Back to Reference Library
behaviour
nutrition
riding science
2024
Cohort Study

Objective Assessment of Equine Locomotor Symmetry Using an Inertial Sensor System and Artificial Intelligence: A Comparative Study.

Authors: Calle-González Natalie, Lo Feudo Chiara Maria, Ferrucci Francesco, Requena Francisco, Stucchi Luca, Muñoz Ana

Journal: Animals : an open access journal from MDPI

Summary

# Editorial Summary Inertial measurement units (IMUs) have become established tools for objective gait analysis in equine practice, yet their expense, setup demands, and sensitivity to sensor placement create practical barriers for widespread adoption. Researchers compared IMU systems against a newer artificial intelligence marker-less motion tracking system (AI-MTS) alongside traditional visual clinical assessment across 20 horses evaluated on varied surfaces and curved tracks, measuring symmetry parameters including vertical head and pelvis displacement. The AI-MTS demonstrated greater sensitivity than IMUs, flagging asymmetry in more limbs overall, though agreement between the two systems varied considerably—strongest for vertical head displacement on straight hard ground but notably weaker for pelvis assessment on soft footing, highlighting the persistent challenge of quantifying hindlimb asymmetry. Whilst clinicians identified more horses as sound via visual inspection than either gait analysis system classified as symmetric, neither technology yet has established clinical relevance thresholds, meaning subtle asymmetries detected by AI-MTS may or may not warrant therapeutic intervention. Implementing these systems systematically throughout training could help the profession determine which degrees of asymmetry correlate with performance impairment or injury risk, potentially refining early detection protocols and individualised conditioning programmes.

Read the full abstract on PubMed

Practical Takeaways

  • AI-based motion tracking offers a more sensitive, marker-less alternative to traditional IMU systems for objective gait assessment in practice, potentially reducing setup time and cost
  • Hindlimb asymmetry assessment remains challenging across all systems; head position analysis in hard ground conditions provides most reliable data for symmetry evaluation
  • Visual clinical assessment alone may miss subtle but potentially significant asymmetries; objective gait analysis should complement rather than replace clinical judgment

Key Findings

  • AI marker-less tracking system (AI-MTS) demonstrated greater sensitivity in detecting asymmetry than IMU systems, identifying asymmetry in more limbs
  • Greatest agreement between AI-MTS and IMU systems was for vertical head position difference in straight hard conditions; lowest agreement for vertical pelvis position in straight soft conditions
  • More horses were classified as sound by clinical visual assessment compared to those identified as symmetric by either gait analysis system
  • Clinical relevance thresholds for asymmetry remain undefined, requiring further standardization through regular training use

Conditions Studied

gait asymmetrylocomotor symmetry assessment