On-field player workload exposure and knee injury risk monitoring via deep learning

Abstract

In sports analytics, an understanding of accurate on-field 3D knee joint moments (KJM) could provide an early warning system for athlete workload exposure and knee injury risk. Traditionally, this analysis has relied on captive laboratory force plates and associated downstream biomechanical modeling, and many researchers have approached the problem of portability by extrapolating models built on linear statistics. An alternative approach would be to capitalize on recent advances in deep learning. In this study, using the pre-trained CaffeNet convolutional neural network (CNN) model, multivariate regression of marker-based motion capture to 3D KJM for three sports-related movement types were compared. The strongest overall mean correlation to source modeling of 0.8895 was achieved over the initial 33 % of stance phase for sidestepping. The accuracy of these mean predictions of the three critical KJM associated with anterior cruciate ligament (ACL) injury demonstrate the feasibility of on-field knee injury assessment using deep learning in lieu of laboratory embedded force plates. This multidisciplinary research approach significantly advances machine representation of real-world physical models with practical application for both community and professional level athletes.

Estimation of Gait Mechanics Based on Simulated and Measured IMU Data Using an Artificial Neural Network

“…They reported normalized root-mean-squared errors of <20% and correlation coefficients ranging from 0.84 to 0.96. Johnson et al (2018Johnson et al ( , 2019 used pre-trained convolutional neural networks for the prediction of the GRF and the knee joint moment during walking, running and sidestepping based on marker trajectories. They achieved a mean correlation higher than 0.85 for the knee joint moments and GRF.…”

Targeted Achilles Tendon Training and Rehabilitation Using Personalized and Real-Time Multiscale Models of the Neuromusculoskeletal System

“…For example Martin et al (2018) and Keuler et al (2019) described a non-invasive method that involves measuring the speed of the of the shear wave generated in response to tapping the tendon to estimate Achilles tendon loading during locomotion. Further, Johnson and colleagues used reduced motion capture marker sets or wearable technology and big data/machine learning to accurately estimate ground reaction forces and moments ( Johnson et al, 2018, 2019a,2020 ) and knee joint moments ( Johnson et al, 2019b ) from kinematic motion capture data alone. Developments such as these are important because of their potential make direct measurement of ground reaction forces redundant.…”

Personalization improves the biomechanical efficacy of foot progression angle modifications in individuals with medial knee osteoarthritis

“…Traditionally, measuring the KAM requires a force plate, motion capture, and a trained gait analyst. The KAM has been computed outside of the gait laboratory, however, using inertial measurement units with and without force-instrumented shoes 43,44 , as well as with machine learning models that use ground reaction forces or simplified kinematic measures as inputs 45–47 . Further work is necessary to validate the ability of these technologies to accurately select personalized modifications.…”