Kinematic state feedback is important for neuroprostheses to generate stable and adaptive movements of an extremity. State information, represented in the firing rates of populations of primary afferent neurons, can be recorded at the level of the dorsal root ganglia (DRG). Previous work in cats showed the feasibility of using DRG recordings to predict the kinematic state of the hind limb using reverse regression. Although accurate decoding results were attained, these methods did not make efficient use of the information embedded in the firing rates of the neural population.;This dissertation proposes new methods for decoding limb kinematics from primary afferent firing rates. We present decoding results based on state-space modeling, and show that it is a more principled and more efficient method for decoding the firing rates in an ensemble of primary afferent neurons. In particular, we show that we can extract confounded information from neurons that respond to multiple kinematic parameters, and that including velocity components in the firing rate models significantly increases the accuracy of the decoded trajectory.;This thesis further explores the feasibility of decoding primary afferent firing rates in the presence of stimulation artifact generated during functional electrical stimulation. We show that kinematic information extracted from the firing rates of primary afferent neurons can be used in a real-time application as a feedback for control of FES in a neuroprostheses. It provides methods for decoding primary afferent neurons and sets a foundation for further development of closed loop FES control of paralyzed extremities.;Although a complete closed loop neuroprosthesis for natural behavior seems far away, the premise of this work argues that an interface at the dorsal root ganglia should be considered as a viable option.;Keywords: bioengineering, muscle spindle, primary afferent, nervous system, closed loop control, state-space modeling, neuroprostheses, FES .
展开▼
