Investigating ACL length, strain and tensile force in high impact and daily activities through machine learning

Roldán Ciudad, Elisa, Reeves, Neil D., Cooper, Glen and Andrews, Kirstie ORCID: https://orcid.org/0000-0002-0860-0604 (2025) Investigating ACL length, strain and tensile force in high impact and daily activities through machine learning. Computer Methods in Biomechanics and Biomedical Engineering. ISSN 1025-5842

Preview

Published Version Available under License Creative Commons Attribution. Download (3MB) | Preview

Abstract

Anterior cruciate ligament (ACL) reconstruction rates are rising, particularly among female athletes, though causes remain unclear. This study: (i) identify accurate machine learning models to predict ACL length, strain, and force during six high-impact and daily activities; (ii) assess the significance of kinematic and constitutional parameters; and (iii) analyse gender-based injury risk patterns. Using 9,375 observations per variable, 42 models were trained. Cubist, Generalized Boosted Models (GBM), and Random Forest (RF) achieved the best R2, RMSE, and MAE. Knee flexion and external rotation strongly predicted ACL strain and force. Female athletes showed higher rotation during cuts, elevating ACL strain and risk.