Title:  Finite Element Analysis and Experimental Evaluation of Magnetic Stimulation of Neurons using Micro-scale Coils toward and Improved Cochlear Implant

Committee:

Dr. Pamela Bhatti, ECE, Chair , Advisor

Dr. Robert Butera, ECE

Dr. Andrew Peterson, ECE

Dr. Manos Tentzeris, ECE

Dr. Suresh Sitaraman, ME

Abstract:

Cochlear implants are one of the most successful neural prosthetic devices in the world. With an estimated user base of 320,000 patients across the globe as of 2012, this implant is considered the best option for patients with profound deafness. The state-of-the-art cochlear implants use electrical stimulation to convey sound information to the auditory neurons. However, this electrical stimulation leads to inflammation of tissue, negative electrochemical reactions, and non-focused stimulation due to current spread; resulting in sub-optimal performance of the device. Additionally, patients complain about abnormal pitch perception and inability to enjoy music. The objective this research is evaluation of magnetic stimulation as an alternative to direct electrical stimulation in the cochlea. A set of parametric studies are performed to examine a range of factors such as power consumption, heating, as well as the respective response of neurons to electrical and magnetic stimulation through experiments and finite-element modeling. For experiments, micro-scale coils are used to assemble a magnetic stimulator, and an electric stimulator is assembled using a commercially available thin-film array. A set of dry experiments (without neurons) are performed to assess the ability of multielectrode arrays to characterize magnetic stimulation. Then experiments with dissociated cortical neurons of embryonic day 18 rats are performed on these planar multielectrode arrays. When the firing rate per channel of these plated arrays are compared before and during magnetic stimulation, a statistically significant change is observed, but no specific change in the rate, in terms of its increase or decrease, is seen. On an average, power consumption for magnetic stimulation (60.4 mW for 3.5 V, and 310 µW for 250 mV) is higher than that for electrical stimulation (40 µW at ±500 µA stimulation amplitude). Additionally, we calculate that at a stimulation rate of 2 Hz, magnetic stimulation leads to a temperature change of 0.23ºC when 3.5 V is the input to the amplifier. Electrical stimulation on the other hand leads to a temperature change of 0.1ºC at ±500 µA stimulation amplitude. While comparing the effects of stimulation due to both electrical and magnetic stimulation, these results highlight a need to further investigate micro-scale magnetic stimulation. This work also presents a novel setup for studying this stimulation, allowing the analysis of a global response from the neurons.