The Advancement of Artificial Retinas

For individuals experiencing visual impairments, the development of an artificial retina represents a significant hope, and progress towards this goal is continually being made. Recent innovations are exploring a novel and encouraging strategy, utilizing microscopic light-converting dots. Virtual reality technology has played a crucial role in demonstrating the potential viability of this approach.

Photovoltaic Retinal Prostheses: A New Direction

These innovative prosthetic devices originate from the École polytechnique fédérale de Lausanne, where Diego Ghezzi has dedicated several years to researching this concept. The core principle involves converting light directly into electrical signals within the retina.

Early Retinal Prosthetics and Their Limitations

Initial retinal prosthetics were developed decades ago, employing a system where an external camera transmits signals to a microelectrode array implanted in the eye. This array stimulates functioning retinal cells directly. However, this method presents challenges.

A primary concern is the need for a physical wire connecting the external camera to the implant, which is generally undesirable in prosthetic applications. Furthermore, the number of electrodes within the array is limited by their size, resulting in relatively low resolution – typically ranging from dozens to hundreds of “pixels.”

Ghezzi’s Approach: Overcoming Existing Challenges

Ghezzi’s innovative solution utilizes photovoltaic materials, which transform light into electrical current. This process mirrors that of a digital camera, but instead of recording an image, the generated current directly stimulates the retina. This eliminates the need for a power or data cable, as light itself provides both.

The EPFL prosthesis consists of thousands of these tiny photovoltaic dots. These dots would be illuminated by an external device projecting light patterns based on input from a camera. Engineering such a system remains a complex undertaking.

Progress and Refinements Since 2018

Initial research into this approach began in 2018, and significant advancements have been made since then, as detailed in a recent publication.

“We increased the number of pixels from approximately 2,300 to 10,500,” Ghezzi explained. “The dots are now so closely packed that they appear as a continuous film.”

However, when these dots are in direct contact with the retina, the resolution remains limited – roughly 100x100 pixels. The goal isn’t to perfectly replicate human vision, but to provide a functional level of sight.

The Trade-off Between Pixel Size and Current Generation

“It is technically feasible to create smaller and denser pixels,” Ghezzi clarified. “However, the electrical current produced diminishes with decreasing pixel area.”

Increasing the number of pixels presents challenges in maintaining functionality, and there's a risk of adjacent dots stimulating the same retinal network. A balance must be struck between pixel density and image intelligibility. While 10,500 pixels seems substantial, its effectiveness remains unproven.

Utilizing Virtual Reality for Testing

To assess the potential of the implant, the research team turned to virtual reality. Directly testing an experimental retinal implant on humans is impractical, so VR provided a means to evaluate whether the device’s dimensions and resolution would be sufficient for everyday tasks.

VR Simulations: Replicating the Visual Experience

Participants were immersed in dark VR environments containing simulated “phosphors” – pinpricks of light mimicking the expected retinal stimulation. The team varied the number of phosphors, their spatial arrangement, and the duration of their illumination, assessing participants’ ability to recognize objects and letters.

Key Findings: The Importance of Visual Angle

The study revealed that visual angle – the overall size of the image area – was the most critical factor. Even a clear image is difficult to interpret if confined to the center of vision. A wider field of view is preferable, even if it means some loss of overall clarity.

The brain’s visual system effectively interprets edges and motion from limited input, demonstrating the robustness of visual perception.

Moving Towards Human Trials

These findings validate the implant’s theoretical design, allowing the team to proceed towards human trials. While this process will take time, this approach holds significant promise compared to earlier wired prosthetics. Widespread availability is still several years away, but the prospect of a functional retinal implant is an exciting development.