Evaluation of discriminant analysis based on dorsoventral symmetry indices to quantify hindlimb lameness during over ground locomotion in the horse.
Authors: Church E E, Walker A M, Wilson A M, Pfau T
Journal: Equine veterinary journal
Summary
Church and colleagues investigated whether inertial sensor data collected from the pelvis during trotting could reliably detect and grade hindlimb lameness in horses without requiring sophisticated laboratory equipment. Using accelerometers mounted on the withers, tuber sacrale, and bilateral tubera coxae of six sound and nine lame horses (graded 1/10 to 2/10), the team analysed 889 strides and applied discriminant analysis to identify which parameters best distinguished lameness severity and affected limb. The pelvic energy ratio emerged as the strongest single indicator of lameness degree, whilst directional symmetry of the tubera coxae markers proved most effective at identifying which hindlimb was affected. These findings matter to practice because they suggest that portable, wearable sensor systems can quantify subtle hindlimb lameness during routine trotting in clinical or yard settings, potentially providing objective data to complement visual assessment and track therapeutic progress without the need for treadmill or force plate facilities. The work lays groundwork for developing practical, evidence-based parameters that equine practitioners could eventually use to standardise lameness grading in the field.
Read the full abstract on PubMed
Practical Takeaways
- •Inertial sensors placed on the pelvis and withers can objectively quantify mild hindlimb lameness during trotting without requiring expensive motion capture laboratories or specialized arena facilities.
- •Pelvic drop asymmetry is a reliable indicator of hindlimb lameness severity, providing a quantifiable metric to track progression or improvement during treatment.
- •This trunk-mounted sensor system could enable practitioners to obtain objective lameness data in field conditions, improving diagnostic confidence and allowing evidence-based tracking of therapeutic outcomes.
Key Findings
- •Pelvic energy ratio from vertical displacement data of inertial sensors provided the best indication for grading severity of hindlimb lameness.
- •Directional symmetry index of tubera coxae sensors showed the highest discriminative power for identifying which hindlimb was lame.
- •Inertial sensor system mounted on trunk (tuber sacrale, bilateral tubera coxae, withers) successfully collected representative gait data during over-ground trotting in clinical settings.
- •Linear discriminant analysis effectively differentiated lame from nonlame horses and identified side of lameness using dorsoventral symmetry parameters.