Article: Modelling the dynamic response of the human spine to shock and vibration using a recurrent neural network Journal: International Journal of Heavy Vehicle Systems (IJHVS) 1997 Vol.4 No.2/3/4 pp.145 - 165 Abstract: The ability to model the spine|s response to mechanical shock and vibration is an important step in assessing the health hazards of repeated impacts to vehicle passengers. Current linear models, such as the Dynamic Response Index (DRI) and the British Standard 6841 filter (BS 6841), perform poorly when the input consists of large-magnitude shocks typical of those experienced by personnel in military vehicles. In this study, a recurrent neural network (RNN) was developed which models the spinal acceleration response of the seated passenger at the LA vertebra to vertical accelerations applied at the seat. ARNN is a universal nonlinear approximator that can, in theory, model any system with memory if trained with a representative set of measured input-output data. The seat-spine system was modelled as a network with four inputs and one output. The back propagation algorithm was used to train the network by adjusting network parameters to minimise the square of the prediction error. The inputs to the network were delayed values of the inputs and outputs. The trained network significantly outperformed the two linear models examined for predicting the z-axis acceleration at the L4 vertebra. Inderscience Publishers - linking academia, business and industry through research

Title: Modelling the dynamic response of the human spine to shock and vibration using a recurrent neural network

Authors: J. Nicol, J. Morrison, G. Roddan, A. Rawicz

Addresses: School of Engineering Science, Simon Fraser University, Burnaby, B.C. V5A 1S6, Canada. ' School of Kinesiology, Simon Fraser University, Burnaby, B.C. V5A 1S6, Canada. ' Advanced Systems Engineering , B. C. Research Inc., 3650 Wesbrook Mall, Vancouver, B.C. V6S 2L2, Canada. ' School of Engineering Science, Simon Fraser University, Burnaby, B.C. V5A 1S6, Canada

Abstract: The ability to model the spine|s response to mechanical shock and vibration is an important step in assessing the health hazards of repeated impacts to vehicle passengers. Current linear models, such as the Dynamic Response Index (DRI) and the British Standard 6841 filter (BS 6841), perform poorly when the input consists of large-magnitude shocks typical of those experienced by personnel in military vehicles. In this study, a recurrent neural network (RNN) was developed which models the spinal acceleration response of the seated passenger at the LA vertebra to vertical accelerations applied at the seat. ARNN is a universal nonlinear approximator that can, in theory, model any system with memory if trained with a representative set of measured input-output data. The seat-spine system was modelled as a network with four inputs and one output. The back propagation algorithm was used to train the network by adjusting network parameters to minimise the square of the prediction error. The inputs to the network were delayed values of the inputs and outputs. The trained network significantly outperformed the two linear models examined for predicting the z-axis acceleration at the L4 vertebra.

Keywords: artificial neural networks; ANNs; nonlinear modelling; repeated impacts; system identification; whole body vibration; dynamic response; human spine; mechanical shocks; health hazards; vehicle passengers; spinal acceleration response; seat-spine system.

DOI: 10.1504/IJHVS.1997.054584

International Journal of Heavy Vehicle Systems, 1997 Vol.4 No.2/3/4, pp.145 - 165

Published online: 18 Jun 2013 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Modelling the dynamic response of the human spine to shock and vibration using a recurrent neural network

Keep up-to-date