A Nerve Fiber Model and Prediction of Electrode-Fiber Coupling Simulation for Design of Peripheral Nerve Interfaces

Michael Ryne Horn, Purdue University

Abstract

The peripheral nervous system (PNS) relays information between the central nervous system (CNS) and end organs. The ability to record and stimulate the PNS would provide a means to both detect anomalies and control them. The resulting neuromodulatory therapies are the outcomes of the emerging field of bioelectronics. It is hoped that therapies can be contrived based off of modeled data along with collected data to help provide a minimally invasive surgery option for many diseases, even heart diseases. However, current nerve models lack the granularity to resolve the subtle complex ion channel interaction, temperature effects, cable properties based off of axon geometries and the interaction of the nerve axon and the recording electrode. To aid in the development of recording and stimulating neural interfacing electrodes a physiologically accurate model for small C- and A-type autonomic nerve fibers which allows for estimation of conduction velocities and the spatial wavelength was developed. A qualitative observation showed that smaller electrodes are needed to record and selectively stimulate these small diameter, slow conducting fibers to compensate for the averaging effect that is seen on the active site of the electrode structure. If the active site of the electrode is on the order of the spatial wavelength of the traveling action potential (AP), the single fiber action potential (SFAP) becomes too small to detect. This is a possible issue with the recording of the elusive C-type fibers. To test the hypothesis, a physiological model of an active axon was modeled based off of ion channels that have been genetically sequenced and whose currents were experimentally measured using somatic patch clamp techniques. These ion channels were placed on an axon whose transmission properties was dictated by a standard cable equation. The impedance of the membrane, impedance of the myelin and resistance of the axoplasm was accounted for by the axons geometric dimensions. Temperature coefficients were incorporated such that the traveling AP conduction velocity would decrease with lower temperatures and increase with higher temperatures. This traveling transmembrane action current (TMAC) was then coupled with a recording electrode to see the effects a recording electrode would have on accurately portraying the SFAP. Results of the simulations proved that the electrode-fiber coupling is greatly effected by the slow conducting nerve fibers of the autonomic system. With such slow conduction the spatial wavelength of the traveling action potential (AP) is on the same order or much smaller than traditional electrodes used. With this insight, new, smaller electrodes were simulated to determine there viability in recording from the small diameter population of nerve fibers in the autonomic system. The method for predicting an estimated SFAP was validated against recorded vagal SFAP recordings in a rat. The results showed a high degree of corelation between the recorded and simulated SFAP waveforms. This thesis builds on the continual work by many modelers prior to me to simulate a physiological accurate nerve fiber and couple that with stimulating and recording electrodes. With the continual increase in sophistication of these models it has become possible to predict the population of nerve fibers an electrode can interact with and at what distances, as well as the waveform that an electrode will record from a nerve fiber. All before the electrode is fabricated and implanted. The future for model based electrode design will help catapult the field of bioelectronics by reducing the time and cost of fabricating expensive electrodes that are not able to do the task they were intended to perform. It will also aid in pattern recognition by providing the template waveforms that are hoped to be found.

Degree

M.S.B.M.E.

Advisors

Yoshida, Purdue University.

Subject Area

Biomedical engineering

COinS