Inverse Dynamics and Artificial Neural Network Applications in Gait Analysis of the Disabled Subjects

Abstract

The aim of this study is to review the achievements of the mathematical modeling of muscles force contribution during walking. In order to determine the contributions of individual muscles to the net force or net muscle torque at a given joint, the external forces acting on lower extremity joints during gait ought to be identified. The solution of this problem, called in biomechanics the inverse dynamics, is now regarded as a classical method of movement modelling. In the hypothesis put forward in this paper, it is considered if the artificial neural network method could be applied to the muscle contraction prediction during gait analysis in normal and disabled subjects. Artificial neural network (ANN) is an artificial intelligence method used in mathematical modelling and its applications in diverse areas, especially in biology and medicine, are steadily progressing. The achievements and possibilities of ANN in biomechanics were presented previously by others authors. For example, Liu and Lockhart [13] attempted at creating a network capable of reproducing muscle forces during gait from EMG signals recorded in working muscles. The objective of our study was to make use of the experience gained in the construction of ANN and to apply advanced mathematical procedures to identical experimental conditions of gaint analysis.