The eye is a miracle as it is. Even though we have a rough understanding of its basic anatomy, we are confronted with a much more complex miracle when we venture into the intricacies of its anatomy and the physiology of the act of seeing. We have theories about the many details – such as the perception and representation of mental images and their storage in the memory – but we still do not exactly know how the act of seeing works.

One of the most mysterious layers of the eye, the retina has an elaborate structure with a slew of functions. Not every eye surgeon dares touch the retina, which houses the most sensitive and special cellular layers. It is a three dimensional, crescent-shaped structure located in the back of the eye and is made up of ten super thin layers of cells that span the exterior part of the eye abutting the veins and its interiors.

Retinal layers

The ten layers that form the retina are:

1. Pigment (coloring matter) layer
2. Layer of rods and cones
3. External limiting membrane
4. Outer nuclear layer comprising rod and cone cells
5. Outer plexiform layer
6. Inner nuclear layer
7. Inner plexiform layer
8. Ganglion layer
9. Nerve fiber layer
10. Inner limiting membrane

These layers are incredibly sensitive and elaborate, and studying their intricate structure give a sense of awe.

Light for sight

Light first arrives at and penetrates through the cornea, the living glassy layer at the outermost layer of the eye. It then goes through the frontal fluid (aquesous humor) and the aperture called the pupil (pupilla). It hits the internal wall of the retina (the inside of the crescent) after passing through the lens in the eye and then the optic fluid filling up the chamber in the back. This alone is an interesting fact because it is much later that the light that gets to the retina reaches the layer of sensitive cone and rod cells, which perceive light. As these cone and rod cells are lined one after another for their protection, light reaches this outer layer of the retina after the ganglion cells, retinal layers and nuclear layers. Such an alignment leads to a reduction of acuity in the peripheral regions of the retina.

We have theories about the many details – such as the perception and representation of mental images and their storage in the memory – but we still do not exactly know how the act of seeing works

The central pit (fovea centralis)

The inner layers at the center of the retina, on the other hand, are drawn to the sides to prevent any loss in visual acuity. Resembling a pit, this section is much thinner than the periphery of the retina, so the layers that are likely to obstruct the passage of the light, and hence reduce visual acuity, are aligned specifically to allow light to directly hit cone and rod cells. Besides, cone cells, which are in charge of exact, colored sight, exist in this region, whereas rod cells, which are in charge of rough and uncolored (black and white) sight, do not.  The central pit where visual acuity is at its highest is for keen, colored, and exact sight. Why then is the rest of the retina not created for acute sight and why is this small section equipped with this ability?

As it turns out, if the entire retina had the ability to see keenly then it would not be possible for the eye to focus on a spot and accurately distinguish it from surrounding objects. If we could see the entire page of a book at a glance, for example, the lines would mix up in our brain. We would not be able to understand what we are reading. We normally start reading from the top of the written page and continue line by line as we focus on and take in each word. Our brain then can focus and perceive a single word accurately by restricting keen perception of the surrounding area.

The retinal pigment layer

The color black is known to absorb, not reflect, light. Thanks to such absorption, the layer made up of black pigments (melanin) or dyes prevents the reflection of light, which is crucial for visual acuity. This black substance functions like the black dye in the bellows of old cameras. If there were not any layer to absorb light, light would scatter off the wall inside the eyeball, thereby obscuring the sharpness between light and dark spots, which is essential for the formation of a clear image, and producing a blurry image due to the overall illumination of the retina.

People who lack this melanin pigment as a result of a genetic defect (Albinism disorder) have white hair, and they are oversensitive to light because the colored iris layer of the eye does not contain the melanin pigment, which refracts light. When an albino person enters a bright area, the light that hits the retina is reflected in all directions through the pigment-lacking retina and the white surfaces of the rigid layer underneath (sclera). Therefore, a ray of light that would normally stimulate a few cones or rods is scattered everywhere, stimulating all or most of the light receivers. As a result, visual acuity in albinos can only be between 20/100 and 20/200, even with the help of the best optical correction, which is a low value compared to 20/20 in normal sight. The joke that rabbits do not wear glasses because they eat carrots is based on the high concentration of vitamin A in this pigment, or black dye, layer of the eye. Indeed, vitamin A is a crucial factor for the health of these pigments.

Layer of rods and cones

Composed of 127 million light-sensitive receivers (photoreceptors), the retina is 0.2 mm thick at the yellow spot (macula lutea), where the image forms most clearly, and 0.1 mm thick at the edges of this area. The 120 million cylindrical rods in the retina are in charge of black and white sight (at twilight), and the 7 million tapered cones, of colored, colored and exact sight. The more common cylindrical rods are 50 µm (microns) in length and 1-5 µm in thickness. The less common cones are 40 µm (microns) in length and 3-5 µm in thickness.

The concentration of the cones increases toward the center of the retina and decreases toward the edges, to be outnumbered by the rods. In the cytoplasm of the cones and rods are stored substances that are sensitive to light (photosensitive) which break up when light contacts them and produce electricity in the cones and rods. In rods this chemical is called rhodopsin. In cones, on the other hand, are three substances corresponding to the colors red, green, and blue that are sensitive to the wavelengths of colored light. To be more exact, there are three separate cone cells that include one of these three substances. Chemically, the photosensitive substances in the cones are a little different from rhodopsins.

The destruction of rhodopsin by light energy

Rhodopsin, along with the color substances, fills up about 40% of rods and cones. They are made up of a protein called scotopsin and a molecule called retinene that is derived from vitamin A. When light energy is absorbed by rhodopsin or the color substances, rhodopsin starts to fade in as fast as one trillionth of a second. The underlying reason for this is that the electrons in the retinene (vitamin A) part of rhodopsin is activated by light, which alters the shape of the retinal molecule at a mind-boggling speed (one trillionth of a second). Extremely complicated and precise chemical and physical changes then take place. It is very difficult to monitor all of these biochemical changes, and they require specialization [1].

The regeneration of the rhodopsin destroyed by light

Rhodopsin destroyed by light is regenerated in the dark. Only in the twentieth century did we manage to partially identify the mechanism by which molecules with very specific geometric shapes decay at incredible speeds only to be regenerated later. It is wondrous that such knowledge and power are present in the cell allowing the light energy to destroy the molecule and the regeneration process is launched. The circulation between destruction and regeneration of the molecule continues throughout a lifetime [2]. Vitamin A is assigned a crucial task in the generation of rhodopsin in the dark. All-trans retinal is converted in the retina into all-trans retinol, which is then converted into 11-cys retinol with the help of an isomerase enzyme. Finally, 11-cys retinol is converted into 11-cys retinal, which in turn combines with scotopsin to form rhodopsin. Vitamin A is present both in the cytoplasm of rods and in the pigment layer of the retina. In this way, vitamin A is kept in reserve to be used for generations of new retinal. If, on the other hand, there is an excess of vitamin A in the retina, the excess amount is converted into retinal, by which the amount of light-sensitive pigment in the retina is lowered.

Night blindness

Night blindness appears in anyone who suffers a serious deficiency of vitamin A. Because there is a lack of vitamin A to be converted into retinal, rhodopsin amounts decrease dramatically. This disease is called night blindness as it is noticeable only in the dark or at night and does not affect sight during the day, due to the reduction of light for proper seeing. In daylight, however, cones can still be stimulated despite a similar decrease in color pigments. Night blindness typically only appears in people that have a low vitamin A diet because huge reserves of vitamin A are normally stored in the liver to be used for the eyes.

The human retina contains 400,000 light receptive cells per square millimeter. For comparison, this number is 680,000 in the retina of the owl, which needs to perceive even the slightest glow when hunting in the night. As another example, with 397,000 such cells the amount in the cat’s retina is almost the same as ours.

An average of 130 sight cells in the retina are connected to a ganglion (nerve node) cell. Constituting the nerve of sight, or the optic nerve (nervus opticus), every nerve fiber is connected to a ganglion cell. It takes about 15-60 seconds for the retinal optic nerves to adapt from dark to light, while it takes as long as 30-45 minutes to adapt from light to dark. The visual range of optic cells, i.e. the lowest and highest amount of light the eye can perceive, is between 10^-7 and 10⁶ nanometers. Light that is below or above this range is invisible to us because of its insufficient or overwhelming wavelength.

All the colors we see in the world are named according to the wavelengths absorbed and reflected by optic cells. The spectrum of the optic cells lies between red and violet, which is the limit of visible light for humans. Light with a wavelength of 400 nm is perceived as violet, while light with a wavelength of 760 nm is perceived as red. Our eyes cannot see infrared and ultraviolet light. Some animals, however, are known to see light beyond these limits. For example, we know that bees can see certain shades of ultraviolet light, which helps them find flowers to pollenate.

Clearly, it is not an easy job to elaborate on the divine art manifested in such a small area as the retina. Complicated structures and reactions that require specialization even to comprehend keep taking place smoothly every moment we look around. At least we can be thankful for this amazing gift of vision given to us free of charge so that we can recognize the universe.

Notes

[1] It causes the cys form of the molecule to change into the all-trans form. Although the all-trans form has the same chemical structure as the cys form, its chemical structure is different in that it is a flat rather than angular molecule. With the impact of light photons, the position of the molecule in space changes, yet its chemical composition remains the same. Because the position of the reactive regions of all-trans retinal no longer fit in with the reactive regions on scotopsin protein, the molecule is pulled off. The product that forms at that moment is batorhodopsins, which is a partly destroyed combination of all-trans retinal and scotopsin. An extremely unstable compound, batorhodopsin turns into lumirhodopsin in nanoseconds, into metarhodopsin I in microseconds, into metarhodopsin II in about one millisecond, and finally into decomposed products, scotopsin and all-trans retinal much more slowly (in seconds).

[2] The first step in the reformation of rhodopsin is the recycling of all-trans retinal into 11-cys retinal, which requires ATP energy and is catalyzed by retinal isomerase enzyme. Once created, 11-cys retinal combines with scotopsin to form rhodopsin.

The discipline in which medicine and engineering truly meet and face new challenges is the field of “biomedical engineering,” an emerging discipline that is only a few decades old. In each sub-specialty of biomedical engineering, researchers study the human body, develop new materials and structures using engineering sciences, either as a treatment method for disease (e.g. artificial bone implants, artificial blood, vascular stents, cardiac valves, etc.) or to diagnose them (e.g. imaging methods and other diagnostic instruments in hospitals). Biomedical engineers face the incredible challenge of developing materials and devices that are compatible with biological systems and capable of working inside the human body to substitute bodily functions. Needless to say, the extreme complexity of the human body makes it impossible to mimic the original system or function of the organs in any way. However, even a poor replacement part or a functional improvement provides great benefit to the patients.

One of the most complex systems of the human body is the nervous system, which consists of the central area (the brain and the spinal cord) and the peripheral parts. The branch of biomedical engineering that deals with the nervous system is “neural engineering.” In this article, we will touch upon a specific subject in the broader area of neural engineering, that is, “neural prosthetics.”1 As the name implies, neural prosthetics is an area where engineering knowledge is utilized to treat neural disorders.

The building blocks of the nervous system are called “neurons.” Neurons generate electric pulses to communicate with each other. The fact that these electric pulses can be elicited by artificial means, i.e. by applying small electric currents to the neurons externally, forms the very foundation of the field of neural prosthetics. Neural engineers can input information into the nervous system by taking advantage of this phenomenon, called “neural stimulation.” Likewise, the information content of neuronal activity can be deciphered by recording the electrical pulses from the neurons and interpreting them according to neuronal function. This two way traffic, monitoring and controlling the neural activity, allows researchers in this field to develop methods of treatment for some sensory, motor, and psychological disorders.

Some of the most successful neural prosthetic applications have been in deep brain stimulation in Parkinson’s disease,

cochlear prosthesis in hearing impairment, bladder emptying and respiration in spinal cord injury, and vagus nerve (10th cranial nerve) stimulation in epilepsy and psychological depression. These are neural prostheses that are readily available as a treatment method for the given ailments. There is a whole host of others that are in the research and development phase. We will review a couple of examples.

In certain diseases of the inner ear hearing is lost as a result of damage to the hair cells inside the cochlea. In normal cochlea the sound information reaches these hair cells after traveling through the ear drum (tympanic membrane) and the structures of the middle ear, causing them to vibrate. This vibration of the hair cells is mechanically transported to the spiral ganglion cells that form the hearing (auditory) nerve. The hearing nerve carries the sound information to the brain in the form of electric pulses. The ganglion cells are healthy and functional even if the entire population of hair cells has been lost as a result of disease. Neural engineers take advantage of the fact that the spiral ganglion cells (which normally accept input from the hair cells) can be electrically stimulated, thus mimicking the function of the hair cells and producing the sensation of sound.2 During a simple surgical operation, the surgeon inserts an electrode into the ear canal which spirals into the lumen of the cochlea so that the sites where the electric current emits from the electrode are adjacent to the spiral ganglion cells (Figure 1). To summarize the principle of the operation; the audio signals are captured by a microphone, processed, converted into electric pulses (A in Figure 1), and transmitted to the implant over a transmitting antenna (B in Figure 1), or headpiece, held in place by magnets. The implant (C in Figure 1) applies the signals to the ganglion cells in the cochlea through tiny electrodes (D in Figure 1). The hearing nerve (auditory nerve) carries the sound information to the brain, where it is “heard.”

Even though the human spiral ganglion has tens of thousands of nerve cells that provide a rich sense of hearing, the cochlear implant, using only six stimulation contacts, can produce auditory perception with sufficient fidelity to enable a deaf individual to use an ordinary telephone.3 Individuals with cochlear implants can also improve their hearing with practice. Thousands of patients have been implanted with cochlear prostheses to date, including children.

The second neural prosthesis application we will review is the retinal prosthesis, which, unlike the cochlear implants, is still in the research phase. Retinitis pigmentosa and age-related macular degeneration both lead to photoreceptor degeneration in the eye and result in a significant visual deficit or blindness.4 A growing body of research supports the feasibility of replacing the function of the photoreceptors with an electronic device. 5–7 A retinal prosthesis is analogous to the cochlear implant in many ways. In a healthy retina, the photoreceptors initiate a neural signal in response to light. In a retinal prosthesis, electrical pulses are utilized to initiate a neural response in the remaining cells of the retina, the bipolar and ganglion cells. It is hypothesized that the perception of shapes and images will be possible through pattern stimulation of the retina. Initial results are encouraging, but the quality of vision that can be attained with this approach is still a question to be answered. A conceptual retinal prosthesis system is shown in Figure 2. The system consists of an external unit coupled to an implanted stimulator with a wireless link. A video camera in the external unit captures an image and converts it to digital data. The implanted unit receives the signal, recovers power and data from the signal, and generates the stimulating current. The stimulus pattern is applied to the retina via the electrode array, which contains distinct electrodes that interface at many locations on the retinal surface. Recent implants in human subjects suggest the feasibility of this approach where individuals attain perception of bright dots in the visual field called ‘phosphenes.’ Furthermore, blind subjects are able to perceive edges when a few of these bright dots are lined up in their visual field.

Figure 3: A conceptual diagram of a Brain-Computer Interface for high level spinal cord injury or patients with ‘locked-in syndrome’ (adapted from Wolpaw et al., see note 9). The recorded neural activity from the motor cortex is processed and converted into command signals to control, for instance, a wheelchair, or to generate electrical signals to activate hand muscles for grasping an object.

Both cochlear and retinal implants are sensory prostheses, i.e. aids for sensory impairments. Another family of neural prostheses deals with motor impairments. Severe motor disability results from high level spinal cord injuries (quadriplegia) where most of the body is paralyzed, sparing only some face, neck, and shoulder muscles. Quadriplegic individuals are in extreme need of a means to control their environment; they need to be in control of their wheelchairs, bed, the room temperature, lights, TV, etc. Because of the level of paralysis it is impossible for them to generate any control signal, except perhaps by sipping or puffing on the end of a tube, which produces a very poor control signal. In the case of a ‘locked-in syndrome’ the condition of the patient is even more serious, with only some functions remaining in the facial muscles. The term ‘brain-computer interface’ has been coined to refer to attempts whereby the motor output of the brain is recorded and interpreted to generate the control signals needed by these patients (Figure 3,9). The ultimate objective of this research will be accomplished when the patients are able to control anything they need to control in their environment, including a computer. The brain-computer interfaces vary in the invasiveness of the approach. The least invasive methods utilize the electroencephalogram (EEG) signals recorded from the scalp. Unfortunately, the signal quality is poor and only ‘on/off’ type of command signals can be generated using this method. In the most invasive, yet most successful applications, an array of electrodes is implanted directly into the motor cortex of the brain at a depth of a couple of millimeters. The recorded signals contain volitional information as the patient makes intentions to move their arms or legs. These signals can be controlled by the patient, and they can in turn be used to control their environment. The current level of success in this type of BCI allows the user to have three dimensional control of a robot arm. This is of invaluable benefit to a quadriplegic individual.

Concluding Remarks: Reflections on Divine Wisdom

Even the subtlest parts of the nervous system are extremely complex. Just to name a few examples, from the highest centers in the brain down to the skeletal muscles in a descending order; the neural circuits of the short-term memory in the hippocampus, fine motor control circuits of the cerebellum, central pattern generators in the spinal cord, and even the control of skeletal muscles in graceful movements of the limbs are impossible to reproduce by artificial means. The Seal of Divine Design is clearly visible in these neural systems, as they are far more complex, far more compact, and far more functionally efficient than any system engineered by mankind. If anything, the growing experience in neurosciences teaches us that the vertebrate nervous system is full of wonders of engineering design. Therefore, it is a great blessing to be a student of both neurosciences and engineering disciplines. This bestows neural engineers with a unique perspective to understand the beauty embroidered into the human nervous system and contemplate on the Divine Wisdom. In spiritual terms, we may think of the human nervous system as a window opening to the works of Divine Wisdom, with manifestations of His Beautiful Names at the brightest level. It is an overwhelming joy to be able to open this window a crack, once in a while, and take a little peek.

References

1. Wise, K.D. ‘Silicon microsystems for neuroscience and neural prostheses,’ IEEE Engineering in Medicine and Biology Society Magazine, vol. 25, no. 5, pp. 22- 29, Sept.-Oct., 2005.
2. G.E. Loeb, ‘Cochlear prosthetics,’ Annu. Rev. Neurosci., vol. 13, pp. 357–371, 1990.
3. J. Helms, V. Weichbold, U. Baumann, H. von Specht, F. Schon, J. Muller, B. Esser, M. Ziese, I. Anderson, and P. D”Haese, ‘Analysis of ceiling effects occurring with speech recognition tests in adult cochlear-implanted patients,’ ORL J. Otorhinolaryngol Relat. Spec., vol. 66, no. 3, pp. 130–135, 2004.
4. E.L. Berson, ‘Retinitis pigmentosa. The friedenwald lecture,’ Invest Ophthalmol. Vis .Sci., vol. 34, no. 5, pp. 1659–1676, Apr. 1993.
5. E. Zrenner, ‘Will retinal implants restore vision?,’ Science, vol. 295, no. 5557, pp. 1022–1025, Feb. 2002.
6. J.F. Rizzo III, J. Wyatt, J. Lowenstein, S. Kelly, and D. Shire, ‘Perceptual efficacy of electrical stimulation of human retina with a microelectrode array during short term surgical trials,’ Invest. Ophthalmol. Vis. Sci., vol. 44, no. 12, pp. 5362–5369, 2003.
7. M.S. Humayun, J. Weiland, G. Fujii, R.J. Greenberg, R. Williamson , J. Little, B. Mech, V. Cimmarusti, G. van Boemel, G. Dagnelie, and E. de Juan, Jr., ‘Visual perception in a blind subject with a chronic microelectronic retinal prosthesis,’ Vision Res., vol. 43, no. 24, pp. 2573–2581, 2003.
8. Weiland, J.D. and Humayun, M.S., ‘A biomimetric retinal stimulation array,’ IEEE Engineering in Medicine and Biology Society Magazine, vol. 25, no. 5, pp. 14-21, Sept.-Oct., 2005.
9. Wolpaw J.R. et al., ‘Brain-computer interfaces for communication and control,’ Clinical Neurophysiolology, vol. 113(6), pp. 767-791, 2002.