A Wireless Battery-Free Multichannel Neural Recorder

Junseok Chae1, Shiyi Liu1

  • 1Arizona State University

Details

08:30 - 19:30 | Wed 26 Oct | Auditorium Foyer | WePOS.22

Session: Poster Session

08:30 - 19:30 | Wed 26 Oct | Main Auditorium | WePOS.22

Session: Ignite Session 2

Abstract

Background: Electrocorticography (ECoG) is a promising technique to measure neural activities from the brain for the purpose of treating neurological disorder and developing neuro prostheses. Current clinical operation requires the implanted ECoG grid electrodes to be wire connected to external electrical instrument for data storage and processing. This setup severely restrains patient movement, and the transcranial wiring raises the possibility of infection and physical trauma. To address these challenges, numerous wireless ECoG systems have been developed, including application-specific integrated circuit (ASIC) technology1,2. However, these systems require extremely sophisticated circuits operating coordinately, which may result in electronics failure. Moreover, these systems consume power proportional to the number of recording channels, limiting system maximum achievable channel number. Purpose: The purpose of this study is to develop a wireless passive multichannel neural recorder, using microwave backscattering effect, featuring simple circuit structure, high reliability, and low power consumption, independent of the number of channels. Method: The fundamental operating principle relies on microwave backscattering effect. The neural recorder utilizes a custom designed antenna to receive microwave energy from an external 2.35/4.7 GHz dual band antenna. The implant antenna is designed and simulated using FEM softwares and then fabricated on a PDMS substrate with medically-approved titanium foil and silicone adhesive. The microwave hits the implant antenna, propagates to a varactor diode, where the microwave mixes with target neural signal. The mixing product is then re-radiated as backscattering microwaves. The external antenna receives the backscattered signal, and demodulates to recover the original neural signal. The varactors diodes form an array to collect multichannel neural signal, using a time multiplexing method. The array also contains inductor/capacitor time delay circuits. A pulse signal propagates through the time delay network and arrives each circuit node at specific time point, turning on each individual channel. The pulse signal is generated with a second PDMS antenna together with a PMOS. No digital logics / circuits or timing clock exists in this method. This allows the power consumption of multichannel recorder to be independent of the number of channels. Result: A 16 channels prototype was fabricated on a PVDF film and was integrated with a custom made ECoG electrodes, using discrete components. A data acquisition card generated multichannel 50~200 Hz emulated neural signals with amplitude as low as 1mVpp. The wireless system was able to detect all 16 channels signals. The minimum detectable signal of a single channel showed 100μVpp. The operating distance between the implant recorder and external antenna reached upto 5cm. Conclusion: Wireless battery-free multichannel neural recorder demonstrated the capability of recording 16 channels as low as 1mVpp, upto 5cm. The wireless recorder may be a promising tool for future neurological research. Presentation Format: I wish my abstract to be considered for Oral Presentation.