Quantification of NeuroMuscular Properties during Torque Tracking Tasks

Quantification of NeuroMuscular Properties during Torque Tracking Tasks

Van der Burg, A.A.

De Vlugt, E. (mentor)
Van Eesbeek, S. (mentor)

Mechanical, Maritime and Materials Engineering

BioMechanical Engineering

BMD

2013-07-01

Human movement control is frequently studied by in vivo quantification of NeuroMuscular System (NMS) properties, e.g. intrinsic joint stiffness and joint damping and reflexive feedback gains, during functional tasks. An important measure in movement control is joint admittance, the relation between joint angle and torque. The NMS properties can be used to indicate the specific contributions to joint admittance by parts of the system. Furthermore, dependencies of properties with system states can also be used for physiological interpretation, returning fundamental knowledge on healthy and pathological movement. This knowledge has proven useful for medical diagnostics and aid. In common functional tasks like walking, the variation in states of the NMS is often large. These large variations in states may cause time-variance of the properties. This study compared quantified properties between tasks that required subjects to track a varying or non-varying level of torque using the ankle dorsiflexor. A novel Linear Parameter Varying (LPV) system identification technique was used to estimate the time varying NMS properties. The results indicated that the admittance for tracking a non-varying torque was higher compared to tracking a sinusoidal varying (0.1, 0.2, 0.3 and 0.4Hz) level of torque. Admittance also increased with increasing frequency of the variation in required torque. The parameters were estimated using a parametric model that was fitted on the Frequency Response Function (FRF) of identified LPV models. Intrinsic ankle stiffness and ankle damping and reflexive position- and force-feedback tested significantly different between tasks. Ankle stiffness was lower and ankle damping and reflexive feedbacks were higher while tracking a varying torque. These properties respectively decreased and increased further with increasing frequency of the variation. Position- and velocity-feedback changed from inhibitory to excitatory with increasing background torque, while force-feedback was solely inhibitory. The results indicated that when tracking a fast varying torque lower ankle admittance was reached by lower intrinsic ankle stiffness and increased inhibitory reflexive feedback, compared to tracking slow varying and non-varying torque levels.

neuromuscular properties
joint admittance
linear parameter varying
torque tracking task

http://resolver.tudelft.nl/uuid:9f48ba1d-9a0c-4b23-9e14-220c1eccf624

2013-09-22

Student theses

master thesis

(c) 2013 Van der Burg, A.A.