A drop landing screening approach to monitor an individual using functional data analysis: An ACL injury case study

Posted on April 30, 2019 by

This case study assesses whether landing behaviour in vertical ground reaction force time-series changes relative to healthy baseline data during the rehabilitation process post ACL reconstruction.

At a glance:

  • Dimension reduction approaches such as functional principal components analysis (fPCA) can detect important characteristics of movement that would otherwise go undetected.
  • Applying fPCA in conjunction with z-scores was demonstrated to be a valuable tool for clinicians when monitoring athletic movement during a rehabilitation process.
  • Clinicians and researchers who are interested in this approach can find software (and a tutorial) for applying fPCA at functionaldata.org.

Stephens, J. M., Chapman, D. W., Tate, K. & Warmenhoven, J. A drop landing screening approach to monitor an individual using functional data analysis: An ACL injury case study. Journal of Science and Medicine in Sport. Epub ahead of print


OBJECTIVES: To explore the practicality of using functional principal components analysis (fPCA) and intra-athlete z-score changes for individual athlete monitoring post-ACL injury and surgery.

DESIGN: A single athlete case study using within-athlete repeated measures in the context of applied athlete monitoring.

METHODS: Using single leg (left) drop landing (3 landings per session) onto a force plate, the athlete completed 6 sessions prior (healthy) and 3 sessions post-ACL injury/surgery. Maximum vertical ground reaction force (vGRF), time to stabilisation (TTS) and outputs from fPCA (fPC scores) for the healthy sessions were used to develop intra-athlete means and standard deviations for each variable. Post-surgery measures were given z-scores relative to the healthy mean and standard deviation for each variable. The standard normal deviate (z = 1.96) was used as a threshold to flag landings that could be indicative of changes in movement behaviour.

RESULTS: Maximum vGRF revealed no post-surgery trials that exceeded the standard normal deviate threshold based on the healthy data. TTS identified one landing post-surgery that exceeded the threshold. Scores for fPC2, fPC3 and fPC4 revealed landings that exceeded the threshold, with fPC4 demonstrating landings greater than the threshold for every trial (except two) post-surgery.

CONCLUSIONS: Including fPCA identified significant and stable changes to the landing strategy (particularly within fPC4). When used in conjunction with other measures such as maximum vGRF and TTS, fPCA has the potential to provide meaningful insights into athlete monitoring models regarding changes to movement characteristics after injury.

KEYWORDS: Force; Injury; Landings; Monitoring

No Comments

Leave a Reply