Evaluation of discriminant analysis based on dorsoventral symmetry indices to quantify hindlimb lameness during over ground locomotion in the horse.

The research article investigates the effectiveness of using gait analysis techniques to identify and quantify lameness in horses, specifically focusing on the movements of the bilateral tubera coxae and pelvis. The study found that the best indicators of hindlimb lameness are the vertical displacement data obtained from the equine’s pelvic midline during locomotion.

Research Methods

• The study involved a group of fifteen horses, with six being healthy and nine being unilaterally lame in their hindlimbs. The lameness severity ranged from 1/10 to 2/10.
• Inertial sensors were attached to four parts of each horse’s body: the withers, the tuber sacrale, and both tubera coxae, which are anatomical landmarks on the trunk and proximal hindlimbs of horses.
• Each horse was made to trot for six trials for data collection, during which 889 strides in total were collected.
• Each horse’s lameness was assessed through video recordings by a qualified equine orthopaedic surgeon.
• The vertical displacement data from each sensor was calculated to determine symmetry indices as well as parameters using Fourier analysis.
• Furthermore, the research team used linear discriminant analysis to find the parameters that would most accurately classify the severity of the lameness and identify the affected limb.

Research Findings

• The study found that the most reliable indicator of the degree of lameness was the “pelvic energy ratio” derived from the vertical displacement data of the pelvic midline.
• The study also found that the tubera coxae sensors’ Directional Symmetry Index provided the highest discriminative power for identifying the lame limb.

Research Conclusions

• Using a trunk-mounted inertial sensor system during ground locomotion could efficiently gather a representative data sample of horses with different grades and types of lameness in practical, clinical settings. This could be a viable and practical method to assess horse hindlimb lameness.
• The authors suggest that these findings are critical for future research aiming to define robust parameters for classifying lameness under practical conditions.