Publication: The use of artificial neural networks to approximate the centre of mass of the body from plantar-surface pressure.

All || By Area || By Year

Title	The use of artificial neural networks to approximate the centre of mass of the body from plantar-surface pressure.
Authors/Editors*	Perry SD, Beattie C, Tung J
Where published*	Journal of Biomechanics
How published*	Journal
Year*	2007
Volume	Under Review
Number
Pages
Publisher
Keywords
Link
Abstract	This study was undertaken to determine if an Artificial Neural Network based model could be used to approximate the transverse plane location of the bodyâs center of mass (COM), during dynamic movements in upright stance and gait, given only pressure data originating from pressure sensing insoles. This type of modelling may provide insight into how the human balance control system uses the sensory information from plantar-surface pressure to control balance. The modeled activities were voluntary leaning in four directions (forward, right, left and backwards), all held for 10 seconds, and normal gait. The model demonstrated good prediction of the COM in the anterior/posterior direction such that the predicted COM approximations were within 10 mm of the measured COM. Extension of this model to two-dimensional space, incorporating medial/lateral information, has also yielded a good prediction of the COM location. Pilot work has also begun on modeling the anterior/posterior COM and it relationship to the base of support during gait using the pressure insoles; initial data is presented here and has shown encouraging results as we continue development of this modelling technique.