Student Presentation -- Ryan Caldwell
Home → Student Presentations Index → Student Presentation -- Ryan Caldwell

ICS import...
ICal Link

Ph.D. Dissertation Defense, Friday July 28, 2017 -- Strategies towards the mitigation of shunting in implanted neural arrays to improve device stability for chronic applications

SMBB 2660, 11:00 am

Speaker: Ryan Caldwell. Advisor: Dr. Florian Solzbacher


By enabling neuroprosthetic technologies, neural microelectrodes can greatly improve diagnostic and treatment options for millions of individuals living with limb loss, paralysis, and sensory and autonomic neural disorders. However, clinical use of these devices is restricted by the limited functional lifetimes of implanted electrodes, which are commonly less than a few years. One cause is the evolution of damage to dielectric encapsulation which insulates microelectrodes from the physiological environment. Fluid penetration and exposure to an aggressive immunological response over time may weaken encapsulating films and cause electrical shunting. This reduces electrode impedance, diverts electrical signal away from target tissue, and causes multi-channel crosstalk. To date, no neural microelectrode encapsulating material or design approach has reliably resolved this issue. We employ the parylene C-encapsulated Utah Electrode Array (UEA), a silicon-micromachined neural interface FDA-cleared for human use, to execute three aims that address this challenge through investigations of new materials, electrode designs, and testing methods.

We first evaluate a novel bilayer encapsulating film comprised of atomic layer deposited Al2O3 and parylene C, testing this film using UEAs and devices with UEA-relevant topography. Contrasting with previous work employing simplified planar structures, the incorporation of neural electrode features on test structures revealed failure modes pointing to the dissolution of Al2O3 over time. Our results emphasize the need for dielectric coatings resistant to water degradation as well as test methods which take electrode features into account. In our second aim, we show through finite element modeling and aggressive in vitro testing that use of degenerately doped silicon as a conductive neural electrode material can mitigate the consequences of encapsulation damage, owing to the high electrochemical impedance of silicon. Our final aim compares oxidative in vitro aging to long term in vivo material damages and finds clear evidence that such in vitro testbeds may help predict certain in vivo damage modes. By pairing this testing with absorption and emission spectroscopic characterization modalities, we identify contributors to material damage and future design solutions. Our results will inform future material and testing choices, to improve the resilience of neural electrode dielectric encapsulation and enhance the longevity of neuroprostheses.