Sensitivity of the OLGA and VCM models to erroneous marker placement: effects on 3D-gait kinematics.

Gait data need to be reliable to be valuable for clinical decision-making. To reduce the impact of marker placement errors, the Optimized Lower Limb Gait Analysis (OLGA) model was developed. The purpose of this study was to assess the sensitivity of the kinematic gait data to a standard marker displacement of the OLGA model compared with the standard Vicon Clinical Manager (VCM) model and to determine whether OLGA reduces the errors due to the most critical marker displacements. Healthy adults performed six gait sessions. The first session was a standard gait session. For the following sessions, 10mm marker displacements were applied. Kinematic data were collected for both models. The root mean squares of the differences (RMS) were calculated for the kinematics of the displacement sessions with respect to the first session. The results showed that the RMS values were generally larger than the stride-to-stride variation except for the pelvic kinematics. For the ankle, knee and hip kinematics, OLGA significantly reduced the averaged RMS values for most planes. The shank, knee and thigh anterior-posterior marker displacements resulted in RMS values exceeding 10°. OLGA reduced the errors due to the knee and thigh marker displacements, but not the errors due to the ankle marker displacements. In conclusion, OLGA reduces the effect of erroneous marker placement, but does not fully compensate all effects, indicating that accurate marker placement remains of crucial importance for adequate 3D-gait analysis and subsequent clinical decision-making.