Analysis of μECoG Design Elements for Optimized Signal Acquisition

Limited Access
This item is unavailable until:
2024-09-16

Date

2022

Authors

Williams, Ashley Jerri

Advisors

Viventi, Jonathan V

Journal Title

Journal ISSN

Volume Title

Repository Usage Stats

57
views
0
downloads

Abstract

High density electrode arrays that can record spatially and temporally detailed neural information provide a new horizon for the scientific exploration of the brain. Chief amongst these new tools is micro-electrocorticography (µECoG), which has grown in usage over the past two decades. As µECoG arrays increase in contact number, density, and complexity, the form factors of arrays will also have to change in tandem – particularly the size and spacing (pitch) of electrode contacts on the array. The continued growth of the field of µECoG research and innovation is hampered by a lack of understanding of how fundamental design aspects of the arrays may impact the information obtained from µECoG in different recording bands of interest and animal models. Utilizing thin-film fabrication to create novel experimental arrays and novel analysis techniques, the work in this dissertation provides an understanding of how differences in electrode contact size and spacing can impact neural metric acquisition in four experimentally and clinically relevant frequency bands of local field potential (LFP), high gamma (HGB), spike band power (SBP), and high frequency broadband (HFB). This dissertation provides innovative arrays that allow for experimental variation within a recording session, unlike much of the work previously published comparing contact size and pitch.

This dissertation shows my work of designing, testing, and implementing novel designs of μECoG arrays to explore the questions of how contact size and pitch may impact neural metrics in rodents and non-human primates (NHPs). In Chapter 2, I used a novel 60-channel array with four different contact size diameters in rodents to explore how contact size, impedance, and noise may impact neural metrics we collect in auditory experiments. We determined that contact size may selectively play a role in neural metric information content acquisition, and that the factors of impedance and noise can impact them significantly in higher frequency bands. This work also showed the ability to resolve multi-unit spiking activity from the surface of the brain. In Chapter 3, I show results obtained using a 61-channel array with different contact pitch in rodents, giving clarity to how the spatial sampling of the neural field may be impacted by the pitch of the electrode contacts used. These results suggest the neural field in higher frequency bands show greater changes at shorter field lengths than lower frequency bands. In Chapter 4, I utilized a larger 244-channel array in a NHP with varied contact sizes to explore how contact size may impact information content obtained from NHPs in the motor-related areas of the brain. Chapter 5 concludes the investigation of how design characteristics may impact neural information content by using an array with a local reference electrode contact to explore how local re-referencing can improve the neural metrics obtained.

The results from this dissertation provide a comprehensive understanding to how the information in the neural field may be impacted by the electrode designs chosen. The utilization of novel in-house fabricated arrays provides a method to explore these neuroscience questions rapidly and at low-cost.

Description

Provenance

Citation

Citation

Williams, Ashley Jerri (2022). Analysis of μECoG Design Elements for Optimized Signal Acquisition. Dissertation, Duke University. Retrieved from https://hdl.handle.net/10161/25817.

Collections


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.