Computational models of upper limb movement during functional reaching tasks for application in electrical stimulation based stroke rehabilitation

Abstract

Introduction: Functional electrical stimulation (FES) has been shown to be an effective approach to upper limb stroke rehabilitation, where it assists patients' arm and shoulder movement. Model-based FES controllers have recently confirmed significant potential to improve accuracy of functional reaching tasks, but they typically require a reference trajectory to track. No upper limb FES control scheme has yet embedded a computational model of the task, however this is critical to ensure the controller reinforces the intended
movement with high accuracy. This paper derives a computational motor control model of the task which can be embedded in FES control schemes, removing the need for a predefined reference trajectory.
Methods: Kinematic data were collected using a Vicon motion capture system from unimpaired (N = 14) participants while they performed two functional reaching tasks in which they: 1) pushed a light switch, and 2) closed a drawer. In each case they starting and finished the movement with their hand on their knee. Dynamic models of each patient's arm were derived using estimated mass, inertial and stiffness parameters.Each task was posed as an optimization problem with position and velocity boundary constraints, and these were solved using iterative algorithms to yield computational models of movement.
Results: For the case of unimpaired participants, the experimentally recorded joint angles were compared with those derived in simulation using the model, and were found to fit closely (mean fitting > 85%).
Conclusion: Functional movements have been accurately modelled as constrained optimization problems involving dynamic models of unimpaired participants' arms. This extends previous computational models of human movement, and shows that they can be solved using iterative methods. Moreover, these methods are suitable to be employed experimentally in future stroke rehabilitation trials using FES to assist task completion in a manner corresponding to unimpaired movement. This hence ensures that assistance is aligned with voluntary intention, and in-so-doing maximizes the potential effectiveness of treatment.

More information

Identifiers

ORCID for Katie Meadmore: orcid.org/0000-0001-5378-8370

ORCID for Ann-Marie Hughes: orcid.org/0000-0002-3958-8206

ORCID for J.H. Burridge: orcid.org/0000-0003-3497-6725

Catalogue record

Export record