Post-stroke rehabilitation, particularly in restoring function to the wrist and hand (W/H), faces significant challenges due to compensatory movement patterns that develop in the shoulder and elbow joints. These compensations help individuals with daily tasks but often result in the learned disuse of distal muscles, hindering motor recovery. Effective rehabilitation requires not only restoring motor control but also improving sensorimotor integration (SMI) between the brain and the targeted muscles. However, current robotic rehabilitation systems often fail to adequately address both the motor pathways for movement control and the sensory pathways necessary for feedback. "Traditional rehabilitation techniques, such as neuromuscular electrical stimulation (NMES) and vibratory stimulation, provide sensory feedback but are limited by issues like muscle fatigue or discomfort." said the author Legeng Lin, a researcher at The Hong Kong Polytechnic University, "To address these challenges, we propose an electromyography (EMG)-driven robot-assisted system with electro-vibro-feedback (EVF), which enhances both the voluntary motor control and the sensory feedback to the wrist and hand muscles. The aim is to improve motor function by modulating the ascending and descending neural pathways, thereby enhancing motor control and neuroplasticity in the affected limb."
The robot-assisted system designed in this study integrates electromyography (EMG) with electro-vibro-feedback (EVF) to aid in the rehabilitation of wrist and hand (W/H) function post-stroke. The system consists of a soft robotic device equipped with five pneumatic fingers that assist in wrist and hand movements by inflating and deflating to provide mechanical assistance. This robot is controlled by the residual EMG signals detected from the forearm extensor (EX) and flexor (FX) muscles of the affected limb. The system operates in two key ways: voluntary motor control (VME) and somatosensory priming. Voluntary motor control refers to when the user voluntarily activates the EX or FX muscles, the electromyographic signal triggers robotic assistance to help complete wrist extension (accompanied by hand opening) or flexion (accompanied by hand closing). somatosensory priming involves the system applying neuromuscular electrical stimulation (NMES) to the EX muscles and focal vibratory stimulation (FVS) to the FX muscles. NMES helps activate weaker extensor muscles and promote muscle contraction, while FVS provides sensory feedback to patients by activating mechanoreceptors without causing spasms in the flexor muscles. This combination of EMG-driven control and sensory feedback aims to improve sensorimotor integration, modulate neural pathways, and ultimately enhance voluntary control and coordination in the paretic upper limb.
The experimental results demonstrate the efficacy of the EMG-driven electro-vibro-feedback (EVF) robot in improving wrist and hand (W/H) motor control and enhancing sensorimotor integration in chronic stroke patients. A single-arm clinical trial with 15 participants was conducted, and significant improvements were observed across multiple assessments. Significant increases were noted in the Fugl-Meyer Assessment (FMA) scores, specifically for the upper extremity (FMA-UE) and W/H subscales, as well as in the Action Research Arm Test (ARAT) scores, particularly for fine motor tasks such as grasping and pinching. The monofilament test, which assesses tactile sensation, showed significant improvements in sensory perception, especially in areas innervated by the median and ulnar nerves, which correspond to the FX muscles. These improvements were maintained at a 3-month follow-up, with participants showing enhanced motor control in W/H tasks. The experiment showed shifts in corticomuscular coherence (CMC) towards the contralateral hemisphere for the EX and FX muscles, indicating that the EVF robot helped to restore more balanced motor control by enhancing neural connections between the cortex and the muscles. The results suggest that the EMG-driven EVF robot significantly enhances both motor function and sensory feedback in stroke patients, leading to improved W/H control, reduced compensation, and long-lasting neuroplastic changes.
The study demonstrated the effectiveness of the EMG-driven electro-vibro-feedback (EVF) robot in improving motor control and enhancing sensorimotor integration in chronic stroke patients. The robot successfully combines voluntary motor control with targeted sensory feedback, addressing both the descending motor pathways for movement control and the ascending sensory pathways for feedback. Despite promising results, the study's sample size was small, and the intervention duration was relatively short. Future studies should explore larger populations, longer training periods, and varying levels of impairment to further assess the robot's long-term efficacy. "Meanwhile, this study only assessed changes in sensorimotor function at three time points. In the future, we will conduct a more detailed assessment across the entire period to better understand the dynamic changes that occur during the rehabilitation process." said Legeng Lin.
Authors of the paper include Legeng Lin, Yanhuan Huang, Wanyi Qing, Man-Ting Kuet, Hengtian Zhao, Fuqiang Ye, Wei Rong, Waiming Li, and Xiaoling Hu.
This work was supported by the University Grants Committee Research Grants Council, Hong Kong (GRF15207120 and GRF15218324); the Innovation and Technology Fund (ITT/012/23GP); the Strategic Topics Grant (STG1/M-401/24-N); and The Hong Kong Polytechnic University (1-ZVVP and 1-CD74).
The paper, "Sensorimotor Integration by Targeted Priming in Muscles with Electromyography-Driven Electro-vibro-feedback in Robot-Assisted Wrist/Hand Rehabilitation after Stroke" was published in the journal Cyborg and Bionic Systems on Jan 27, 2026, at DOI: 10.34133/cbsystems.0507.
