Can bionic eyes restore vision? Breakthroughs, challenges, and future frontiers in ophthalmology – A comprehensive review

Abstract

Bionic eye technology aims to restore partial vision in individuals with severe visual impairment by artificially stimulating the visual system. Primary approaches involve the use of retinal implants, optic nerve stimulation, and cortical visual prostheses. Retinal implants have shown the most clinical progress, with devices such as Argus II and Alpha AMS demonstrating the ability to restore basic visual function in patients with retinitis pigmentosa. These implants use electrode arrays to stimulate the remaining retinal neurons, thereby enablingusers to perceive light, detect motion, and recognize large objects. However, the current technologies have limitations in terms of resolution, field of view, and long-term stability. Ongoing research has focused on developing high-density electrode arrays, improving biocompatibility, and enhancing signal processing. Clinical trials have shown that bionic eyes can provide meaningful improvements in visual function and quality of life in some patients, although outcomes vary. Although regulatory approvals have been achieved for certain devices, accessibility remains limited because of high costs. Future research directions include optogenetic approaches, development of wireless systems for power and data transmission, and integration of artificial intelligenceto optimize device performance. Overall, bionic-eye technology shows promise for restoring partial vision. However, significant challenges remain in achieving high-resolution, naturalistic vision.

Highlights

Bionic eye technology has demonstrated the ability to restore partial vision in individuals with severe visual impairment, particularly in conditions such as retinitis pigmentosa and age-related macular degeneration.

Clinical trials of devices such as Argus II and PRIMA have shown improvements in light perception, motion detection, and object recognition in some patients, although the outcomes have varied.

Current limitations include low resolution, narrow field of view, and challenges with long-term biocompatibility and stability of implants.

Advanced materials, artificial intelligence integration, and hybrid biological-electronic approaches show promise for enhancing device performance and patient outcomes.

Although regulatory approvals have been achieved for certain devices, high costs and limited accessibility remain significant barriers to their widespread adoption.

Introduction

Bionic eye technology constitutes a significant advancement in ophthalmology, offering hope to millions of individuals experiencing blindness and severe visual impairment. The term “bionic eye” denotes an artificial visual prosthesis designed to restore partial vision by interfacing with the nervous system. Unlike traditional corrective methods such as eyeglasses or contact lenses, bionic eyes function by bypassing damaged or nonfunctional photoreceptor cells and directly stimulating the remaining viable neurons in the retina or visual cortex [1]. The importance of bionic eye technology lies in its potential to significantly enhance the quality of life for those who have lost sight due to various ocular conditions. With the advent of these technologies, individuals who were previously unable to perceive their surroundings can regain some level of visual function, enabling them to navigate the world with increased independence and confidence [2].

Visual impairment and blindness are significant global health issues impacting millions across various demographics. The World Health Organization (WHO) reports that over 2.2 billion individuals worldwide suffer from some degree of visual impairment, with around 43 million being classified as blind [3]. The primary causes of blindness include age-related macular degeneration (AMD), glaucoma, diabetic retinopathy, and RP [4]. These conditions often lead to irreversible damage to the photoreceptors, rendering traditional treatments ineffective in restoring sight. The prevalence of blindness varies by region, with low- and middle-income countries facing greater challenges due to limited access to healthcare, early diagnostic services, and treatment options [5]. In contrast, high-income countries are seeing advancements in medical interventions, such as retinal implants and gene therapies, providing new hope for those affected. The economic and social impact of visual impairment is substantial, affecting both individuals and the healthcare system. Consequently, the development of innovative solutions, like bionics, is essential to tackle this escalating public health challenge.

A critical evaluation of the literature highlights the significant differences among the available bionic eye technologies. While epiretinal prostheses such as the Argus II have demonstrated moderate success in restoring basic visual functions, their resolution and limited visual field restrict practical utility [6]. Subretinal implants, notably Alpha AMS and PRIMA, provide improved visual acuity through direct stimulation of inner retinal layers but face issues related to long-term electrode stability and immune responses [7]. Suprachoroidal implants, like the Australian prosthesis, have emerged as a favorable alternative with fewer biocompatibility issues, although electrode density and visual resolution remain areas needing enhancement [8]. Cortical implants such as Orion bypass the retina completely, showing promise for cases of extensive retinal or optic nerve damage; however, their invasive nature and potential risks of CNS reactions require rigorous long-term assessment [9]. Continued interdisciplinary research focusing on electrode material improvements, biocompatibility enhancements, and patient-specific outcomes is critical for optimizing future device performance [10].

This review aims to provide a comprehensive analysis of the advancements in bionic eye technology, covering the latest developments, current challenges, and future prospects in the field. Recent comprehensive reviews have highlighted both the engineering advances and remaining challenges in retinal prosthesis development, emphasizing the need for continued research to overcome technical limitations and improve patient outcomes [11, 12]. The key objectives include exploring the evolution of bionic eye technology, examining the different types of bionic eye implants analyzing the current clinical applications and outcomes, identifying challenges and limitations, forecasting future developments, and assessing the impact of socioeconomic factors. By addressing these objectives, this review seeks to contribute to the ongoing discourse in ophthalmology and biomedical engineering by offering insights into how bionic eye technology shapes the future of vision restoration.

Structure of the human eye and visual pathway

The human eye functions like a camera, converting light into electrical signals that the brain interprets as images. The eye’s outer layer contains the cornea and sclera. The cornea is a transparent dome that bends light, while the sclera provides structural support [13]. The anterior chamber’s aqueous humor maintains pressure. The iris controls light through pupil adjustment. The lens focuses light onto the retina, changing shape to focus on objects at different distances [14]. Figure 1 illustrates the eye’s anatomy.

Fig. 1.

Human Eye Anatomy

The retina, a light-sensitive layer at the back of the eye, contains photoreceptor cells—rods and cones—that convert light into electrical signals. Rods detect motion and function in low light, while cones enable color vision and detail. The macula contains dense cones for high-acuity vision, with the fovea being the point of sharpest vision [15]. These signals are processed by retinal neurons before ganglion cells’ axons form the optic nerve, which exits at the optic disc creating a blind spot [16]. At the optic chiasm, nerve fibers partially cross, with nasal retina fibers crossing sides while temporal retina fibers remain unchanged [17]. The optic tract carries information to the lateral geniculate nucleus, which relays signals to the primary visual cortex (V1). V1 processes basic visual information before relaying it to higher-order areas for coherent visual perception [18].

Role of the retina, optic nerve, and brain in vision

Natural vision depends on intact neural transmission from retina to brain through the optic nerve. In conditions like glaucoma and optic neuropathy, this transmission can be disrupted, causing visual impairment. Bionic eyes attempt to restore partial vision by directly stimulating visual neural pathways and bypassing damaged natural pathways. However, current implants face limitations in resolution compared to natural vision [19]. The brain plays a central role in processing visual information. The primary visual cortex (V1) processes basic visual elements like edges, shapes, and motion. Visual information is then relayed to extra-striate areas V2, V3, V4, and V5, each specializing in different aspects of vision. V4 processes color, while V5 handles motion perception. The dorsal and ventral streams (“where” and “what” pathways) further process visual information. The dorsal stream, projecting to the parietal lobe, guides spatial awareness and movements, while the ventral stream, projecting to the temporal lobe, handles object recognition [20]. The integration of visual information involves feedback connections between visual areas, creating a coherent visual world despite fragmented retinal information. The brain fills the blind spot from the optic disc to maintain a continuous visual field [21].

Common causes of blindness and visual impairment

Blindness and visual impairment can result from conditions affecting the eye, optic nerve, or brain. Common causes include retinitis pigmentosa, AMD, and optic neuropathy. These conditions often lead to irreversible vision loss, highlighting the need for bionic eye technology. RP is a group of inherited retinal disorders causing progressive degeneration of photoreceptor cells. The condition begins with night blindness and peripheral vision loss, leading to tunnel vision and potentially complete blindness. RP is caused by mutations affecting photoreceptor cell function. While there is no cure, bionic eye technologies offer hope for restoring vision by stimulating remaining retinal neurons [22]. AMD is a leading cause of vision loss in older adults, affecting the macula. AMD exists in two forms: dry and wet. Dry AMD involves drusen accumulation beneath the retina and retinal pigment epithelium (RPE) cell atrophy. Wet AMD involves abnormal blood vessel growth beneath the retina, causing rapid vision loss. While anti-VEGF injections can slow wet AMD progression, there is no cure for dry AMD. Bionic eye technologies aim to restore central vision in advanced AMD cases [23]. Optic neuropathy refers to optic nerve damage from causes including glaucoma, ischemic optic neuropathy, and inflammatory conditions. Glaucoma involves progressive optic nerve damage due to elevated intraocular pressure. Ischemic optic neuropathy occurs when blood flow to the optic nerve is compromised. While treatments focus on underlying causes, bionic eye technologies that stimulate the visual cortex are being explored [24]. Other causes of blindness and visual impairment include diabetic retinopathy, cataracts, and traumatic eye injury. Diabetic retinopathy, a complication of diabetes, results from damage to blood vessels in the retina, leading to vision loss if left untreated. Cataracts, characterized by the clouding of the lens, are a common cause of reversible blindness, particularly in older adults. Traumatic eye injuries, such as those caused by accidents or penetrating wounds, can damage the eye structures and lead to permanent vision loss [25].

Evolution of bionic eye technology

The advancement of bionic eye technology signifies an extraordinary intersection of neuroscience, engineering, and medicine. In recent decades, there has been notable progress in creating devices capable of partially restoring vision in those with severe visual impairments. The idea of using artificial means to restore sight dates back to the 18th century, when scientists began to grasp the electrical nature of nerve impulses. Luigi Galvani’s experiments with animal electricity in the late 1700s laid the foundation for the concept that electrical stimulation could activate neural tissue. However, it wasn’t until the 20th century that serious attempts were made to apply this knowledge to practical vision restoration [26]. One of the earliest efforts in artificial vision was undertaken in the 1960s by British physiologist Giles S. Brindley. Brindley experimented with electrical stimulation of the visual cortex in blind patients by implanting electrodes on the brain’s surface. Although his work showed that electrical stimulation could produce phosphenes (perceived spots of light), the technology was rudimentary and the results were inconsistent. Nonetheless, Brindley’s experiments proved that artificial vision is feasible [27]. During the 1960s and 1970s, researchers investigated retinal stimulation as an alternative to cortical stimulation. The retina, being the initial stage of visual processing, offers a more focused approach for vision restoration. Early animal model experiments involved placing electrodes on the retina’s surface and stimulating the remaining retinal neurons. These studies demonstrated that electrical stimulation of the retina could trigger neural responses, paving the way for more sophisticated retinal prostheses [28]. As illustrated in Fig. 2, bionic eyes bypass damaged photoreceptors by utilizing external cameras and electrode arrays to stimulate the retina and send signals to the brain.

Fig. 2.

Mechanism of bionic eye

Milestones in artificial vision development

The evolution of bionic eye technology has been characterized by several pivotal achievements, each marking a substantial advancement in vision restoration capabilities. These achievements encompass the development of effective retinal implants, progress in electrode design and materials, and the successful clinical testing of bionic eye devices. The Argus I Retinal Prosthesis (2002): Created by Second Sight Medical Products, the Argus I was among the initial retinal prostheses to be tested in clinical trials. This device included an external camera mounted on glasses, a video processing unit, and an implanted electrode array connected to the retina. The Argus I proved that electrically stimulating the retina could enable blind individuals to perceive light and shapes, representing a significant breakthrough in artificial vision [29]. The Argus II Retinal Prosthesis (2011): Launched in 2011 by Second Sight Medical Products, the Argus II retinal prosthesis made considerable advancements over earlier models like the Argus I. Approved for patients with RP , it featured an enhanced electrode array with 60 electrodes, improving visual resolution and functionality [26]. The Argus II system captures visual data through an external camera, processes it with a video processing unit, and wirelessly transmits signals to directly stimulate the retina. Clinical results showed improved abilities to recognize basic shapes, detect motion, and navigate independently, although issues such as device degradation and limited resolution eventually led to its commercial discontinuation in 2019 [22]. The Alpha-IMS Subretinal Implant (2013): Developed by Retina Implant AG, the Alpha-IMS was a subretinal implant that directly stimulated the retina’s bipolar cells. Unlike epiretinal implants like the Argus II, the Alpha-IMS did not require an external camera, as it utilized a photodiode array to capture light and generate electrical signals. Clinical trials showed that the Alpha-IMS could restore some level of functional vision, including the ability to recognize faces and read large letters [30]. Cortical Visual Prostheses: While most bionic eye technologies focus on retinal stimulation, researchers have also investigated cortical visual prostheses that directly stimulate the visual cortex. The Orion Visual Cortical Prosthesis System, developed by Second Sight, is a recent example. This system completely bypasses the retina and optic nerve, making it suitable for patients with damage to these structures. Early trials have shown promising results, with patients reporting the perception of light and basic shapes [24]. Advances in Materials and Miniaturization: Recent advancements in materials science and microfabrication have enabled the development of more sophisticated electrode arrays and implantable devices. For example, flexible electrodes made from biocompatible materials such as graphene and conductive polymers have improved the safety and efficacy of retinal implants. Additionally, miniaturization of components has allowed for less invasive surgical procedures and more comfortable devices for patients [31].

Comparison with other assistive technologies

The advancement of bionic eye technology can be likened to other assistive devices, such as cochlear implants in the field of audiology. Both technologies strive to restore sensory functions by connecting with the nervous system and have seen considerable progress over recent decades. Cochlear implants, which enable individuals with profound hearing loss to perceive sound, are similar to bionic eyes in their use of electrical stimulation on neural tissues and their combination of external and internal parts. Nonetheless, there are significant differences in the complexity of the sensory systems they aim to restore. Vision requires the processing of spatial, temporal, and color information, making it more intricate than hearing, which mainly involves temporal processing. Consequently, bionic eye technology encounters more challenges in achieving high-resolution vision compared to the relatively high-quality sound provided by cochlear implants. Furthermore, while cochlear implants are applicable to a broad range of hearing loss causes, bionic eyes are currently restricted to specific conditions like RP and AMD.The success of cochlear implants provides valuable lessons for the bionic eye sector, especially in areas such as patient selection, post-implantation rehabilitation, and ongoing device enhancements [32–34].

Types of bionic eye technologies

Bionic eye technologies have evolved significantly over the past few decades, with retinal implants emerging as one of the most promising approaches for restoring vision in individuals with severe visual impairment. Retinal implants are designed to bypass damaged photoreceptor cells and directly stimulate the remaining retinal neurons, thereby creating artificial vision. Comprehensive Comparative Overview of Bionic Eye Technologies is presented in Table 1.

Table 1.

Comprehensive comparative overview of bionic eye technologies

Technology	Implant Type	Mechanism	Clinical Efficacy (Key outcomes)	Long-Term Stability Issues	Biocompatibility/Adverse Events	Current Regulatory Status
Argus II	Epiretinal	Direct ganglion cell stimulation via external camera	Basic shape recognition, motion detection, limited resolution.	Electrode impedance increase, device degradation. [35]	Glial scarring, immune reactions [36]	U.S. Food and Drug Administration(FDA)/Conformite Europeenne(CE) approved (discontinued)
Alpha AMS	Subretinal	Photodiode array stimulates bipolar cells	Moderate vision recovery, shapes, letter reading [37]	Encapsulation, electrode corrosion, stability degradation [38]	Foreign body reaction, inflammation [39]	CE marked (discontinued)
PRIMA	Photovoltaic Subretinal	Wireless photovoltaic silicon diodes stimulate bipolar cells	Significant improvement (up to 23 Early Treatment Diabetic Retinopathy Study (ETDRS) letters), face/letter recognition [40]	Long-term stability promising [41]	Mild adverse events [42]	Ongoing pivotal trials
Australian Suprachoroidal	Suprachoroidal	Electrodes placed between sclera-choroid, minimal retinal trauma	Good object localization, stable electrode functionality [23]	Excellent stability (~ 97% electrodes stable over 2.7 yrs) [8]	Minimal adverse reactions [43]	FDA Breakthrough status, ongoing trials
Orion	Cortical	Direct visual cortex stimulation bypassing visual pathway	Perception of basic shapes and motion, suitable for optic nerve damage [9]	Invasive, limited long-term clinical data	CNS-related immune responses, potential intracranial complications [24]	Early trials, investigational

Retinal implants

Retinal implants are electronic devices that aim to restore vision by stimulating the retinal neurons in individuals with degenerative retinal diseases, such as RP and AMD. These implants are broadly categorized into two types: subretinal and epiretinal. Each type has a unique design, mechanism of action, and clinical applications.

Subretinal implants

Subretinal implants are placed between the RPE and the photoreceptor layer, directly stimulating bipolar cells in the inner retina. One of the most well-known subretinal implants is the Alpha AMS, developed by Retina Implant AG. The Alpha IMS required an external power supply system for operation [35, 36]. The light-sensitive CMOS chip was powered via an inductive link to hermetically housed electronics in a subdermal ceramic housing, with a silicone power cable routed under the temporal muscle connecting the intraocular chip to the external power supply [35]. The required energy to power the 1,500 individual micro photodiode-amplifier-electrode elements reached the chip via a polyimide foil in the eye and the silicone power cable, which runs under the temporal muscle to a ceramic housing behind the ear [35]. This ceramic housing contained a magnet, coil for inductive coupling, and electric circuits to generate the desired currents. The external device allowed patients to adjust control signals, stimulation frequency, pulse duration, gain, and sensitivity to adapt stimulation to actual light conditions [37, 38]. Patients could control stimulation frequency and pulse duration via control software, while gain and sensitivity could be controlled by the patient to adapt the stimulation to the actual light conditions [38]. The voltages that regulate the sensitivity (Vgl) and gain (Vbias) of the amplifiers), which are related to the perception of contrast and brightness respectively, were adjusted manually on a handheld power supply device. The external power supply’s external coil attached trans dermally and magnetically to the implanted ceramic housing behind the ear [35]. The Alpha AMS is a wireless device that consists of a micro photodiode array (MPDA) containing approximately 1,600 electrodes. MPDA captures light entering the eye and converts it into electrical signals, which are then delivered to bipolar cells. This design eliminates the need for an external camera, because the implant itself acts as a light sensor [30].

PRIMA subretinal implant

The PRIMA system, developed by Pixium Vision and now Science Corp., is a photovoltaic subretinal implant designed for patients with geographic atrophy due to AMD . PRIMA uses an array of miniature silicon photodiodes to directly convert incident light into electrical stimulation, eliminating the need for external power sources. Recent clinical trials, including the PRIMAvera study, have shown significant improvements in visual acuity, with some patients regaining the ability to read letter sequences and recognize faces. PRIMA is currently under review for CE marking and represents a promising advancement in subretinal implant technology [39–41]. The results showed that the device could restore some degree of functional vision, including the ability to detect light, recognize shapes, and even read large letters. Patients have reported improvements in their daily activities, such as navigating indoor environments and identifying objects. However, the resolution provided by Alpha AMS is limited and the quality of vision is far from normal. Additionally, the long-term stability of the device remains a concern, as the implant performance can degrade over time owing to factors such as tissue encapsulation and electrode corrosion [42].

Epiretinal implants

Epiretinal implants are placed on the surface of the retina, directly stimulating ganglion cells. Argus II, developed by the Second Sight Medical Products, is one of the most widely used epiretinal implants. The Argus II system consists of three main components: an external camera mounted on glass, a video processing unit (VPU), and an implanted electrode array with 60 electrodes. The camera captures visual information that is processed by the VPU and transmitted wirelessly to the electrode array. The electrodes then stimulate the ganglion cells, creating patterns of light perception known as phosphenes [29].

Mechanism of action and effectiveness

The mechanism of action of the retinal implants depends on their placement and design. Subretinal implants, such as Alpha AMS, rely on the natural optics of the eye to capture light and stimulate bipolar cells. This approach mimics the function of photoreceptors and provides a natural stimulation pattern. However, the effectiveness of subretinal implants is limited by the density of the electrode array and the health of the remaining retinal neurons. In patients with advanced retinal degeneration, bipolar cells may also be damaged, thereby reducing the efficacy of the implant [28].

Epiretinal implants, such as Argus II, bypass the photoreceptors and bipolar cells entirely, directly stimulating ganglion cells. This approach allows for greater flexibility in image processing, as the external camera and VPU can enhance the visual information before it is transmitted to the retina. However, the effectiveness of epiretinal implants is limited by the number of electrodes and the ability of the brain to interpret generated phosphenes. The resolution of the Argus II, for example, is significantly lower than that of natural vision, and patients often report difficulty in recognizing complex shapes and faces [31].

Both subretinal and epiretinal implants have shown promise in restoring limited vision in patients with retinal degenerative disease. However, the quality of vision provided by these devices is far from normal, and their effectiveness varies depending on the extent of retinal damage and patient’s ability to adapt to artificial vision. Ongoing research aims to improve the resolution, stability, and usability of retinal implants with the goal of providing more functional and naturalistic vision to patients [24].

Advantages and limitations

Retinal implants offer several advantages over other approaches to vision restoration. One of their primary advantages is their ability to target the retina, which is the first stage of visual processing. By stimulating the remaining retinal neurons, retinal implants can provide a more natural and localized form of artificial vision compared to cortical implants, which stimulate the visual cortex directly. Additionally, retinal implants are less invasive than cortical implants because they do not require brain surgery [43].

Subretinal implants, such as Alpha AMS, have the advantage of using the natural optics of the eye, eliminating the need for an external camera. This design allows a more compact and user-friendly device. However, subretinal implants are limited by the density of the electrode array and the health of the remaining retinal neurons. In patients with advanced retinal degeneration, bipolar cells may also be damaged, thereby reducing the efficacy of the implant [30].

Epiretinal implants, such as Argus II, offer the advantage of greater flexibility in image processing, as the external camera and VPU can enhance visual information before they are transmitted to the retina. This approach allows for more sophisticated image-processing algorithms that can improve the quality of artificial vision. However, epiretinal implants are limited by the number of electrodes and the ability of the brain to interpret generated phosphenes. The resolution of the Argus II, for example, is significantly lower than that of natural vision, and patients often report difficulty in recognizing complex shapes and faces [44].

Despite these limitations, retinal implants represent a significant advancement in bionic eye technology. They offer hope to patients with retinal degenerative diseases, providing them with the ability to perceive light, recognize objects, and navigate their environment. Ongoing research aims to address the limitations of current retinal implants with the goal of developing devices that can provide higher resolution, more naturalistic vision, and greater long-term stability [34].

Lateral geniculate nucleus (LGN) stimulation

LGN stimulation represents an alternative approach for vision restoration, targeting the thalamic relay center that processes visual information before it reaches the visual cortex. Experimental studies in animals and limited human research have demonstrated that electrical stimulation of the LGN can elicit patterned visual perception, potentially bypassing both retinal and optic nerve damage. While this approach remains in early preclinical and clinical exploration, it offers a promising avenue for patients with extensive damage to the visual pathway [45].

Advanced retinal interface technologies

Suprachoroidal retinal prostheses

Suprachoroidal retinal prostheses represent a significant advancement in the field of vision restoration, particularly for individuals with profound vision loss due to inherited retinal diseases, such as RP. Unlike epiretinal and subretinal implants, suprachoroidal devices are surgically positioned in the potential space between the sclera and the choroid, which offers several clinical and surgical advantages. This anatomical location allows for a less invasive implantation procedure, reduces the risk of direct retinal trauma, and provides a more stable long-term implant position, as confirmed by optical coherence tomography in clinical trials [23].

The most notable clinical development in this area is Australia’s second-generation 44-channel suprachoroidal retinal prosthesis, which was developed through collaborative efforts at the Bionics Institute and Centre for Eye Research Australia. In a prospective, single-arm clinical trial, four participants with advanced RP and bare-light perception received the implant and were followed for up to 2.7 years [23]. The results demonstrated that the device maintained stable positioning under the macula with minimal movement and no serious device-related adverse events. Importantly, 97% of electrodes remained functional throughout the study period, indicating robust device durability and biocompatibility [46, 47].

Functionally, all participants experienced significant improvements in their ability to perform orientation and mobility tasks such as locating objects, identifying movement, and navigating unfamiliar environments. Screen-based localization accuracy, motion discrimination, and spatial discrimination were all significantly enhanced when the device was active compared to when it was off. These improvements translated into greater independence and confidence in daily life, as confirmed by both objective assessments and self-reported outcomes [23, 47]. The implant also had a positive impact on emotional well-being, with participants reporting increased willingness to explore new environments and interact socially [46, 47]. These findings underscore the device’s potential not only to restore functional vision but also to enhance the quality of life of individuals living with severe vision impairment.

The success of this trial has paved the way for larger, multi-center studies aimed at regulatory approval in the United States and Australia. The suprachoroidal approach is now recognized as a clinically viable and safe alternative for patients with outer retinal degeneration, with ongoing research focused on increasing electrode density, refining stimulation algorithms, and expanding indications to other retinal diseases [23, 46, 47].

Photovoltaic retinal implants

Photovoltaic retinal implants, such as the PRIMA system, are at the forefront of next-generation wireless vision restoration technologies. Unlike traditional retinal prostheses that require trans-scleral cabling or external power sources, photovoltaic implants are composed of arrays of miniature silicon photodiodes that directly convert incident light into electrical stimulation of the inner retinal neurons. This fully wireless design reduces surgical complexity, minimizes the risk of infection, and enhances patient comfort and device longevity [39, 48]. A significant breakthrough in photovoltaic retinal prosthesis technology was demonstrated by the POLYRETINA system, a wide-field, high-density photovoltaic epiretinal prosthesis containing 10,498 physically and functionally independent photovoltaic pixels with 120 μm pitch. Single-pixel illumination reproducibly induced network-mediated responses from retinal ganglion cells at safe irradiance levels, enabling high spatial resolution equivalent to the pixel pitch. This approach represents a major advancement in achieving wide retinal coverage while maintaining high-resolution stimulation capabilities for artificial vision applications [10].

The PRIMA system, developed by Pixium Vision and now Science Corp, has been evaluated in a series of clinical trials targeting patients with GA secondary to AMD—a population for whom no restorative treatments previously existed. In the PRIMAvera pivotal trial (NCT04676854), 38 patients with advanced dry AMD received the subretinal PRIMA implant and were followed for at least 12 months [49, 50]. The device is paired with smart glasses equipped with a camera and a projection system that transmits processed images onto the subretinal implant, which then stimulates the remaining retinal neurons.

Clinical outcomes have been promising: patients demonstrated a mean improvement of 23 ETDRS letters (equivalent to 4.6 lines) in visual acuity at 12 months post-implantation, with the best-performing patient achieving an 11.8-line gain [49, 50]. Many participants regained the ability to read sequences of letters and recognize faces, representing a meaningful restoration of “form vision” that directly impacts daily activities and independence [49–51]. Importantly, the natural visual acuity of patients (without the device) remained stable, indicating a favorable safety profile. The most common adverse events were manageable and included retinal breaks, transient increases in intraocular pressure, and subretinal hemorrhage, all of which were addressed without long-term sequelae [49].

From an engineering perspective, the PRIMA implant is designed for durability, with in vitro reliability testing suggesting a lifespan of up to 27 years at 90% reliability under simulated intraocular conditions [48]. The implant features high-density pixel arrays and advanced biocompatible coatings, such as silicon carbide, to ensure long-term stability and resistance to corrosion [48]. Recent innovations include the development of 3D honeycomb-shaped pixels and fractal electrode geometries, which have the potential to further improve spatial resolution and selective stimulation of bipolar cells, theoretically enabling visual acuity up to 20/80 in future iterations [39, 48].

The PRIMA system’s wireless, self-powered operation, combined with its demonstrated safety and efficacy, positions it as a leading candidate for the restoration of central vision in patients with advanced retinal degeneration. Ongoing research is focused on increasing pixel density, optimizing image processing algorithms, and expanding the technology to additional patient populations [39, 49, 50].

Bioelectronic and genetic approaches

In addition to traditional bionic eye technologies, bioelectronic and genetic approaches are emerging as promising strategies for vision restoration. These methods leverage advancements in stem cell therapy, optogenetics, and hybrid models that combine biological and electronic components. By addressing the root causes of vision loss at the cellular and molecular levels, these approaches offer the potential for more natural and long-lasting solutions.

Stem cell therapy and optogenetics for vision restoration

Stem cell therapy aims to restore vision by replacing damaged or degenerated retinal cells with healthy ones derived from stem cells. This approach is particularly relevant for conditions like RP and AMD, where photoreceptor cells or RPE cells are lost. Stem cells, which have the potential to differentiate into various cell types, can be used to generate photoreceptors or RPE cells for transplantation into the retina [52].

Several types of stem cells are being explored for vision restoration, including embryonic stem cells (ESCs), induced pluripotent stem cells (iPSCs), and adult stem cells. ESCs and iPSCs are particularly promising because they can be differentiated into any cell type, including photoreceptors and RPE cells. Clinical trials using stem cell-derived RPE cells have shown encouraging results in patients with AMD, with some patients experiencing improvements in visual acuity and retinal function. However, challenges remain, such as ensuring the survival, integration, and functionality of transplanted cells in the host retina. Additionally, there are concerns about immune rejection and the potential for tumor formation [53].

Optogenetics is a cutting-edge technique that involves genetically modifying neurons to make them sensitive to light. This approach is particularly useful for restoring vision in patients with advanced retinal degeneration, where photoreceptor cells are lost but other retinal neurons remain intact. By introducing light-sensitive proteins, such as channel rhodopsin, into the remaining retinal neurons, optogenetics can restore light sensitivity to the retina. When exposed to light, these proteins open ion channels, triggering neural activity and sending visual signals to the brain [54].

One of the key advantages of optogenetics is its potential to restore vision without the need for complex electronic devices. However, the technique faces several challenges. First, the light sensitivity of optogenetic proteins is lower than that of natural photoreceptors, requiring high-intensity light for activation. Second, delivering the genetic material to the retina in a safe and efficient manner is a significant hurdle. Viral vectors are commonly used for gene delivery, but they have limitations in terms of payload capacity and immune response. Despite these challenges, early clinical trials of optogenetic therapies have shown promise, with patients reporting improvements in light perception and visual function [55].

Hybrid models combining biology with electronics

Hybrid models represent an innovative approach to vision restoration by combining biological and electronic components. These models aim to leverage the strengths of both fields, using biological cells to enhance the performance of electronic devices or vice versa. One example of a hybrid model is the use of stem cell-derived RPE cells in conjunction with retinal implants. In this approach, stem cell-derived RPE cells are transplanted into the retina to improve its health and function, while a retinal implant provides electrical stimulation to the remaining retinal neurons [56, 57]. This combination has the potential to enhance the efficacy and longevity of retinal implants by providing both biological support through viable RPE cell monolayers and electronic stimulation capabilities [57].

Another example of a hybrid model is the integration of optogenetics with electronic devices. In this approach, ontogenetically modified retinal neurons are stimulated using a retinal implant that provides both light and electrical stimulation. The implant can be designed to deliver precise patterns of light and electrical signals, optimizing the activation of ontogenetically modified neurons. This hybrid approach has the potential to overcome some of the limitations of optogenetics, such as the need for high-intensity light, while also improving the resolution and functionality of retinal implants [58].

Hybrid models also extend to the development of biohybrid sensors, which combine biological components with electronic sensors to create more sensitive and adaptable devices. For example, researchers are exploring the use of living cells, such as photoreceptors or RPE cells, in combination with electronic sensors to create biohybrid retinal implants. These implants can potentially provide more naturalistic vision by mimicking the function of the retina more closely. Additionally, biohybrid sensors can be designed to respond to changes in the retinal environment, such as inflammation or oxidative stress, providing real-time feedback and improving the long-term stability of the implant [59].

While bioelectronic and genetic approaches hold great promise, they also face significant challenges. For stem cell therapy, ensuring the survival, integration, and functionality of transplanted cells remains a major hurdle. Advances in cell delivery methods, such as 3D bioprinting and scaffold-based approaches, are being explored to address these challenges. For optogenetics, improving the light sensitivity and efficiency of optogenetic proteins is a key area of research. Additionally, developing safer and more efficient gene delivery methods is critical for the success of optogenetic therapies [60].

Hybrid models also face challenges related to the integration of biological and electronic components. Ensuring the biocompatibility and long-term stability of these devices is essential, as is developing methods to optimize the interaction between biological and electronic components. Advances in materials science, nanotechnology, and bioengineering are expected to play a key role in overcoming these challenges and enabling the development of more effective hybrid models [61].

Bioelectronic and genetic approaches, including stem cell therapy and optogenetics, offer promising avenues for vision restoration. However, challenges such as immune rejection, limited cell survival, and efficient gene delivery remain. Future research should focus on improving safety, efficacy, and integration with electronic devices for more effective clinical translation.

Technological innovations and components

Electrode count serves as a proxy for potential spatial resolution in bionic eye systems. While Argus II offers just 60 electrodes, Alpha AMS and PRIMA provide substantially higher counts (1600 and approximately 378, respectively), enhancing visual detail and potential functional vision [29, 35, 39]. Suprachoroidal devices like the Australian implant currently use a 44-channel electrode array, offering stable and effective stimulation with excellent biocompatibility [8]. Orion cortical systems, also comprising around 60 electrodes, bypass retinal circuitry but face other limitations related to cortical complexity and invasiveness. Higher electrode counts typically enable finer visual pattern recognition, although they necessitate increasingly complex hardware and sophisticated signal-processing approaches [19]. Comparison of electrode counts for different bionic eye technologies, indicating potential visual resolution capabilities is shown in Fig. 3.

Fig. 3.

Comparison of Electrode Count and Visual Resolution in Bionic Eye Systems

Advanced electrode materials

Recent innovations in advanced biomaterials and electronics have significantly advanced bionic eye technologies. Advanced flexible electrodes and bio-integrative materials improve electrode-neural interfaces, providing stable long-term stimulation [62]. The Argus II Retinal Prosthesis System, which has been clinically tested in over 350 patients worldwide, utilizes platinum disc electrodes embedded in a polyimide substrate [63, 64]. Clinical studies have demonstrated that these platinum electrodes maintain functionality over extended periods, with the device showing 97% electrode functionality even after years of implantation in human patients [47, 65]. The polyimide substrate material (specifically Durimide) has been extensively tested for biocompatibility with human retinal cell lines and has shown excellent safety profiles in clinical applications [66].

Human clinical trials of the Australian second-generation bionic eye have demonstrated that platinum disc electrodes arranged in a suprachoroidal configuration provide effective stimulation with minimal complications [8, 23, 67–69]. In a clinical trial involving four participants with RP, the 44-electrode platinum array showed remarkable stability, with 97% of electrodes remaining functional after 2.7 years of follow-up [65].

Diamond electrodes for enhanced durability

While diamond electrodes show promise in laboratory settings, their clinical application in human retinal prosthetics remains in development [70]. Current FDA-approved devices like the Argus II continue to rely on proven platinum-iridium electrode materials that have demonstrated safety and efficacy in human clinical trials [63, 71].

Biocompatible substrate materials

Polyimide substrates have been the cornerstone of clinically approved retinal prosthetics. The Argus II system uses a thin-film polyimide substrate that has undergone extensive biocompatibility testing according to ISO 10,993 standards [72]. Human clinical trials have confirmed that polyimide-based electrodes maintain their integrity and biocompatibility over extended implantation periods [64, 71].

In clinical applications, polyethylene terephthalate (PET) has emerged as a promising substrate material for conjugated polymer retinal prostheses. Unlike electro-spun silk fibroin which caused inflammatory responses in large animal models, PET demonstrated excellent biocompatibility in preclinical studies preparing for human trials [73].

Nanoparticle-based optical interfaces

The most significant advancement in nanoparticle-based retinal interfaces for humans comes from quantum dot clinical trials. A first-in-human clinical trial involving 20 patients with severe RP demonstrated the safety of intravitreal quantum dot injections [74]. In this study, 78% of patients reported subjective improvements in visual function, with no serious adverse events attributed to the quantum dots [74].

Conjugated polymer nanoparticles represent another clinical advancement, with “liquid retinal prostheses” made of poly(3-hexylthiophene) (P3HT) nanoparticles showing promise for human applications [75, 76]. These nanoparticles can be directly injected into the eye and have demonstrated sustained vision restoration in animal models, paving the way for human clinical trials [77].

Clinical studies have also explored polysialic acid-decorated nanoparticles for retinal applications. These nanoparticles have completed comprehensive safety evaluations including studies in non-human primates, demonstrating no adverse effects and supporting their advancement to human clinical trials for AMD [78].

Technological innovations in microelectronics, wireless power, and artificial intelligence (AI) have significantly improved the functionality and user experience of bionic eye devices. Nonetheless, limitations such as power consumption, device complexity, and the need for adaptive signal processing persist. Future progress will depend on advances in microfabrication, energy efficiency, and robust manufacturing.

Development of advanced materials

The development of advanced materials has played a critical role in improving the safety, durability, and functionality of bionic eye devices. Traditional materials, such as silicon and metals, are often rigid and prone to corrosion, limiting their suitability for long-term implantation. Advanced materials, such as biocompatible polymers, graphene, and conductive hydrogels, offer superior properties, including flexibility, biocompatibility, and resistance to degradation [79]. Biocompatible electrodes are a key area of innovation in bionic eye technology. Traditional electrodes are often made of rigid materials that can cause tissue damage or inflammation over time. Flexible electrodes, made from materials such as conductive polymers or graphene, can conform to the shape of the retina or brain, reducing the risk of damage and improving the stability of the implant. Additionally, these materials have excellent electrical properties, enabling efficient and precise stimulation of neural tissue [80]. Flexible implants are another important advancement enabled by advanced materials. Traditional implants are often rigid and bulky, making them difficult to implant and increasing the risk of complications. Flexible implants, made from materials such as silicone or polyimide, can be rolled or folded during implantation, reducing the size of the incision and minimizing tissue damage. Once implanted, these devices can conform to the shape of the surrounding tissue, improving their stability and functionality [81]. The development of advanced materials has also enabled the creation of more durable and energy-efficient devices. For example, materials with high thermal conductivity can dissipate heat more effectively, reducing the risk of tissue damage and improving the longevity of the implant. Similarly, materials with low electrical resistance can improve the efficiency of energy transmission, extending the battery life of bionic eye devices [82, 83].

Materials science innovations in retinal prosthetics

The advancement of retinal prosthetics has been significantly driven by innovations in materials science, which have enabled the development of electrodes with enhanced biocompatibility, charge injection capacity, and long-term stability [12]. These material innovations are critical for creating high-resolution, durable implants that can effectively restore vision while minimizing tissue damage and inflammatory responses.

Novel intelligent materials for vision restoration

Recent advances emphasize intelligent materials that actively interface with neural tissue to enhance prosthetic vision. For example, tellurium nanowire–based retinal implants have been shown in animal models to restore visible and near‑infrared perception, demonstrating high biocompatibility and no adverse effects in macaques, offering prospects for next‑generation prostheses [84]. Implantable hydrogel electrodes—engineered as smart, soft, hydrated materials—have emerged as exceptionally biocompatible neural interfaces. Their mechanical compliance and conductive properties help minimize inflammatory responses while enabling stable chronic neural modulation [85]. Furthermore, AI‑designed smart hydrogel systems show promise in adaptive biomedical applications, adjusting responsiveness based on environmental stimuli to optimize neural interface performance [86]. Finally, graphene‑based composites and hybrid soft electronics combine high electrical conductivity with tissue‑like flexibility, promoting efficient neuron stimulation and minimizing foreign body response—indicating promising directions for future bionic eye electrode design [87].

Comparative analysis of electrode materials

Platinum-based electrodes

Platinum has been the traditional material of choice for clinical retinal prostheses due to its excellent biocompatibility and established safety profile [19, 88]. The Argus II and Australian suprachoroidal systems utilize platinum disc electrodes that have demonstrated stable functionality in human patients for extended periods [19]. Standard platinum electrodes have a relatively low charge injection capacity (CIC) of 0.05–0.15 mC/cm², which limits their ability to deliver sufficient charge through miniaturized electrodes [88, 89]. To overcome this limitation, modified forms such as platinum gray have been developed, which can achieve CICs up to 1 mC/cm² while maintaining biocompatibility. This enhancement is achieved through increased surface roughness, which expands the effective surface area available for charge transfer without increasing the geometric footprint [12, 19].

Iridium oxide electrodes

Iridium oxide (IrOx) electrodes represent a significant advancement over platinum, offering CICs of 1–5 mC/cm² through reversible faradaic reactions [89, 90]. This material exists in several forms, including sputtered iridium oxide film (SIROF), electrodeposited iridium oxide film (EIROF), and activated iridium oxide film (AIROF), each with distinct properties [89, 91]. EIROF and AIROF demonstrate higher charge storage capacity due to their more porous structure, while SIROF exhibits superior mechanical and electrochemical stability [91]. The IRIS V2 retinal prosthesis utilizes iridium oxide electrodes, which enable more efficient stimulation with lower power requirements compared to platinum-based systems [90]. Recent innovations include combining iridium oxide with other materials, such as platinum gray, to create composite coatings with enhanced mechanical stability and electrochemical performance.

PEDOT-based electrodes

Poly(3,4-ethylenedioxythiophene) (PEDOT) and its derivatives represent a promising class of conductive polymers for neural interfaces, offering exceptional charge injection capacities of 2–15 mC/cm² [92, 93]. PEDOT: PSS (polystyrene sulfonate) electrodes demonstrate significant advantages over metallic electrodes, particularly for microelectrodes with diameters below 200 μm, where they can provide up to 9.5 times higher charge injection capacity and 88% lower power requirements for the same charge density [93]. The flexibility and biocompatibility of PEDOT make it particularly suitable for conforming to the curved surface of the retina [92, 94]. Recent advancements include the development of PEDOT coatings with enhanced adhesion to underlying metal electrodes through surface roughening techniques, which have demonstrated 100% stability under ultrasonic treatment [92]. While PEDOT-based electrodes remain experimental for retinal applications, their exceptional performance metrics make them promising candidates for next-generation high-density retinal prostheses [93, 94].

Graphene-based electrodes

Graphene and its derivatives offer unique advantages for neural interfaces due to their exceptional electrical properties, mechanical flexibility, and biocompatibility. Reduced graphene oxide (rGO) electrodes have demonstrated CICs of 1–10 mC/cm² with low impedance, making them suitable for high-density retinal devices [95, 96]. The two-dimensional structure of graphene enables the creation of ultra-thin, flexible electrodes that can conform to the curvature of the retina while maintaining excellent electrical properties [95]. Recent studies have shown that amphiphilic rGO , functionalized with methoxy poly(ethylene glycol) (mPEG) chains, provides enhanced double-layer charging capacitance and improved neural cell stimulation [96]. While graphene-based electrodes remain in preclinical development, they show significant promise for next-generation retinal prosthetics due to their ability to achieve high spatial resolution with minimal tissue damage [95, 96].

Diamond electrodes

Boron-doped diamond (BDD) electrodes represent a breakthrough in long-term stability for neural interfaces, with projected longevity exceeding 10 years. These electrodes offer CICs of 0.5-3 mC/cm² while maintaining exceptional corrosion resistance and biocompatibility [97]. The wide electrochemical window of diamond electrodes (up to 2.47 V) allows for higher voltage stimulation without inducing water electrolysis, enabling more efficient neural activation [89]. Three-dimensional diamond electrode arrays have been developed to achieve high electrode density while maintaining spatial resolution, with pillar electrodes that reduce the distance between electrodes and retinal ganglion cells. While diamond electrodes remain in preclinical development for retinal applications, their exceptional stability and biocompatibility make them promising candidates for long-term implantation [97].

Substrate materials engineering for flexibility and biocompatibility

The choice of substrate material significantly impacts the overall performance and biocompatibility of retinal prostheses. Flexible substrates that can conform to the curvature of the retina are essential for maintaining stable electrode-tissue contact while minimizing mechanical stress on the surrounding tissue [73].

Polyimide substrates

Polyimide has emerged as a preferred substrate material for retinal prostheses due to its excellent biocompatibility, mechanical flexibility, and electrical insulation properties. The Argus II system utilizes a thin-film polyimide substrate that has undergone extensive biocompatibility testing according to ISO 10,993 standards. Polyimide-based electrodes have demonstrated stable functionality and biocompatibility over extended implantation periods in human clinical trials [98]. Advanced polyimides, such as BPDA-PPD based formulations, exhibit very low water uptake (0.045%) and maintain excellent mechanical properties after long-term storage in simulated body conditions [73]. Microfabrication techniques allow for the creation of high-density electrode arrays on polyimide substrates with feature sizes as small as 35 μm, enabling the development of complex, multi-channel retinal interfaces.

Polyethylene terephthalate (PET)

Polyethylene terephthalate (PET) has demonstrated excellent biocompatibility as a substrate for conjugated polymer retinal prostheses. In comparative studies, PET has shown superior performance compared to electro-spun silk fibroin, which caused inflammatory responses in animal models [73]. PET’s mechanical properties allow it to be easily shaped and positioned during surgical implantation while maintaining structural integrity over time. Recent clinical studies have confirmed that PET-based active devices do not evoke significant inflammatory responses when implanted in the subretinal space, making them suitable candidates for human applications [73].

Parylene C

Parylene C has emerged as another promising substrate material for flexible retinal devices due to its excellent biocompatibility, chemical inertness, and mechanical flexibility. When combined with bio ceramics as base materials, parylene C enables high in vivo safety through flexible device structures that can conform to the curvature of the retina. Recent advancements in thin-film encapsulation for parylene-based devices have achieved stable performance for extended periods under continuous bias conditions. The vapor deposition process used for parylene coating allows for the creation of ultra-thin, pinhole-free insulating layers, which is critical for preventing current leakage in high-density electrode arrays [99].

Nanomaterials for enhanced electrode-tissue interfaces

Nanomaterials offer unique opportunities to enhance the interface between electrodes and retinal tissue, improving both the efficiency of electrical stimulation and the long-term biocompatibility of implants [100].

Quantum dots

Quantum dots represent a revolutionary approach to retinal neuromodulation. These photovoltaically active nanoparticles, when injected intravitreally, diffuse throughout the retina and trigger action potentials in neural cells when stimulated by visible light. A first-in-human clinical trial involving 20 patients with severe RP demonstrated the safety of intravitreal quantum dot injections, with 78% of patients reporting subjective improvements in visual function and no serious adverse events attributed to the quantum dots. These nanoparticles offer the potential for a minimally invasive approach to vision restoration without requiring complex electronic implants [100].

Carbon nanotubes

Carbon nanotubes (CNTs) have shown remarkable potential for enhancing the performance of retinal electrodes. CNT-coated microelectrodes provide a three-dimensional interface with retinal tissue, significantly increasing the effective surface area and improving charge transfer efficiency. Studies have demonstrated that CNT electrodes enable recording of spontaneous spikes from retinal tissue with exceptionally high signal-to-noise ratios, reaching values as high as 75. Additionally, CNTs exhibit strong affinity for neural tissue, promoting improved cell-electrode coupling over time. When combined with other materials such as iridium oxide, CNT composites can achieve enhanced charge injection capacity while maintaining excellent biocompatibility [101].

Nanoparticle-based optical interfaces

Emerging nanoparticle approaches offer alternatives to traditional electrical stimulation methods for retinal prosthetics. Light-sensitive semiconducting polymer nanoparticles, such as poly(3-hexylthiophene) (P3HT), can be directly injected into the eye to functionally replace damaged photoreceptors. These “liquid retinal prostheses” convert natural light into electrical signals with high spatial resolution, potentially offering a less invasive alternative to conventional implants. Perovskite-based nanoparticle systems have demonstrated neuromorphic capabilities for artificial vision applications, exhibiting photonic synaptic functionalities including paired-pulse facilitation and short/long-term memory analogous to the human eye [100].

Advanced fabrication techniques for high-density electrode arrays

The development of high-density electrode arrays is critical for achieving high-resolution vision with retinal prostheses, requiring advanced fabrication techniques to create miniaturized, closely-spaced electrodes while maintaining electrical performance and mechanical stability [99].

Photolithography and microfabrication

Photolithography remains the foundation for fabricating high-density electrode arrays, enabling the creation of precisely patterned electrodes and interconnects with feature sizes in the micrometer range [102]. Advanced optical contact lithography (OCL) can achieve metal feedlines with widths as small as 520 nm and interconnect spaces of 280 nm on flexible substrates, while electron beam lithography (EBL) allows for even smaller features, with feedlines as narrow as 50 nm. These techniques enable the fabrication of high-density arrays with minimal footprint, reducing tissue damage during implantation [99]. Multi-layer fabrication processes, involving sequential deposition and patterning of insulating and conductive materials, allow for the creation of complex, three-dimensional electrode structures with enhanced surface area and improved charge injection capacity [102].

Variable depth dicing and wet isotropic etching

A novel combination of variable depth dicing and wet isotropic etching has been developed for creating high-density, out-of-plane, high aspect ratio electrode arrays with convoluted shapes that can conform to the curved surface of the retina. This maskless method enables the fabrication of trough-shaped, spherical, and saddle-shaped electrode arrays with controlled radius of curvature, allowing for better matching to the geometry of the target tissue. The process involves creating a matrix of variable depth kerfs using a dicing saw, filling them with glass frit, and then using wet isotropic etching to form sharp electrode tips. This approach allows for the creation of electrode arrays with densities up to 6.25 electrodes/mm², suitable for high-resolution retinal stimulation [102].

3D printing and additive manufacturing

Emerging additive manufacturing techniques offer new possibilities for creating customized, patient-specific retinal prostheses. These approaches allow for the precise control of electrode geometry and spacing, as well as the integration of multiple materials with different properties. Recent advancements in multi-material 3D printing enable the fabrication of devices with both rigid and flexible components, optimizing the mechanical interface with retinal tissue. Additionally, 3D printing facilitates the creation of complex, three-dimensional electrode structures with increased surface area, enhancing charge injection capacity without increasing the geometric footprint [103].

Table 2 provides a comprehensive comparison of the key properties of electrode materials used in retinal prosthetics, highlighting their relative advantages and current clinical status. The charge injection capacity values are based on experimental measurements under physiologically relevant conditions, providing a quantitative metric for comparing the efficiency of different materials in delivering electrical stimulation to retinal tissue. The biocompatibility ratings reflect the tissue response observed in both in vitro and in vivo studies, with excellent indicating minimal inflammatory response and good cell viability. Mechanical properties describe the flexibility and conformability of the materials, which is critical for maintaining stable contact with the curved surface of the retina while minimizing mechanical stress on the surrounding tissue. Longevity estimates are based on accelerated aging studies and clinical observations, with diamond electrodes showing the greatest potential for long-term stability due to their exceptional corrosion resistance. Recent advances in materials science have enabled the development of safer, more durable, and higher-resolution retinal prostheses. However, challenges remain in achieving long-term biocompatibility, minimizing immune responses, and ensuring stable electrode-tissue interfaces. Further research is needed to optimize material properties for chronic implantation and to address device degradation and foreign body reactions.

Table 2.

Comparative materials properties

Material	Charge Injection Capacity (mC/cm²)	Biocompatibility	Mechanical Properties	Longevity	Current Clinical Use
Platinum	0.05–0.15	Excellent	Rigid	5–10 years	Argus II, Australian Suprachoroidal
Iridium Oxide	01-May	Good	Semi-flexible	3–7 years	IRIS V2
PEDOT	Feb-15	Very Good	Flexible	Under investigation	Experimental
Graphene	01-Oct	Excellent	Highly flexible	Under investigation	Preclinical
Diamond (Boron-doped)	0.5-3	Excellent	Rigid	> 10 years	Preclinical

Clinical trials and current applications

Overview of major clinical trials and their outcomes

Bionic eye technology has undergone extensive clinical evaluation, with several landmark trials demonstrating the potential for partial vision restoration in individuals with severe visual impairment due to retinal degenerative diseases. The most rigorously studied devices are the Argus II (epiretinal), Alpha AMS (subretinal), PRIMA (photovoltaic subretinal), and the Australian suprachoroidal retinal prosthesis. (Table 3)

Table 3.

Detailed clinical trials & critical outcomes

Trial Name & Technology	Disease Targeted	Trial Results (Clinical Efficacy)	Long-term Follow-up	Critical Adverse Events	Regulatory & Commercial Status	Critical Remarks
Argus II (Epiretinal) NCT00407602 [22, 36]	Retinitis Pigmentosa	Improved basic visual tasks, object detection, low resolution, limited visual field	5 years (long-term safety established)	Electrode degradation, fibrosis, significant immune reaction	FDA/CE approved, discontinued	Limited efficacy, major challenges in long-term stability
Alpha AMS (Subretinal) NCT01024803 [37, 46, 108]	Retinitis Pigmentosa	Moderate reading ability, letter recognition, partial shape detection	Up to 3 years (declining functionality)	Significant encapsulation, electrode corrosion, immune reaction	CE marked, discontinued	Biocompatibility issues major cause of decline
PRIMAvera (Photovoltaic Subretinal) NCT04676854 [40, 42, 54, 111, 112]	AMD (geographic atrophy)	Major visual acuity improvement (23 ETDRS letters), face recognition, form vision	12 months ongoing, promising durability data	Minor retinal hemorrhage, intraocular pressure (manageable)	Ongoing pivotal trials, pending regulatory approval	Strong candidate, awaiting long-term data
Australian Suprachoroidal Implant NCT01603576 & NCT03406416 [8, 23, 51, 52, 68]	Retinitis Pigmentosa	Substantial improvement in mobility, object localization, stable electrode functionality	Stable at 2.7 years	Minimal adverse events, excellent biocompatibility	FDA Breakthrough designation, larger trials planned	Excellent safety profile, moderate resolution
Orion (Cortical Implant) NCT03344848 [9, 24]	Severe visual pathway damage	Early reports of basic visual perception, ongoing trials	Short-term only, limited long-term outcomes	CNS complications risk, highly invasive surgery	Investigational, early trials stage	Promising, but invasive and risks substantial

Argus II was the first retinal prosthesis to receive both FDA and CE approval, following multi-center trials (NCT00407602) involving patients with advanced RP . Over 200 patient-years of follow-up demonstrated significant improvements in light perception, motion detection, and object recognition, with 5-year safety data confirming the durability of the implant and manageable adverse events. However, the device is now discontinued for new patients, though follow-up continues for existing users [22, 104, 105].

Alpha AMS, a subretinal implant developed by Retina Implant AG, was CE marked and tested in several European centers. Clinical trials showed that patients could regain the ability to detect light, recognize shapes, and even read large letters, though outcomes varied and the device has since been discontinued [35, 38, 106].

IRIS V2 was evaluated in open-label trials in Europe for patients with retinal dystrophies, showing improved light perception and object localization, but limited commercial uptake [107, 108].

PRIMAvera, the pivotal trial of the PRIMA photovoltaic subretinal implant, recently completed enrollment of 38 patients with GA due to dry AMD. Preliminary results show a mean improvement of 23 ETDRS letters at 12 months, with regulatory submission planned for 2024–2025 [50, 109, 110].

The Australian second-generation suprachoroidal retinal prosthesis demonstrated substantial improvements in functional vision, navigation, and quality of life over 2.7 years in a phase I/II trial. All four participants experienced significant benefits, with 97% of electrodes remaining functional and no serious device-related adverse events. The team has achieved FDA breakthrough device designation and is planning larger, multi-center trials for regulatory approval [46, 47, 65].

Clinical trials have demonstrated that bionic eye technologies can restore partial vision and improve quality of life. However, outcomes vary with disease progression and neural adaptation, and current devices offer limited resolution and field of view. Larger, long-term studies and standardized outcome measures are needed, along with strategies to improve device accessibility and reduce costs.

Regulatory approvals and commercialization status

The Argus II is the first and only bionic eye system to receive FDA approval, as well as CE marking in Europe. It has been commercially available since 2011, though its high cost (approximately $150,000 per device) limits accessibility [63]. Furthermore, insurance coverage issues, high pricing barriers, and limited patient acceptance due to modest visual restoration levels also contributed significantly to commercial difficulties [63]. The Argus II experience highlights essential commercialization lessons: emphasizing cost-effective scalable manufacturing, robust business modeling, and more effective market engagement strategies is crucial for future bionic eye technologies. The Alpha AMS implant has received CE marking but is not yet FDA-approved, while the Orion system remains in the investigational stage, with ongoing trials to refine its safety and efficacy. Despite these advancements, commercialization remains a challenge due to high development costs, regulatory hurdles, and the need for long-term patient support. Both Second Sight Medical Products and Retina Implant AG faced significant financial difficulties leading to discontinuation of their devices [111, 112]. Second Sight announced winding down operations in March 2020 amid the COVID-19 pandemic, laying off 84 of its 108 employees. The company stated that “against a background of unprecedented economic shock caused by the COVID-19 pandemic and inability to secure additional financing, the company’s Board of Directors has evaluated strategic alternatives and decided to pursue an orderly wind down of the company’s operations” [113, 114]. Second Sight laid off 90% of its employees and subsequently merged with Nano Precision Medicine [115].

Retina Implant AG dissolved in March 2019, with the Alpha AMS work continuing only within academic partners at the University of Tübingen [112]. The shareholders of Retina Implant AG resolved to dissolve the company at an extraordinary general meeting held on March 19, 2019. The company cited two main reasons for dissolution: despite 16 years of intensive research, groundbreaking medical progress had unfortunately failed to materialize, and their work had been hampered by the innovation-hostile climate of Europe’s rigid regulatory and health systems [112].

These failures highlight the ongoing challenges of commercializing bionic eye technology, including high development costs, limited patient populations, and inadequate reimbursement structures. The high cost of bionic eye devices, with the Argus II priced between $115,000 to $150,000 for the device alone and total treatment costs reaching nearly $500,000 per patient, created significant barriers to accessibility and widespread adoption [116]. Limited reimbursement coverage in healthcare systems further compounds the financial challenges for patients, while navigating the approval and commercialization process requires extensive clinical trials and compliance with stringent safety and efficacy standards .

The discontinuation of these devices left more than 350 patients worldwide with implanted Argus II technology essentially obsolete, as upgrades and repairs were no longer available. This situation underscores the critical need for sustainable business models and comprehensive long-term support systems in the development of implantable medical devices [111].

Real world case studies and patient experiences

Real-world applications of bionic eye technology have provided valuable insights into its impact on patients’ lives. For instance, a case study of an Argus II user with RP reported improved navigation in familiar environments and the ability to recognize large objects, such as furniture and doorways [29]. However, users often require extensive training to interpret the visual signals provided by the device, and some report frustration with the limited resolution.

Another case involved a patient with the Alpha AMS implant who regained the ability to read large letters and recognize faces, significantly improving their independence [42]. Despite these successes, patients often face challenges with adapting to the limited resolution and the need for frequent device adjustments. These real-world experiences highlight both the potential and limitations of current bionic eye technologies.

Biocompatibility and long-term implant stability

One of the major challenges facing bionic eye technologies is the long-term biocompatibility and structural integrity of implanted materials. Chronic implantation of electrodes frequently triggers inflammatory responses characterized by glial cell proliferation, fibrosis, electrode encapsulation, and subsequent impedance increases, all of which significantly degrade device functionality over time [81, 117]. For instance, longitudinal studies of the Argus II implant have documented progressive electrode impedance increases linked directly to foreign body reactions and glial encapsulation processes, severely restricting stimulation effectiveness [63, 118]. Similarly, subretinal implants like the Alpha AMS reported electrode corrosion and encapsulation leading to gradual performance decline [119].

Recent developments in advanced biocompatible materials such as conductive polymers (PEDOT), graphene, and diamond-based electrodes aim to minimize inflammatory responses, prolong device longevity, and enhance electrical performance [120]. Suprachoroidal approaches have demonstrated promising biocompatibility, with clinical studies reporting minimal adverse tissue responses and stable long-term electrode functionality [8, 23]. However, these technologies are still in early clinical phases, necessitating further detailed histological and functional assessments.

Challenges and limitations

As illustrated in Fig. 4, the challenges and limitations of bionic eye technology encompass technical constraints, biocompatibility issues, ethical considerations, and accessibility concerns.

Fig. 4.

Overview of Challenges and Limitations of Bionic Eye Technology

Technical and engineering constraints

Bionic eye technology faces several technical challenges, including limited resolution and field of view. Current devices provide only coarse vision, far from the high-resolution images required for tasks like reading or facial recognition [43]. For example, the Argus II offers a visual field of 20 degrees and a resolution of 60 pixels, which is insufficient for detailed vision. Additionally, the complexity of the visual system makes it difficult to replicate natural vision, as the brain must interpret artificial signals that differ significantly from natural neural inputs.

Biocompatibility and long-term stability of implants

Biocompatibility remains a critical concern, as implants must function reliably within the human body for extended periods. Issues such as electrode degradation, inflammation, and immune responses can compromise device performance over time [118]. For instance, some patients with retinal implants have experienced tissue scarring or device failure after several years of use. Long-term stability is also a challenge, as the delicate nature of retinal and cortical tissues makes them susceptible to damage from prolonged implantation.

Ethical and psychological considerations for patients

Ethical concerns include the potential for unrealistic patient expectations and the psychological impact of adapting to artificial vision. Patients may experience frustration or disappointment if the device does not meet their expectations [119]. For example, some individuals report difficulty adjusting to the limited visual input provided by bionic eyes, which can lead to feelings of isolation or depression. Additionally, the invasive nature of implantation surgeries raises questions about patient consent and the risks involved.

Cost and accessibility issues

The high cost of bionic eye devices and associated surgeries limits their accessibility, particularly in low- and middle-income countries. For example, the Argus II costs approximately $150,000, excluding surgery and post-operative care [63]. Insurance coverage is often inadequate, leaving many patients unable to afford the technology. Addressing these cost barriers is essential to ensure equitable access to bionic eye solutions.

Future prospects and research directions

The future of bionic eye technology holds immense promise, with ongoing research focused on overcoming current limitations and expanding the scope of vision restoration. Contemporary analyses of visual prostheses development emphasize that while significant progress has been made over the past two decades, major challenges remain in achieving the ambitious clinical goals of functional vision restoration [11]. These challenges span various aspects including electrode-retina interfaces, biocompatibility, power transmission, and the complexity of replicating natural visual processing [11, 12]. A key goal is to enhance resolution and field of view, as current devices like the Argus II and Alpha AMS provide only coarse vision. Advances in microfabrication and nanotechnology are paving the way for high-density electrode arrays that can deliver more detailed visual information. Improving biocompatibility and longevity is another critical area, with researchers exploring advanced materials like conductive polymers and graphene to reduce immune responses and extend implant lifespan. The integration of AI and machine learning is set to revolutionize bionic eye technology by enabling adaptive and personalized vision restoration, optimizing stimulation patterns, and predicting patient outcomes. Nanotechnology offers unique opportunities for developing next-generation devices, such as quantum dots and nanowires, which can improve efficiency and resolution. Gene editing and optogenetics are emerging as powerful tools, with the potential to correct genetic mutations or make retinal neurons light-sensitive, offering a more naturalistic approach to vision restoration. Hybrid models that combine biological and electronic components, such as stem cell-derived retinal cells integrated with implants, are also being explored to enhance functionality. Addressing cost and accessibility barriers remains crucial, with efforts focused on cost-effective manufacturing, expanded insurance coverage, and global partnerships to ensure equitable access. Ethical and regulatory considerations will play a vital role in shaping the responsible development and deployment of these technologies. By addressing these challenges and exploring new frontiers, bionic eye technology is poised to transform the lives of millions affected by blindness and severe visual impairments, offering hope for a brighter future.

Recent studies highlight significant advancements in bionic eye technology through the integration of artificial intelligence, improved neural interfaces, and emerging surgical methods. Beyeler and Sanchez-Garcia (2022) proposed AI-powered bionic eyes that leverage deep learning algorithms to optimize stimulation patterns, significantly improving perceptual outcomes in patients [121]. Such integration of AI could potentially revolutionize visual prostheses, making them adaptive and personalized. Moreover, current research is deeply invested in understanding neural circuitry involved in vision restoration, critically influencing future implant designs and effectiveness [122]. Emerging ophthalmologic frontiers also include minimally invasive implantation procedures, advanced biocompatible materials, and enhanced visual resolution [123, 124]. Additionally, Lin et al. (2025) underscore emerging therapeutic approaches, highlighting innovative materials and new surgical techniques aiming to drastically reduce post-implantation complications and improve patient quality of life [125]. George (2023) provided insights into clinical efficacy of various bionic eye models, stressing patient-specific customization as a vital future development [126].

Conceptual applications and future integration

Virtual reality simulation of bionic vision

Recent advances in immersive virtual reality (VR) technology have enabled highly realistic simulation of bionic vision, providing invaluable tools for research, device development, and user training. The VR-SPV (Virtual Reality Simulated Prosthetic Vision) and BionicVisionXR toolboxes are open-source platforms that allow sighted participants to “see through the eyes” of a bionic eye user by using psychophysically validated computational models of prosthetic vision [127]. These systems simulate the limited field of view, phosphene patterns, and perceptual distortions experienced by bionic eye recipients, supporting the design and optimization of new stimulation strategies and device configurations. Studies using VR-SPV have systematically evaluated how clinically reported visual distortions affect performance in tasks such as letter recognition and obstacle avoidance, highlighting the importance of realistic phosphene modeling for predicting user outcomes [127]. Embedding simulated prosthetic vision in immersive VR also enables rapid, iterative testing of new algorithms and device concepts before clinical deployment, and can help set realistic expectations for patients and clinicians.

Metaverse integration potential

The integration of bionic eye technology with metaverse platforms is an emerging area of research, with the potential to enhance both the functionality and user experience of visual prostheses. Smart contact lens technologies—such as Innovega’s iOptik^®—are being developed to serve as lightweight, stylish interfaces for augmented and virtual reality (AR/VR) applications, including metaverse environments. These devices can overlay digital information onto the user’s visual field, potentially providing bionic eye users with real-time augmented cues for navigation, object recognition, and social interaction. As XR (extended reality) technologies mature, they are increasingly incorporating multimodal interfaces, including haptics, eye- and face-tracking, and brain–computer interfaces, which could further enhance the immersive experience for users of visual prostheses [128]. However, this integration also raises important ethical and privacy concerns, including data security, equitable access, and the psychological impact of digital augmentation, which must be addressed as the technology evolves.

Training and rehabilitation applications

Virtual reality simulators are also revolutionizing training and rehabilitation for both clinicians and bionic eye users. RetinaVR, a portable and affordable VR platform validated on the Meta Quest 2 headset, enables immersive simulation of complex vitreoretinal surgical procedures, such as vitrectomy and membrane peeling. Validation studies have demonstrated construct validity and user learning curves, indicating that VR-based training can significantly improve surgical skills and safety, especially in resource-limited settings [129]. These platforms can be adapted to provide tailored rehabilitation environments for bionic eye recipients, allowing them to practice navigation, object localization, and daily living tasks in a safe, controlled, and repeatable manner. Such training is critical for maximizing the functional benefit of bionic eye devices, as users must learn to interpret artificial visual signals and adapt to the unique perceptual experiences provided by their implants [121].

AI-augmented vision and patient-centered design

Looking forward, the next generation of bionic eye systems is expected to incorporate artificial intelligence (AI) for real-time scene understanding and visual augmentation. AI-powered “Smart Bionic Eye” concepts propose the use of deep learning to enhance object recognition, facial identification, and navigation, tailored to the specific needs of visually impaired users [121]. Virtual prototyping in VR environments allows researchers and designers to iteratively refine these visual augmentation strategies based on direct feedback from users, ensuring that new features address real-world challenges [121, 127]. Such patient-centered design approaches are essential for developing widely adopted, effective neuroprosthetic devices.

Conclusion

While bionic eye technology significantly advances visual restoration, future therapeutic approaches will increasingly integrate stem cell therapies, optogenetics, and intelligent materials for greater efficacy. Stem cell-based therapies promise regeneration of retinal tissues, potentially reversing certain blindness conditions and complementing or even replacing electronic implants. Optogenetics offers the potential to restore higher-resolution vision by genetically sensitizing retinal neurons to controlled stimulation. Moreover, intelligent adaptive materials and bioelectronics that dynamically modulate neural interfaces in real-time present a paradigm shift, enabling personalized stimulation and enhanced biocompatibility. Interdisciplinary research at the intersection of biotechnology, neuroscience, and materials science will be crucial to overcoming current limitations, ultimately paving the way for revolutionary clinical breakthroughs and more naturalistic vision restoration.

Author contributions

K.-Y. C. contributed to the conceptualization, methodology, software development, investigation, data validation, formal analysis, visualization, and drafting of the original manuscript. H.-C. C. was responsible for conceptualization, methodology, software implementation, and contributed to data acquisition and analysis. C.-M. C. supervised the research process, provided methodological guidance, ensured data validation, and managed project administration. C.-M. C. is the guarantor of the work and accepts full responsibility for the finished manuscript, had access to all data, and made the final decision to submit for publication. All authors participated in the critical review and revision of the manuscript, approved the final version, and agree to the submission and publication of this work.

Funding

No specific funding was received from any funding bodies in the public, commercial, or not-for-profit sectors to conduct the work described in this manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable. This study does not involve human participants, human data or human tissue.

Consent for publication

Not applicable. This study does not involve individual data.

Competing interests

The authors declare no competing interests.