Pixel Characterization of a Protein-Based Retinal Implant Using a Microfabricated Sensor Array

Retinal degenerative diseases are characterized by the loss of photoreceptor cells within the retina and affect 30-50 million people worldwide. Despite the availability of treatments that slow the progression of degeneration, affected patients will go blind. Thus, there is a significant need for a prosthetic that is capable of restoring functional vision for these patients. The protein-based retinal implant offers a high-resolution option for replacing the function of diseased photoreceptor cells by interfacing with the underlying retinal tissue, stimulating the remaining neural network, and transmitting this signal to the brain. The retinal implant uses the photoactive protein, bacteriorhodopsin, to generate an ion gradient in the subretinal space that is capable of activating the remaining bipolar and ganglion cells within the retina. Bacteriorhodopsin can also be photochemically driven to an active (bR) or inactive (Q) state, and we aim to exploit this photochemistry to mediate the activity of pixels within the retinal implant. In this study, we made use of a novel retinomorphic foveated image sensor to characterize the formation of active and inactive pixels within a protein-based retinal implant, and have measured a significant difference between the output frequencies associated with the bR and Q states.