Agreement between two inertial sensor gait analysis systems for lameness examinations in horses

Abstract

Summary 25Gait analysis is gaining in popularity for quantification of lameness and two commonly used inertial 26 sensor systems assess trunk movement symmetry: can these be used interchangeably in multi 27 centre studies? 28We compared head and pelvic movement symmetry between two inertial sensor gait analysis 29 systems in 13 horses equipped simultaneously with the two systems. The first system quantified 30 dorso-ventral movement in the local reference frame (System A), the other system global vertical 31 movement (System B). Widths of limits of agreement were calculated employing a well-established 32 regression method dealing with systematically changing differences over the range of measured 33 values. 34Widths of limits of agreement between system A and system B were narrower for pelvic movement 35 than for head movement. For head movement, they ranged from 6.4 to 6.9mm for in-hand trot and 36 from 7.3 to 9.7mm in the lunge and for pelvic movement from 2.5 to 4.4mm in-hand and from 3.6 37 to 5.3mm on the lunge. 38

Comparison of visual lameness scores to gait asymmetry in racing Thoroughbreds during trot in‐hand

“…; Pfau et al . ) ( Table , column ORIG THRESS +HW). Three of these horses had been visually assessed as concurrently forelimb and hindlimb lame.…”

“…) were adapted to >|7 mm| for HDmin and >|4 mm| for PDmin based on correction equations presented in (Pfau et al . ). Horses with HDmin and WDmin values showing same signs were categorised as forelimb lame, horses with HDmin and WDmin of opposite sign were categorised as hindlimb lame (Persson‐Sjodin et al .…”

“…The ability to use the relationship between HDmin and WDmin to differentiate between forelimb and hindlimb lame horses was investigated for all horses that were found to be concurrently outside 'normal ranges' for both pelvic and head movement asymmetry. For categorising existence and side of lameness, previously reported clinical thresholds values of >|6 mm| for head movement asymmetry and >|3 mm| for pelvic movement asymmetry (McCracken et al 2012) were adapted to >|7 mm| for HDmin and >|4 mm| for PDmin based on correction equations presented in (Pfau et al 2016a). Horses with HDmin and WDmin values showing same signs were categorised as forelimb lame, horses with HDmin and WDmin of opposite sign were categorised as hindlimb lame (Persson-Sjodin et al 2016;Pfau et al 2017).…”

Ambulatory Assessment of Instantaneous Velocity during Walking Using Inertial Sensor Measurements

“…The performance assessment for the cyclical component of velocity was based on the procedure devised for method comparison studies [27,28]. The Mean Difference (MD) and upper and lower Limit of Agreement (LA), namely MD ± 1.96 Standard Deviation (SD) of differences, were computed for the cyclical component of velocity in the ML, VT and AP directions, as it was obtained from IMU and OMCS data.…”

Repeatability of gait analysis measurements in Thoroughbreds in training

“…We have reported a difference in the amount of movement asymmetry quantified between the two IMU systems [17] and are speculating that this may have to do with the different filtering approaches applied to the underlying acceleration data: a Fourier and polynomial approach [18] vs. a high-pass filter [19]. We have reported a difference in the amount of movement asymmetry quantified between the two IMU systems [17] and are speculating that this may have to do with the different filtering approaches applied to the underlying acceleration data: a Fourier and polynomial approach [18] vs. a high-pass filter [19].…”