Our goal is to develop a new class of electrophysiological recording devices, on the same physical scale as the neurons they record, by combining modern imaging with implanted optoelectronics. Although conventional multi-electrode recording can monitor neural activity with high temporal precision, it is chronically invasive due to the wires/shanks used to bring signals out. Optical imaging techniques, conversely, provide finer spatial resolution, easing neural identification, but at the cost of significantly worse temporal resolution. such techniques also require either chemical (e.g., fluorescent dyes) or genetic modification of the tissue, and a sufficiently non-scattering light path to image single cells. To best harness the synergy between these modalities, we are developing untethered microscale optoelectronically transduced electrodes (MOTEs) which combine optoelectronic elements for power and communication with custom CMOS circuits for low-noise amplification and encoding of electrophysiological signals. Each MOTE is powered by a photovoltaic (PV) cell, providing 1-2W of electrical power to measure, amplify, and encode electrophysiological signals (Vnoise ~ 10's V, BW > 10 KHz), up-linking this information through an LED, where a single device can act as both PV and LED (PVLED). As the proposed system is wireless and can be confined within an extremely small volume (cross-section ~ 10m), MOTEs are free of the drawbacks found in the standard wire- and shank-based neural electrodes, while maintaining high temporal resolution. To be most useful, MOTEs' PV will be designed to harvest power from the optical stimulus used in fluorescence imaging. Similarly, the integrated LED for uplink will be designed to emit at wavelengths and intensities consistent with a fluorescent imaging system. Such an approach will enable both down- and up-link of optics to be handled by existing imaging systems with minimal modification, and will also allow simultaneous imaging and electrical recording of neural activity from the same volume of neural tissue. Furthermore, because MOTEs are both tetherless and insensitive to the scattering of their power-supply and signal-carrying light, they can be used even in situations where motion of an experimental animal, and/or the scattering by intervening tissue disallow multi-site electrical recording or activity imaging. Our preliminary work proves the feasibility of MOTEs, but significant improvements remain to shrink them to the minimally invasive scale (~10m wide) and to enable simultaneous read-out from 100's to 1000's of MOTEs at once.

Public Health Relevance

We will develop free-floating implantable devices on the scale of 10mx10mx200m, able to measure electrophysiological signals, but use light to harvest power, synchronize, and uplink measured data. Appropriate choice of optical wavelengths and intensity, coupled with efficient coding schemes, will allow the system to support simultaneous imaging and electrical recording of neural activity from 100's of sites. This technology will not only enable neurobiological experiments currently unattainable, and will also provide a new, minimally invasive platform for measuring electrical signals deep in live tissue of moving animals.

Agency
National Institute of Health (NIH)
Institute
National Institute of Neurological Disorders and Stroke (NINDS)
Type
Multi-Year Funded Research Project Cooperative Agreement (UF1)
Project #
1UF1NS107687-01
Application #
9588702
Study Section
Special Emphasis Panel (ZNS1)
Program Officer
Langhals, Nick B
Project Start
2018-09-30
Project End
2021-09-29
Budget Start
2018-09-30
Budget End
2021-09-29
Support Year
1
Fiscal Year
2018
Total Cost
Indirect Cost
Name
Cornell University
Department
Engineering (All Types)
Type
Biomed Engr/Col Engr/Engr Sta
DUNS #
872612445
City
Ithaca
State
NY
Country
United States
Zip Code
14850