Wearable Sensor-Based Prediction of Knee Joint Moments in Competitive Rugby Players During Multidirectional Movements.
- 1Human Performance Laboratory, Faculty of Kinesiology, University of Calgary, Calgary, Canada.
- 2Biomedical Engineering Graduate Program, University of Calgary, Calgary, Canada.
Abstract
This study aimed to develop an interpretable machine learning framework using plantar-pressure sensors to estimate knee moments during rugby-specific movements. A total of 31 North American competitive rugby players were recruited to perform four common rugby movements on an infilled artificial turf surface in a motion capture laboratory. Synchronized data acquisition was performed for plantar pressure, motion capture, and ground reaction forces. Knee moments were calculated via inverse dynamics. Genetic Algorithm-Deep Forest Regression (GA-DFR) was used to select plantar-pressure sensors and predict resultant moments within participant-level nested cross-validation. Four alternative models predicted the resultant using the same layouts and participant splits. Complementary deep forest regression (DFR) models predicted the three moment components using these layouts. SHapley Additive exPlanations (SHAP) interpreted held-out predictions. Layouts contained 7-18 sensors. Pooled resultant R² was 0.795, with root mean square error of 0.0512 N·m/N. DFR performed better than linear regression and the convolutional neural network, with no significant differences from the deep neural network or long short-term memory network. Pooled R² was 0.806, 0.662, and 0.675 for flexion-extension, adduction-abduction, and internal-external rotation moments, respectively. Resultant temporal SHAP contributions exceeded their stance-wide mean during 9%-55% of stance, peaking at 24%. Reduced plantar-pressure inputs supported estimation of resultant and direction-specific knee-moment waveforms in unseen rugby players.
Related
- Comparison of Cumulative Dissipated Energy Between Two Phacoemulsification Systems Using a Modified Flip-and-Chop Technique With Multidirectional Jerking Movement From a Phacoemulsification Handpiece.
- A multidirectional extradural corridor for pretemporal endoscope-assisted orbitozygomatic–anterior petrosectomy
- Isokinetic Knee Strength and Multidirectional Jump-Landing Asymmetry in Professional Football Players After Anterior Cruciate Ligament Reconstruction: An Exploratory Cross-Sectional Study
- Maternal Recognition of Ultrasound-Observed Fetal Movements Among Pregnancies in the Third Trimester: Concordance, Directional Discrepancies, and Associated Factors
- Bickerstaff Brainstem Encephalitis with a Decorticate-like Posture, Involuntary Movements, Impaired Visual Acuity, and Prolonged Motor Disability
BibTeX
@article{gao2026wearable,
title = {{Wearable Sensor-Based Prediction of Knee Joint Moments in Competitive Rugby Players During Multidirectional Movements.}},
author = {Zixiang Gao and Robin Trama and David McNish and Michael Esposito and Jake Ruschkowski and Darren J Stefanyshyn and John W Wannop},
journal = {Scandinavian journal of medicine \& science in sports},
year = {2026},
month = oct,
doi = {10.1111/sms.70391},
url = {https://europepmc.org/article/MED/42961009},
note = {CC BY},
}
RIS
TY - JOUR
TI - Wearable Sensor-Based Prediction of Knee Joint Moments in Competitive Rugby Players During Multidirectional Movements.
AU - Zixiang Gao
AU - Robin Trama
AU - David McNish
AU - Michael Esposito
AU - Jake Ruschkowski
AU - Darren J Stefanyshyn
AU - John W Wannop
PY - 2026
DA - 2026/10/30
JO - Scandinavian journal of medicine & science in sports
DO - 10.1111/sms.70391
UR - https://europepmc.org/article/MED/42961009
AB - This study aimed to develop an interpretable machine learning framework using plantar-pressure sensors to estimate knee moments during rugby-specific movements. A total of 31 North American competitive rugby players were recruited to perform four common rugby movements on an infilled artificial turf surface in a motion capture laboratory. Synchronized data acquisition was performed for plantar pressure, motion capture, and ground reaction forces. Knee moments were calculated via inverse dynamics. Genetic Algorithm-Deep Forest Regression (GA-DFR) was used to select plantar-pressure sensors and predict resultant moments within participant-level nested cross-validation. Four alternative models predicted the resultant using the same layouts and participant splits. Complementary deep forest regression (DFR) models predicted the three moment components using these layouts. SHapley Additive exPlanations (SHAP) interpreted held-out predictions. Layouts contained 7-18 sensors. Pooled resultant R² was 0.795, with root mean square error of 0.0512 N·m/N. DFR performed better than linear regression and the convolutional neural network, with no significant differences from the deep neural network or long short-term memory network. Pooled R² was 0.806, 0.662, and 0.675 for flexion-extension, adduction-abduction, and internal-external rotation moments, respectively. Resultant temporal SHAP contributions exceeded their stance-wide mean during 9%-55% of stance, peaking at 24%. Reduced plantar-pressure inputs supported estimation of resultant and direction-specific knee-moment waveforms in unseen rugby players.
SN - 0905-7188
SN - 1600-0838
ER -