Biodynamic verification of an estimated muscular activity model for orthosis prescription support systems

In this paper, we performed biodynamic verification of a muscular activity model using Bayes estimation. In creating this model, we aimed to enable quantitative selection of lower foot orthoses based on a patient's muscular activity in the lower foot. Because physical models require the use of large-scale measurement systems, which cannot be used clinically, they are not suitable for making these measurements. Therefore, we chose Bayes estimation to construct a model for estimating the muscular activity from parameters that can be measured easily, such as joint angle and sole pressure. This model allows for not only the estimation of muscle activity, but also another closely related parameter through the change in muscular activity, which is a parent node to the muscle activity node. The three advantages of our model are that it 1) reports the influences on muscle activity, which change throughout the gait cycle, by using 10% level nodes for each factor; 2) expresses the influence of those factors, which are different at low and high muscular activity levels; and 3) compensates for missed predictions by estimating muscle activity in 10% increments. Here, we verify the biodynamic validity of the model parent node for four foot muscles.