System identification of evoked mechanomyogram in isometric contraction

The purpose of this study was to clarify the elasticity of the anterior tibial muscle depending on the contraction level. System identification technique was applied to an evoked mechanomyogram (MMG) and dorsiflex torque at a low level of isometric contraction. Electrical stimulation was applied to the common peroneal nerve when the subject maintained contraction levels at 0 (resting state), 5, 10, 20, and 30% of the maximum voluntary contraction. The evoked MMG and dorsiflex torque were measured with an acceleration sensor and a strain gauge load cell, respectively. Stimulation was conducted using 31 monopolar rectangle pulses with an interpulse interval of 1 s, and at supramaximal strength. The evoked MMGs and dorsiflex torques were averaged synchronously. System identification was performed using a singular value decomposition method. The undamped natural frequency of the system was calculated from the poles of the transfer function. The evoked MMG and dorsiflex torque in isometric contraction approximated well with a sixth-and a second-order model, respectively. The MMG peak-to-peak and the dorsiflex torque amplitude during isometric contraction decreased as the contraction level increased. The highest and intermediate natural frequencies of the sixth-order model of the evoked MMG (f₁, and f₂, respectively) tended to decrease as the contraction level increased. These decreases might reflect extension of the subcutaneous tissue. The lowest undamped natural frequency (f₃) increased as the contraction level increased. This increase might reflect an increase in muscle stiffness. In conclusion, the muscle elasticity during isometric contraction was elucidated by the proposed method.