FREE SHIPPING over $40
OVER 9000 FIVE-STAR REVIEWS
5% OFF ALL SUBSCRIPTIONS
100% HAPPINESS GUARANTEE
Analyze Diet
0
Home / Equine Research Bank / Equine Vet J / Validation of a handheld smartphone markerless gait-analysis...
Equine veterinary journal2026; doi: 10.1002/evj.70149

Validation of a handheld smartphone markerless gait-analysis tool using an estimated groundline in horses.

Abstract: A handheld smartphone-based computer vision algorithm (RealHorse® [RH]) offers accessible alternatives for equine gait analysis but requires validation against a gold-standard three-dimensional multicamera optical motion capture system (Qualisys® [QS]). Objective: To evaluate the accuracy and precision of RH in measuring vertical displacement signals (VDS) at the eye, withers, back and croup in horses trotting on a straight line and on a circle. Methods: Cross-sectional comparative validation study of a markerless computer vision algorithm. Methods: Fifty-nine horses were recorded while trotting on a straight line and 24 were lunged on a circle. RH detected two-dimensional anatomical keypoints on each frame, which were used to estimate a dynamic groundline and compute ground relative VDS with stride-based difference in maxima (Maxdiff) and minima (Mindiff). QS provided synchronous ground-relative VDS reference values. Agreement was evaluated using mean signed error, mean absolute error and Bland-Altman analysis. Results: On the straight line (n = 2620 strides), the pooled stride-level MAE for Maxdiff and Mindiff was 3.8 mm. Keypoint-specific errors were 5.1 mm (eye), 4.3 mm (withers) and 3.0 mm (croup). On the circle (n = 2419 strides), pooled stride-level error increased to 5.5 mm. Trial-level analysis (n = 58 trials) showed much lower errors: 1.4 mm for both eye and withers and 1.1 mm for croup. On the circle (n = 24 trials), trial-level errors were higher, with 2.8 mm for the eye, 1.8 mm for the withers and 3.3 mm for the croup. The back keypoint consistently showed the lowest errors across both stride and trial levels. Conclusions: RH measurements of the croup Mindiff during circling resulted in higher values and showed the largest error. Conclusions: RH measured vertical displacement of all keypoints with high accuracy and precision (trial-level MAE 1.1-1.4 mm straight, 1.8-3.3 mm circle), supporting its use for equine gait analysis.
© 2026 The Author(s). Equine Veterinary Journal published by John Wiley & Sons Ltd on behalf of EVJ Ltd.
Get Full Text
Publication Date: 2026-01-16 PubMed ID: 41546456DOI: 10.1002/evj.70149Google Scholar: Lookup
The Equine Research Bank provides access to a large database of publicly available scientific literature. Inclusion in the Research Bank does not imply endorsement of study methods or findings by Mad Barn.
LinkedInCopy
  • Journal Article

Summary

This research summary has been generated with artificial intelligence and may contain errors and omissions. Refer to the original study to confirm details provided. Submit correction.

Overview

  • This study evaluated the accuracy and precision of a handheld smartphone-based gait analysis tool (RealHorse®) by comparing it to a high-quality, established motion capture system (Qualisys®) when measuring vertical movements of horse body parts during trotting.

Introduction

  • The RealHorse® (RH) tool uses smartphone cameras and computer vision to track horse gait without markers, making it potentially accessible and practical for routine use.
  • Validating RH against the gold-standard Qualisys® (QS) 3D motion capture system is necessary to confirm if RH reliably measures vertical displacements in key anatomical points during horse locomotion.

Objective

  • To assess how accurately and precisely RH measures vertical displacement signals (VDS) at four anatomical points: eye, withers, back, and croup in horses trotting in two conditions — straight-line and circular paths.

Methods

  • Participants:
    • Fifty-nine horses recorded trotting on a straight line.
    • Twenty-four horses lunged on a circular track.
  • Data Collection:
    • RH recorded video using smartphone cameras and detected 2D anatomical keypoints frame-by-frame.
    • The keypoints were used to estimate a dynamic groundline, allowing calculation of vertical displacement relative to the ground.
    • Displacement metrics were stride-based differences in maxima (Maxdiff) and minima (Mindiff).
    • QS simultaneously captured 3D ground-relative VDS reference data to compare with RH results.
  • Data Analysis:
    • Errors between RH and QS measurements were calculated using mean signed error (bias), mean absolute error (MAE), and Bland-Altman agreement analysis.
    • Analysis was performed both at the stride-level (individual steps) and trial-level (entire run) for more generalized accuracy assessment.

Results

  • Straight-Line Trotting:
    • Total of 2,620 strides analyzed.
    • Pooled stride-level MAE was 3.8 mm for Maxdiff and Mindiff combined.
    • Errors by keypoint:
      • Eye: 5.1 mm
      • Withers: 4.3 mm
      • Croup: 3.0 mm
    • At the trial level (58 trials), errors were smaller:
      • Eye and withers: 1.4 mm
      • Croup: 1.1 mm
  • Circular Trotting (lunging):
    • 2,419 strides analyzed.
    • Stride-level pooled MAE increased to 5.5 mm.
    • Trial-level error (24 trials) was higher than straight line:
      • Eye: 2.8 mm
      • Withers: 1.8 mm
      • Croup: 3.3 mm
    • The back keypoint consistently showed the lowest error across all analyses.
    • The croup Mindiff during circling had the largest errors and higher displacement values, indicating this was the most challenging measurement for RH.

Conclusions

  • The smartphone-based RH tool demonstrated high accuracy and precision in measuring vertical displacements of key anatomical points of trotting horses compared to the 3D motion capture gold standard.
  • Errors at the trial level were generally very low (1.1 to 3.3 mm), indicating reliable measurements sufficient for clinical or research gait analysis applications.
  • Greater errors during circular trotting, especially at the croup, suggest some limitations in measurement under certain conditions but still within an acceptable range.
  • Overall, RH presents a feasible, accessible option for markerless equine gait analysis outside of specialized lab environments.

Cite This Article

APA
Key K, Kirkegaard J, Berg K, Andresen KR, Skov Hansen S. (2026). Validation of a handheld smartphone markerless gait-analysis tool using an estimated groundline in horses. Equine Vet J. https://doi.org/10.1002/evj.70149

Publication

Equine veterinary journal
ISSN: 2042-3306
NlmUniqueID: 0173320
Country: United States
Language: English

Researcher Affiliations

Key, Karsten
  • Keydiagnostics (RealHorse®), Fredensborg, Denmark.
Kirkegaard, Jakob
  • Keydiagnostics (RealHorse®), Fredensborg, Denmark.
Berg, Katja
  • Keydiagnostics (RealHorse®), Fredensborg, Denmark.
Andresen, Kristian Ringkjær
  • iKeyVet®, Fredensborg, Denmark.
Skov Hansen, Sabrina
  • Department of Veterinary Clinical Sciences, University of Copenhagen, Copenhagen, Denmark.

References

This article includes 43 references
  1. The prevalence of lameness in FEI equine athletes and its correlation to performance – AAEP. [cited 2025 September 23]. Available from: https://aaep.org/post/the‐prevalence‐of‐lameness‐in‐fei‐equine‐athletes‐and‐its‐correlation‐to‐performance/
  2. Arkell M, Archer RM, Guitian FJ, May SA. Evidence of bias affecting the interpretation of the results of local anaesthetic nerve blocks when assessing lameness in horses. Vet Rec 2006;159:346–349.
  3. Fuller CJ, Bladon BM, Driver AJ, Barr ARS. The intra‐ and inter‐assessor reliability of measurement of functional outcome by lameness scoring in horses. Vet J 2006;171:281–286.
  4. Hammarberg M, Egenvall A, Pfau T, Rhodin M. Rater agreement of visual lameness assessment in horses during lungeing. Equine Vet J 2016;48:78–82.
  5. Parkes RSV, Weller R, Groth AM, May S, Pfau T. Evidence of the development of ‘domain‐restricted’ expertise in the recognition of asymmetric motion characteristics of hindlimb lameness in the horse. Equine Vet J 2009;41:112–117.
  6. Hewetson M, Christley RM, Hunt ID, Voute LC. Investigations of the reliability of observational gait analysis for the assessment of lameness in horses. Vet Rec 2006;158:852–857.
  7. New global survey: 95% of orthopaedic vets say earlier detection could prevent more serious lameness cases. [cited 2025 September 23]. Available from: https://sleip.com/news/new‐global‐survey‐95‐of‐orthopaedic‐vets‐say‐earlier‐detection‐could‐prevent‐more‐serious‐lameness‐cases
  8. Buchner HH, Savelberg HH, Schamhardt HC, Barneveld A. Head and trunk movement adaptations in horses with experimentally induced fore‐ or hindlimb lameness. Equine Vet J 1996;28:71–76.
  9. Rhodin M, Smit IH, Persson‐Sjodin E, Pfau T, Gunnarsson V, Björnsdóttir S. Timing of vertical head, withers and pelvis movements relative to the footfalls in different equine gaits and breeds. Animals 2022;12:3053.
  10. Starke SD, May SA, Pfau T. Understanding hind limb lameness signs in horses using simple rigid body mechanics. J Biomech 2015;48:3323–3331.
  11. Weishaupt MA, Wiestner T, Hogg HP, Jordan P, Auer JA. Compensatory load redistribution of horses with induced weightbearing hindlimb lameness trotting on a treadmill. Equine Vet J 2004;36:727–733.
  12. England JT, Rutherford D. A cross‐sectional survey investigating challenges in lameness identification among equine owners. Equine Vet J 2024;56:34–35.
  13. Serra Bragança FM, Rhodin M, van Weeren PR. On the brink of daily clinical application of objective gait analysis: what evidence do we have so far from studies using an induced lameness model?. Vet J 2018;234:11–23.
  14. Roepstorff C, Dittmann MT, Arpagaus S, Serra Bragança FM, Hardeman A, Persson‐Sjödin E. Reliable and clinically applicable gait event classification using upper body motion in walking and trotting horses. J Biomech 2021;114:110146.
  15. Serra Bragança FM, Roepstorff C, Rhodin M, Pfau T, van Weeren PR, Roepstorff L. Quantitative lameness assessment in the horse based on upper body movement symmetry: the effect of different filtering techniques on the quantification of motion symmetry. Biomed Signal Process Control 2020;57:101674.
  16. Keegan KG. Evidence‐based lameness detection and quantification. Vet Clin North Am Equine Pract. 2007;23:403–423.
  17. Pfau T, Witte TH, Wilson AM. A method for deriving displacement data during cyclical movement using an inertial sensor. J Exp Biol. 2005;208:2503–2514.
  18. Keegan KG, Yonezawa Y, Pai PF, Wilson DA, Kramer J. Evaluation of a sensor‐based system of motion analysis for detection and quantification of forelimb and hind limb lameness in horses. Am J Vet Res. 2004;65:665–670.
  19. Key K, Berg K, Kirkegaard J, Ringkjær Andresen K, Skov Hansen S. Evaluating the accuracy of a vision‐based algorithm for groundline estimation in trotting horses using multiple camera angles, including handheld applications. Vet Med Sci. 2026;12. https://doi.org/10.1002/vms3.70739
  20. Scott B, Seyres M, Philp F, Chadwick EK, Blana D. Healthcare applications of single camera markerless motion capture: a scoping review. PeerJ. 2022;10:e13517.
  21. Pereira TD, Aldarondo DE, Willmore L, Kislin M, Wang SS‐H, Murthy M, et al. Fast animal pose estimation using deep neural networks. Nat Methods. 2019;16:117–125.
  22. Mathis A, Mamidanna P, Cury KM, Abe T, Murthy VN, Mathis MW, et al. DeepLabCut: markerless pose estimation of user‐defined body parts with deep learning. Nat Neurosci. 2018;21:1281–1289.
  23. Cao Z, Hidalgo G, Simon T, Wei S‐E, Sheikh Y. OpenPose: Realtime multi‐person 2D pose estimation using part affinity fields. 2018 arXiv.org. [cited 2024 October 3]. Available from: https://arxiv.org/abs/1812.08008v2
  24. Esteva A, Chou K, Yeung S, Naik N, Madani A, Mottaghi A, et al. Deep learning‐enabled medical computer vision. NPJ Digit Med. 2021;4:1–9.
  25. Wang Y, Li J, Zhang Y, Sinnott RO. Identifying lameness in horses through deep learning. Proceedings of the 36th Annual ACM Symposium on Applied Computing. New York, NY, USA: Association for Computing Machinery; 2021. p. 976–985. https://doi.org/10.1145/3412841.3441973
  26. Lawin FJ, Byström A, Roepstorff C, Rhodin M, Almlöf M, Silva M, et al. Is markerless more or less? Comparing a smartphone computer vision method for equine lameness assessment to multi‐camera motion capture. Animals. 2023;13:390.
  27. Key KT, Berg K, Kirkegaard J, Hansen SS. Reliability, agreement and variability of a markerless computer vision algorithm for equine gait analysis under field conditions. Equine Vet J. 2025. https://doi.org/10.1111/evj.70109
  28. Hardeman AM, Egenvall A, Serra Bragança FM, Swagemakers J‐H, Koene MHW, Roepstorff L, et al. Visual lameness assessment in comparison to quantitative gait analysis data in horses. Equine Vet J. 2022;54:1076–1085.
  29. Starke SD, Clayton HM. A universal approach to determine footfall timings from kinematics of a single foot marker in hoofed animals. PeerJ. 2015;3:e783.
  30. Gan Z, Wiestner T, Weishaupt MA, Waldern NM, David Remy C. Passive dynamics explain quadrupedal walking, trotting, and Tölting. J Comput Nonlinear Dyn. 2016;11:0210081–02100812.
  31. Butterworth S. On the theory of filter amplifiers. Wirel Eng. 1930;7:536–541.
  32. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet Lond Engl. 1986;1:307–310.
  33. Giavarina D. Understanding Bland Altman analysis. Biochem Medica. 2015;25:141–151.
  34. Bland JM, Altman DG. Agreement between methods of measurement with multiple observations per individual. J Biopharm Stat. 2007;17:571–582.
  35. van Weeren PR, Barneveld A. Quantification of skin displacement in the walking horse. In: Bergmann G, Kölbel R, Rohlmann A, editors. Biomechanics: basic and applied research: selected Proceedings of the Fifth Meeting of the European Society of Biomechanics, September 8–10, 1986, Berlin, F.R.G. Dordrecht: Springer Netherlands; 1987. p. 159–164. https://doi.org/10.1007/978‐94‐009‐3355‐2_17
  36. Bergh A, Egenvall A, Olsson E, Uhlhorn M, Rhodin M. Skin displacement in the equine neck. Equine Vet J. 2014;46:36.
  37. Wade L, Needham L, McGuigan P, Bilzon J. Applications and limitations of current markerless motion capture methods for clinical gait biomechanics. PeerJ. 2022;10:e12995.
  38. Bosch S, Serra Bragança F, Marin‐Perianu M, Marin‐Perianu R, Van der Zwaag BJ, Voskamp J, et al. EquiMoves: a wireless networked inertial measurement system for objective examination of horse gait. Sensors. 2018;18:850.
  39. Kallerud AS, Marques‐Smith P, Bendiksen HK, Fjordbakk CT. Objective movement asymmetry in horses is comparable between markerless technology and sensor‐based systems. Equine Vet J. 2025;57(1):115–125.
  40. Daniel WW. Biostatistics: a Foundation for Analysis in the health sciences. 6th ed. New York, NY, USA: John Wiley & Sons; 2018.
  41. Serra Bragança FM, Hernlund E, Thomsen MH, Waldern NM, Rhodin M, Byström A, et al. Adaptation strategies of horses with induced forelimb lameness walking on a treadmill. Equine Vet J. 2021;53:600–611.
  42. Hardeman AM, Serra Bragança FM, Swagemakers JH, van Weeren PR, Roepstorff L. Variation in gait parameters used for objective lameness assessment in sound horses at the trot on the straight line and the lunge. Equine Vet J. 2019;51:831–839.
  43. Serra Bragança FM, Broomé S, Rhodin M, Björnsdóttir S, Gunnarsson V, Voskamp JP, et al. Improving gait classification in horses by using inertial measurement unit (IMU) generated data and machine learning. Sci Rep. 2020;10:17785.

Citations

This article has been cited 0 times.
Subscribe to our newsletter
Join 99,101 horse owners like you who receive our email updates on horse health and nutrition.