EMG-controlled knee orthosis lowers effort in sit-to-stand.

Objective: Pilot study with ten healthy adults, testing whether a lightweight, low-cost knee orthosis equipped with EMG-driven impedance control reduces quadriceps muscle effort during the sit-to-stand (STS) transition. Methods: Ten able-bodied adults performed 15 paced STS repetitions under three conditions: without orthosis (No-Ortho), orthosis worn unpowered (Ortho-OFF; friction-compensated), and orthosis actively powered (Ortho-ON). Surface electromyography (EMG) was recorded using 8-channel thigh bracelets on both legs. EMG signals from the braced leg were processed using ridge regression and slew-rate limiting to generate a normalized control signal that dynamically scales knee stiffness while maintaining constant damping. Median values and trial-to-trial variance of the average rectified EMG (ARV) were analyzed across four distinct movement phases (SIT, UP, STAND, DOWN) using linear mixed-effects models with log-transformed data and Bonferroni-adjusted planned contrasts. Results: Powered assistance significantly reduced median bilateral ARV by 11% during the UP phase and 15% during the DOWN phase ( p adj < 0.001 ) , with greater reductions (up to 21%) observed on the braced limb. Variance in muscle activation decreased substantially (by up to 44%) on the braced leg during the DOWN phase, suggesting more repeatable activation patterns and neuromuscular consistency across trials. No significant compensatory activation was observed in the contralateral limb. Additionally, within-session adaptation trends were observed as participants progressively increased preparatory torque during the SIT phase, while UP-phase ARV trended downward. Conclusion: A lightweight, affordable knee orthosis employing a rapid ( ≈ 10 s), minimally calibrated EMG-driven impedance controller effectively reduces quadriceps muscle activation during STS without compromising natural movement coordination. Torque capacity limitations (16 Nm) may limit effectiveness for heavier users, and further research is needed to evaluate kinematic fidelity fully.