Skip to main content
eScholarship
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

High-Resolution μECoG: Design, Fabrication, and Applications

Abstract

Since the early 1950s, electrocorticography (ECoG), the measurement of electrical potentials on the surface of the brain has become an invaluable tool in neurosurgery for the localization of epileptic foci before resection. The ECoG electrodes used in clinical practice are made in an archaic serial process that involves hand-soldering wires to a stiff, coarse grid of electrodes with a spatial resolution >1 cm, and a tangle of transcranial wires. In this thesis we report a modern microfabrication process based on photolithographic patterning of conductor thin films and of Parylene C, a biocompatible, transparent polymer. We used that process to make very thin and flexible ECoG arrays with electrode densities exceeding those of their clinical counterparts by more than two orders of magnitude, and addressed interconnect and noise performance issues. We constructed devices with multiple interconnected conductor layers, and used transparent conductors for integration of ECoG with optical neural stimulation techniques. Moreover, we developed a microscale ECoG with integrated loop antenna for a fully implantable, wireless system.

To show that such high-resolution devices have practical utility, we conducted acute and chronic in vivo studies in rats. We found that sufficiently small ECoG electrodes were able to register superficial multi-unit activity. We computed high-resolution tonotopic maps of the auditory cortex in anesthetized rats, and demonstrated that functional mapping using signal power in the 70 Hz - 170 Hz band (high-&gamma) is consistent with but much quicker and often more robust than functional mapping using action potentials recorded intracortically with penetrating electrodes. Finally we demonstrated that &muECoG can serve as a less invasive alternative to penetrating electrodes in a brain-machine interface (BMI) paradigm by training awake behaving rats, chronically implanted with &muECoG, to perform a 1D center-out task with auditory feedback by differentially modulating the high-&gamma signal on electrodes separated by as little as 200 &mum.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View