Development of Artificial Neural Network Based Active Ankle Prosthesis Algorithm Using Gait Analysis Data

Abstract

Amputation is the removal of a part or all of a limb due to a disease, accident or trauma. In the United States, an average of 500 people loses at least one limb every day, of which about 65% is due to lower limb amputations. Since energetically active prostheses are costly, amputees usually continue to their daily lives using a wheelchair or a passive prosthesis. The aim of this study is to determine the optimum sensor needs of active ankle prostheses and to develop algorithm suitable for this sensor infrastructure, thus a device that is both easy to use and financially accessible to anyone can be built up. In this context, three neural networks structures with different inputs were developed and their performances were evaluated. As a result, it is determined that if a device in which only EMG signals are used as network inputs, a total of 5 signals should be collected from muscles that are responsible for hip, knee and wrist movements. It has been found that, when it is wanted to work with less EMG sensors, it is necessary to perform a force or torque feedback into the system. Three EMG signals collected from the dorsi and plantar flexor muscles of the ankle were found to be sufficient for such a system. These findings are important in order to determine the amount of the sensor needed by an active ankle prosthesis whose requirements can be variable according to the amputation level of the patient and the mechanical design flexibility.