SECTION I

Assistive technologies to enhancesensorimotor performance

CHAPTER 1

Building the bionic eye: an emerging reality andopportunity

Lotfi B. Merabet{,{,*

{ Department of Ophthalmology, Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston,Massachusetts, USA

{ The Boston Retinal Implant Project, Center for Innovative Visual Rehabilitation, Boston VA Medical Center, Boston,Massachusetts, USA

Abstract: Once the topic of folklore and science fiction, the notion of restoring vision to the blind is nowapproaching a tractable reality. Technological advances have inspired numerousmultidisciplinary groupsworldwide to develop visual neuroprosthetic devices that could potentially provide useful vision andimprove the quality of life of profoundly blind individuals. While a variety of approaches and designsare being pursued, they all share a common principle of creating visual percepts through the stimulationof visual neural elements using appropriate patterns of electrical stimulation. Human clinical trials arenow well underway and initial results have been met with a balance of excitement and cautiousoptimism. As remaining technical and surgical challenges continue to be solved and clinical trials moveforward, we now enter a phase of development that requires careful consideration of a new set of issues.Establishing appropriate patient selection criteria, methods of evaluating long-term performance andeffectiveness, and strategies to rehabilitate implanted patients will all need to be considered in order toachieve optimal outcomes and establish these devices as viable therapeutic options.

Keywords: retinal implant; neuroprosthesis; blind; retina; rehabilitation.

We can rebuild him. . .we have the technology.—from the television series “The Six MillionDollar Man”

Introduction: history, an unmet demand, andstate of the art

Our fascination with building a bionic humanmirrors the technological advances that ubiquitouslycharacterize the modern era. Today, this idea hasbecome less the subject of science fiction and morethe pursuit of intense scientific research. Advances

*Corresponding author.Tel.: þ (617) 573-4130; Fax: þ (617) 573-4178E-mail: lotfi_merabet@meei.harvard.edu

A. M. Green, C. E. Chapman, J. F. Kalaska and F. Lepore (Eds.)Progress in Brain Research, Vol. 192ISSN: 0079-6123Copyright � 2011 Elsevier B.V. All rights reserved.

3DOI: 10.1016/B978-0-444-53355-5.00001-4

within the realms of microfabrication, microelec-tronics, material science, wireless technology, andhigh-speed computer processing power haveallowed for the development of neuroprostheticdevices designed to assist individuals living with sen-sory loss and/or motor impairment. The basic prem-ise underlying all neuroprosthetic approaches is thattargeted and controlled delivery of electrical stimu-lation to nerves or muscles can potentially restore(to a certain degree) the physiological function of adamaged organ or limb (Marbach et al., 1982). Thesuccess of cochlear implants, developed over30 years ago, serves as a well-known example. Thisneuroprosthetic device has helped thousands of pro-foundly deaf individuals regain hearing and developspeech communication (Jones et al., 2008; Loeb,1990). Similarly, sophisticated artificial limbs haveled to improved walking mobility and even graspingskills for amputees (Allin et al., 2010; Craelius, 2002;Laferrier andGailey, 2010). The continueddevelop-ment of brain–machine interfaces (BMIs) is alsoproviding exciting hope for paralyzed patients. Byrecording neuronal signals from the brain that codefor movement, these signals can be converted andused to control external devices such as a roboticlimb prosthesis (Donoghue, 2002; Hochberg et al.,2006; Nicolelis, 2003). Rapid progress in all of thesearenas continues and inmanyways serves as inspira-tion for the development of a visual neuroprosthesisfor the blind. Today, several device designs andapproaches are being developed and human clinicaltrials are well underway (for extensive reviews see:Chader et al., 2009; Dagnelie, 2006; Dowling, 2005;Humayun, 2007; Javaheri et al., 2006; Merabetet al., 2005).

According to the World Health Organization,there are 314 million visually impaired individualsworldwide (2009 WHO fact sheet; http://www.who.int/mediacentre/factsheets/fs282/en/). Whilean astonishingly large number, it is worthy to notethat only a minority of individuals (�45 million)are actually considered profoundly blind(defined as best-corrected visual acuity worse than20/400 Snellen acuity) and have some degree ofresidual visual function. Further, a tragic reality

exists. The greatest portion of these individualslive in developing countries and the majority ofthe leading causes of blindness are actually avoid-able and/or treatable (e.g., surgery for cataractsor antibiotic treatment for trachoma). Thus, therestoration of functional vision through a visualprosthesis will likely target only a restrictedsegment of the blind population. Moreover, it isimportant to realize that not all individuals andall forms of visual impairment could potentiallybenefit from a visual neuroprosthesis. Aspresently conceived, visual prosthetic devices havebeen designed for individuals with profound visionloss and who have had normal visualdevelopment (as opposed to congenital causes ofblindness). Further, as these devices are designedto interface with viable neuronal tissue, the siteof damage and nature of pathology will largelydictate whether a prosthetic device can be feasiblyimplemented. For example, in conditions wherethe overall functional and structural integrity ofthe retina is compromised (e.g., trauma, glaucoma,or retinal complications related to diabetes), a ret-inal-based visual prosthesis is unlikely to be effec-tive in restoring visual function (see discussion onvarious visual prosthesis approaches). Theselimitations notwithstanding, it is also important tohighlight advances being made in other areas ofbiomedical research such as gene therapy and celltransplantation. Thesemolecular-based approachesmay in time provide new treatments and helphalt the progression of vision loss particularly withrespect to hereditary causes of blindness (Aclandet al., 2001; MacLaren et al., 2006). At the sametime, blind individuals will continue to benefitfrom the use of sensory-substitution devices (alsodiscussed in this edition). These devices are spe-cially designed to leverage sensory informationobtained from the intact senses (e.g., hearing andtouch) to substitute for the vision. This allows ablind user to interact with his or her surroundingenvironment (Bach-y-Rita, 2004; Bach-y-Rita andKercel, 2003). Thus, the future rehabilitation ofindividuals with visual impairment will likely con-tinue to encompass multidisciplinary approaches

4

and include molecular-based therapies designed tohalt the progression of vision loss, the use ofsensory-substitution devices, and potentiallyrestore a certain level of functional visionthrough the use of visual neuroprostheticdevices. Here, we will highlight advances anddiscuss future perspectives relating to visualneuroprosthesis development.

Summary of visual neuroprosthetic approaches:from retina to cortex

Generally speaking, the operating premiseunderlying a visual neuroprosthesis is to artificiallyreplace the function of damaged neuronal elements

that make up the visual pathway (Fig. 1). Typically,patterned microelectrical stimulation is deliveredthrough an array of tiny microelectrodes to elicitthe perception of organized patterns of light (how-ever, see also the development of submillimeter,geometrically constrained microfluidic channels todeliver targeted and controlled release ofneurotransmitters; Peterman et al., 2004). The elec-trical stimulation of these surviving visual neuronalelements evokes the subjective sensation of dis-crete points of light (referred to as “phosphenes”;Gothe et al., 2002;Marg andRudiak, 1994). In prin-ciple, by delivering appropriate multisite patternsof electrical stimulation (i.e., characterizing theshape of the intended visual target and reflectingthe neural structure's retinotopic organization),

Cortex

Optic nerve

Retina

Mioro-electrode

array

Spiral cuffelectode

Ganglion

Bipolar

Photoreceptor SUBRETINAL

EPIRETINAL

Micro-electrode

array

(c)

(a)

(b)

Fig. 1. Summary diagram illustrating various neuroprosthesis approaches to restore vision. Theoretically, any point along the visualpathway can be electrically stimulated to generate the perception of phosphenes and thus represents a potential site to implant avisual prosthesis. At the level of the retina, an implanted device generates electrical current to stimulate cells of the inner retina(i.e., ganglion and bipolar cells). Two approaches are possible: (i) epiretinal; in which the device is attached to the inner surfaceof the retina, and (ii) subretinal; in which the device is placed within the underside of the retina. The optic nerve can bestimulated by implanting a cuff electrode around the nerve. In the cortical approach, electrodes are placed either intracorticallyor on the cortical surface in order to stimulate the visual cortex directly and thus bypassing afferent visual structures.

5

geometrical visual percepts can be generated. Thisallows for the perception of visual images (muchakin to viewing a stadium electronic scoreboard orthe images generated by an ink jet printer). Thepattern of electrical stimulation delivered is deter-mined by analyzing an image captured by a digitalcamera or in response to the images captured bythe optics of the eye itself. With regards to visualperception, this “scoreboard” approach certainlyrepresents a great oversimplification. It is clear thatmany attributes characterize a visual scene such ascolor, motion, and form. However, ascurrently conceived, visual prostheses are designedto address only oneof themost basic components ofvision, that is, spatial detail.

Among the biggest challenges of prostheticvision is the puzzle of the neural code for percep-tion. The complexity of the neural code suggeststhat prosthetic devices should rely on intactneural circuitry whenever possible in order totake advantage of any intact sensory processingavailable (Dagnelie and Schuchard, 2007). Thus,reducing the complexity of neural codingnecessary could potentially be achieved byimplanting the prosthetic device at the earliestpoint along the visual pathway that retainsfunctional integrity. Following to this premise,the retina would represent the earliest site ofpotential neuronal interface.

Retinitis pigmentosa (RP) and age-relatedmacular degeneration (AMD) are two retinaldisorders that contribute greatly to the incidenceof inherited blindness and blindness in theelderly, respectively (Bunker et al., 1984; Kleinet al., 1997). Profound vision loss results largelydue to the progressive degeneration of the light-capturing component of the outer-segment ofthe retina, that is, the photoreceptor cells.However, the remaining retinal elements withinthe inner retinal layers (e.g., the bipolar andganglion cells that converge to form the opticnerve) appear to survive in large numbers.Further, these elements remain responsive toelectrical simulation even in highly advancedstages of the disease (Humayun et al., 1996). In

essence, a retinal-based visual prosthesis wouldreplace the function of the degenerated photore-ceptor cells by stimulating the surviving retinalneuronal machinery. A set of pivotal humanexperiments demonstrated that electrical stimula-tion of the retina of RP patients (Humayun et al.,1996; Rizzo et al., 2003b) as well as one patientwith AMD (Humayun et al., 1999) led to thegeneration of phosphenes despite the fact thatpatients were profoundly blind for many years.Experiments lasted minutes to hours whilepatients remained awake in order to describetheir visual experiences. Following electricalstimulation, patients reported visual patternedperceptions that were initially relatively crude.However, the gross geometric structure of thephosphene patterns could be altered in acontrolled fashion by varying the position andnumber of the stimulating electrodes and thestrength or duration of the delivered current(Humayun et al., 1996; Rizzo et al., 2003a,b). Thisdemonstration of proof of principle has led manygroups worldwide to pursue development of avariety of retinal-based designs and approaches.Currently, the retinal-based approach is arguablyreceiving the most attention as evidenced by sizeand number of on-going human clinical trials.

Two retinal-based approaches are beingpursued that are largely differentiated by theirlocation of implantation with respect to the retina.In the subretinal approach, the implant is placedin the region of degenerated photoreceptors bycreating a pocket between the sensory retinaand retinal pigment epithelium (RPE) layer. Inthe epiretinal approach, the implant device isattached to the inner surface of the retina, closeto the ganglion cell side (Fig. 1a).

The subretinal visual neuroprosthesis design iscurrently being pursued by the Boston RetinalImplant Project (a large joint collaborative effortthat includes the Massachusetts Eye and EarInfirmary and Harvard Medical School, theMassachusetts Institute of Technology, theBoston Veterans Affairs Healthcare System, andother partnering institutions; see Fig. 2; Shire

6

et al., 2009). By virtue of being placed injuxtaposition to the nearest layer of survivingneurons (i.e., bipolar cells), the subretinalapproach affords greater inherent mechanicalstability. This is due to the fact that the ultrathinelectrode array is effectively “sandwiched”between the inner-segment of retina and theRPE layer. Further, this approach has thetheoretical advantage of not only being closer tosurviving neuronal elements (thus potentiallyrequiring lower amounts of electrical stimulatingcurrent) but also exploiting retinal signalpreprocessing inherent to the bipolar cell layer.The placement of a subretinal device does requireelaborate and complex surgical methods. For theBoston Retinal Implant device, this includesinserting an ultrathin flexible microelectrodearray through an incision made on the outsidescleral wall of the ocular globe. This surgicalapproach allows the device to reside within thesubretinal space created (referred to as the “ab

externo approach” as opposed to “ab interno”;where ones passes through the vitreous humorof the eye and inserts the device through an inci-sion made directly in the retina; see Javaheriet al., 2006). Another feature of thisconfiguration is that it leaves the bulk of theelectronic hardware outside of the eye thusavoiding complications related to heat generationand corrosion and facilitates the exchange ofelectronic components as needed. For its opera-tion, a miniature camera mounted on a pair ofeyeglasses is used for image capture. Theseimages are then analyzed by an externally wornportable microprocessor used to convert theimage data into an electronic signal. Theappropriate signal pulses (delivering data andpower) are transferred to the implant wirelesslyvia radio frequency (RF) coils. The resultingsignal is transmitted to the subretinal microelec-trode array driving the surviving retinal neuralelements (i.e., bipolar and ganglion cells) with

Fig. 2. Artist conception of the Boston Retinal Implant device. (a) Specially designed glasses house a miniature camera used for thecapture of images. The image is analyzed by an image processing unit and appropriate stimulus pulse information and power aresent via a transmission coil. A secondary receiving coil (sutured around the iris of the eye) captures the wireless informationtransmitted. (b) The transmitted information is relayed through a series of electronic components (hermetically sealed in atitanium case) and then ultimately to the stimulating electrode array that is inserted into subretinal space through a scleral flapcreated behind the eye. (c) Cross-sectional view of the eye and implant device. Note that only the electrode array penetrates thesclera and that the bulk of the implant components lie outside the eye.

7

appropriate patterned electrical stimulation. It ishere that the signal processing begins and isfurther integrated as it passes down the opticnerve on to the visual cortex for final perceptionof the visual image. All electronic parts arehermetically sealed in a titanium case connectedto an external flex circuit and the microelectronicarray (Kelly et al., 2009). To date, the group hassucceeded in developing a wireless retinalprosthesis prototype as the first step toward ahuman subretinal prosthesis implant. Initial stud-ies in animal models have been successful inimplanting active versions of the device andrefining surgical techniques and mechanicaldesign (Kelly et al., 2009). Human clinical trialsare now being planned.

Variations of the subretinal implant designhave also been pursued by several large consortiaefforts. The Artificial Silicone Retina (ASR)developed by Optobionics Corporation contains�5000 microphotodiodes, each containing itsown stimulating electrode (Chow et al., 2004).When implanted under the retina, photocurrentsgenerated by absorbed light stimulate adjacentretinal neurons in a multisite fashion. In a phase1 trial of safety and efficacy carried out in sixpatients with profound vision loss from RP(followed from 6 to 18 months after implanta-tion), patients reported an improvement in visualfunction after implantation. These reports wereevidenced by an increase in visual field size andthe ability to name more letters using astandardized visual acuity chart (Chow et al.,2004). While the relatively simple design of thisdevice was intuitively appealing (note that nocamera and subsequent image processing isrequired with this device), the apparent improve-ment in vision was not attributed to true prostheticvision per se, but rather to a potential neurotrophic(or “cell rescue”) effect related to microelectriccurrents generated by the device (Pardue et al.,2005a,b). With this limitation in mind, a multilay-ered subretinal chip device incorporating signalamplification is now being pursued by a Germanconsortium (Retina Implant AG). This device has

recently been implanted in profoundly blind RPpatients and recent results have been encouraging.Early human clinical trial data suggests that stablevisual percepts can be obtained and implantedpatients profoundly blind with RP have been ableto identify objects and letters (Besch et al., 2008;Zrenner, 2002).

As a contrasting design approach, the epiretinalstrategy entails placing an electrode array alongthe inner surface of the retina to stimulate theunderlying ganglion cells. This procedureemploys more typical vitreoretinal surgerytechniques so as to affix the microelectrode arrayon to the retinal surface (e.g., using a retinaltack). The Artificial Retina Project has beenactively pursued by a collaborative effort betweenthe Doheny Eye Institute (University of SouthernCalifornia) and Second Sight Medical Products.Like the Boston Retinal Implant design, thisdevice incorporates a digital camera mounted ona pair of eyeglass capturing an image that in turnis converted into an electrical signal that isdelivered to the retina (Humayun et al., 2003).Initial testing with a 16 electrode device(Argus I) in human volunteers with advancedRP has been successful. A large-scale multicen-tered phase II FDA-sponsored clinical trial is cur-rently underway to evaluate a second generationimplant (Argus II; 60 electrodes) in the largestcohort of visual prosthesis recipients to date.Results suggest that patients chronicallyimplanted with this device can detect phosphenesat individual electrodes, discriminate crudeshapes upon multiple electrode stimulation, andrecognize simple stimuli presented via a head-mounted camera (Humayun et al., 2009; Weilandet al., 2004). Very recently, the group reportedthat implanted subjects showed a significantimprovement in accuracy in a spatial visual-motortarget localization task comparing performance inpatients implanted with their second generationdevice. Subjects were instructed to locate andtouch a high contrast square target presented ona monitor. Nearly all subjects (26/27) showed asignificant improvement in accuracy (Ahuja

8

et al., 2011). This is consistent with the observa-tion that implanted subjects were able to developappropriate head-scanning techniques and good“camera-hand” coordination in using their visualprosthetic device (Ahuja et al., 2011).Other groups are also pursuing the epiretinal

approach including a variety of German-basedconsortiums. While still in earlier stages of devel-opment, early results have also been encouraging(Eckmiller et al., 2005; Gerding, 2007; Gerdinget al., 2007).Other notable downstream approaches have

been developed. A Belgian consortium hasdeveloped a prosthesis designed to stimulate theoptic nerve using a four-electrode cuff electrodedesign and driven by stimuli captured by anexternal camera (Fig. 1b). Two patients havebeen chronically implanted to date. Reports fromone blind volunteer demonstrated that electricalstimulation evoked the perception of localized,and often colored, phosphenes throughout thevisual field (Veraart et al., 2003). After 4 monthsof psychophysical testing, the patient couldrecognize and distinguish orientations of lines,some shapes and even certain letters (Brelenet al., 2005; Veraart et al., 2003).Finally, there have also been attempts to

deliver electrical stimulation to the visual cortexitself (Fig. 1c). Historically, this represents theoldest approach in developing a visualneuroprosthesis. By stimulating the visual cortexdirectly (thus bypassing earlier visual structures),this strategy has the appealing feature ofpotentially helping all forms of blindnessregardless of ocular pathology. Early seminalwork in a profoundly blind volunteer demonstra-ted that electrical stimulation delivered to the cor-tex (using surface electrodes) evoked theperception of discrete phosphenes (Brindley andLewin, 1968). While the phosphene perceptionswere rather crude, their spatial location approxi-mately corresponded to the known corticalretinotopic representation of visual space. Laterefforts incorporated a digital video cameramounted onto a pair of glasses interfaced with a

cortical stimulating array via a cable attached inthe patient's skull (Dobelle and Mladejovsky,1974). Several blind volunteers have beenimplanted and reportedly, one patient could dis-tinguish the outline of a person and identify theorientation of certain letters using this device(Dobelle et al., 1974). While certainly apioneering effort, the cortical approach still facesseveral technical challenges. These includedetermining the appropriate encoding strategiesthat are necessary to generate patterns ofstimulation, safety concerns due to the inherentinvasiveness of surgical implantation and the riskof focal seizures induced by direct cortical stimu-lation. However, new electrode designs (such asthe 100-electrode array developed at theUniversity of Utah; Normann et al., 1999) andadvances in wireless technology have stimulatedrenewed interest and several groups are furtherpursuing this approach (Fernandez et al., 2005;Normann et al., 2009; Tehovnik et al., 2009;Troyk et al., 2003, 2005).

Current technical challenges

As with all neuroprosthesis efforts, the develop-ment and realization of a visual neuroprostheticdevice will require continued and extensivecollaborative effort among basic scientists,engineers, and clinicians. Despite great technicalprogress, certain technical challenges are immedi-ately apparent and must be solved before a visualneuroprosthesis can be considered a viableclinical therapy (for further discussion see Chaderet al., 2009; Cohen, 2007a,b; Dagnelie andSchuchard, 2007; Winter et al., 2007).

For example, electrode geometry posesinherent limitations that must be carefully consid-ered. This is particularly true with regards to howclosely electrodes can be placed next to eachother thus impacting the theoretical resolutionthe visual prosthesis can provide. Further,electrode geometry is intimately related to theamount of current that can be delivered safely

9

to the target neuronal tissue (i.e., charge/densitylimits). As these neuroprosthetic devices aremeant to be implanted and used for very long-periods of time, the effect of prolonged and focalelectrical stimulation delivered to delicate (andeven potentially further degenerating) neuronaltissue remains unknown. In this direction, newelectrode designs, materials, and coatings arebeing actively pursued in order to improve andexpand safety profiles. One intriguing possibilityis the development of pillar electrode arrays.Implanting this electrode array design has shownthat penetrating pillars are able to attractneuronal elements (e.g., ganglion cells of theretina). The closer electrode–neuron interfacemay allow for lower currents to be used and thussafer injection of current for prolonged electricalstimulation (Butterwick et al., 2009).

There also exists the issue of how a capturedimage is coregistered with the natural movementof the eye. Inappropriate compensatory eyemovements may lead to perceptual mismatch,causing the patient wearing the implant tomislocalize objects in the external world. Thispotential confound is particularly true of implantdesigns that incorporate the use of an externalmounted camera. To solve this issue, sophisti-cated eye-tracking mechanisms have beenproposed and designed to generate appropriateshifts in the image (e.g., Palanker et al., 2005).However, these solutions await furtherdevelopment. Interestingly, recent work withvisual simulations suggests that following training,implanted patients may learn to carry outappropriate compensatory head and cameramovements to generate more stable percepts(Chen et al., 2009; Srivastava et al., 2009).

Identifying appropriate candidates for implanta-tion and determining the optimal placement ofthe implant are also of crucial importance(Merabet et al., 2007). Diagnostic techniquestypically found in the clinical setting (such as theelectroretinogram, visual evoked potentials, andvisual field perimetry) are certainly intuitivechoices to help characterize the profoundness of

an individual's visual impairment. Establishingpredictive testing methods that allow for cor-relations between objective measures of visualfunction and eventual implant success would behighly desirable (Bach et al., 2010; Dagnelie,2008). Along these lines, work has been done todevelop extensive methodologies aimed atdetermining “best candidates” for long-termimplantation of a microelectronic retinal implant(specifically defined as those requiring lowestcurrent levels delivered to the retina to elicit visualperceptions; Yanai et al., 2003). This includes aseries of preoperative visual, psychophysical, andelectrophysiological tests (including dark-adaptedbright flash and flicker electroretinograms andelectrical-evoked responses; Yanai et al., 2003).Novel applications of other imaging techniquessuch as functional magnetic resonance imaging(fMRI), diffusion tensor imaging (DTI), and trans-cranial magnetic stimulation (TMS) may alsoprove helpful for the direct evaluation of overallvisual cortical function and excitability (Fernandezet al., 2002; Merabet et al., 2007). With regard toretinal implants, the use of high-resolution opticalcoherence tomography (OCT) provides detailedanalysis and characterization of retinal laminaranatomy (Matsuo and Morimoto, 2007; Fig. 3a).This is particularly relevant in considering morerecent detailed anatomical findings indicating thatthere is extensive retinal reorganization ofcellular components and interconnections inpatients with longstanding retinal pathologies suchas RP (Marc et al., 2003). It would follow that adegenerating retina may respond verydifferently to electrical stimulation over time(Dagnelie, 2006). Implantation of a retinal pros-thesis during stages of complete photoreceptorloss, but with minimally altered inner retinal struc-ture, may prove beneficial in increasing the likeli-hood that a visual prosthesis will function. Thus,continued assessment of retinal structural andfunctional viability could assist in not only selectingappropriate candidates, but also identifying theoptimal location and timing of implantation as afunction of disease progression (Fig. 3b).

10

Moving forward: new challenges and remainingquestions

Experimental evidence from numerous groupshas demonstrated, at least in principle, that pat-terned electrical stimulation can evoke patternedlight perceptions. However, as human clinicaltrials expand and patients continue to interact

with these visual neuroprosthetic devices on amore long-term basis, we now move away fromthe goal of simply demonstrating proof of princi-ple toward establishing the fact that a visualneuroprosthesis can indeed improve the qualityof life of an implanted patient. In that direction,we also need to define what are tractablemilestones of success. Overall success can cer-tainly be interpreted differently particularly whentaken from the perspective of the device user.Moving forward, the implementation and poten-tial benefit of a visual prosthesis needs to consideroutcome measures and performance assessmentsthat translate directly into improvements in thequality of life of blind individuals (e.g., accuraterecognition and grasping of objects or skillfulnavigation in an unfamiliar environment andcarrying out activities of daily living; Dagnelie,2008). There is a clear need for new standardizedtesting and assessment of device efficacy that canbe quantified in a manner that is scientificallytestable and verifiable. In addition, the selectioncriteria for potential candidates must be clear.Not only would this allow for easier comparisonof results across design efforts but also establishand evaluate patients’ visual demands and needswithin the context of what a visual prosthesiscan ultimately deliver.

A review of human testing reveals thatimplanted recipients experience difficulties infully understanding the visual informationprovided by these visual prosthetic devices. Inthe initial studies, the reported patterns of visualpercepts often did not correspond to what waspredicted based on the patterns of electrical stim-ulation delivered (e.g., Rizzo et al., 2003a,b). Thiskey observation suggests that our intuitive senseas to how to generate patterned visual percepts(i.e., the “scoreboard approach”) may not proveto be an adequate strategy (Fernandez et al.,2005). This might be related to the fact thatstimulation is carried out on neuronal tissue thatis severely degenerated and, therefore, physiolog-ically compromised. Certainly, the quality ofvisual percepts is likely to improve as remaining

Fig. 3. Use of advanced imaging methodology to ascertainimplant positioning and retinal integrity. (a) Using opticalcoherence tomography (OCT), the cross-sectional position ofthe implanted microelectrode array (arrow) can be viewed insubretinal space. (b) Three-dimensional combined OCT withretinal microperimetry allows for simultaneous assessment ofstructure and function at specific points of retina. Theresulting topographic map (values indicate luminance levelsdetected in decibels; dB) could potentially be used forpostoperative evaluation as well as preoperative assessmentof candidate implant locations. Image generated using anOPKO/OTI combined optical coherence tomography andscanning laser ophthalmoscope with microperimetry feature(Opko Health, Inc. Miami, FL).

11

technical challenges continue to be solved.However, there may arrive a point whenengineering and surgical issues will no longerrepresent the greatest impediment to futureprogress. Rather, the greatest barrier will likelylie in our ignorance of how to introduce visualinformation that is meaningful to the visuallydeprived brain. It is a misconception that simpleperceptions generated from patterned light aresufficient to generate meaningful vision. Further,increasing the resolution of images (e.g., byincreasing the number of stimulating electrodes)with the goal of generating more complexperceptions would initially be perceptuallymeaningless rather than helpful.

Several studies have highlighted that followingthe loss of vision, the brain undergoes profoundneuroplastic transformation and that the occipitalvisual cortex is a major site of these changes(Bavelier and Neville, 2002; Fine et al., 2003;Merabet et al., 2005; Pascual-Leone et al., 1999).The extent and magnitude of these neuroplasticchanges is likely to be influenced by such factorsas the cause, onset, and duration of blindness.The plasticity of the visual system may allow fora considerable degree of adaptation. However,understanding the precise constraints of theseneuroplastic processes will be crucial and haveclear implications for rehabilitative training andprogress in device development (Fernandezet al., 2005).

A better understanding of how the brain adaptsto the loss of sight and how the remaining sensesprocess information in the visually deprivedbrain are necessary to appropriately modulaterestored visual input and, ultimately, to allowmeaningful vision with a neuroprosthetic device.One possibility might be to envision a patient-controlled system that coordinates and registersthe visual perceptions generated by a visualprosthesis with the identification of objectsperceived through other senses (such as touchand audition). Patients could learn to integratethese concordant sources of sensory stimuli intomeaningful percepts and, ultimately, gain the

ability to identify objects in the visual world(Merabet et al., 2005). Related studies in thedevelopment of sensory-substitution devices willlikely contribute greatly to our knowledge in thisarena. Further, these issues of training animplanted patient to “see again” are directlyrelated to the realm of visual rehabilitation. Theadaptive strategies and structured training neces-sary to interpret newly acquired visual perceptsthat ultimately translate to useful functionalvision should not be left to chance (Dagnelieand Schuchard, 2007). This should be carriedout within the context of a patient's current reha-bilitation program (such as sensory-substitutiondevices as well as mobility aids such as a guide-dog) to ensure an appropriate functional overlap.Clearly, any functional advantage gained throughthe use of a visual prosthetic device should meet,if not exceed, current rehabilitative options.Ultimately, the implementation of a visualprosthesis should not interfere with anindividual's on-going rehabilitative program.

Conclusion

The loss of sight can have a devastatinglynegative impact on the quality of life of anindividual. The goal of restoring functional visionto blind, while certainly valiant, still facesformidable challenges before it will ever becomea tractable reality. However, there are groundsto be cautiously optimistic and there is everyreason to believe we are on the path to achievethis goal. It is also important to realize that therehabilitation of the blind is a very complex prob-lem, requiring extraordinarily diverse, lengthy,and intimate collaborations among basicscientists, engineers, clinicians, educators, andrehabilitative experts.

As technical challenges continue to be solved,there also remains the issue of understandinghow the brain adapts to the loss of vision itself.Success in restoring functional vision depends onour understanding of how blindness affects the

12

brain and what it means to “see” again. Theneural changes that result from loss of vision needto be addressed if the restoration of visual input isto lead to functional vision. These issues ofneuroplasticity also lead to questions regardingthe feasibility of the visual prosthesis approachand its potential to benefit blind individuals. There-fore, it is essential that future research explores themechanisms that underlie brain plasticity followingthe loss of vision. Such insight could help to developand refine strategies for merging visual sensationsthat are generated by the prosthesis. Uncoveringthese adaptive strategies may ultimately assist inthe rehabilitation process itself.

References

Acland, G. M., Aguirre, G. D., Ray, J., Zhang, Q.,Aleman, T. S., Cideciyan, A. V., et al. (2001). Gene therapyrestores vision in a canine model of childhood blindness.Nature Genetics, 28, 92–95.

Ahuja, A. K., Dorn, J. D., Caspi, A., McMahon, M. J.,Dagnelie, G., Dacruz, L., Stanga, P., Humayun, M. S., &Greenberg, R. J. (2011). Blind subjects implanted with theArgus II retinal prosthesis are able to improve performancein a spatial-motor task. British Journal of Ophthalmology, 95(4), 539–543.

Allin, S., Eckel, E., Markham, H., & Brewer, B. R. (2010).Recent trends in the development and evaluation of assis-tive robotic manipulation devices. Physical Medicine andRehabilitation Clinics of North America, 21, 59–77.

Bach, M., Wilke, M., Wilhelm, B., Zrenner, E., & Wilke, R.(2010). Basic quantitative assessment of visual performancein patients with very low vision. Investigative Ophthalmologyand Visual Science, 51, 1255–1260.

Bach-y-Rita, P. (2004). Tactile sensory substitution studies.Annals of the New York Academy of Sciences, 1013, 83–91.

Bach-y-Rita, P., & Kercel, W. S. (2003). Sensory substitutionand the human-machine interface. Trends in CognitiveSciences, 7, 541–546.

Bavelier, D., & Neville, H. J. (2002). Cross-modal plasticity:Where and how? Nature Reviews Neuroscience, 3, 443–452.

Besch, D., Sachs, H., Szurman, P., Gulicher, D., Wilke, R.,Reinert, S., et al. (2008). Extraocular surgery for implanta-tion of an active subretinal visual prosthesis with externalconnections: Feasibility and outcome in seven patients. TheBritish Journal of Ophthalmology, 92, 1361–1368.

Brelen, M. E., Duret, F., Gerard, B., Delbeke, J., &Veraart, C. (2005). Creating a meaningful visual perception

in blind volunteers by optic nerve stimulation. Journal ofNeural Engineering, 2, S22–S28.

Brindley, G. S., & Lewin, W. S. (1968). The sensations pro-duced by electrical stimulation of the visual cortex. Journalof Physiology, 196, 479–493.

Bunker, C. H., Berson, E. L., Bromley, W. C., Hayes, R. P., &Roderick, T. H. (1984). Prevalence of retinitis pigmentosa inMaine. American Journal of Ophthalmology, 97, 357–365.

Butterwick, A., Huie, P., Jones, B. W., Marc, R. E.,Marmor, M., & Palanker, D. (2009). Effect of shape andcoating of a subretinal prosthesis on its integration withthe retina. Experimental Eye Research, 88, 22–29.

Chader, G. J., Weiland, J., & Humayun, M. S. (2009).Artificial vision: Needs, functioning, and testing of a retinalelectronic prosthesis. Progress in Brain Research, 175,317–332.

Chen, S. C., Suaning, G. J., Morley, J. W., & Lovell, N. H.(2009). Simulating prosthetic vision: II Measuring functionalcapacity. Vision Research, 49, 2329–2343.

Chow, A. Y., Chow, V. Y., Packo, K. H., Pollack, J. S.,Peyman, G. A., & Schuchard, R. (2004). The artificial siliconretina microchip for the treatment of vision loss fromretinitis pigmentosa. Archives of Ophthalmology, 122,460–469.

Cohen, E. D. (2007a). Prosthetic interfaces with the visual sys-tem: Biological issues. Journal of Neural Engineering, 4,R14–R31.

Cohen, E. D. (2007b). Safety and effectiveness considerationsfor clinical studies of visual prosthetic devices. Journal ofNeural Engineering, 4, S124–S129.

Craelius, W. (2002). The bionic man: Restoring mobility. Sci-ence, 295, 1018–1021.

Dagnelie, G. (2006). Visual prosthetics 2006: Assessment andexpectations. Expert Review of Medical Devices, 3, 315–325.

Dagnelie, G. (2008). Psychophysical evaluation for visual pros-thesis. Annual Review of Biomedical Engineering, 10,339–368.

Dagnelie, G., & Schuchard, R. A. (2007). The state of visualprosthetics—Hype or promise? Journal of RehabilitationResearch and Development, 44, xi–xiv.

Dobelle, W. H., & Mladejovsky, M. G. (1974). Phosphenesproduced by electrical stimulation of human occipital cortex,and their application to the development of a prosthesis forthe blind. Journal of Physiology, 243, 553–576.

Dobelle, W. H., Mladejovsky, M. G., & Girvin, J. P. (1974).Artificial vision for the blind: Electrical stimulation of visualcortex offers hope for a functional prosthesis. Science, 183,440–444.

Donoghue, J. P. (2002). Connecting cortex to machines:Recent advances in brain interfaces. Nature Neuroscience,5(Suppl.), 1085–1088.

Dowling, J. (2005). Artificial human vision. Expert Review ofMedical Devices, 2, 73–85.

13

Eckmiller, R., Neumann, D., & Baruth, O. (2005). Tunableretina encoders for retina implants: Why and how. Journalof Neural Engineering, 2, S91–S104.

Fernandez, E., Alfaro, A., Tormos, J. M., Climent, R.,Martinez, M., Vilanova, H., et al. (2002). Mapping of thehuman visual cortex using image-guided transcranial mag-netic stimulation. Brain Research. Brain Research Protocols,10, 115–124.

Fernandez, E., Pelayo, F., Romero, S., Bongard, M., Marin, C.,Alfaro, A., et al. (2005). Development of a cortical visualneuroprosthesis for the blind: The relevance ofneuroplasticity. Journal of Neural Engineering, 2, R1–R12.

Fine, I., Wade, A. R., Brewer, A. A., May, M. G.,Goodman, D. F., Boynton, G. M., et al. (2003). Long-termdeprivation affects visual perception and cortex. NatureNeuroscience, 6, 915–916.

Gerding, H. (2007). A new approach towards a minimal inva-sive retina implant. Journal of Neural Engineering, 4,S30–S37.

Gerding, H., Benner, F. P., & Taneri, S. (2007). Experimentalimplantation of epiretinal retina implants (EPI-RET) withan IOL-type receiver unit. Journal of Neural Engineering,4, S38–S49.

Gothe, J., Brandt, S. A., Irlbacher, K., Roricht, S.,Sabel, B. A., & Meyer, B. U. (2002). Changes in visual cor-tex excitability in blind subjects as demonstrated by trans-cranial magnetic stimulation. Brain, 125, 479–490.

Hochberg, L. R., Serruya, M. D., Friehs, G. M., Mukand, J. A.,Saleh, M., Caplan, A. H., et al. (2006). Neuronal ensemblecontrol of prosthetic devices by a human with tetraplegia.Nature, 442, 164–171.

Humayun, M. S. (2007). Artificial sight: Basic research, bio-medical engineering, and clinical advances. New York:Springer.

Humayun, M. S., de Juan, E., Jr, Dagnelie, G.,Greenberg, R. J., Propst, R. H., & Phillips, D. H. (1996).Visual perception elicited by electrical stimulation ofretina in blind humans. Archives of Ophthalmology, 114,40–46.

Humayun, M. S., de Juan, E., Jr, Weiland, J. D., Dagnelie, G.,Katona, S., Greenberg, R., et al. (1999). Pattern electricalstimulation of the human retina. Vision Research, 39,2569–2576.

Humayun, M. S., Dorn, J. D., Ahuja, A. K., Caspi, A.,Filley, E., Dagnelie, G., et al. (2009). Preliminary 6 monthresults from the Argus II epiretinal prosthesis feasibilitystudy. Conference Proceedings of IEEE Engineering in Med-icine and Biology Society, 2009, 4566–4568.

Humayun, M. S., Weiland, J. D., Fujii, G. Y., Greenberg, R.,Williamson, R., Little, J., et al. (2003). Visual perceptionin a blind subject with a chronic microelectronic retinalprosthesis. Vision Research, 43, 2573–2581.

Javaheri, M., Hahn, D. S., Lakhanpal, R. R., Weiland, J. D., &Humayun, M. S. (2006). Retinal prostheses for the blind.Annals of the Academy of Medicine, Singapore, 35, 137–144.

Jones, S.,Harris,D.,Estill,A.,&Mikulec,A.A. (2008). Implant-able hearing devices.Missouri Medicine, 105, 235–239.

Kelly, S. K., Shire, D. B., Chen, J., Doyle, P., Gingerich, M. D.,Drohan, W. A., et al. (2009). Realization of a 15-channel,hermetically-encased wireless subretinal prosthesis for theblind. Conference Proceedings of IEEE Engineering in Med-icine and Biology Society, 2009, 200–203.

Klein, R., Klein, B. E., Jensen, S. C., & Meuer, S. M. (1997).The five-year incidence and progression of age-relatedmaculopathy: The Beaver Dam Eye Study. Ophthalmology,104, 7–21.

Laferrier, J. Z., & Gailey, R. (2010). Advances in lower-limbprosthetic technology. Physical Medicine and RehabilitationClinics of North America, 21, 87–110.

Loeb, G. E. (1990). Cochlear prosthetics. Annual Review ofNeuroscience, 13, 357–371.

MacLaren, R. E., Pearson, R. A., MacNeil, A., Douglas, R. H.,Salt, T. E., Akimoto, M., et al. (2006). Retinal repair bytransplantation of photoreceptor precursors. Nature, 444,203–207.

Marbach, W. D., Zabarsky, M., Hoban, P., & Nelson, C.(1982). Building the bionic man. Newsweek, 100, 78–79.

Marc, R. E., Jones, B. W., Watt, C. B., & Strettoi, E. (2003).Neural remodeling in retinal degeneration. Progress in Reti-nal and Eye Research, 22, 607–655.

Marg, E., & Rudiak, D. (1994). Phosphenes induced by mag-netic stimulation over the occipital brain: Description andprobable site of stimulation. Optometry and Vision Science,71, 301–311.

Matsuo, T., & Morimoto, N. (2007). Visual acuity and per-imacular retinal layers detected by optical coherence tomog-raphy in patients with retinitis pigmentosa. The BritishJournal of Ophthalmology, 91, 888–890.

Merabet, L. B., Rizzo, J. F., Amedi, A., Somers, D. C., &Pascual-Leone, A. (2005). What blindness can tell us aboutseeing again: Merging neuroplasticity and neuroprostheses.Nature Reviews Neuroscience, 6, 71–77.

Merabet, L. B., Rizzo, J. F., 3rdPascual-Leone, A., &Fernandez, E. (2007). 'Who is the ideal candidate?': Dec-isions and issues relating to visual neuroprosthesis develop-ment, patient testing and neuroplasticity. Journal of NeuralEngineering, 4, S130–S135.

Nicolelis, M. A. (2003). Brain-machine interfaces to restoremotor function and probe neural circuits. Nature ReviewsNeuroscience, 4, 417–422.

Normann, R. A., Greger, B., House, P., Romero, S. F.,Pelayo, F., & Fernandez, E. (2009). Toward the develop-ment of a cortically based visual neuroprosthesis. Journalof Neural Engineering, 6, 035001.

14

Normann, R. A., Maynard, E. M., Rousche, P. J., &Warren, D. J. (1999). A neural interface for a cortical visionprosthesis. Vision Research, 39, 2577–2587.

Palanker, D., Vankov, A., Huie, P., & Baccus, S. (2005).Design of a high-resolution optoelectronic retinal prosthesis.Journal of Neural Engineering, 2, S105–S120.

Pardue, M. T., Phillips, M. J., Yin, H., Fernandes, A.,Cheng, Y., Chow, A. Y., et al. (2005). Possible sources ofneuroprotection following subretinal silicon chip implanta-tion in RCS rats. Journal of Neural Engineering, 2, S39–S47.

Pardue, M. T., Phillips, M. J., Yin, H., Sippy, B. D., Webb-Wood, S., Chow, A. Y., et al. (2005). Neuroprotective effectof subretinal implants in the RCS rat. Investigative Ophthal-mology and Visual Science, 46, 674–682.

Pascual-Leone, A., Hamilton, R., Tormos, J. M., Keenan, J. P.,& Catala, M. D. (1999). Neuroplasticity in the adjustment toblindness. In J. Grafman & Y. Christen (Eds.), Neuronalplasticity: Building a bridge from the laboratory to the clinic.Berlin Heidelberg, New York: Springer-Verlag.

Peterman, M. C., Noolandi, J., Blumenkranz, M. S., &Fishman, H. A. (2004). Localized chemical release from anartificial synapse chip. Proceedings of the National Academyof Sciences of the United States of America, 101, 9951–9954.

Rizzo, J. F., 3rd, Wyatt, J., Loewenstein, J., Kelly, S., &Shire, D. (2003a). Methods and perceptual thresholds forshort-term electrical stimulation of human retina with micro-electrode arrays. Investigative Ophthalmology and VisualScience, 44, 5355–5361.

Rizzo, J. F., 3rd, Wyatt, J., Loewenstein, J., Kelly, S., &Shire, D. (2003b). Perceptual efficacy of electrical stimula-tion of human retina with a microelectrode array duringshort-term surgical trials. Investigative Ophthalmology andVisual Science, 44, 5362–5369.

Shire, D. B., Kelly, S. K., Chen, J., Doyle, P., Gingerich, M. D.,Cogan, S. F., et al. (2009). Development and implantation ofa minimally invasive wireless subretinal neurostimulator.IEEE Transactions on Biomedical Engineering, 56,2502–2511.

Srivastava, N. R., Troyk, P. R., & Dagnelie, G. (2009).Detection, eye-hand coordination and virtual mobility per-formance in simulated vision for a cortical visual prosthesisdevice. Journal of Neural Engineering, 6, 035008.

Tehovnik, E. J., Slocum, W. M., Smirnakis, S. M., &Tolias, A. S. (2009). Microstimulation of visual cortex torestore vision. Progress in Brain Research, 175, 347–375.

Troyk, P., Bak, M., Berg, J., Bradley, D., Cogan, S.,Erickson, R., et al. (2003). A model for intracortical visualprosthesis research. Artificial Organs, 27, 1005–1015.

Troyk, P. R., Bradley, D., Bak, M., Cogan, S., Erickson, R.,Hu, Z., et al. (2005). Intracortical visual prosthesisresearch—Approach and progress. Conference Proceedingsof IEEE Engineering in Medicine and Biology Society, 7,7376–7379.

Veraart, C., Wanet-Defalque, M. C., Gerard, B.,Vanlierde, A., & Delbeke, J. (2003). Pattern recognitionwith the optic nerve visual prosthesis. Artificial Organs, 27,996–1004.

Weiland, J. D., Yanai, D., Mahadevappa, M., Williamson, R.,Mech, B. V., Fujii, G. Y., et al. (2004). Visual task perfor-mance in blind humans with retinal prosthetic implants.Conference Proceedings of IEEE Engineering in Medicineand Biology Society, 6, 4172–4173.

Winter, J. O., Cogan, S. F., & Rizzo, J. F. 3rd. (2007). Retinalprostheses: Current challenges and future outlook. Journalof Biomaterials Science, Polymer Edition, 18, 1031–1055.

Yanai, D., Lakhanpal, R. R., Weiland, J. D.,Mahadevappa, M., Van Boemel, G., Fujii, G. Y., et al.(2003). The value of preoperative tests in the selection ofblind patients for a permanent microelectronic implant.Transactions of the American Ophthalmological Society,101, 223–228, discussion 228–230.

Zrenner, E. (2002). The subretinal implant: Canmicrophotodiode arrays replace degenerated retinalphotoreceptors to restore vision? Ophthalmologica, 216(Suppl. 1), 8–20, discussion 52–53.

15