Visual prosthesis or prosthetic eyes is a type of implantable device, designed based on the concept of functional electrical stimulation (FES) to partly rehabilitate the vision function of the blind eyes. Optical image acquisition, signal processing and nerve stimulation can be realized with this device (Zhou et al., 2006).
To design an effective prosthetic eye firstly requires a thorough understanding of various aspects such as the physiology of the human eye, the processes involved in the formation of visual perception, the causes of vision loss, the different approaches in restoring vision as well as the design methods, materials and other considerations.
The human eye consists of various components which can be classified according to their functions. It is a pressurized sphere with components that perform three main functions (Weiland et al., 2006):
Optical component which consists of the cornea, pupil, vitreous body and lens to focus an image on the retina.
Get your grade
or your money back
using our Essay Writing Service!
Sensory component which consists of rods and cones at the back of the retina near the choroid that functions as photoreceptors that captures incident photons.
Neural components which consists of bipolar neural cells in the inner retina to convert light information into neural electrical signals.
Figure 1: The Human Eye [J3]
It is known that retinitis pigmentosa (RP) and age-related macular degeneration (AMR) are the two leading causes of blindness (Liu et al., 2000). These diseases result in the degeneration of the photoreceptors of the retina while a substantial amount of retinal ganglion cells remain unaffected. Since most of the visual impairment problem begin in the retina, a thorough understanding of retina physiology is crucial in the design of a device that can potentially restore some degree of vision to the blind patient.
The retina is the neuronal front end of the visual system - the image sensor. It has a thickness of about 0.5 mm and it lines the back of the eye. It offers the first step in edge detection and colour analysis of an image and consists of five distinct layers (Stetten, 2000):
Photoreceptor layer - These are light-sensitive pigments, to convert incident photons into chemical energy, made up of rod and cone cells. Cones function in bright light and rods vice versa. There are about 125 million receptors in each eye.
Bipolar cells - First level of information processing. Its function is to respond at the border between dark and light areas. Hence, it demonstrates "centre-surround" receptive field, meaning that if a small dot on the retina elicit a bright response, the surrounding elicit an opposite response. The "centre" response comes from the photoreceptors while the "surround" response from the horizontal cells.
Horizontal cells - Provide surround response to the bipolar cells.
Amacrine cells - Involved in the detection of motion.
Ganglion cells - Here is where most of the data compression occurs. These cells are triggered by bipolar cells, and the neurons only fire when there is meaningful information (detection of border between light and dark). The ganglion cell axons (about 1 million of them) collectively form the optic nerve.
Visual input begins when light enters the eye, passes through a series of transparent layers (cornea, aqueous humor, lens and vitreous body) and encounters the photoreceptors at the back of the retina. The receptors connect through an intermediate layer of transparent nerve cells to the ganglion cells and leave the retina through the ganglion cells axons or optic nerve (as illustrated in Figure 2) which extends to the visual cortex of the brain. The optic nerve is the only source of visual input to the brain (Wang et al., 2005). However, how the brain processes the input into complex and detailed visual perceptions is still not fully understood to this day.
Figure 2: The Pathway in the Process of Visual Perception (Wyatt et al., 1996)
As there are over 100 million photoreceptors and ganglion cells in the eye, it is impossible to replace them. However, there is a possibility of using whatever that remains to restore some degree of vision.
Approaches to the Design of Prosthetic Eye Implants
It has been known since the late 1960s that blind humans can perceive electrically elicited phosphenes in response to either retinal or cortical stimulations (Weiland et al., 2006). A phosphene is an entoptic phenomenon (visual effects whose source is within the eye itself) characterized by the experience of seeing light without light actually entering the eye. Phosphenes can be directly induced by mechanical, electrical, or magnetic stimulation of the retina or visual cortex as well as by random firing of cells in the visual system.
Always on Time
Marked to Standard
There are four different approaches (Zhou et al., 2006) used in the design of visual prosthesis to stimulate nerves in different locations of the eye:
Optic nerve prosthesis
The first experimental work started with electrical stimulation of the visual cortex using a grid of large surface electrode. It was performed by Giles Brindley in 1968. An 80-electrode device was implanted on the visual cortical surface of a 52-year old subject, blind from severe glaucoma and retinal detachment in the left eye. With this system, the patient was able to see light points in 40 different locations in the visual field, demonstrating that at least 50 % of the electrodes were functional. However, there are several drawbacks - large stimulus currents were needed and the phosphenes observed by the subject were quite large. Optic nerve has also been experimented but without producing any favourable results because of the complex structure of the optic nerve which is about 1 - 2 mm in diameter but packed with 1.2 million fibers within it. Stimulation of a single electrode can produce multiple percepts in the visual field and result in the unwanted perception of large blur (Weiland et al., 2006).
Hence, recent researches on the design of visual prosthesis have been narrowed down to two approaches - sub-retinal and epi-retinal prosthesis. An epi-retinal implant is placed on the inner limiting membrane of the retina while a subretinal implant is inserted in the photoreceptor layer. Because of its thin surface and wide area, the retina allows for the implant of a microelectrode array and more localized electrical stimulation to produce a more detailed phosphene pattern. The retinal receptive field properties are well understood. Therefore, electrical stimulation of an area in the retina produces the perception of light in predictable locations.
Table 1 compares the two approaches and details the advantages and disadvantages in terms of image processing, device size, surgery and attachment of device to the retina (Weiland et al., 2006).
Disadvantage: If ganglion cells are stimulated, then the nonlinear image processing performed in the retina may need to be replicated
Advantage: If the bipolar cells can be activated and retinal network is intact, the sub-retinal stimulator is closer to the bipolar cells and can use more of retinal processing
Advantage: Vitreous cavity provides ample space to accommodate device with minimal disruption to retina
Disadvantage: Sub-retinal space is limited, hence large components must be placed outside eye orbit
Advantage: Surgical procedure of the vitreous cavity can be performed easily
Disadvantage: Surgery may be complicated as it involves approaching the retina from the back of the eye
Attachment of device to the retina
Disadvantage: Design of epi-retinal electrode array must conform to curvature of retina without causing damage to the retinal inner linings
Advantage: It has been experimentally proven that the retina can hold a 3-mm-diameter sub-retinal device in place without causing retinal detachment
Table 1: Comparison between Epi-retinal And Sub-retinal Implants in Terms of Design Consideration
Aspects of Design and Models of Prosthetic Eyes
Basic Concept of A Retinal Prosthesis
Basically, all retinal prostheses (both epi-retinal and sub-retinal) have three main systems:
Optical image acquisition - a micro camera made from biocompatible materials, which can be either charge-coupled device (CCD) which is an electronic array shifts and accumulate charges at one corner of the array or complementary metal-oxide semiconductor technology (CMOS) image sensor with transistors at each pixel to amplify and move electrical charges using traditional wires. This is where light from the image is captured and converted into electrical signals and functions as substitute for the loss of photoreceptor cells.
Signal (Data) processing - this acts like a microprocessor in the retinal prosthesis and it is capable of performing functions such as conversion, amplification, rectification and modulation of the electrical signals acquired from the optical image acquisition system.
Nerve stimulation - in the retinal prosthesis, visual perception is formed when the retinal neurons - ganglion cells (epi-retinal) or photoreceptors (sub-retinal) are electrically stimulated using an implantable microelectrode array. The presence of more electrodes will produce a more distinct image.
This Essay is
a Student's Work
This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.Examples of our work
These three components can be either extra-ocular or intra-ocular, depending on the design of the prosthesis. An extra-ocular device is placed outside of the eye whereas an intra-ocular device is implanted within the eye. Data transmission between the three systems mentioned above is done either through cable or wireless connection. Besides, there is also a need for power supply, which could be an external battery or generated through other means.
Figure 3: Basic Concept of a Retinal Prosthesis
Examples of Retinal Prosthesis
Three different models which are taken from on-going research in this field will be discussed.
Model 1: The MIT-Harvard Device [J2]
Optical image acquisition components: CCD camera
Signal processing components: signal processing chip mounted on a pair of sunglasses
Nerve stimulation components: stimulator chip powered by a photodiode array
Microelectrode array implant location: inner limiting membrane (epi-retinal)
Power source: portable external batteries
Principle of operation:
The image in front of the subject will be captured by the tiny CCD mounted on a pair of sunglasses.
The output from the CCD will be modulated onto a carrier signal from a small, fixed-direction laser on the glasses.
The laser beam powers the stimulator chip with the aid of the photodiode array and simultaneously conveys the visual information.
Stimulator chip directs current to the electrode array to electrically stimulate specific locations of the ganglion cells.
Advantages: ganglion cells are well-arranged; easy to access surgically
Disadvantages: axons could interfere because of its location between the electrodes and the ganglion cells and may result in unwanted perception of a large blur
Figure 4: The MIT-Harvard Device (Collaboration between Massachusetts Institute of Technology Research Lab of Electronics and Harvard Medical School Eye and Ear Infirmary)
Model 2: EPI-RET-3 Implant System [J4]
Optical image acquisition components: CMOS image sensor
Signal processing components: extra-ocular portable computer system with transmitter unit mounted on eyeglass frame
Nerve stimulation components: receiver unit implanted in an artificial lens and a set of 25 three-dimension stimulation electrodes
Microelectrode array implant location: ganglion cell layer (epi-retinal)
Power source: rechargeable internal batteries inside eyeglass frame
Principle of operation:
The image acquired is processed by the external computer system to generate simple patterns.
The patterns are transformed into stimulation pulse sequence.
These data will modulate the carrier frequency and transmitted wirelessly using inductive coupling method. Power is also transmitted in the same manner.
The electromagnetic signal is acquired by the receiver coil and forwarded to the stimulator chip.
Based on the data received, the stimulator chip activates the selected electrodes in the upper ganglion cell layer and subsequently triggers phosphenes in the visual cortex to create a visual sensation.
Advantages: no cable connection needed; 3-D electrode provide good contact to the ganglion cells;
Disadvantages: an array of 25 electrodes is only sufficient to evoke images of simple patterns
Figure 5: EPI-RET-3 retina implant system (A project funded by the German Ministry of Education and Research)
Model 3: Hermetically-Encased Wireless Sub-retinal Prosthesis [J7]
Optical image acquisition components: separate external hardware components required (not described in this design but is similar to previous two models mentioned above)
Signal processing components: controller microchip
Nerve stimulation components: micro-fabricated array of sputtered iridium oxide film (SIROF) electrodes
Microelectrode array implant location: sub-retinal space
Power source: transmitted wirelessly from external power supply
Principle of operation:
Image is acquired from external optical device.
A computer-based controller with a user interface determines which electrodes to drive and the stimulation current level.
Data signal and power is transmitted wirelessly to the implant (shown in Figure 6)
The implant receives the power and data via the receiver coils right behind the conjunctiva and processes the data through the circuitry enclosed in the hermetic case.
The implant sends electrical stimulation current to the sub-retinal nerve cells (i.e. photoreceptor layer cells) through a thin-film micro-fabricated SIROF to create a visual perception.
Advantages: large receiver coils allows for better data and power telemetry; hermetic titanium case enclosing the circuitry has 10-year survivability and suitable for longer term implantation; serpentine electrode array that extends to the back of the eye ball provides easier access into the sub-retinal space.
Disadvantages: risk of infection due to placement of receiver coils on the conjunctiva, which is delicate and sensitive
Figure 6: Hermetically-encased wireless sub-retinal prosthesis
Important Factors in the Design of A Prosthetic Eye
External hardware such as eyeglass causes few problems to the user. Usually, it is the internal hardware that is implanted in the eye that creates most complications especially because of its invasive nature and contact with delicate tissues. Besides, certain signal processing aspects also has to be considered. These are some of the crucial factors [J9] that need to be taken into account in the design process of a functional prosthetic eye:
Number of channels: The number of channels in a retinal prosthesis depends on the size of the stimulation electrodes. Each channel functions similarly to a pixel in a digital image. Due to the limited space in the retina, bigger electrodes will result in fewer available channels. However, most of the recent experiments use less than 25 electrodes which are only sufficient to create simple visual perception such as alphabets and patterns.
Size and shape of stimulator electrode: With smaller electrodes there can be more channels. However, the smaller the size of the electrode, the higher the impedance becomes. This means that the stimulator will need to withstand higher compliance voltage of up to 10V. As for shape, it needs to have rounded corners to minimize retinal tissue damage when implanted.
Stimulus waveforms: There needs to be large tuning range and programmability to cater individuals with varied threshold. A study based on rabbits suggests that shorter pulse train in sub-milliseconds is more effective.
Power and data telemetry: The transmission of power or data from the between the different components of the eye prosthesis such as power supply, image sensor, microcontroller and electrode array is best done wirelessly, using radio frequency (RF) waves to avert potential tissue infection or injury due to transcutaneous wires inside the eye ball. Power input of 100 mW to the implant and data rate of 2 Mbps is sufficient to support high-density stimulation.
Stimulator current injection: Recent clinical trials [J3] have found that stimulus current in the range of 300 to 500 ÂµA per millisecond is enough to elicit response from the retinal neurons. Safe current limit decreases with electrode size and often expressed in coulombs. It is between 0.9 to 50 nC.
Flexibility of microelectrode array: A retinal microelectrode array must be flexible to conform to the spherical shape of the retina. This can be achieved with the use of polymer substrate such as polydimethyl siloxane (PDMS) and parylene.
Biocompatibility of fabrication materials: The most critical issue in the design of the eye prosthesis is the interface between the electrodes and the neural tissues. Traditionally, rare metals such as gold, platinum, iridium and titanium nitride are used for making the stimulation electrodes. However, prolonged use of metal parts on these tissues results in glioma aggregation and scar formation. To overcome such problems, research [J10] is under way to replace conventional electrodes with conductive polymers. These polymers should have high affinity to biological tissues and can stimulate the neurons intracellularly, hence decreasing the threshold current value.
Surgical procedures: All eye implants require some form of surgery, usually on the lens or retina of the eye. The surgical procedure to insert an epi-retinal implant is usually done through the vitreous cavity which is relatively easier as compared to a sub-retinal prosthesis which requires surgery through the back sclera of the eye. Another factor to be considered is the risk during or post-operation such as infection, inflammatory reaction, neo-vascularization or hemorrhage.
To work towards a fully-functional eye prosthesis, firstly it requires a thorough understanding of the visual system such as the anatomical structure of the eye and the functions of the different component or layers in the eye. Next, the approach to focus on has to be decided, i.e. the location of the implant. It could be the visual cortex, optic nerves, epi-retinal layer or sub-retinal layer.
A strong knowledge in the concept of prosthetic eye is also required. Basically, the prosthetic eye should be able to perform three basic functions i.e. optical image acquisition, signal processing and electrical nerve stimulation. A complete and functional eye prosthesis usually consists of both extra-ocular (such as eyeglass frame, image sensor, power supply, processor circuitry and transmitter unit) and intra-ocular (such as power/data receiver and microelectrode array) components. Because of its complicated nature, most researchers work on a specific aspect of the design, and work collaboratively, with each focusing on his or her field of expertise such as circuit design, hardware design, software programming, telemetry, material selection and implant method. Some of the factors that need to be taken into consideration in the design process include number of channels, size and shape of stimulator electrode, stimulus waveform, power and data telemetry, stimulator current injection, flexibility of microelectrode array, biocompatibility of fabrication materials and surgical procedures.
To date, most of the research in this area is still in the experimental stage, with mostly field trials done on non-human subjects. There are also some clinical trials done on human subjects with some promising results. Experimental results show that functional electrical stimulation of the retinal nerve is a possible way to restore some degree of useful vision in blind subjects but there are still lots of fine-tuning to be done at the design level before such a device can be implemented successfully.