What do visual prosthesis users see, and why? Clinical studies have shown that the vision provided by current devices differs substantially from normal sight.

Rather than predicting perceptual distortions, one needs to solve the inverse problem: What is the best stimulus to generate a desired visual percept?

pulse2percept is an open-source Python simulation framework used to predict the perceptual experience of retinal prosthesis patients across a wide range of implant configurations.

We show that a neurologically-inspired decoding of CNN activations produces qualitatively accurate phosphenes, comparable to phosphenes reported by real patients.

What is the required stimulus to produce a desired percept? Here we frame this as an end-to-end optimization problem, where a deep neural network encoder is trained to invert a known, fixed forward model that approximates the underlying biological system.

We present a phenomenological model that predicts phosphene appearance as a function of stimulus amplitude, frequency, and pulse duration.

We propose HBA-U-Net: a U-Net backbone with hierarchical bottleneck attention to highlight retinal abnormalities that may be important for fovea and optic disc segmentation in the degenerated retina.