Reyes Ryan-David, Kobetic Rudolf, Nandor Mark, Makowski Nathaniel, Audu Musa, Quinn Roger, Triolo Ronald
Advanced Platform Technology Center, Department of Veterans Affairs, Louis Stokes Cleveland Department of Veterans Affairs Medical Center, Cleveland, OH, United States.
Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States.
Front Neurorobot. 2020 Dec 11;14:588950. doi: 10.3389/fnbot.2020.588950. eCollection 2020.
This study assessed the metabolic energy consumption of walking with the external components of a "Muscle-First" Motor Assisted Hybrid Neuroprosthesis (MAHNP), which combines implanted neuromuscular stimulation with a motorized exoskeleton. The "Muscle-First" approach prioritizes generating motion with the wearer's own muscles via electrical stimulation with the actuators assisting on an as-needed basis. The motorized exoskeleton contributes passive resistance torques at both the hip and knee joints of 6Nm and constrains motions to the sagittal plane. For the muscle contractions elicited by neural stimulation to be most effective, the motorized joints need to move freely when not actively assisting the desired motion. This study isolated the effect of the passive resistance or "friction" added at the joints by the assistive motors and transmissions on the metabolic energy consumption of walking in the device. Oxygen consumption was measured on six able-bodied subjects performing 6 min walk tests at three different speeds (0.4, 0.8, and 1.2 m/s) under two different conditions: one with the motors producing no torque to compensate for friction, and the other having the motors injecting power to overcome passive friction based on a feedforward friction model. Average oxygen consumption in the uncompensated condition across all speeds, measured in Metabolic Equivalent of Task (METs), was statistically different than the friction compensated condition. There was an average decrease of 8.8% for METs and 1.9% for heart rate across all speeds. While oxygen consumption was reduced when the brace performed friction compensation, other factors may have a greater contribution to the metabolic energy consumption when using the device. Future studies will assess the effects of gravity compensation on the muscular effort required to lift the weight of the distal segments of the exoskeleton as well as the sagittal plane constraint on walking motions in individuals with spinal cord injuries (SCI).
本研究评估了使用“肌肉优先”型运动辅助混合神经假体（MAHNP）外部组件行走时的代谢能量消耗，该假体将植入式神经肌肉刺激与电动外骨骼相结合。“肌肉优先”方法优先通过电刺激利用佩戴者自身的肌肉产生运动，而致动器则根据需要提供辅助。电动外骨骼在髋关节和膝关节处产生6牛米的被动阻力扭矩，并将运动限制在矢状面内。为了使神经刺激引发的肌肉收缩最有效，电动关节在不主动辅助期望运动时需要自由移动。本研究分离了辅助电机和传动装置在关节处增加的被动阻力或“摩擦力”对使用该装置行走时代谢能量消耗的影响。在两种不同条件下，对六名身体健康的受试者进行了6分钟步行测试，测量了他们在三种不同速度（0.4、0.8和1.2米/秒）下的耗氧量：一种条件是电机不产生扭矩来补偿摩擦力，另一种条件是电机根据前馈摩擦模型注入动力以克服被动摩擦力。以代谢当量（METs）衡量，在所有速度下未补偿条件下的平均耗氧量与摩擦补偿条件下在统计学上存在差异。在所有速度下，METs平均下降了8.8%，心率平均下降了1.9%。虽然当支架进行摩擦补偿时耗氧量降低，但在使用该装置时，其他因素可能对代谢能量消耗有更大的影响。未来的研究将评估重力补偿对抬起外骨骼远端部分重量所需肌肉力量的影响，以及矢状面约束对脊髓损伤（SCI）患者步行运动的影响。
