A Multi-Layered Quantitative Assessment Approach for Hand Spasticity Based on a Cable-Actuated Hand Exoskeleton

Abstract

Over the past two decades, numerous hand exoskeletons have been developed for rehabilitation scenarios, yet very few have the capability to assess spasticity. This paper introduces a quantitative assessment approach for hand spasticity developed on a cable-driven hand exoskeleton that was specifically designed with spastic hand in mind. The exoskeleton features an adaptive cable-linkage transmission mechanism equipped with bi-directional force transducers, enabling constant-velocity extension and flexion of individual finger joints while simultaneously recording joint torque and angle. Based on the exoskeleton, a multi-layered quantitative assessment approach is proposed to evaluate spasticity on individual finger joints. The basis of the approach is collecting joint resistance information automatically following Modified Tardieu Scale, serving for the computation of six middle-level parameters related to the properties of a spastic joint. The parameters include the range of motion, average resistance torque, joint stiffness, joint viscosity, catch angle, and reflex-induced resistance torque. The first four mechanical parameters are finally combined into a biomechanical metric, whereas the rest two reflex-related parameters result in a neurological metric, to neatly describe spasticity level. The performance of the exoskeleton to measure the parameters was tested on six healthy subjects. The assessment approach applied with the developed exoskeleton shows good reliability, repeatability, and validity to capture the features of spastic hands, providing strong evidence for further validation on real patients.