A Safe CMOS Stimulation and Communications Chip for Retinal Implants

Abstract

 

Purpose:

 

This work is related to the efforts of the Boston Retinal Implant Project to develop a sub-retinal prosthesis to restore vision to the blind. We report here on the design and testing in experimental implants of an ASIC CMOS To design, build, test and use in experimental implants a single microchip ASIC CMOS (application specific integrated circuit complementary metal-oxide semiconductor) wireless control and communication system that is flexible, safe, extremely low-power and able to drive at least 256 electrodes.

 

Methods:

 

Bill Ellersick, CEO of Analog Circuit Works, headed the design effort, and MOSIS generously gave us access to the IBM 0.18 micron fabrication process at no cost. This fabrication process has both standard 1.8 volt CMOS FET’s (complementary metal-oxide semiconductor field-effect transistors) as well as 20 V. LDDMOS (lightly doped drain) FET’s. The system we designed receives (FSK) frequency-shift keyed inputs at 6.78 MHz from an implanted secondary coil (see figure below) and transmits data back through a lower-bandwidth channel using load-shift keying. Safety features in the design include implant-based electrode voltage monitoring, stimulus charge limits, error checking on data transmission to the implant, and comprehensive self-test and performance monitoring. Each stimulus cycle is initiated by a transmitted word with a full 32-bit error check code, reducing the probability of accepting an erroneous message to about 10¹⁰.

 

Results:

 

We have fabricated three versions of the chip to date. The first two versions had a number of problems that were isolated and fixed, and the third version, submitted November 28,2011,is expected to work well enough to use in animal trials for the FDA. Operation of electrode drivers to within 5% of commanded stimulus levels, of the radio receiver in the presence of reverse telemetry with the full range of modulation from 10% to 80%, and of the on chip logic have been verified. The chip works well enough for use in animal trials for the FDA, and a follow-on version has been designed that should be usable in human trials.

 

Conclusions:

 

We have built and tested a low-power, safe, flexible chip for the Boston retinal prosthesis that fits in our 11mm diameter hermetic package andthat can drive 256+ electrodes.  

 

Keywords: retina