Bionic Eyes and Retinal Prostheses

The idea of restoring sight with electronics sounds like science fiction until a patient who has been blind for decades perceives light, shapes, and movement through a tiny array of electrodes implanted against the retina. Retinal prostheses — commonly called bionic eyes — represent one of the most tangible intersections of neuroscience and biomedical engineering. The field has delivered real, FDA-approved devices, but it has also exposed hard truths about commercial viability, patient support, and the gap between laboratory promise and lived experience.

How Retinal Prostheses Work

A retinal prosthesis bypasses damaged photoreceptors by electrically stimulating the surviving retinal neurons — typically the ganglion cells — that still connect to the brain via the optic nerve. The general architecture follows a consistent pattern across devices:

1. External camera mounted on a pair of glasses captures the visual scene.
2. Video processing unit (worn on the body or integrated into the glasses) converts the image into a pattern of electrical stimulation.
3. Wireless transmitter sends the stimulation data across the sclera to an implanted receiver.
4. Electrode array, positioned epiretinally or subretinally, delivers small electrical pulses to retinal neurons.

The brain interprets those pulses as phosphenes — discrete points of light. With training, recipients learn to associate patterns of phosphenes with objects, doorways, and movement. The resolution depends heavily on electrode count: a 60-electrode array produces a coarse, pixelated percept quite different from the roughly 1 million ganglion cell axons in a healthy human optic nerve (National Eye Institute).

The Argus II: Landmark and Cautionary Tale

Second Sight Medical Products' Argus II remains the most widely recognized retinal prosthesis. The FDA granted it a humanitarian device exemption (HDE) in 2013 for the treatment of severe retinitis pigmentosa (RP) in adults with bare or no light perception (FDA). Approximately 350 patients worldwide received the 60-electrode epiretinal implant.

Clinical results were meaningful but modest. In the Argus II pivotal trial, 89% of implanted subjects performed significantly better on a door-finding task and 56% improved on a direction-of-motion task compared to the device-off condition (ClinicalTrials.gov, NCT00407602). Recipients reported improved orientation and mobility, though none achieved anything close to functional reading vision.

The cautionary dimension arrived when Second Sight entered financial distress and effectively ceased operations in 2020. Patients were left with implanted hardware and no manufacturer to provide ongoing technical support, replacement parts, or software updates. This situation raised urgent bioethics questions about long-term corporate responsibility for implanted neurotechnology — questions that remain largely unanswered by existing regulatory frameworks.

Other Devices and Approaches

The Argus II was not the only entrant. A broader landscape includes:

• Alpha AMS / Alpha IMS (Retina Implant AG, Germany): A subretinal device with 1,600 photodiodes that used ambient light to generate stimulation, eliminating the external camera. The higher electrode density offered theoretically finer resolution. Retina Implant AG also ceased commercial operations, illustrating a recurring business-model challenge.
• PRIMA (Pixium Vision, France): A photovoltaic subretinal implant activated by near-infrared light projected from specialized glasses. Each 2mm-wide chip contains 378 electrodes. A feasibility study reported that subjects with geographic atrophy from dry age-related macular degeneration (AMD) could perceive letters and patterns (Palanker et al., Nature Medicine, 2020). Pixium Vision pursued further trials, though the company also faced financial restructuring.
• Cortical prostheses bypass the eye entirely, stimulating the visual cortex directly. The Orion device (also developed by Second Sight) placed a 60-electrode array on the occipital cortex, potentially serving patients whose optic nerves are damaged. Research-stage cortical implants at institutions including Baylor College of Medicine and the Netherlands Institute for Neuroscience have demonstrated that patterned cortical stimulation can produce recognizable shapes (Chen et al., Science, 2020).

Candidate Selection and Limitations

Retinal prostheses are not universal solutions. Candidacy for epiretinal or subretinal devices requires that the inner retinal circuitry — ganglion cells and bipolar cells — remain at least partially intact. This restricts the primary population to conditions like RP and, potentially, advanced dry AMD. Diseases that destroy the ganglion cells or optic nerve (advanced glaucoma, optic neuritis) fall outside the scope of retinal-level implants.

Functional outcomes remain limited by electrode count, electrode-tissue interface stability, and the brain's ability to learn phosphene-based vision. Even the highest-density implants tested to date produce visual acuity far below the 20/200 threshold for legal blindness.

What Comes Next

Research trajectories aim to close the resolution gap. Approaches include:

• Higher-density electrode arrays — designs targeting 1,000+ electrodes to improve spatial resolution.
• Optogenetic strategies — genetically modifying surviving retinal cells to become light-sensitive, bypassing electronics altogether. GenSight Biologics' GS030 therapy, combining an optogenetic vector with light-amplifying goggles, demonstrated partial visual recovery in a patient with RP (Sahel et al., Nature Medicine, 2021).
• Stem cell–derived photoreceptor transplantation — a biological rather than electronic replacement strategy under investigation at the NEI and multiple academic centers (NEI Strategic Plan).

The commercial failures of early device makers underscore a critical lesson: the engineering problem and the sustainability problem are separate challenges. Building a device that works is necessary but not sufficient; maintaining a support ecosystem for patients with permanent implants requires durable institutional backing, not just startup funding.

Frequently Asked Questions

Who is eligible for a retinal prosthesis?

Candidacy has historically been limited to adults with severe retinitis pigmentosa who retain some inner retinal function. Investigational devices are expanding eligibility to include dry AMD. A thorough evaluation by a retinal specialist and electrophysiological testing are standard prerequisites.

Can bionic eyes restore normal vision?

No device tested to date restores anything resembling normal visual acuity. The best-reported outcomes allow perception of light, shapes, movement, and high-contrast edges — useful for orientation and mobility but not for reading standard print or recognizing faces.

What happened to Argus II patients after Second Sight stopped operations?

Patients with functioning Argus II implants faced uncertainty regarding hardware repairs and software support. Some advocacy groups and academic centers have worked to maintain limited technical assistance, but no formal manufacturer-backed support structure exists.

Are cortical visual prostheses available?

Cortical implants remain investigational. No cortical visual prosthesis has received FDA marketing authorization. Research devices have demonstrated proof-of-concept phosphene perception and simple shape recognition in small cohorts.

References

• FDA Humanitarian Device Exemption — Argus II Retinal Prosthesis System
• National Eye Institute — Retinitis Pigmentosa
• NEI Strategic Plan for Vision Research
• Sahel et al., "Partial Recovery of Visual Function in a Blind Patient after Optogenetic Therapy," Nature Medicine, 2021
• Palanker et al., "Photovoltaic Restoration of Sight with High Visual Acuity," Nature Medicine, 2020
• Chen et al., "Shape Perception via a High-Channel-Count Neuroprosthesis," Science, 2020
• ClinicalTrials.gov — Argus II Retinal Prosthesis System Pivotal Trial (NCT00407602)

The law belongs to the people. Georgia v. Public.Resource.Org, 590 U.S. (2020)