Towards a Smart Bionic Eye: AI-Powered Artificial Vision for the Treatment of Incurable Blindness How can we return a functional form of sight to people who are living with incurable blindness? Despite recent advances in gene and stem cell therapies are showing great promise, there are no effective treatments for many people blinded by severe degeneration or damage to the retina, the optic nerve, or cortex. In such cases, an electronic visual prosthesis (“bionic eye”) may be the only option. However, the quality of current prosthetic vision is still rudimentary and does not differ much across different device technologies. A major outstanding challenge is translating electrode stimulation into a code that the brain can understand. Rather than aiming to one day restore natural vision with visual prostheses, we might be better off thinking about how to create practical and useful artificial vision. Specifically, a visual prosthesis has the potential to provide visual augmentations through the means of artificial intelligence (AI) based scene understanding, tailored to specific real-world tasks that are known to affect the quality of life of people who are blind (e.g., face recognition, outdoor navigation, self-care). The goal of this proposal is thus to address fundamental questions at the intersection of neuroscience, computer science, and human-computer interaction that will enable the development of a Smart Bionic Eye; that is, a visual neuroprosthesis that functions as an AI-powered visual aid by providing visual augmentations to support specific everyday tasks. To enable such a technology, we first need to 1) understand how visual prostheses interact with the human visual system to shape perception, 2) identify visual augmentation strategies that best support specific real-world tasks, and 3) develop a prototyping system that allows us to validate as well as iteratively improve upon our augmentation strategies with the bionic eye recipient in the loop. This work will further our understanding of how brain stimulation leads to perception, and the insights gained from identifying optimal visual augmentation strategies may be broadly applicable to different visual aids and sensory substitution devices, therefore potentially benefitting both people who are blind and people with low vision. Last but not least, the ability of a visual prosthesis to support everyday tasks might make the difference between abandoned technology and a widely adopted next-generation neuroprosthetic device.

Project Narrative This project addresses fundamental questions at the intersection of computational neuroscience, artificial intelligence (AI), and human-computer interaction that will enable the development of an AI-powered visual prosthesis for people who are blind. Furthermore, this work will elucidate the perceptual effects of electrical stimulation of the retina and visual cortex as well as offer rigorous computational and data-driven models that will be broadly applicable to a variety of visual aids and neural stimulation technologies.