Short communicationCalculation of joint moments following foot contact across two force plates
Introduction
Force plates are commonly used in 3-Dimensional Gait Analysis (3DGA) to provide measurements to allow the computation of joint moments using inverse dynamics. One issue that affects the data captured from force plates is that it is often impossible for a subject to get the whole of his or her foot onto the force plate. Often both feet touch the force plate or a foot touches multiple adjacent force plates. This study aims to determine the validity of combining the measurements generated by two force plates when a subject's foot comes in contact with more than one force plate to calculate the moments about ankle, knee and hip joints.
A particular issue is in determining how to compare results given the stride to stride variability in force plate measurements. The overall approach is to take multiple measurements of strides in which the foot either makes a clean contact with a single force plate or spans two force plates and use analysis of variability components to determine the relative proportions of the overall variability coming from stride to stride variability and from difference in measurement technique.
Section snippets
Materials and methods
A 10 camera motion capture system (VICON, Oxford, UK) was used to collect kinematic data. The system consisted of MX40 cameras sampling at 100 Hz using VICON Nexus® software. Kinetic data was collected using 2 AMTI SGA6-4 force plates (47 cm × 51 cm) sampling at 1000 Hz. The laboratory walkway was 10 m long and the calibrated measurement volume was 2 m × 1.5 m × 1.5 m.
Given that the focus of the study is on the accuracy of force plate measurements rather than the characteristics any given subject, the study
Results
Fig. 1 shows the kinetics traces coming from the customised BodyLanguage® model from gait trials where 1 foot was in contact with 1 force plate and the trials where 1 foot was in contact with both force plates. It can be seen that the results are very similar.
Table 1 shows the estimates of variance results from the multi-level random effects linear regression model. Stride to stride variance or force plate contact variance were similar across all joints for a given plane but coronal plane
Discussion
Both the graphical results and the statistical analysis suggest that differences arising from computing joint moments from measurements made by either one or two force plates are indistinguishable from the stride to stride variability of an unimpaired adult subject. This confirms that combining the force plate measurements to calculate joint moments is unlikely to have any effect on the analysis and interpretation of clinical gait data.
An issue that can arise from two force plates contact is
Conflicts of interest
None.
References (3)
- et al.
Do strain gauge force platforms need in situ correction?
Gait & Posture
(2002)
Cited by (4)
The effect of joint translation constraint on within-participant variability of kinematics and kinetics during running in cerebral palsy
2019, Clinical BiomechanicsCitation Excerpt :Allowing a constrained amount of joint translation may permit the joint centre to be more accurately located during kinetic calculations, thereby reducing extrinsic variability. Joint moment SDs were larger than previously published studies of walking in healthy participants using three DoF joint models (Sangeux et al., 2016; Wong et al., 2010) but this is not surprising given joint moments are larger and more variable in running compared to walking (Estep et al., 2016; Novacheck, 1998), and more variable in children with cerebral palsy than typically developing children (Klejman et al., 2010; Steinwender et al., 2000). Potvin and colleagues (Potvin et al., 2017) reported a smaller effect of joint model DoF on joint moments compared to joint angles in walking, whereas the present study has found a significant reduction in all joint moment SDs using the IK6Constrained model.
A review of musculoskeletal modelling of human locomotion
2021, Interface FocusOutputs of human walking for bipedal robotic controller design
2012, Proceedings of the American Control ConferenceAmbulatory hip angle estimation using multiple model hybrid dynamic Bayesian networks
2011, Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology