Real-Time Closed-Loop Functional Electrical Stimulation Control of Muscle Activation with Evoked Electromyography Feedback for Spinal Cord Injured Patients

Functional electrical stimulation (FES) is a neuroprosthetic technique to help restore motor function of spinal cord-injured (SCI) patients. Through delivery of electrical pulses to muscles of motor-impaired subjects, FES is able to artificially induce their muscle contractions. Evoked electromyography (eEMG) is used to record such FES-induced electrical muscle activity and presents a form of M-wave. In order to monitor electrical muscle activity under stimulation and ensure safe stimulation configurations, closed-loop FES control with eEMG feedback is needed to be developed for SCI patients who lose their voluntary muscle contraction ability. This work proposes a closed-loop FES system for real-time control of muscle activation on the triceps surae and tibialis muscle groups through online modulating pulse width (PW) of electrical stimulus. Subject-specific time-variant muscle responses under FES are explicitly reflected by muscle excitation model, which is described by Hammerstein system with its input and output being, respectively, PW and eEMG. Model predictive control is adopted to compute the PW based on muscle excitation model which can online update its parameters. Four muscle activation patterns are provided as desired control references to validate the proposed closed-loop FES control paradigm. Real-time experimental results on three able-bodied subjects and five SCI patients in clinical environment show promising performances of tracking the aforementioned reference muscle activation patterns based on the proposed closed-loop FES control scheme.

功能性电刺激（FES）是一种神经修复技术，用于帮助恢复脊髓损伤（SCI）患者的运动功能。通过向运动功能受损者的肌肉传递电脉冲，FES能够人为地诱导其肌肉收缩。诱发肌电图（eEMG）用于记录这种FES诱导的肌肉电活动，并呈现一种M波形式。为了监测刺激下的肌肉电活动并确保安全的刺激配置，需要为失去自主肌肉收缩能力的SCI患者开发具有eEMG反馈的闭环FES控制。这项工作提出了一种闭环FES系统，通过在线调节电刺激的脉冲宽度（PW）来实时控制小腿三头肌和胫骨肌肌群的肌肉激活。肌肉兴奋模型明确反映了FES下特定受试者的时变肌肉反应，该模型由哈默斯坦系统描述，其输入和输出分别为PW和eEMG。采用模型预测控制基于肌肉兴奋模型计算PW，该模型可在线更新其参数。提供了四种肌肉激活模式作为期望的控制参考，以验证所提出的闭环FES控制范例。在临床环境中对三名健全受试者和五名SCI患者进行的实时实验结果表明，基于所提出的闭环FES控制方案跟踪上述参考肌肉激活模式具有良好的性能。