Jim Loudin

Optoelectronic retinal prosthesis: system design and performance.

The design of high-resolution retinal prostheses presents many unique engineering and biological challenges. Ever smaller electrodes must inject enough charge to stimulate nerve cells, within electrochemically safe voltage limits. Stimulation sites should be placed within an electrode diameter from the target cells to prevent blurring and minimize current. Signals must be delivered wirelessly from an external source to a large number of electrodes, and visual information should, ideally, maintain its natural link to eye movements. Finally, a good system must have a wide range of stimulation currents, external control of image processing and the option of either anodic-first or cathodic-first pulses. This paper discusses these challenges and presents solutions to them for a system based on a photodiode array implant. Video frames are processed and imaged onto the retinal implant by a head-mounted near-to-eye projection system operating at near-infrared wavelengths. Photodiodes convert light into pulsed electric current, with charge injection maximized by applying a common biphasic bias waveform. The resulting prosthesis will provide stimulation with a frame rate of up to 50 Hz in a central 10 degrees visual field, with a full 30 degrees field accessible via eye movements. Pixel sizes are scalable from 100 to 25 microm, corresponding to 640-10,000 pixels on an implant 3 mm in diameter.

Strength–Duration Relationship for Extracellular Neural Stimulation: Numerical and Analytical Models

The strength-duration relationship for extracellular stimulation is often assumed to be similar to the classical intracellular stimulation model, with a slope asymptotically approaching 1/τ at pulse durations shorter than chronaxy. We modeled extracellular neural stimulation numerically and analytically for several cell shapes and types of active membrane properties. The strength-duration relationship was found to differ significantly from classical intracellular models. At pulse durations between 4 μs and 5 ms stimulation is dominated by sodium channels, with a slope of -0.72 in log-log coordinates for the Hodgkin-Huxley ion channel model. At shorter durations potassium channels dominate and slope decreases to -0.13. Therefore the charge per phase is decreasing with decreasing stimulus duration. With pulses shorter than cell polarization time (∼0.1-1 μs), stimulation is dominated by polarization dynamics with a classical -1 slope and the charge per phase becomes constant. It is demonstrated that extracellular stimulation can have not only lower but also upper thresholds and may be impossible below certain pulse durations. In some regimes the extracellular current can hyperpolarize cells, suppressing rather than stimulating spiking behavior. Thresholds for burst stimuli can be either higher or lower than that of a single pulse, depending on pulse duration. The modeled thresholds were found to be comparable to published experimental data. Electroporation thresholds, which limit the range of safe stimulation, were found to exceed stimulation thresholds by about two orders of magnitude. These results provide a biophysical basis for understanding stimulation dynamics and guidance for optimizing the neural stimulation efficacy and safety.

A curvable silicon retinal implant

We have developed a curvable photovoltaic monolithic retinal implant that requires no electrical power or data connection. The implant consists of a two-dimensional network of miniature silicon solar cells that directly stimulate the retina when illuminated by a goggle system. A MEMS process isolates adjacent pixels and makes the arrays curvable allowing them to conform to the shape of the retina.

Electronic enhancement of tear secretion.

OBJECTIVEnTo study electrical stimulation of the lacrimal gland and afferent nerves for enhanced tear secretion, as a potential treatment for dry eye disease. We investigate the response pathways and electrical parameters to safely maximize tear secretion.nnnAPPROACHnWe evaluated the tear response to electrical stimulation of the lacrimal gland and afferent nerves in isofluorane-anesthetized rabbits. In acute studies, electrical stimulation was performed using bipolar platinum foil electrodes, implanted beneath the inferior lacrimal gland, and a monopolar electrode placed near the afferent ethmoid nerve. Wireless microstimulators with bipolar electrodes were implanted beneath the lacrimal gland for chronic studies. To identify the response pathways, we applied various pharmacological inhibitors. To optimize the stimulus, we measured tear secretion rate (Schirmer test) as a function of pulse amplitude (1.5-12 mA), duration (0.1-1 ms) and repetition rate (10-100 Hz).nnnMAIN RESULTSnStimulation of the lacrimal gland increased tear secretion by engaging efferent parasympathetic nerves. Tearing increased with stimulation amplitude, pulse duration and repetition rate, up to 70 Hz. Stimulation with 3 mA, 500 μs pulses at 70 Hz provided a 4.5 mm (125%) increase in Schirmer score. Modulating duty cycle further increased tearing up to 57%, compared to continuous stimulation in chronically implanted animals (36%). Ethmoid (afferent) nerve stimulation increased tearing similar to gland stimulation (3.6 mm) via a reflex pathway. In animals with chronically implanted stimulators, a nearly 6 mm increase (57%) was achieved with 12-fold less charge density per pulse (0.06-0.3 μC mm(-2) with 170-680 μs pulses) than the damage threshold (3.5 μC mm(-2) with 1 ms pulses).nnnSIGNIFICANCEnElectrical stimulation of the lacrimal gland or afferent nerves may be used as a treatment for dry eye disease. Clinical trials should validate this approach in patients with aqueous tear deficiency, and further optimize electrical parameters for maximum clinical efficacy.

High resolution optoelectronic retinal prosthesis

Electronic retinal prostheses seek to restore sight in patients with retinal degeneration by delivering pulsed electric currents to retinal neurons via an array of microelectrodes. Most implants use inductive or optical transmission of information and power to an intraocular receiver, with decoded signals subsequently distributed to retinal electrodes through an intraocular cable. Surgical complexity could be minimized by an integrated prosthesis, in which both power and data are delivered directly to the stimulating array without any discrete components or cables. We present here an integrated retinal prosthesis system based on a photodiode array implant. Video frames are processed and imaged onto the retinal implant by a video goggle projection system operating at near-infrared wavelengths (~ 900 nm). Photodiodes convert light into pulsed electric current, with charge injection maximized by specially optimized series photodiode circuits. Prostheses of three different pixel densities (16 pix/mm2, 64 pix/mm2, and 256 pix/mm2) have been designed, simulated, and prototyped. Retinal tissue response to subretinal implants made of various materials has been investigated in RCS rats. The resulting prosthesis can provide sufficient charge injection for high resolution retinal stimulation without the need for implantation of any bulky discrete elements such as coils or tethers. In addition, since every pixel functions independently, pixel arrays may be placed separately in the subretinal space, providing visual stimulation to a larger field of view.

Strength-duration relationship for extracellular neural stimulation

Understanding the mechanisms and dynamics of extracellular neural stimulation is very important for the development of electro-neural interfaces in general, and for the design of stimulation waveforms and electrode configurations for neural prosthetic implants, in particular. The most common type of neural stimulation is extracellular, and yet its mechanisms and dynamics have scarcely been explored and described. Its strength-duration relationship is often assumed to be similar to the classical intracellular dependence, with a slope asymptotically approaching 1/τ at pulse durations τ shorter than chronaxy. The current study explores the basic mechanisms of extracellular stimulation and derives its strength-duration curve in a wide range of stimulus durations for various waveforms and cell shapes. We used two different models of active membrane properties: the Hodgkin-Huxley model of the squid giant axon [1], and a six-channel salamander retinal ganglion cell model [2]. Three cell geometries were analyzed: an idealized planar cell with two uniformly polarized flat surfaces, and more realistic spherical and cylindrical shapes corresponding to the soma and unmyelinated axon or axon hillock. The strength-duration relationship was found to differ significantly from classical intracellular models. For the Hodgkin-Huxley model at pulse durations between 4 μs and 5 ms stimulation is dominated by sodium channels, and has a slope of approximately –0.72 in log-log coordinates. At shorter durations it is dominated by the potassium channels, and has a much lower slope of about –0.13. With pulses shorter than cell polarization time (typically about 0.1-1 μs), it is dominated by polarization dynamics, and asymptotically approaches the classical –1 slope. For retinal ganglion cells we demonstrate that extracellular stimulation can have not only lower but also upper thresholds, and may be impossible below certain pulse durations. For both cell models we have found that in some stimulation regimes the stimulus can hyperpolarize cells, suppressing rather than stimulating spiking behavior. Thresholds for burst stimuli can be either higher or lower than that of a single pulse, depending on pulse duration. The modeled thresholds were found to be comparable to published experimental data obtained with rabbit retinal ganglion cells. These results provide a biophysical basis for understanding stimulation dynamics, and guidance for optimizing the efficacy and safety of extracellular neural stimulation.

Progress Towards a High-Resolution Retinal Prosthesis

Electronic retinal prostheses represent a potentially effective approach for restoring some degree of sight in blind patients with retinal degeneration. Functional restoration of sight would require hundreds to thousands of electrodes effectively stimulating remaining neurons in the retina. We present a design of an optoelectronic retinal prosthetic system having 3mm diameter retinal implant with pixel sizes down to 25 micrometers, which allows for natural eye scanning for observing a large field of view, as well as spatial and temporal processing of the visual scene to optimize the patient experience. Information from a head mounted video camera is processed in a portable computer and delivered to the implanted photodiode array by projection from the LCD goggles using pulsed IR (810 nm) light. Each photodiode converts pulsed light (0.5 ms in duration) into electric current with efficiency of 0.3 A/W using common bi-phasic power line. Power is provided by the inductively-coupled RF link from the coil on the goggles into a miniature power supply implanted between the sclera and the conjuctiva, and connected to subretinal implant with a thin 2-wire trans-scleral cable. 3-dimensional structures in the subretinal prosthesis induce retinal migration and thus ensure close proximity between stimulating electrodes and the target retinal neurons. Subretinal implantations of the 3-dimentional pillar and chamber arrays in RCS rats with 2 and 6 week follow-up demonstrate achievement of intimate proximity between the stimulation cites and the inner nuclear layer. In some instances formation of a fibrotic seal has been observed.