Stimulation parameter optimization for functional electrical stimulation assisted gait in human spinal cord injury using response surface methodology

Background The aims of this study were to identify the reflex moment induced by flexion withdrawal reflex and to optimize stimulation parameters for restoring swing motion with respect to initial kinematic conditions in human with spinal cord injury. Methods The influence of hip position and passive movement in the reflex moment were tested in six subjects with chronic spinal cord injury. The two-dimensional dynamic models consisted of thigh, shank and foot segments were developed to compute the swing-phase response and the response surface method was also used to optimize stimulation parameters for restoration of gait by functional electrical stimulation. Findings At three different hip positions, significant linear relationship was found between the reflex moment and hip angle (P < 0.05) and hip movement also increased the reflex moment compare to isometric conditions. The hip and knee flexion velocities significantly contributed to the hip and knee flexion angle during the swing-phase (P < 0.05) and increase of initial joint velocity resulted in a decrease of the burst frequency and duration time for optimal swing motion in spinal cord injured patients. Interpretation From dynamic simulation, we concluded that optimal solutions of pulse amplitude, frequency and duration time of burst for electrical stimulation assisted gait were influenced by initial kinematic conditions at toe-off. The reflex model and the results of this study can be applied to the design and control strategies of neuroprosthetic devices using functional electrical stimulation for spinal cord injured patients.