Patterns (Jul 2021)
A machine learning framework to optimize optic nerve electrical stimulation for vision restoration
Abstract
Summary: Optic nerve electrical stimulation is a promising technique to restore vision in blind subjects. Machine learning methods can be used to select effective stimulation protocols, but they require a model of the stimulated system to generate enough training data. Here, we use a convolutional neural network (CNN) as a model of the ventral visual stream. A genetic algorithm drives the activation of the units in a layer of the CNN representing a cortical region toward a desired pattern, by refining the activation imposed at a layer representing the optic nerve. To simulate the pattern of activation elicited by the sites of an electrode array, a simple point-source model was introduced and its optimization process was investigated for static and dynamic scenes. Psychophysical data confirm that our stimulation evolution framework produces results compatible with natural vision. Machine learning approaches could become a very powerful tool to optimize and personalize neuroprosthetic systems. The bigger picture: Electrical stimulation of the optic nerve can allow the restoration of lost visual functions in an effective and clinically exploitable way. To achieve this goal, it is crucial to develop a suitable approach to target selectively nerve fiber subpopulations that mediate different sensations but share similar locations in the nerve. In the present work, we use a simple computational model of the primate visual system to show that it is possible to optimize the stimulation at the level of the optic nerve to replicate a pattern of activity in a cortical region, producing, at the same time, reliable sensations. This result could produce nerve stimulation patterns that exploit the convergent nature of the visual system to “correct” the representation error introduced at the nerve level. In the long term, this would lead to eliciting naturalistic sensations from non-intuitive protocols that exploit machine learning to overcome the technological limits of nerve interfaces.
