Automatic detection of break-over phase onset in horses using hoof-mounted inertial measurement unit sensors.
Authors: Tijssen M, Hernlund E, Rhodin M, Bosch S, Voskamp J P, Nielen M, Serra Braganςa F M
Journal: PloS one
Summary
# Editorial Summary: Break-over Detection Using Hoof-Mounted Sensors Prolonged break-over—the interval when the hoof begins to rotate over the toe during the late stance phase—can signal musculoskeletal pathology, yet accurate measurement of its onset remains challenging in clinical settings. Tijssen and colleagues developed and validated two algorithms capable of automatically detecting break-over initiation from acceleration and angular velocity data captured by wireless inertial measurement units (IMUs) mounted on the lateral hoof wall, testing these in walk and trot against the gold-standard optical motion capture system and force plate measurements across seven Warmblood horses. The acceleration-based algorithm proved superior, achieving accuracy within −9 to 23 milliseconds (precision ~24 ms) when compared to optical motion capture, with slightly larger margins against force plate data (−37 to 20 ms accuracy, ~29 ms precision), with performance varying by gait and whether the front or hind limb was assessed. Whilst these performance metrics are promising for quantifying break-over timing objectively, the authors appropriately note that clinical validation is essential before practitioners can confidently apply IMU-based break-over detection to lameness assessment or to evaluate the efficacy of trimming and shoeing interventions. This technology could eventually offer farriers and veterinarians a practical, real-time tool for monitoring hoof mechanics and detecting gait abnormalities that might otherwise go unrecognised during routine examinations.
Read the full abstract on PubMed
Practical Takeaways
- •Hoof-mounted IMU technology can objectively measure break-over timing, providing quantitative data to assess whether trimming/shoeing changes are achieving their intended biomechanical goals
- •This tool may help detect early musculoskeletal issues by identifying abnormally prolonged break-over phases before lameness becomes clinically obvious
- •Further validation is still needed before this becomes a reliable clinical diagnostic tool—currently suitable for research and gait analysis development, not yet routine practice
Key Findings
- •Acceleration-based algorithm achieved accuracy of -9 to 23 ms (against OMC) and -37 to 20 ms (against force plate) for detecting break-over phase onset
- •Hoof-mounted IMU sensors can automatically detect break-over phase onset in walk and trot with precision around 24-29 ms depending on validation method
- •Algorithm performance varied by gait and hoof location, with front hooves showing different characteristics than hind hooves
- •IMU-based detection shows promise for clinical applications but requires further validation for lameness detection and shoeing evaluation