Estimation of vertical tuber coxae movement in the horse from a single inertial measurement unit

Pelvic movement is altered in hind limb lame horses. A simple method using one inertial measurement unit (IMU) attached to the sacrum, which objectively measures 'hip hike', will help detecting hind limb lameness in large scale studies. In this study, IMUs over the sacrum and the left and right tuber coxae (TC) quantified vertical movement symmetry (MS) in 10 clinically lame horses. A simple geometrical model was used to estimate TC movement from the sacrum IMU. The differences between measured and estimated MS values (mean±SD) and MS changes during different exercise conditions (straight, circle, flexion) were quantified using established MS parameters to assess the performance of the sacral IMU for estimating TC movement. Mean differences between measured and estimated MS values were variable between horses and ranged from 5 to 30 mm, SD of differences ranged from 7 to 14 mm. However, the difference between measured and estimated change in MS induced by lungeing or flexion test, was lower, with a value <1 mm for one MS measure, which assesses differences between left and right TC movement. Estimating TC movement from sacral movement does not accurately quantify the true state of TC MS since limits of agreement (LoA) overlap the decision boundaries for detecting mild lameness. The LoA of changes in vertical TC movement between exercises compared favourably with the average changes between exercise conditions. While in moderate to severe cases, changes in TC movement may be estimated from a single sensor over the sacrum, in mildly lame horses it should be measured with additional sensors.