Master's thesis (2016)
Abstract
Introduction: Three-dimensional ground reaction forces (3D-GRF) measurement is an important aspect of gait analysis to compute centers of pressure, spatial-temporal parameters and joint torques for functional evaluation in rehabilitation. To obtain these 3D-GRF on several gait cycles, a long measurement platform is required. However, the difficulty of such measurement systems arise from the fact that the unique force platform provides the resultant of GRF for the two feet combined. Therefore, one needs to decompose the 3D-GRF into components under each foot. Indeed, both feet are placed on the same force platform, which solely provides the global resultant of left and right foot together. Furthermore, no method has been previously validated on reasonable number of participants. Objective: This study aims to develop an automatic method using s-type parametric curve modeling to increase the accuracy of decomposition of 3D-GRF during double stance into left and right profiles and along X, Y and Z axes. Methods: Thirty healthy adults (age: 24.8 (Standard Deviation (SD): 3.1) years, 17 males, weight: 70.6 (SD: 11.4) kg) walked naturally barefoot on a floor equipped with separate force platforms. The decomposition modeling method was applied to global 3D-GRF using different parametric curve functions as in 3rd order polynomial, sine and sine-sigmoid functions. To validate this method, the decomposed 3D-GRF was compared to the 3D-GRF independently recorded for each subject. A global error is calculated based on global GRF, which is the comparison between the decomposed global 3D-GRF and recorded global 3D-GRF. The global 3D-GRF is obtained by computing the square root of the sum of the second exponential of GRF along each axis.
Statistics
Total downloads
Downloads per month in the last year
Origin of downloads
