Summary
Objective: To evaluate the use of inertial measurement units (IMU) for quantification of canine
limb kinematics.
Methods: Sixteen clinically healthy, medium-sized dogs were enrolled. Baseline kinematic data
were acquired using an optical motion capture system. Following this baseline data
acquisition, a harness system was used for attachment of IMU to the animals. Optical
kinematic data of dogs with and without the harness were compared to evaluate the
influence of the harness on gait parameters. Sagittal plane joint kinematics acquired
simultaneously with IMU and the optical system were compared for the carpal, tarsal,
stifle and hip joints. Comparisons of data were made using the concordance correlation
coefficient (CCC) test and evaluation of root mean squared errors (RMSE).
Results: No significant differences were demonstrated in stance duration, swing duration or
stride length between dogs instrumented with or without the harness, however, mean
RMSE values ranged from 4.90° to 14.10° across the various joints. When comparing
simultaneously acquired optical and IMU kinematic data, strong correlations were found
for all four joints evaluated (CCC: carpus = 0.98, hock = 0.95, stifle = 0.98, hip
= 0.96) and median RMSE values were similar across the joints ranging from 2.51° to
3.52°.
Conclusions and Clinical relevance: Canine sagittal plane motion data acquisition with IMU is feasible, and optically
acquired and IMU acquired sagittal plane kinematics had good correlation. This technology
allows data acquisition outside the gait laboratory and may provide an alternative
to optical kinematic gait analysis for the carpal, tarsal, stifle, and hip joints
in the dog. Further investigation into this technology is indicated.
Keywords
Gait analysis - kinematics - inertial sensors - inertial measurement units - dog -
canine