision added meaningful, previously uncaptured information to the analysis.
  • The two-dimensional and three-dimensional performance scores were consistently and positively correlated with each other at each time point, indicating that the three-dimensional approach extended, rather than contradicted, the conventional analysis while adding greater sensitivity to detect the group difference.

What This Means for Movement Science Researchers and Athlete Monitoring and Concussion-Recovery Teams

For researchers and performance staff involved in monitoring athletes with a concussion history, this study offers a methodologically important finding: a season-long, repeated-measures approach combined with a more sensitive, three-dimensional movement-variability framework was able to detect a persistent motor control difference that neither a single-timepoint snapshot nor the conventional two-dimensional version of the same analysis picked up. This suggests that teams relying on standard screening tools or isolated, one-off assessments after a concussion may be missing subtle but real, lasting differences in how an athlete regulates movement variability, even well after they've been cleared to return to play and are no longer reporting symptoms. The markerless motion-capture approach used here also illustrates a practical advantage for this kind of repeated, season-long athlete monitoring: capturing detailed, whole-body kinematic data across multiple testing sessions throughout a competitive season without the repeated marker setup and placement time that a marker-based approach would otherwise require.

The Role of Theia3D in Movement Analysis

Theia3D served as the markerless motion-capture tool in this study, enabling detailed, whole-body kinematic tracking of a simple, repeatable squat task across three separate testing sessions spread throughout a competitive soccer season. Its use here supported a novel, more sensitive mathematical movement-analysis framework rather than serving as the subject of validation itself, illustrating how markerless capture's practicality for frequent, repeated testing can enable this kind of longitudinal, season-long research design that would be considerably more burdensome to carry out with a marker-based system.

Study Limitations and Future Directions

The authors note that their relatively small sample size limited the statistical power of the study and the generalizability of its findings, and they recommend a larger sample in future research to build further confidence in the robustness of the group differences observed. Participants with a concussion history varied in how recently that concussion had occurred, within the prior three years, with some within the past year, and this variation may have introduced variability that masked any time-dependent recovery patterns that might otherwise have been detectable. The squat task was performed at a single, fixed pace set by the metronome, chosen based on a pilot study and practical testing constraints, which may not have reflected each individual athlete's natural movement rhythm or true maximal motor control capacity. The authors identify testing at a range of movement speeds, following athletes from the acute period after a concussion through later stages of recovery, and combining this kind of movement-variability analysis with other biomechanical and neurological assessment methods as important next steps for understanding the underlying mechanisms behind these persistent motor control differences.

Citation: Zhao, X., Rahimi, R., Cournoyer, J., Clouthier, A., & Graham, R. B. (2026). Detection of season-long post-concussion motor deficits in female soccer players using a 3D goal equivalent manifold framework. Scientific Reports, 16, 23415. https://doi.org/10.1038/s41598-026-49059-0

Link: Read the full study

Can Motion Capture Detect Lasting Motor Effects of Concussion in Soccer Players?

This study of 25 female varsity soccer players uses Theia3D markerless motion capture and a novel three-dimensional movement-variability framework to show that players with a history of concussion exhibit persistently weaker motor control regulation across a full competitive season, a difference a conventional two-dimensional version of the same analysis was not sensitive enough to detect. If your work involves athlete monitoring or concussion-recovery research, talk to our team about applying Theia3D to your own studies.

Recent Posts