Hybrid robotic and electrical stimulation-based neurorehabilitation after stroke
File(s)
Author(s)
Cazenave, Lucille Madeleine Marie
Type
Thesis
Abstract
Stroke is a leading cause of adult disability, with over 70% of survivors experiencing upper-limb impairments. While rehabilitation can reduce these deficits, its intensity and availability often remain insufficient, motivating the development of technology-assisted approaches.
Rehabilitation robots enable task-specific motor training, while functional electrical stimulation (FES) supports movement by eliciting muscle contraction and providing sensory feedback. However, FES is limited by rapid muscle fatigue and can be uncomfortable over time. Hybrid systems combining robotics and FES may address these limitations by allowing consistent kinematic guidance from the robot while preserving the neurophysiological engagement provided by FES. This could enable rehabilitation training to be effective and tolerable across a range of impairment levels.
This thesis investigated the effects of hybrid robot-FES assistance and training on motor behaviour, physiological responses, and user experience in unimpaired individuals and stroke survivors, with a focus on clinical applicability. To this end, a simple and portable one-degree-of-freedom robot-FES system was developed to assist wrist flexion/extension, with both systems acting on the same joint. A series of experiments explored how robotic, FES, and hybrid assistance affect performance, effort, fatigue and user experience. Moreover, the effect of a three-week hybrid robot-FES rehabilitation training on functional and physiological recovery after stroke was evaluated.
Overall, the findings indicate that the hybrid system can effectively assist movement while remaining comfortable in unimpaired and stroke individuals, with limited fatigue increase. A three-week intervention of daily hybrid training, in addition to usual care, produced clinically meaningful improvements in impairment, function, and movement quality, alongside high patient engagement and satisfaction. Together, these results support the clinical relevance of such hybrid robot-FES rehabilitation as a safe and effective complement to conventional physiotherapy.
This thesis advances our understanding of hybrid robot-FES assistance and training, informing the development and implementation of clinically viable and efficient neurorehabilitation technologies.
Rehabilitation robots enable task-specific motor training, while functional electrical stimulation (FES) supports movement by eliciting muscle contraction and providing sensory feedback. However, FES is limited by rapid muscle fatigue and can be uncomfortable over time. Hybrid systems combining robotics and FES may address these limitations by allowing consistent kinematic guidance from the robot while preserving the neurophysiological engagement provided by FES. This could enable rehabilitation training to be effective and tolerable across a range of impairment levels.
This thesis investigated the effects of hybrid robot-FES assistance and training on motor behaviour, physiological responses, and user experience in unimpaired individuals and stroke survivors, with a focus on clinical applicability. To this end, a simple and portable one-degree-of-freedom robot-FES system was developed to assist wrist flexion/extension, with both systems acting on the same joint. A series of experiments explored how robotic, FES, and hybrid assistance affect performance, effort, fatigue and user experience. Moreover, the effect of a three-week hybrid robot-FES rehabilitation training on functional and physiological recovery after stroke was evaluated.
Overall, the findings indicate that the hybrid system can effectively assist movement while remaining comfortable in unimpaired and stroke individuals, with limited fatigue increase. A three-week intervention of daily hybrid training, in addition to usual care, produced clinically meaningful improvements in impairment, function, and movement quality, alongside high patient engagement and satisfaction. Together, these results support the clinical relevance of such hybrid robot-FES rehabilitation as a safe and effective complement to conventional physiotherapy.
This thesis advances our understanding of hybrid robot-FES assistance and training, informing the development and implementation of clinically viable and efficient neurorehabilitation technologies.
Version
Open Access
Date Issued
2025-09-17
Date Awarded
2026-02-01
Copyright Statement
Attribution-Non Commercial-No Derivatives 4.0 International Licence (CC BY-NC-ND)
Advisor
Burdet, Etienne
Bentley, Paul
Sponsor
UKRI
European Commission
Grant Number
ICT 871767
EP/S023283/1
Publisher Department
Department of Computing
Department of Bioengineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)