Robustness of Combined sEMG and Ultrasound Modalities Against Muscle Fatigue in Force Estimation

It is evident that surface electromyography (sEMG) based prosthesis is constrained due to sensitivity to muscle fatigue. This paper investigated the muscle fatigue robustness for sEMG, ultrasound and the fusion sEMG/ultrasound signals towards the proportional force prediction. The linear regression model is developed, and evaluated on the non-fatigue state and fatigue state. Seven able-bodied subjects participated in the experiment to validate the model. The results demonstrate that sEMG outperforms ultrasound in force estimation accuracy, but ultrasound is more robust against muscle fatigue than sEMG. Furthermore, the fusion sEMG/ultrasound signal shows comparable force prediction accuracy to sEMG and better muscle fatigue robustness than sEMG. The fusion sEMG/ultrasound modality overcomes the defect of sEMG modality, making it a promising modality for the long-term use of prosthetic force control.