The goal of this research project is to develop an intracortical visual prosthesis (ICVP) that will compensate for blindness by stimulating the visual centers within the brain. Cortical stimulation has the potential to be used for a large fraction of the population with blindness because the approach does not require an intact retina or optic nerve. A large percentage of individuals with blindness have damaged eyes or optic nerves, and a brain-based approach is their only prosthetic solution.
The current approach for the ICVP consists of multiple stimulation modules implanted into the brain’s occipital lobe, controlled and powered, transcutaneously, by an extracoporeal transmitter unit placed outside of the head. These stimulation modules are about a tenth the size of a penny, and each contains an array of 16 electrodes that serve to transmit information to and from the brain.
Visual prostheses function by sending artificial images directly into the brain through stimulator control signals, via a transcutaneous link. In order to send these signals, the research team at IIT has developed subminiature autonomous wireless stimulator modules. The stimulator module derives its power and communications, wirelessly, from the external transmitter. This approach has numerous advantages over those that use a central implanted package and a system of interconnecting cables which cross the dura and tether the electrode arrays.
Research for the development Technology development for the Intracortical Visual Prosthesis (ICVP) has spanned several decades, and dates back to the Neural Prosthesis program at the National Institutes of Health (NIH).
Although it has long been presumed that the ICVP was technically feasible, actual implementation of a multi-channel system in a human has remained out of grasp due to limitations in the technology that could assure safe stimulation of the visual cortex. After assuming leadership of the project in 2000, our team embarked upon a path to finalize technology so that a multichannel ICVP could be tested in humans.
To accomplish this, engineering developments were required in electrode fabrication, wireless stimulators, and physical configuration of a system that might deploy up to 1000 intracortical electrodes. This project has now reached the point at which the requisite technology is available.
Communicating artificial images to the brain requires an understanding of how to exploit the artificial cortical neural interface. While significant knowledge is available about normal visual processing, much less in known about how to use an artificial cortical neural interface to communicate visual information that can be decoded by the brain. Learning how to encode the artificial image information so that the natural tuning of the brain can enhance the sparse visual information will be essential to the ICVP being used as a visual sensory replacement.
Of the 37 million people who suffer from blindness worldwide, a staggering 90% live in underdeveloped countries; in many of these, the Gross National Income (GNI) per capita amounts to less than $2000 (US). In the United States, the GNI in 2012 was $52,610 (US) per capita – this means that the average person of an underdeveloped has an income 26 times less than the average North American. This economic disparity poses a great challenge to the widespread dispersal of visual prosthesis – the fact that the areas of the world that need this technology the most also happen to be the most economically strained is a significant hurdle that will need to be crossed in time. One potential advantage of the IIT ICVP system is its modularity. Using fewer implantable stimulators, albeit with more sparse visual information, could make the ICVP economically viable, worldwide.
Presently our team is preparing the ICVP for clinical testing, and configuring the ICVP for testing in humans requires multiparametric system assessment, and estimation of sensory benefits to a recipient volunteer. To minimize the surgical risks, we plan to limit implantation of the electrodes to the dorso-lateral surface of the occipital lobe. Stimulation of the visual cortex produces dot-like perceptions called phosphenes. Although the lateral surface is not optimal from the standpoint of spatial visiotopic mapping, we estimate that the ICVP system will produce a somewhat randomly positioned and sized set of phosphenes that can be utilized by the volunteer as relevant visual perceptions. Considering that our 16-electrode array design has a diameter of approximately 5 mm, we estimate that we will be able to place at least 600 to 650 electrodes in this lateral area.
It may be desirable to approach the surgery in two phases. In phase one, a small number of 16-electrode modules would be implanted for a period of 2-4 months to demonstrate the stability of the electrode-neural interface, and answer critical questions related to electrode sizing and stimulation parameters. In phase two, up to sixty modules might be implanted in the same volunteer. However, even with 1000 electrodes, it is uncertain whether or not there would be a sufficient density of phosphenes to rely upon a bit-map approach to vision. While only a portion of the primary visual cortex (V1) would have implanted electrodes, preliminary studies suggest that we can compensate for this lack of spatial coverage through user scanning, allowing the user to perform at least elementary visual tasks such as navigation and object recognition.