Examination of a muscular activity estimation model using a Bayesian network for the influence of an ankle foot orthosis
Examination of a muscular activity estimation model using a Bayesian network for the influence of an ankle foot orthosis
复制标题
使用贝叶斯网络检查肌肉活动估计模型对踝足矫形器的影响
DOI：
10.1109/embc.2012.6347470
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Masakatsu G. Fujie
中科院分区：
文献类型：
--
作者：
Jun Inoue;Kazuya Kawamura;Masakatsu G. Fujie
In the present paper, we examine the appropriateness of a new model to examine the activity of the foot in gait. We developed an estimation model for foot-ankle muscular activity in the design of an ankle-foot orthosis by means of a statistical method. We chose three muscles for measuring muscular activity and built a Bayesian network model [1] to confirm the appropriateness of the estimation model. We experimentally examined the normal gait of a non-disabled subject. We measured the muscular activity of the lower foot muscles using electromyography, the joint angles, and the pressure on each part of the sole. From these data, we obtained the causal relationship at every 10% level for these factors and built models for the stance phase, control term, and propulsive term. Our model has three advantages. First, it can express the influences that change during gait because we use 10% level nodes for each factor. Second, it can express the influences of factors that differ for low and high muscular-activity levels. Third, we created divided models that are able to reflect the actual features of gait. In evaluating the new model, we confirmed it is able to estimate all muscular activity level with an accuracy of over 90%.
