17.4 ELECTRICAL CONSIDERATIONS IN RETINAL PROSTHETIC DEVICESThe effectiveness of an electrical stimulation for an intraocular retinal prosthesis, whether epiretinal or subretinal, is gov
Trang 1interface between the electronics and neurons, and the matrix to enable survival of the cellularcomponents while being housed in microelectronics.
17.2.1 Simulations of Prosthetic Vision
One of the major arguments supporting the concept of a retinal prosthesis is the fact that cochlearimplant patients can understand speech with only six input channels Simulations of cochlear implantaudition have shown that speech reduced to as few as four frequencies provides enough informationfor the human brain to understand language Similarly, it is hoped that visual prostheses will be able
to transmit useful information without replacing the input from all 100 million photoreceptors.Several experiments were done to define the minimum acceptable resolution for useful vision Earlystudies in this area focused on simulating prosthetic vision from a cortical implant The points ofstimulation (pixels) required for specific activities varied from 80 to more than 600, depending onthe activity being performed (Brindley, 1965) Most recent studies show that 625 pixels is a betterestimate for certain tasks It was concluded that 625 electrodes implanted in a 1 cm2area near thefoveal representative of the visual cortex could produce a phosphene image with a visual acuity ofapproximately 20/30 Such acuity could provide useful restoration of functional vision for theprofoundly blind (Cha et al., 1992a–c)
Although these studies began to delineate the number of electrodes needed, the fact that all thepixels were projected on a very small area of the retina, made it impractical to translate to the design
of a retinal prosthesis, in which the electrodes would be spread over the entire macular region Thus,
a low vision enhancement system (LVES) has been modified to filter images on a head mounteddisplay in order to simulate pixelized prosthetic vision and to produce an array of dots The resultssuggested that a fair level of visual function can be achieved for facial recognition and reading largeprint text using pixelized vision parameters such as a 25 25 grid in a 108 field, with high contrast
imaging and four or more gray levels
17.3 MECHANICAL EFFECTS OF IMPLANTATION OF RETINAL PROSTHESISRetinal tissue is delicate and can easily tear or detach from the back of the eye The delicate nature
of the retinal tissue can also predispose it to pressure necrosis by a chronic implant being placed on
it Increased intraocular pressure, typical in glaucoma, can lead to damage to retinal ganglion cellsand significant visual loss Also, there is an abundant blood supply within and underneath the retina.Disruption of this vasculature can lead to chronic inflammation or new blood vessel formation, both
of which can lead to retinal damage Studies have shown that an epiretinal array can be secured tothe inner retinal surface in a safe and secure manner, is mechanically stable, and biologicallytolerated over a 6-month period (Majji et al., 1999)
Any intraocular implantable device has to be tested for biocompatibility Since these devices are
to remain within the intraocular environment for many years, they have to continue to be ally effective, and also not cause mechanical damage over time Moreover, the device should alsonot undergo long term degradation, like corrosion, in the ocular environment
electric-17.3.1 Infection and Inflammation
The eye, as is the central nervous system, has been described as immunological or partiallyimmunological privileged (Rocha et al., 1992) Despite this fact, the inflammatory course isidentical to that occurring elsewhere in the body once an incitement for inflammation has occurred(Oehmichen, 1983) Mere surgical manipulation, any infection, biodegradation or any release oftoxic substances from a foreign body can provoke a severe inflammatory response Bacterialinfections are often delayed and appear to be due in part to the host’s inability to respond properly
Trang 2to infections Their origins are frequently distant infected sites in the body or skin flora (Doughertyand Simmons, 1982).
17.3.2 Ocular Side-Effects of Long Term Implantation
Since the field of retinal implants is relatively new, there are few reports available on the long-termside-effects or complications related to implantation of a device Sham surgeries have been done,with no electrical stimulation, to simulate prosthetic implantation, to study the mechanical damage
to the eye In one such study, performed in four dogs, mild retinal folds were noticed at one edge ofthe array, which did not progress over time; there was no retinal detachment (RD) seen in any of thedogs Retinal pigment epithelium (RPE) changes were noted near the retinal tacks which are used tofix the epiretinal implant (Majji et al., 1999) In another study (Walter et al., 1999), nine out of tenrabbits were implanted without serious complications The implant was found to be stable at theoriginal fixation site and there was no change noted in retinal architecture underneath the implant
by light microscopy In three cases, mild cataract formation was observed, while in one case, a total
RD was found after a 6-month follow-up In another study, three rabbits were implanted with anelectrode array in the subretinal space No side-effects were reported (Chow and Chow, 1997).The anatomy and physiology of the retina evaluated after implantation of a retinal implant.Vascular integrity was evaluated by injection of fluorescent dye into the blood stream andsubsequent imaging of the dye’s presence in the ocular blood flow (a technique called fluoresceinangiography) Good vascular perfusion was noted during the entire follow-up period of more than 6months (Majji et al., 1999) Also, in the same study, electroretinogram (ERG) findings were found
to be within reasonable limits after the surgery There is histopathological confirmation that theretina underneath an epiretinal array does not undergo any damage over 6 months of follow-up.Light microscopy and electron microscopy have proved that the retinal microstructure does notshow any signs of degradation over this time, though the area around the tack showed localized loss
of retinal and RPE layers
A single volunteer with end-stage RP has been chronically implanted with an optic nerve cuffelectrode connected to an implanted neurostimulator and antenna in February 1998 Chronicfollow-up of this patient has not shown any side-effects to the surgery or the presence of electrodesaround the optic nerve
17.3.3 Attachment of the Implant to the Retina
Any implanted device will be exposed to the ocular movements, especially in cases where vitreoussurgery replaces the vitreous gel with fluid-filled cavity, where counter-currents from the fluid cangenerate forces on the epiretinal implant; hence, it requires a stable fixation to its intended anatomic
location Ocular rotational movements have been recorded to reach 7008 visual angle/sec These
extreme movements can certainly dislodge the epiretinal device and move it away from the requiredlocation The subretinal implant will not face the same counter-current movements as an epiretinalimplant would, since it is expected to stay within the confines of the subretinal space taking theadvantage of the adherence forces between the sensory retina and the retinal pigment epithelium.Even though the likelihood of displacement of such devices is low, they have been known to bedisplaced after implantation (Peyman et al., 1998) Surgical implantation of such a device can beeither through the sclera (ab externo) or intraocularly through a retinotomy site after a vitrectomyprocedure
There have been various approaches to the attachment of the epiretinal implant or device to theretina Bioadhesives, retinal tacks, and magnets have been considered and tested as some of themethods for the array attachment Retinal tacks and the electrode array have been shown to befirmly attached to the retina for up to 1 year of follow-up with no significant clinical or histologicalside-effects (Majji et al., 1999) Similar results were seen in rabbits (Walter et al., 1999)
Trang 3There have been studies on the use of commercially available compounds for their suitability asintraocular adhesives in rabbits One type of adhesive (SS-PEG hydrogel, Shearwater Polymers,Inc.) proved to be strongly adherent and nontoxic to the retina (Margalit et al., 2000) Other groupshave done similar experiments (Lowenstein et al., 1999).
The preferable fixation site for the intracortical microstimulation arrays is the cortex itself; skullwill not be a good site due to the brain’s constant movement in relation to the skull These arrays arecurrently inserted either manually in an individual fashion or in a group of 2 to 3 electrodes normal
to the cortical surface to a depth of 2 mm or by a pneumatic system that inserts 100-electrode arraysinto the cortex in about 200 msec
17.3.4 Hermetic Sealing of the Electronics
Prostheses will be composed of electronic parts within the eye These components will be exposed
to the chemical environment in the eye These implanted parts will have to be sealed, such that theyare not exposed to corrosion of the ocular fluids Also, this protective coat will have to last for someyears or decades for the continued functioning of the implant The requirement of hermeticallysealing a circuit in the case of neural stimulating devices is complicated by the demand thatmultiple conductors (feedthroughs) must penetrate the hermetic package so that the stimulationcircuit can be electrically connected to each electrode site in the array These connections are themost vulnerable leakage points in the system (Margalit et al., 2004)
17.4 ELECTRICAL CONSIDERATIONS IN RETINAL PROSTHETIC DEVICESThe effectiveness of an electrical stimulation for an intraocular retinal prosthesis, whether epiretinal
or subretinal, is governed by a number of parameters characteristic of the electrode array, includingshape and size of the electrodes, spacing between electrodes, electrode materials, current returnpositions, and stimulating current waveform, to name a few Optimal electrode array type andcharacteristics must also take into account other factors that can influence the one or moreparameters, including thermal or electrical safety or ease of surgical implantation
17.4.1 Stimulating Electrodes: General Considerations with Regard to ElectricalStimulation of the Retina
The characteristics of the stimulating electrode array are often of competing nature: for example, itmight be desirable to mechanically position the electrodes as close as possible to the ganglion andbipolar cells, but that would then result in penetrating electrodes that could harm the fragilestructure of the retina Similarly, it may appear natural to develop small electrodes to achievehigh-resolution electrical stimulation of the retina; however, current densities needed to elicitphosphenes may exceed safety limits and potentially cause damage to the retina Further, it is notcompletely clear, to say the least, the relation between size of the electrode and size of the visualspot induced by that electrode
The problem is phenomenally complex, as it simultaneously involves neural activation at themicroscopic level and control of the spread of the current in retinal tissue at the macroscopic level.Both problems are strongly coupled and involve very different scales and methods of analysis,which increases the complexity of solving the problem of optimal stimulation of retinal tissue and,indirectly, the problem of optimal physical characteristics of the stimulating electrode arrays.Besides geometrical considerations that can affect the effectiveness of the electrical stimulation
of the retinal tissue, other aspects of the system design can have a significant impact on the inducedstimulation Among the challenges that must be considered to achieve optimal electrical stimula-tion, in the sense of an electrical stimulation which uses as little current as possible to elicit visual
Trang 4perception, there are the actual characteristics of the ‘‘contact’’ between retina and electrode, whichstrongly impact the current magnitude and direction in retinal tissue In fact, even though each layer
of the retina is characterized by a different conductivity, the vitreous humor is in general cantly more conductive than each of the layers of the retinal tissue The consequences of this caneasily be understood by thinking of the vitreous humor as the ‘‘preferred path’’ of the electricalcurrent as opposed to the retina, if the conditions are such to make this possible Therefore, if astimulating electrode has its surface in contact with the vitreous humor, and not only with the retina
signifi-as it may happen for example with dome-shaped electrodes with only the tip in actual contact withthe retina, most of the current will tend to flow through the vitreous humor without passing throughthe retina when the current return is located in the eyeball This, in turn, may result in highercurrents needed to stimulate the retina and therefore elicit vision It is therefore clear that the choice
of stimulating electrodes in terms of shape, size, and characteristics, as well as the system design inits entirety, including the choice of the current return location for the electrodes, can have asubstantial impact on the effectiveness of the electrical stimulation of the retina This, in turn,has a significant impact on the feasibility of the entire system, since a more effective stimulationwill require less current, which will result in less power dissipation by the stimulating microchip,leading to a lower temperature increase in the eye and surrounding tissue due to the operation of theretinal prosthesis
17.4.2 The Impedance Method for the Solution of Quasi-Static
Electromagnetic Problems
The problem of characterizing the current spread in retinal tissue, which can also lead to a betterunderstanding of the neural activation once coupled with models of the neural cells, can be solvedthrough quasi-static electromagnetic methods A very versatile method that has a number ofbenefits in the modeling of the system is the impedance method (Gandhi et al., 1984) (or admittancemethod [Armitage et al., 1983]), but other methods based on the solution of the quasi-staticelectromagnetic problem can be used as well (finite-element method, finite-difference method,scalar potential finite-difference method [Dawson et al., 1996], to name a few) The impedancemethod is based on the discretization of the physical model that must be modeled into computa-tional cells The edges of these computational cells are impedances (or admittances) which arecomputed using the electrical conductivity of the material in the cell and the width, length, andheight of the computational cell Therefore, the physical model is represented by means of anelectrical network with resistance or admittances derived from the physical properties of thephysical model itself In its basic formulation, the impedance method uses uniform cells todiscretize the physical model; however, nonuniform cells, leading to a multiresolution impedancemethod, can be used to reduce the computational time and computer memory needed to solve theproblem (Eberdt et al., 2003)
The problem of characterizing the current spread in the retina translates, therefore, into theproblem of developing an accurate model of the eye and the retina, with a geometrical resolutionsufficiently high to describe current variations on the geometrical scale of interest (DeMarco et al.,2003) Even with the multiresolution impedance method, however, it is extremely challenging todevelop a model that reaches cellular scales in the retinal tissue and at the same time covers anextended area such as the entire eyeball Therefore, some compromise must be reached in terms ofresolutions vs geometrical scales of interest for the complete characterization of the system
A possible approach is to discretize the fine retinal structure and electrode geometries withresolutions as low as 5 mm, for example, and subsequently use neural models with the currentlevels found in the neural layers in order to model the response to electrical signals Anotherapproach would be the direct coupling of the macro-scale current spread modeling with electricalcircuits to model the neural interaction This is because in methods such as the impedance oradmittance methods, there is no restriction on the circuit element used between two nodes In the
Trang 5simplest case this is impedance related to the electrical properties of the biological tissue orelectrodes: in more complex cases it can be an arbitrarily complex circuit that can be solved withcircuit simulators such as SPICE1 In fact, the entire impedance or admittance network can besolved with such circuit simulators, with subcircuits describing specific functions or particularbehaviors related to the electrical stimulation.
Figure 17.2 shows an example of a multiresolution computational mesh of a retinal section, withits various layers classified and associated to a conductivity specific for each of them (Eberdt et al.,2003) Figure 17.3 shows instead the current spread in this classified model of the retina for twotypes of electrodes, coaxial electrodes and dome electrodes with side current return, respectively, asobtained by two-dimensional multiresolution impedance method simulations It can be qualita-tively seen that the current magnitudes in various layers of the retina depend upon the type ofelectrode Higher resolution and coupling with neural models can also be incorporated in thesemodels It should be noted, however, that there is a degree of uncertainty with respect to a number
of parameters, such as the conductivity of each layer, which is estimated based on water content andaffinity with other tissues, and actual retinal geometric features, which can be significantly distorted
in diseased retinas
17.5 RETINAL PROSTHESIS AND RELATED THERMAL EFFECTS
An implantable device for neural stimulation should generally receive power and data wirelessly(Rucker and Lossinsky, 1999) — through a telemetry link — process the received data, and injectcurrents in the neural tissue by means of a number of stimulating electrodes that in general need toaccommodate desired waveforms, frequency of stimulation, and amplitudes of stimulating signals.Each of these characteristics is generally responsible for power dissipation, which may result inthermal increase in the human body in proximity of the implanted device
A dual-unit epiretinal prosthesis (DeMarco et al., 1999; Liu et al., 2000), consisting of
an extraocular unit with an external camera for image collection, a data encoding chip, and theprimary coil for inductive power and data transfer and an intraocular unit with the secondarycoil, data processing chips, an electrode stimulator chip, and the electrode array for epiretinalstimulation, could potentially lead to significant temperature increase in the eye and surroundingtissues
Figure 17.2 Example of a multiresolution computational mesh of a frog retina.
Trang 6The wireless link causes electromagnetic power deposition in the head and eye tissues, whichcould lead to indirect thermal rise in the tissue, known to be the dominant physiological hazard due
to power deposition in human tissues (Adair and Petersen, 2002) Moreover, the implantedelectronic IC chips will dissipate power in the form of heat, which will directly lead to the thermalelevation in the surrounding tissues It is therefore necessary to quantify these thermal effects inorder to determine the safe limits of operation of the prosthetic system
The temperature rise in the head and eye tissues due to the operation of the prosthesis can beexperimentally determined within vivo experiments or computationally evaluated by means of acomputer code for the solution of the bio-heat equation Preliminary computational predictionshave been performed to evaluate the thermal influence of a dual-unit epiretinal prosthesis system onthe human head and eye tissues and, therefore, provide a quantitative measure of the temperaturerise in human body as a result of the operation of an implantable neurostimulator As an example oftypical methods and results, the following paragraphs and subsections provide a brief account of themethods and model used in such bio-engineering computations
To quantify the thermal impact of the dual-unit epiretinal prosthesis system, the bio-heatequation can be numerically discretized both spatially and temporally using the well-knownfinite-difference time domain (FDTD) method (Sullivan, 2000; Wang and Fujiwara, 1999) Inthis example, the computational prediction was performed on a very high-resolution anatomicallyaccurate three-dimensional human head model obtained from the National Library of Medicine(The National Library of Medicine, The Visible Human Project, 2000) For the computationalstudy, the different tissues in the head model were modeled by their dielectric and thermalproperties (DeMarco et al., 2003) Figure 17.4 shows the head model, which was utilized in thecomputational domain to evaluate the natural steady state (or basal, initial) temperature distribution
in the model (due to the internal tissue metabolism with no implanted heat sources)
(a)
Figure 17.3 Qualitative image of the current spread in the frog retina due to (a) coaxial electrodes and (b) disc electrodes Current density values range from white (max) to black (zero).
Trang 7The bio-heat equation is developed from the well-known heat equation (Necati, 1985) by sidering the additional sources of thermal influence for computations involving the human body(DeMarco et al., 2003; Bernardi et al., 2003; Gosalia et al., 2004) In the presence of implantabledevices and sources of electromagnetic power deposition, the bio-heat equation is given as:
. r . ðKrT Þ: thermal spatial diffusion term, which leads to heat transfer through conduction (K [J/m .
sec. 8C]);
. A: tissue specific internal metabolic heat production, which will lead to an initial natural steadystate temperature distribution (J/m3.sec);
. B: tissue specific capillary blood perfusion coefficient (J/m3.sec. 8C) This has a cooling influence
proportional to the difference in tissue temperature (T ) and blood temperature (TB);
. rSAR and Pdensitychip : external heat sources due to electromagnetic power deposition and powerdissipated by the implanted electronics, which will lead to a thermal rise beyond the initial naturalsteady state temperature distribution in the head model
Besides the bio-heat equation, the heat exchange at the tissue interface with the external ment has to be modeled accurately At this interface, a boundary condition to model the heatexchange with the surrounding environment is imposed on the computations,
Trang 8wheren is perpendicular to the skin surface and the right hand expression models the heat lossesfrom the surface of the skin due to convection and radiation, which is proportional to the differencebetween skin temperature (T(x, y, z)) and external environmental temperature (Ta).
For all the computations performed in the example above, the temperature of blood was
assumed to be constant at 378C, whileHais the heat convection coefficient and is assumed to be10.5 W/(m2. 8C) The thermal parameters for all the tissues in the head model have been directly
obtained from previous studies (DeMarco et al., 2003; Bernardi et al., 2003)
In order to validate the thermal method and model used,in vivo experiments conducted with dogswere simulated, and experimental and computational results were compared The experiment com-prised of mechanically holding a heater probe (1.4 1.4 1.0 mm in size) dissipating 500 mW in the
vitreous cavity of the eye of the dog for 2 h (Gosalia et al., 2004; Piyathaisere et al., 2003) Theexperimental set up included thermocouples to measure the temperature rise at different locations inthe vitreous cavity and the retina during this period Figure 17.5 shows the comparison between theexperimentally observed and the simulated results for temperature rise at the retina and the vitreouscavity The uncertainty in the exact locations of the thermocouples during the actual experiment is thelikely cause of the small difference between simulated and experimental results
17.5.1 Heat and the Telemetry System
As mentioned in the preceding paragraphs, the wireless telemetry system can be a source of thermalrise since it causes deposition of electromagnetic (EM) power in the head and eye tissues Using theFDTD technique, the deposited EM power can be quantified in terms of the specific absorption rate(SAR) and several studies have quantified the thermal effects in the human head and eye tissuesbased on the evaluated SAR using the bio-heat equation (DeMarco et al., 2003; Bernardi et al.,
1998, 2000; Hirata et al., 2000) SAR is expressed as sE * = 2r ð Þ for conductivity s, electric field E *,
Figure 17.5 Comparison between observed experimental results and computationally derived results for an experiment designed to validate the computational models (From Gosalia K, Weiland J, Humayun M, and Lazzi G IEEE Transactions on Biomedical Engineering, 51(8): 1469–1477, 2004 With permission.)
Trang 9and mass density r at each cell (x, y, z) in the computational model In the radiofrequency range, the
IEEE/ANSI (IEEE standard safety levels, 1999) safety limit for peak 1-g EM power deposition is1.6 W/kg for the general population (the reader is encouraged to refer to the standard for a detaileddescription of maximum permissible exposure [MPE], SAR, and effect of the frequency for EMsafety considerations) In general, if the EM power deposition remains well within this limit, thethermal effects induced will be negligible Therefore, it is necessary to quantify the EM powerdeposition in the head tissues due to the wireless telemetry link to establish if there could bepotential hazards As an example and to illustrate the procedure, we have used a circular coil ofapproximately 37 mm diameter modeled at a distance of 20 mm from the eye and excited by a 2 Acurrent at the center operation frequency of 10 MHz Computed peak 1-g SAR observed in the headmodel due to such an excitation was 0.02 W/kg At this currently estimated operating current levelfor the wireless telemetry link, the SAR values do not exceed the IEEE safety limits for powerabsorption (IEEE Standard exposure to RF, 1999) Thus, it can be reasonably concluded that thecontribution of SAR to the final temperature elevation would be negligible compared to the rise intemperature due to power dissipation in the implanted chip In these cases, the power dissipationdue to the implanted chip and coil alone can be considered as the extraneous heat source (besidesthe natural metabolism of the eye)
However, it should be noted that this will not always be the case The peak 1-g SAR valuedirectly depends upon the wireless link employed for supplying power and data to the implanteddevice, the geometrical characteristics of the wireless devices, the frequency of operation, theirplacement with respect to the human body, and their power level In general, one must evaluate theSAR to ensure that it is within guidelines and determine whether such SAR could result in a thermalincrease and therefore would need to be included in the bio-heat equation
17.5.2 Power Dissipation of Implanted Electronics
In order to compute the thermal elevation due to implanted electronics, the implanted chip wasmodeled in the three-dimensional head model The chip was modeled to have a composite thermalconductivityK¼ 60 J/(m sec 8C) and encapsulated in a 0.5-mm thick layer of insulation (K ¼ 60
J/[m sec 8C]) These values of thermal conductivity are very high compared to the values of the
tissues in the human head (Gosalia et al., 2004)
When an actual prosthesis is implanted, there are several parametric options that can beexplored to minimize the thermal elevation in the surrounding tissues In order to characterizethese options, several thermal simulations were performed with the chip modeled with differentsizes, placed at different locations (within the eyeball) and also dissipating different amounts ofpower in order to gain an insight into the best possible configuration (from the point of view of leastthermal elevation) for an implant in the eye
As an example of the impact of the location of the implanted microchip on the temperatureincrease, we considered two locations for positioning the implanted unit within the eyeball of thepatient In the first case, the lens can be removed and the implanted chip hinged between the ciliarymuscles of the eye (referred to as the anterior position) The other considered position is in themiddle of the vitreous cavity parallel to the axis of the eyeball (referred to as the center position).Both these cases were characterized computationally The implanted chip was modeled at boththese locations and thermal simulations were performed to study the variation in temperatureincrease in different human head tissues as a function of the implant location
For both the above cases, the size of the implanted chip was kept constant at 4 4 0.5 mm
and was allowed to dissipate 12.4 mW (anticipated worst case power dissipation from an implantedcurrent stimulator chip driving a 16 electrode array positioned on the retina) The power density foreach cell of the model of the chip was calculated from the total power dissipated (12.4 mW) and waskept uniform throughout the total volume of the chip (it should be noted that uniform powerdissipation is a further simplification since such an implanted device could, in effect, exhibit
Trang 10nonuniform ‘‘hot-spots’’) It was observed that within 26 min of actual stimulation time (because ofthe extremely small time step in the FDTD simulations, the actual simulation time was significantlyhigher), the thermal elevation profiles in the tissues reached to within 5 to 7% of their final values.Since this provided a good indication of the approximate thermal rise, all the simulations wereperformed for approximately 26 min (physical time).
The maximum temperature increase for both chip positions was observed on the surface of
the insulating layer In both cases, the maximum thermal increase was approximately 0.828C In the
first case where the chip was placed in the anterior position, the temperature of the ciliary muscles
rose by 0.368C as compared to 0.198C when the chip was placed in the center position In the vitreous cavity, temperature rise was 0.268C for the chip placed in center of the eye while the anterior chip raised its temperature by 0.168C (Gosalia et al., 2004).
A chip placed in the anterior chamber of the eye raised the temperature of the retina by less than
half the amount that a chip placed in the center did (0.05 8C by anterior chip as compared to 0.128C
by a center chip) (Gosalia et al., 2004) In these simulations, it was observed that the vitreous cavitywas acting as a heat sink since the rise in temperature of tissues beyond the eyeball is very small
A graphic comparison of the thermal elevation observed for the anterior and the center placed chips
is provided in Figure 17.6 The anterior position is certainly preferable for the implanted unit inorder to minimize the temperature rise in the vitreous cavity and on the retina
A similar analysis can be performed to compute the impact of the size of the implant anddissipated power on the temperature increase in the tissue (Gosalia et al., 2004) It is worth pointingout, however, that power dissipation of the implanted microchip is probably the most significantparameter among all to be considered
Two cases were considered in this example: in the first case, the chip dissipated 12.4 mWand in the second case, it dissipated 49.6 mW For both of these cases, the size of the chip was
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Influence of POSITION of Implant on tissue heating
Insulation of the chip
Anterior position of the chip Mild-vitreous position of the chip
Trang 114 4 0.5 mm and it was placed in the center of the eyeball Power density was again kept
uniform throughout the chip The computation was performed for 26 min of simulated physicaltime
Figure 17.7 graphically compares the temperature increase observed on the insulation, in thevitreous cavity and on the retina for both cases From the thermal elevation results, it is observedthat increasing the power dissipation by a factor of 4 does not necessarily lead to a rise in thetemperature by the same factor In the majority of tissues, a temperature rise by a factor of around3.5 to 5 is observed for a four times increase in the power dissipation in the implant
This preliminary investigation provided a qualitative and quantitative estimate of the thermalinfluence of such an implanted prosthetic system in the eye Also, in the actual system, the variousparametric variations can be optimized to yield the least harmful configuration from the point ofview of thermal damage to the tissues of the eye of head Several efforts are currently underway toaccurately quantify the contribution of each aspect of such a prosthetic configuration to the eventualthermal and electromagnetic influence on the human tissues
17.6 FUTURE IMPLICATIONS
A retinal prosthesis will form several interfaces with the eye including thermal, electrical, andmechanical All of these interfaces must be considered simultaneously during the design of a safeand effective retinal prosthesis For example, it may be possible to reduce the thermal concerns byusing a larger electrode that consumes less power However, such an electrode may stimulate alarge area of the retina and not allow fine resolution vision Many other optimization problems arepresented by such a complex interaction Therefore, future designs may well need to use automatedoptimization algorithms to yield the most effective device
Figure 17.7 Variation of the temperature as a function of the power dissipated by the retinal implant chip.
Trang 12While future implants will depend on the continued advances in technology, the success of theseimplants (i.e., helping the blind see) will be jeopardized if we do not understand the neurobiology ofthe electrically stimulated visual system (Weiland and Humayun, 2003) The sense of vision isenormously complex and the nervous system has the ability to remodel in response to new stimuli.The development of prototypes that can be permanently implanted in research animals now gives
us the ability to study these effects by applying advanced microscopy and tissue labeling methodsdeveloped in neuroscience basic research While these studies are absolutely necessary and willyield valuable information, human implant studies are the only way to verify the effectiveness ofthe devices Therefore, a multifaceted effort including technology development, biological re-search, and strict monitored, limited human tests is needed to advance the current artificial visiondevices from proof-of-principle to accepted clinical treatment for blindness
The work in visual prostheses has come a long way from the days of laboratory research and theinitial volunteer experiments Today, we have a few patients implanted with the actual device; thesedevices have shown no major side-effect or complication related to surgery Some of these patientshave shown encouraging responses Artificial visual stimulus is being tried at various levels, fromthe retina all the way to the cortex Each type of implant has its own advantages and problems Theimplant has to be not only biocompatible, but also be able to avoid damage from corrosion in thebiological spaces the device will be implanted in Long term damage from electrical current is anissue, as is the issue with the type of vision generated by the blind patients through these implants.There are several challenges involved and issues to be considered during the design anddevelopment of a retinal prosthetic system, which can restore a limited form of vision Theelectrical considerations of the prosthetic system (size and shape of electrodes, magnitude ofcurrent injection, size and shape of the implanted unit and its power dissipation, frequency, andstrength of the wireless telemetry link) are closely coupled with safety considerations of the entiresystem (maximum allowable current densities and thermal elevation) These issues have to beresolved to realize a safe and effective retinal prosthesis system or any other implantable neuro-stimulator with a large number of channels Several electromagnetic methods and computationaltechniques are being utilized to investigate the electrical performance characteristics of a prostheticimplant The impedance (or admittance) method coupled with the multiresolution meshing scheme(to represent the intricate details of the retinal tissues — with a 5 mm resolution) appears verypromising for characterizing the current spread in the retinal layers for given current stimulationand electrode array parameters The computational implementation of the bio-heat equationthrough the FDTD method has been utilized to characterize the thermal elevation in the eye andhead tissues due to the operation of the wireless telemetry link and power dissipation of the implant.Both these numerical techniques employ a very high spatial resolution and anatomically accuratemodel of the human head and eye Tissues are represented by their dielectric and thermal properties
as required for the specific computational investigation Using these methods, it is possible tooptimize the performance of an implantable neurostimulator such as the epiretinal prosthesissystem with respect to effectiveness of stimulation and power dissipation
Trang 13Bak M, Girvin JP, Hambrecht FT, Kufta CV, Loeb GE, and Schnidt EM Visual sensations produced byintracortical microstimulation of the human occipital cortex.Medical and Biological Engineering andComputing, 28:257–259, 1990.
Bernardi P, Cavagnaro M, Pisa S, and Piuzzi E SAR distribution and temperature increases in an anatomicalmodel of the human eye exposed to the field radiated by the user antenna in a wireless LAN.IEEETransactions on Microwave Theory and Techniques, 46:2074–2082, December 1998
Bernardi P, Cavagnaro M, Pisa S, and Piuzzi E Specific absorption rate and temperature increases in the head
of a cellular phone user.IEEE Transactions on Microwave Theory and Techniques, 48:1118–1126,July 2000
Bernardi P, Cavagnaro M, Pisa S, and Piuzzi E Specific absorption rate and temperature elevation in a subjectexposed in the far field of radio-frequency sources operating in the 10–900 MHz range IEEETransactions on Biomedical Engineering, 50:295–304, March 2003
Berson EL, Rosner B, Sandberg MA, Hayes KC, Nicholson BW, Weigel-DiFranco C, and Willett WA
A randomized trial of vitamin A and vitamin E supplementation for retinitis pigmentosa (see ments).Archives of Ophthalmology, 111:761–772, 1993
com-Brabyn JA New developments in mobility and orientation aids for the blind.IEEE Transactions on ical Engineering, 29:285–289, 1982
Biomed-Button J and Putnam T Visual responses to cortical stimulation in the blind.Journal of Iowa State MedicalSociety, 52:17–21, 1962
Brindley GS The number of information channels needed for efficient reading.Journal of Physiology, 177:44,1965
Brindley GS and Lewin WS The sensations produced by electrical stimulation of the visual cortex.Journal ofPhysiology, 196:479–493, 1968
Brindley G and Rushton D Implanted stimulators of the visual cortex as visual prosthetic devices.Transactions of the American Academy of Ophthalmology and Otolaryngology, 78:OP741–OP745,1974
Cha K, Horch KW, and Normann RA Simulation of a phosphene-based visual field: visual acuity in apixelized vision system.Annals of Biomedical Engineering, 20:439–449, 1992a
Cha K, Horch KW, and Normann RA Mobility performance with a pixelized visual system.Vision Research,32:1367–1372, 1992b
Cha K, Horch KW, Normann RA, and Boman DK Reading speed with a pixelized vision system.Journal ofOptical Society of America, 9:673–677, 1992c
Chen SJ, Humayun MS, and Weiland JD Electrical stimulation of the mouse retina: a study of electricallyelicited visual cortical responses The Association of Research in Vision and Ophthalmology AnnualMeeting, 1999; Fort Lauderdale Abstract 3886 S736
Chiang YP, Bassi JL, and Javitt JC Federal budgetary costs of blindness.Milbank Q, 70:319–340, 1992.Chow AY and Chow VY Subretinal electrical stimulation of the rabbit retina Neuroscience Letters,225:13–16, 1997
Chow AY and Peachey NS The subretinal microphotodiode array retinal prosthesis (letter; comment).Ophthalmic Research, 30:195–198, 1998
Crapper DR and Noell WK Retinal excitation and inhibition from direct electrical stimulation.Journal ofNeurophysiology, 6:924–947, 1963
Curcio CA, Medeiros NE, and Millican CL Photoreceptor loss in age-related macular degenration.gative Ophthalmology Visual Science, 37:1236–1249, 1996
Investi-Dawson TW, De Moerloose J, and Stuchly MA Comparison of magnetically induced ELF fields in humanscomputed by FDTD and scalar potential FD codes.Applied Computational Electromagnetics Society,11:63–71, 1996
DeMarco SC, Clements M, Vichienchom K, Liu W, Humayun M, de Juan EJ, Weiland J, and Greenberg
RJ An epi-retinal visual prosthesis implementation Annual Conference BMES, vol 1, p 475, 1999.DeMarco SC, Lazzi G, Liu W, Weiland JD, and Humayun MS Computed SAR and thermal elevation in a 0.25
mm 2-D model of the humal eye and head in response to an implanted retinal stimulator: parts 1 and 2.IEEE Transactions on Antennas and Propagation, 51(9):2274–2286, 2003
Dobelle WH and Mladejovsky MG Phosphenes produced by electrical stimulation of human occipital cortex,and their application to development of a prosthesis for the blind.Journal of Physiology, 243:553–576,1974
Trang 14Dobelle WH, Mladejovsky MG, Evans JR, Roberts TS, and Girvin JP ‘‘Braille’’ reading by a blind volunteer
by visual cortex stimulation.Nature, 259:111–112, 1976
Dougherty SH and Simmons RL Infections in bionic man: the pathobiology of infections in prosthetic devices
— part II.Current Problems in Surgery, 19:265–319, 1982
Eberdt M., Brown PK, and Lazzi G Two-dimensional SPICE-linked multiresolution impedance method forlow-frequency electromagnetic interactions.IEEE Transactions on Biomedical Engineering, 50, 881–
889, July 2003
Eckmiller R Learning retina implants with epiretinal contacts.Ophthalmic Research, 29:281–289, 1997.Foerster O Beitrage zur pathophysiologie der sehban und der spehsphare Journal of Psychology andNeurology (Lpz), 39:435–463, 1929
Gandhi OP, DeFord JF, and Kanai H Impedance method for calculation of power deposition patterns inmagnetically induced hyperthermia IEEE Transactions on Biomedical Engineering, 31:644–651,1984
Gosalia K, Weiland J, Humayun M, and Lazzi G Thermal elevation in the human eye and head due to theoperation of a retinal prosthesis.IEEE Transactions on Biomedical Engineering, 51(8):1469–1477,2004
Greenberg RJ, Velte TJ, Humayun MS, Scarlatis GN, and de Juan E Jr A computational model of electricalstimulation of the retinal ganglion cell.IEEE Transactions on Biomedical Engineering, 46:505–514,1999
Guenther E, Troger B, Schlosshauer B, and Zrenner E Long term survival of retinal cell cultures on retinalimplant materials.Vision Research, 39:3988–3994, 1999
Hirata A, Matsuyama SI, and Shiozawa T Temperature rises in the human eye exposed to EM waves in thefrequency range 0.6–6 GHz IEEE Transactions on Electromagnetic Compatibility, 42:386–393,2000
Humayun MS, de Juan E Jr, Dagnelie G, Greenberg RJ, Propst RH, and Phillips DH Visual perception elicited
by electrical stimulation of retina in blind humans.Archives of Ophthalmology, 114:40–46, 1996.Humayun MS, Prince M, de Juan EJ, Barron Y, Moskowitz M, Klock IB, and Milam AH Morphometricanalysis of the extramacular retina from postmortem eyes with retinitis pigmentosa InvestigativeOphthalmology and Visual Science, 40:143–148, 1999a
Humayun MS, de Juan EJ, Weiland JD, Dagnelie G, Katona S, Greenberg RJ, and Suzuki S Pattern electricalstimulation of the human retina.Vision Research, 39:2569–2576, 1999b
IEEE Standard for Safety Levels With Respect to Human Exposure to Radio Frequency ElectromagneticFields 3 kHz to 300 GHz IEEE Standard C95.1, 1999
Jones KE and Normann RA An advanced demultiplexing system for physiological stimulation IEEETransactions on Biomedical Engineering, 44:1210–1220, 1997
Kanda H, Morimoto T, Fujikado T, Tano Y, Fukuda Y, and Sawail H Electrophysiological studies of thefeasibility of suprachoroidal–transretinal stimulation for artificial vision in normal and RCS rats.Investigative Ophthalmology and Visual Science, 45:560–566, 2004
Karny H Clinical and physiological aspects of the cortical visual prosthesis Surveys of Ophthalmology,20:47–58, 1975
Kim A, Sadda S, Pearlman J, Humayun M, de Juan EJ, Melia M, and Green WR Morphometric analysis of themacula in eyes with disciform age-related macular degeneration.Retina, 22: 471–477, 2002.Lederman RJ, Noell WK Optic nerve population responses to transretinal electrical stimulation.Vision Re-search, 9:1041–1052, 1969
Liu W, Vichienchom K, Clements M, DeMarco SC, Hughes C, McGucken E, Humayun MS, de Juan EJ,Weiland JD, and Greenberg RJ A neurostimulus chip with telemetry unit for retinal prosthetic device,IEEE Journal of Solid-State Circuits, 35:1487–1497, October 2000
Lowenstein J, Rizzo JF, Shahin M, Coury A Novel retinal adhesive used to attach electrode array to retina.The Association for Research in Vision and Ophthalmology Annual Meeting, 1999; Fort Lauderdale.Abstract 3874
Majji AB, Humayun MS, Weiland JD, Suzuki S, D’Anna SA, and de Juan E Jr Long term histological andelectrophysiological results of an inactive epiretinal electrode array implantation in dogs.InvestigativeOphthalmology and Visual Science, 40:2073–2081, 1999
Margalit E, Fujii G, Lai J, Gupta P, Chen S, Shyu J, Piyathaisere DV, Weiland JD, de Juan E Jr., and Humayun
MS Bioadhesives for intraocular use.Retina, 20:469–477, 2000
Trang 15Margalit E, Dagnelie G, Weiland JD, de Juan E, and Humayun MS Can vision be restored by electricalstimulation? In: Horch KW and Dhillon GS (eds) (Chapter 7.5)Nueroprostheses: Theory and Practice.Singapore: World Scientific, 1067–1102, 2004.
Maynard EM, Nordhausen CT, and Normann RA The Utah intracortical electrode array: a recording structurefor potential brain–computer interfaces Electroencephalography and Clinical Neurophysiology,102:228–239, 1997
Maynard EM Visual prostheses.Annual Review of Biomedical Engineering, 3:145–168, 2001
Nordhausen CT, Maynard EM, and Normann RA Single unit recording capabilities of a 100 microelectrodearray.Brain Research, 726:129–140, 1996
Necati O.Heat Transfer: A Basic Approach New York: McGraw Hill, 1985
Normann RA, Maynard EM, Rousche PJ, and Warren DJ A neural interface for a cortical vision prosthesis.Vision Research, 39:2577–2587, 1999
Oehmichen M Inflammatory cells in the central nervous system: an integrating concept based on recentresearch in pathology, immunology and forensic medicine.Progress in Neuropathology, 5:277–335,1983
Peterman MC, Mehenti NZ, Bilbao KV, Lee CJ, Leng T, Noolandi J, Bent SF, Blumenkranz MS, and Fishman
HA The artificial synapse chip: a flexible retinal interface based on directed retinal cell growth andneurotransmitter stimulation.Artificial Organs, 27(11):975–985, 2003
Peyman G, Chow AY, Liang C, Chow VY, Perlman JI, and Peachey NS Subretinal semiconductor photodiode array.Ophthalmic Surgical Lasers, 29:234–241, 1998
micro-Piyathaisere DV, Margalit E, Chen SJ, Shyu JS, D’Anna SA, Weiland JD, Grebe RR, Grebe L, Fujii G, Kim
SY, Greenberg RJ, de Juan EJ, and Humayun MS Heat effects on the retina.Ophthalmology Surgeryand Lasers Imaging, 34(2):114–120, 2003
Pollen DA Responses of single neurons to electrical stimulation of the surface of the visual cortex.Brain,Behavior and Evolution, 14:67–86, 1977
Potts AM and Inoue J The electrically evoked response (EER) of the visual system II Effect of adaptation andretinitis pigmentosa.Investigative Ophthalmology, 8:605–612, 1969
Potts AM and Inoue J The electrically evoked response of the visual system (EER) III Further consideration tothe origin of the EER.Investigative Ophthalmology, 9:814–819, 1970
Potts AM, Inoue J, and Buffum D The electrically evoked response of the visual system (EER).InvestigativeOphthalmology, 7:269–278, 1968
Ranck JB Jr Which elements are excited in electrical stimulation of mammalian central nervous system:
a review.Brain Research, 98:417–440, 1975
Rita P, Kaczmarek ME, Tyler ME, and Garcia-Lara J Form perception with a 49-point electrotactile stimulusarray on the tongue: a technical note.Journal of Rehabilitation Research and Development, 35:427–
430, 1998
Rizzo J and Wyatt J Prospects for a visual prosthesis.Neuroscientist, 3:251–262, 1997
Rizzo J, Wyatt J, Loewenstein J, and Kelly S Acute intraocular retinal stimulation in normal and blindhumans The Association of Research in Vision and Ophthalmology Annual Meeting, 2000; FortLauderdale Abstract 532 S102
Rocha G, Baines MG, and Deschenes J The immunology of the eye and its systemic interactions.CriticalReviews of Immunology, 12:81–100, 1992
Rucker L and Lossinsky A Percutaneous Connectors 30th Neural Prosthesis Workshop, NINDS, NINCD,NIH, October 1999
Santos A, Humayun MS, de Juan EJ, Greenberg RJ, Marsh MJ, Klock IB, and Milam AH Preservation of theinner retina in retinitis pigmentosa A morphometric analysis.Archives of Ophthalmology, 115:511–