Abstract
The objective of this study is building the systems which can prescript desirable leg orthosis for leg paralyzed patients. In this paper we build leg muscular estimate model by using Bayesian network. Our method can estimate activity level of muscle using only two kinds of parameter. Sole pressure distribution and Joint angle. And this method has three advantages. Our use of 10% increments in the levels of the measured factors enabled changes in these factors during gait to be reflected in the model. Variations in the influence of factors that differ between low and high muscle activity are represented. It is easier to use than physical models; three- dimensional motion analysis is not necessary and the method is convenient for clinical use. We estimate healthy person's leg muscular activity, and leg paralyzed person's leg muscular activity by using our model. The accuracy of healthy person model estimation is over 90%. And the accuracy of leg paralyzed model estimation is over 85%.
