REASONS FOR PERFORMING STUDY: Hindlimb lameness is common and can be difficult to diagnose or quantify in evaluating response to nerve blocks. An objective measure of lameness can also be used to evaluate the effectiveness of the treatment's contribution to evidence-based medicine. The inertial sensor system can be used to capture 6 degree of freedom movement during over ground locomotion and here was used to quantify tuber coxae movement in nonlame and lame horses. HYPOTHESIS: Tuber coxae movement is useful for discriminating between nonlame and lame horses. OBJECTIVES: To measure left and right tuber coxae movement in lame and nonlame horses during over ground locomotion and to implement a linear discriminant analysis to discriminate between lame and nonlame horses. METHODS: Two inertial sensors were attached to the skin over left and right tuber coxae of 21 horses (9 mildly and 12 not lame). Horses were trotted on a hard surface. A total of 1021 strides were collected. For each stride 34 features were extracted from the dorsoventral and craniocaudal movement and used in 2 different classification scenarios (lame vs. nonlame or left lame, right lame and nonlame) using linear discriminant analysis. RESULTS: Six degree of freedom inertial sensors were successfully used to collect kinematic data continuously from left and right tuber coxae in horses during over ground locomotion. These data were used for an automated classification of lameness. In the first scenario, a sensitivity of 89% was achieved with a specificity of 75%. In the second scenario, all horses could be correctly assigned to the correct class in a simple 3 class reclassification test. POTENTIAL RELEVANCE: A mobile system that reliably detects and quantifies hindlimb lameness in horses during unconstrained locomotion could be a valuable tool to perform an evidence-based assessment of lameness in horses in a clinical setting, e.g. before and after nerve blocks or before and after surgery.
REASONS FOR PERFORMING STUDY: Hindlimb lameness is common and can be difficult to diagnose or quantify in evaluating response to nerve blocks. An objective measure of lameness can also be used to evaluate the effectiveness of the treatment's contribution to evidence-based medicine. The inertial sensor system can be used to capture 6 degree of freedom movement during over ground locomotion and here was used to quantify tuber coxae movement in nonlame and lame horses. HYPOTHESIS: Tuber coxae movement is useful for discriminating between nonlame and lame horses. OBJECTIVES: To measure left and right tuber coxae movement in lame and nonlame horses during over ground locomotion and to implement a linear discriminant analysis to discriminate between lame and nonlame horses. METHODS: Two inertial sensors were attached to the skin over left and right tuber coxae of 21 horses (9 mildly and 12 not lame). Horses were trotted on a hard surface. A total of 1021 strides were collected. For each stride 34 features were extracted from the dorsoventral and craniocaudal movement and used in 2 different classification scenarios (lame vs. nonlame or left lame, right lame and nonlame) using linear discriminant analysis. RESULTS: Six degree of freedom inertial sensors were successfully used to collect kinematic data continuously from left and right tuber coxae in horses during over ground locomotion. These data were used for an automated classification of lameness. In the first scenario, a sensitivity of 89% was achieved with a specificity of 75%. In the second scenario, all horses could be correctly assigned to the correct class in a simple 3 class reclassification test. POTENTIAL RELEVANCE: A mobile system that reliably detects and quantifies hindlimb lameness in horses during unconstrained locomotion could be a valuable tool to perform an evidence-based assessment of lameness in horses in a clinical setting, e.g. before and after nerve blocks or before and after surgery.
Authors: Stephan Bosch; Filipe Serra Bragança; Mihai Marin-Perianu; Raluca Marin-Perianu; Berend Jan van der Zwaag; John Voskamp; Willem Back; René van Weeren; Paul Havinga Journal: Sensors (Basel) Date: 2018-03-13 Impact factor: 3.576
Authors: Thilo Pfau; David M Bolt; Andrew Fiske-Jackson; Carolin Gerdes; Karl Hoenecke; Lucy Lynch; Melanie Perrier; Roger K W Smith Journal: Animals (Basel) Date: 2022-03-17 Impact factor: 2.752