Biomedical Engineering Reference
In-Depth Information
These arrays have the main advantage of dissipating the heat from the
electronic components into the vitreous. But because edema can develop under
the array and thicken the retina, retinal microtacks must be used to secure the
array in place (glial cells replacing retinal cells under the tacks). Because these
tacks can break as a result of eye movement, subretinal implantation is a more
desirable strategy. Subretinal arrays are implanted into the space that would be
occupied by photoreceptors in a healthy retina between the pigment epithelial
layer and the outer layer of the retina. Unlike epiretinal implants, the remaining
neural network of the retina is used, without any need of extra signal
processing. The implant is held in place by natural ocular pressure and thus
avoids the need for microtacks. There is direct contact with the neural retina.
This allows the natural neural circuitry to carry out the signal processing and
therefore makes it the most ecient neural-microelectrode interface. The close
proximity to the bipolar cells allows for less intense stimulation and thus
minimizes the destruction of the sensitive retinal layer.
A variation of the subretinal implant is the microphotodiode array. These are
known as artificial silicon retina because they mimic the function of natural
photoreceptors. The silicon array utilizes the same principles that govern the
conversion of solar energy to electrical energy in solar cells.
52
For instance,
when light reaches the array of photodiodes, photons within the diodes are
converted into electrical signals that locally stimulate neuron cells of the retina.
There is no need for an external power source since the incident light is, in
principle, sucient to power the array. However, since existing photodiodes are
relatively inecient, approximately a thousand times less ecient than natural
receptors, only intense light can provide sucient photon flux. Moreover,
the in vivo stability of the silicon is poor and this results in the progressive
degradation of the microarray. Severe corrosion completely dissolves the
silicon oxide passivation layer within 6-12 months after implantation. Once the
passivation layer is degraded, the bulk silicon underneath this layer begins to
corrode. The exact mechanism that is responsible for this extensive corrosion is
unknown. It may be due to local electrochemical processes and the reaction of
strong oxidase enzymes in macrophage cells present at the implant site. The
diculties associated with achieving satisfactory biocompatibility and thus
long-term implant stability must be overcome before these prostheses can be
made available to the visually impaired.
Alternative microphotodiode arrays constructed from ceramic materials and
based on the photoferroelectric effect are being studied
53
in order to circumvent
the problem of silicon degradation, although a fully functional artificial retinal
prosthetic has not yet been achieved. Systems currently being researched
require general anesthesia and a six-hour operation to implant surgically,
construct and connect multiple pieces of hardware in the eye, or alternatively,
to insert surgically an implant into the eye which is connected to a wire passing
through the patient's skull. Patients wear eyeglasses with an external camera
and transmitter as well as a belt with a video processor and battery that charges
the system. The patient is able to see forward, but must move the head to
change the field of view. These systems provide up to 60 pixels of sight capacity,
d
n
4
t
3
n
g
|
2
n
3
.
