Back to Reference Library
farriery
veterinary
biomechanics
anatomy
nutrition
physiotherapy
2025
Cohort Study

Objective movement asymmetry in horses is comparable between markerless technology and sensor-based systems.

Authors: Kallerud Anne S, Marques-Smith Patrick, Bendiksen Helle K, Fjordbakk Cathrine T

Journal: Equine veterinary journal

Summary

# Editorial Summary Markerless AI-based lameness detection represents a significant technological development for equine practitioners, but its reliability under real-world conditions needed rigorous comparison with established inertial measurement unit (IMU) systems. Kallerud and colleagues evaluated a markerless iPhone-based system (SleipAI) alongside two commercial IMU devices (Equinosis Q and EquiMoves) and subjective assessment in 41 client-owned horses during straight-line trotting, classifying limbs as symmetric or asymmetric and comparing the normalised asymmetry data across platforms. The markerless and sensor-based systems demonstrated moderate agreement in identifying asymmetric limbs (kappa 0.59 for forelimbs, 0.44 for hindlimbs), with the two IMU systems showing the strongest concordance; importantly, most discrepancies arose from different asymmetry thresholds rather than fundamentally different measurements of the same asymmetry. A notable limitation emerged with SleipAI analysing significantly fewer hindlimb strides (27±6) compared to the Lameness Locator (45±13) and EquiMoves (53±11), potentially restricting data robustness for hindlimb assessment. For practitioners, these findings suggest that markerless AI technology can provide comparable objective asymmetry detection to IMU systems under field conditions when thresholds are standardised, offering a more accessible entry point to objective gait analysis, though its reduced stride capture for hindlimbs warrants consideration when assessing distal pelvic or tarsal pathologies.

Read the full abstract on PubMed

Practical Takeaways

  • Markerless AI lameness detection technology is sufficiently comparable to established IMU systems for field-based asymmetry classification, offering a potentially more accessible alternative for routine screening
  • Markerless systems may have limitations with hindlimb analysis (fewer strides captured), so practitioners should be aware that forelimb assessments may be more reliable with this technology
  • When choosing between objective systems, understand that discrepancies often reflect different asymmetry thresholds rather than genuine disagreement—define your threshold based on clinical relevance for your use case rather than relying on default settings

Key Findings

  • Markerless AI system (SleipAI) showed comparable classification of asymmetric limbs to two IMU systems (Equinosis LL and EquiMoves) under field conditions with moderate inter-rater agreement (k=0.59 forelimbs; 0.44 hindlimbs)
  • EquiMoves analyzed significantly more strides (53±11) than both SleipAI and Equinosis LL for both forelimbs and hindlimbs (p<0.006)
  • SleipAI analyzed significantly fewer hindlimb strides (27±6) compared to Equinosis LL (45±13; p<0.001), representing a potential limitation of the markerless technology
  • Strongest agreement in normalized asymmetry data was found between the two IMU systems (EquiMoves and Equinosis LL), with discrepancies between systems largely attributable to imposed asymmetry thresholds rather than fundamental disagreement on affected limbs

Conditions Studied

lamenessmovement asymmetry

Related References

Comparing Inertial Measurement Units to Markerless Video Analysis for Movement Symmetry in Quarter Horses.

Pfau, Landsbergen, Davis, Kenny, Kernot, Rochard, Porte-Proust, Sparks, Takahashi, Toth, Scott(2023)Sensors (Basel, Switzerland)
Farriery

Agreement between veterinarians and three objective evaluation systems in naturally occurring equine lameness.

McPeek Jenna L, Menarim Bruno, Sponseller Beatrice, McClendon Margaret, Adam Emma N, Adams Amanda A, Slone Stacy, Page Allen E(2025)Equine veterinary journal

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

Calle-Gonz&#xe1;lez Natalie, Lo Feudo Chiara Maria, Ferrucci Francesco, Requena Francisco, Stucchi Luca, Mu&#xf1;oz Ana(2024)Animals : an open access journal from MDPI

Agreement between subjective evaluations and a markerless AI-based gait analysis system during lungeing assessment in traditional racehorses.

Meistro F, Ralletti M V, Rinnovati R, Spadari A(2025)Journal of equine veterinary science

Investigation of Thresholds for Asymmetry Indices to Represent the Visual Assessment of Single Limb Lameness by Expert Veterinarians on Horses Trotting in a Straight Line.

Macaire Claire, Hanne-Poujade Sandrine, De Azevedo Emeline, Denoix Jean-Marie, Coudry Virginie, Jacquet Sandrine, Bertoni L&#xe9;lia, Tallaj Am&#xe9;lie, Audigi&#xe9; Fabrice, Hatrisse Chlo&#xe9;, H&#xe9;bert Camille, Martin Pauline, Marin Fr&#xe9;d&#xe9;ric, Chateau Henry(2022)Animals : an open access journal from MDPI