Estimation and validation of dynamic ankle impedance for accurate ankle spasticity assessment

Abstract

Spasticity is a symptom of an upper motor neuron syndrome, characterized by resistance change in response to passive stretch depending on the joint velocity and position. For accurate diagnosis, quantification of individual components of resistance is required. For accurate diagnosis of spasticity, quantification of detailed components of resistance is required. Clinical evaluation criteria, known as modified Ashworth scale (MAS) however, rely on evaluator’s subjective sense and lacks analysis on individual components of resistance. This thesis deals with the development and validation of parameters that can quantitatively assess the components of resistance by estimating the dynamic impedance of ankle joints in both healthy people and stroke patients. Resistance (joint angle, resistance torque, electromyogram signal) is quantitatively measured in the ankle joint passive movement of healthy people and stroke patients through experiments. The elastic and damping parameter of muscle based on neuro-muscular model of ankle joint were estimated and validated. Analysis of the dynamic impedance components of the stroke patient group revealed that the changes in passive characteristics (elasticity, viscosity) due to hypertonia and the reflex characteristics (muscle activity and nerve signaling). The reliability of the estimated impedance components was verified through the statistical data of the estimated parameters and variance in estimated torque value. Torque estimates confirmed at least 80% model accuracy in both healthy groups and stroke patient group. Based on the research results, it is possible to provide the necessary information for the ankle rehabilitation of the stroke patients and to provide the information necessary to generate the assistance ankle joint force of the wearable rehabilitation robot.