Park Hyun-Joo, Durand Dominique M
Department of Biomedical Engineering, Neural Engineering Center, Case Western Reserve University, Cleveland, OH , 44106, USA.
Biol Cybern. 2014 Aug;108(4):445-57. doi: 10.1007/s00422-014-0612-8. Epub 2014 Jun 18.
The flat interface nerve electrode (FINE) has demonstrated significant capability for fascicular and subfascicular stimulation selectivity. However, due to the inherent complexity of the neuromuscular skeletal systems and nerve-electrode interface, a trajectory tracking motion control algorithm of musculoskeletal systems for functional electrical stimulation using a multiple contact nerve cuff electrode such as FINE has not yet been developed. In our previous study, a control system was developed for multiple-input multiple-output (MIMO) musculoskeletal systems with little prior knowledge of the system. In this study, more realistic computational ankle/subtalar joint model including a finite element model of the sciatic nerve was developed. The control system was tested to control the motion of ankle/subtalar joint angles by modulating the pulse amplitude of each contact of a FINE placed on the sciatic nerve. The simulation results showed that the control strategy based on the separation of steady state and dynamic properties of the system resulted in small output tracking errors for different reference trajectories such as sinusoidal and filtered random signals. The proposed control method also demonstrated robustness against external disturbances and system parameter variations such as muscle fatigue. These simulation results under various circumstances indicate that it is possible to take advantage of multiple contact nerve electrodes with spatial selectivity for the control of limb motion by peripheral nerve stimulation even with limited individual muscle selectivity. This technology could be useful to restore neural function in patients with paralysis.
扁平界面神经电极（FINE）已展现出在束状和束下刺激选择性方面的显著能力。然而，由于神经肌肉骨骼系统以及神经电极界面固有的复杂性，尚未开发出一种使用如FINE这样的多触点神经袖套电极对肌肉骨骼系统进行功能性电刺激的轨迹跟踪运动控制算法。在我们之前的研究中，针对多输入多输出（MIMO）肌肉骨骼系统开发了一种对系统先验知识要求较少的控制系统。在本研究中，构建了一个更逼真的计算踝关节/距下关节模型，其中包括坐骨神经的有限元模型。通过调制置于坐骨神经上的FINE每个触点的脉冲幅度，对该控制系统进行测试以控制踝关节/距下关节角度的运动。仿真结果表明，基于系统稳态和动态特性分离的控制策略，对于不同的参考轨迹（如正弦和滤波随机信号）能产生较小的输出跟踪误差。所提出的控制方法还展现出对外部干扰和系统参数变化（如肌肉疲劳）的鲁棒性。在各种情况下的这些仿真结果表明，即使个体肌肉选择性有限，利用具有空间选择性的多触点神经电极通过外周神经刺激来控制肢体运动也是可行的。这项技术可能对恢复瘫痪患者的神经功能有用。
