Date of Award

Fall 12-18-2015

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Programs

Medical Sciences Interdepartmental Area

First Advisor

Joseph Ka-Chun Siu

Second Advisor

Nicholas Stergiou

Third Advisor

Vijay Shivaswamy

Abstract

There are nearly 26 million people with diabetes mellitus (DM) in the US, and half of chronic DMs develop somatosensory deficits due to diabetic polyneuropathy or diabetic peripheral neuropathy (DPN). The absence or impaired somatosensory feedback (e.g. touch sensation or joint proprioception) resulted from the damage of large nerve fiber, and motor deficits such as attenuated muscle strength and abnormal plantar pressure of lower extremity have been identified in DPN, and these sensorimotor impairments lead to an increased number of falls. To reduce the risk of falling, a well-coordinated and adapted limb movement driven by the feedforward (anticipatory) and feedback (reactive) control movement strategies are required to deal with forthcoming and instantaneous perturbations during walking respectively. The top-down feedforward control communicates with the central nervous system (CNS) and forms the basis for computing necessary motor output by simultaneously predicting or correcting errors of event information from the bottom-up feedback control. The altered spatiotemporal gait pattern in DM can either be the compensation of somatosensory feedback deficits or the compromised CNS-driven motor command. Exploring the feedforward and feedback controls not only illustrates the potential cause of the DM’s altered gait pattern but also offer the future opportunity to design prospective clinical intervention for DM’s safety and wellness.

The overall objective of this study unveiled the impacts of feedforward and feedback control on DM/DPN’s dynamic balance during walking. This dissertation adopted a virtual reality-based obstacle crossing task to examine our central hypothesis of potential altered sensory and CNS-driven motor command of DPN would be manifested through the adjustment of spatiotemporal gait characteristics compared with healthy controls. In addition, we investigated how the visual guidance played a role to the on-line adjustment of these altered gait measures as the compensation. In results, DM demonstrated the compromised feedback control by lowering their maximal toe elevation during crossing and increasing their step width after crossing; while DPN presented the both compromised feedforward and feedback controls by decreasing the toe elevation during crossing and increasing stride/stance time after crossing of obstacle. Besides, the adjustment of the altered spatiotemporal gait characteristics were observed through the visual guidance. With the combination of virtual obstacle crossing task design with the guidance of visual information, the future virtual obstacle crossing training paradigm can be implemented for training diabetes population to reduce the risk of falling.