Fourier analysis of trunk displacements: a method to identify the lame limb in trotting horses

• Published in: Journal of biomechanics 2002-09, Vol.35 (9), p.1173-1182
• Main Authors: Audigié, Fabrice, Pourcelot, Philippe, Degueurce, Christophe, Geiger, Didier, Denoix, Jean Marie
• Format: Article
• Language: English
• Subjects: Algorithms; Animals; Breeding; Female; Forelimb - physiopathology; Fourier Analysis; Fourier series; Gait; Hindlimb - physiopathology; Horse Diseases - diagnosis; Horse Diseases - physiopathology; Horses; Image Interpretation, Computer-Assisted - methods; Imaging, Three-Dimensional - methods; Imaging, Three-Dimensional - veterinary; Kinematics; Lameness; Lameness, Animal - classification; Lameness, Animal - diagnosis; Lameness, Animal - physiopathology; Lumbar Vertebrae - physiopathology; Male; Models, Biological; Movement; Reference Values; Reproducibility of Results; Sacrum - physiopathology; Sensitivity and Specificity; Spine - physiopathology; Symmetry; Thoracic Vertebrae - physiopathology; Trunk
• ISSN: 0021-9290; 1873-2380
• DOI: 10.1016/S0021-9290(02)00089-1

The aim of this paper is to present a method allowing the identification of the lame limb in trotting horses. Using a 3-D kinematic analysis system, 13 sound and 25 lame horses fitted with 4 skin markers placed on the dorsal midline of their trunk were recorded while trotting on a track in the conditions of the routine lameness examination. The vertical displacements of the trunk markers underwent Fourier analysis. Results indicated that these displacements could be represented using only the first and second harmonics. From these two harmonics, indices were then developed. The sensitivity of these indices to the different types of experimental errors was also studied. Results showed that the values of the indices of the lame horses were relatively unaffected by the experimental errors. In lame horses, these indices allowed the quantification of the degree of the lameness, the identification of lame limb with a reliability >95% and the characterisation of the type of trunk movements. These indices could be easily implemented in a computer program to provide objective information to the clinician or to be used as a first step in the development of an expert system. Moreover, these clinical tools may also be extended to other quadrupedal or bipedal locomotions.