Date of Award

6-2014

Degree Name

MS in Engineering - Bioengineering

Department/Program

Biomedical and General Engineering

Advisor

Robert Szlavik

Abstract

Peripheral neuropathy refers to diseases of or injuries to the peripheral nerves in the human body. The damage can interfere with the vital connection between the central nervous system and other parts of the body, and can significantly reduce the quality of life of those affected. In the US, approximately between 15 and 20 million people over the age of 40 have some forms of peripheral neuropathy. The diagnosis of peripheral neuropathy often requires an invasive operation such as a biopsy because different forms of peripheral neuropathy can affect different types of nerve fibers. There are non-invasive methods available to diagnose peripheral neuropathy such as the nerve conduction velocity test (NCV).

Although the NCV is useful to test the viability of an entire nerve trunk, it does not provide adequate information about the individual functioning nerve fibers in the nerve trunk to differentiate between the different forms of peripheral neuropathy. A novel technique was proposed to estimate the individual nerve fiber diameters using group delay and simulated annealing optimization. However, this technique assumed that the fiber depth is always constant at 1 mm and the fiber activation due to a stimulus is depth independent. This study aims to incorporate the effect of fiber depth into the fiber diameter estimation technique and to make the simulation more realistic, as well as to move a step closer to making this technique a viable diagnostic tool.

From the simulation data, this study found that changing the assumption of the fiber depth significantly impacts the accuracy of the fiber diameter estimation. The results suggest that the accuracy of the fiber diameter estimation is dependent on whether the type of activation function is depth dependent or not, and whether the template fiber diameter distribution contains mostly large fibers or both small and large fibers, but not dependent on whether the fiber depth is constant or variable.

Share

COinS