Biomechanical Markers of Forward Hop-Landing After ACL-Reconstruction: A Pattern Recognition Approach

Prasanna Sritharan, Mario A. Muñoz, Peter Pivonka, Adam L. Bryant, Hossein Mokhtarzadeh, Luke G. Perraton

Research output: Contribution to journalArticleResearchpeer-review

14 Citations (Scopus)

Abstract

Biomechanical changes after anterior cruciate ligament reconstruction (ACLR) may be detrimental to long-term knee-joint health. We used pattern recognition to characterise biomechanical differences during the landing phase of a single-leg forward hop after ACLR. Experimental data from 66 individuals 12-24 months post-ACLR (28.2 ± 6.3 years) and 32 controls (25.2 ± 4.8 years old) were input into a musculoskeletal modelling pipeline to calculate joint angles, joint moments and muscle forces. These waveforms were transformed into principal components (features), and input into a pattern recognition pipeline, which found 10 main distinguishing features (and 8 associated features) between ACLR and control landing biomechanics at significance α= 0.05. Our process identified known biomechanical characteristics post-ACLR: smaller knee flexion angle; less knee extensor moment; lower vasti, rectus femoris and hamstrings forces. Importantly, we found more novel and less well-understood adaptations: smaller ankle plantar flexor moment; lower soleus forces; and altered patterns of knee rotation angle, hip rotator moment and knee abduction moment. Crucially, we identified, with high certainty, subtle aberrations indicating landing instability in the ACLR group for: knee flexion and internal rotation angles and moments; hip rotation angles and moments; and lumbar rotator and bending moments. Our findings may benefit rehabilitation and assessment for return-to-sport 12–24 months post-ACLR.

Original languageEnglish
Pages (from-to)330-342
Number of pages13
JournalAnnals of Biomedical Engineering
Volume50
Issue number3
DOIs
Publication statusPublished - Mar 2022

Keywords

  • Anterior cruciate ligament
  • Feature selection
  • Knee osteoarthritis
  • Machine learning
  • Musculoskeletal modelling
  • Principal component analysis

Cite this