Intracortical neural interface systems
Abstract/Contents
- Abstract
- Acquisition of neuronal electrical activity via chronically implanted electrode arrays has enabled a wide range of advances in electrophysiological experimentation towards basic neuroscience as well as neural prosthetics. These tools are used to explore the function of the central and peripheral nervous systems and make an impact in the way we diagnose, treat, and understand a broad range of neurological ailments such as epilepsy, chronic pain, obsessive compulsive disorder, and chronic neurodegenerative diseases. There is research at Stanford involving the study of motor / premotor cortical activity and its application towards prostheses, enabled in part by the "Hermes" series of mobile neural acquisition systems, which aim to provide high-quality recording of ensemble neuronal activity in freely behaving primates over long periods of time. Physical size and battery-life constraints have thus far limited our capabilities to record only from a subset of the available channels, or to record a compressed version of neural activity in the form of threshold crossings. This work presents the final generation "HermesE" wireless multichannel neural recording system, which overcomes the aforementioned limitations through the design and use of a high channel count, high fidelity neural sensor interface IC that achieves state-of-the-art power consumption and noise performance while providing instrument grade neural recordings for the primary purpose of enabling neuroscience research. Switched-capacitor signal conditioning provides well-controlled frequency response and good channel-to-channel matching, obviating the need for calibration and easing its use by scientists and engineers in experimental settings. A 2.2 uVrms input-referred noise in a 10 kHz bandwidth is achieved, and the conditioned signals are digitized at 31.25 kSa/s by 10-bit SAR ADCs with 60.3 dB SNDR and 42 fJ /conv-step. The sensor interface occupies 25 mm^2 in 0.13 um CMOS and consumes 6.4 mW from 1.2 V while providing simultaneous access to 96 channels of broadband neural data. High data rate wireless communication is enabled by a custom, power efficient pulse ultra-wideband transmitter in 65 nm CMOS using simple OOK modulation. A low voltage FPGA frames and scrambles the data prior to transmission to aid the clock and data recovery process in the receiver. The mobile system consumes a total of 15.1 mW, more than an order of magnitude lower than the previous HermesD system, and at 3x the channel count. The system was validated by recording in vivo neural signals from rhesus macaques implanted with 96-channel Utah Electrode Arrays, successfully demonstrating its viability for a broad set of long duration and untethered experiments that are unachievable with bulky commercial off-the-shelf systems.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2013 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Gao, Hua |
|---|---|
| Associated with | Stanford University, Department of Electrical Engineering. |
| Primary advisor | Meng, Teresa H |
| Thesis advisor | Meng, Teresa H |
| Thesis advisor | Murmann, Boris |
| Thesis advisor | Poon, Ada Shuk Yan |
| Advisor | Murmann, Boris |
| Advisor | Poon, Ada Shuk Yan |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Hua Gao. |
|---|---|
| Note | Submitted to the Department of Electrical Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2013. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2013 by Hua Gao
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...