Carnegie Mellon University
Browse
ashwatik_ECE_2016.pdf (40.54 MB)

Residual Voltage in Biphasic Electrical Stimulation: Cause, Clues, and Control

Download (40.54 MB)
thesis
posted on 2016-02-04, 00:00 authored by Ashwati KrishnanAshwati Krishnan
Electrical stimulation of neural tissue has been known to evoke functional responses in ani-
mals. Stimulation is primarily performed by passing controlled, symmetric biphasic current
pulses to an electrode placed near the neural tissue. The biphasic current pulse consists of a
negative pulse, followed by a positive pulse to maintain charge neutrality. A theoretical anal-
ysis on a rst order electrode-electrolyte (tissue) interface model has shown that perfectly
balanced input current signals do not ensure net neutrality at the interface, due to an unre-
coverable loss of charge via the faradaic impedance. In chronically implanted devices, there is
currently no practical way to quickly identify changes that occur at the electrode-tissue inter-
face, especially in high-density electrode arrays. This work explores the extent to which the
residual voltage can act as a preliminary indicator of electrode degradation, because residual
voltage is essentially a characteristic of the interface. While residual voltage provides timely
feedback when sampled at regular intervals, it can accumulate and a DC voltage on a stimu-
lation electrode can be potentially unsafe in a chronically implanted device. The method of
delivering current signals has traditionally been implemented as an open loop system. This
work also demonstrates that one can control the residual voltage by correcting the positive
pulse of the biphasic signal based on the existing state of the electrode-electrolyte/tissue
interface. While the resulting biphasic stimulation waveform is imbalanced, the interface
is electrochemically neutral. The updated value of the imbalanced anodic pulse width can
provide us meta-data about the interface, in case there is any degradation. By controlling
the residual voltage at the electrode actively with feedback, the proposed closed loop system
will improve the safety of neural stimulator systems.

History

Date

2016-02-04

Degree Type

  • Dissertation

Department

  • Electrical and Computer Engineering

Degree Name

  • Doctor of Philosophy (PhD)

Advisor(s)

Shawn K. Kelly

Usage metrics

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC