Real-time estimation of FES-induced joint torque with evoked EMG

• DOI: 10.1186/s12984-016-0169-y
• Publication Date: 2016-06-23T02:21:10Z
• Authors: Zhan Li; David Guiraud; David Andreu; Mourad Benoussaad; Charles Fattal; Mitsuhiro Hayashibe
• Keywords: Functional electrical stimulation

Functional electrical stimulation (FES) is a neuroprosthetic technique for restoring lost motor function of spinal cord injured (SCI) patients and motor-impaired subjects by delivering short pulses to their paralyzed muscles or nerves. FES induces action potentials respectively on nerves so that muscle activity can be characterized the synchronous recruitment units with its compound electromyography (EMG) signal called M-wave. The recorded evoked EMG (eEMG) employed predict resultant joint torque, modeling FES-induced torque based eEMG an essential step provide necessary prediction expected response before achieving accurate control FES. Previous works tracking issues were mainly offline analysis. However, toward personalized clinical rehabilitation applications, real-time systems are essentially required considering subject-specific responses against stimulation. This paper proposes wireless portable stimulator used estimating/predicting real time processing eEMG. Kalman filter recurrent neural network (RNN) embedded into system identification estimation. Prediction results 3 able-bodied SCI demonstrate promising performances. As estimators, both RNN approaches show clinically feasible estimation/prediction signals only, moreover requires less computational requirement. proposed establishes platform estimating assessing mechanical output, electromyographic recordings associated models. It will contribute open new modality consolidated personal healthcare patients.