High-density Electrode Arrays Research Articles

Microelectronic packaging for retinal prostheses

An innovative parylene-based high-density chip-level integrated interconnect (CL-I/sup 2/) packaging system for retinal implants id discussed. The implications of this CL-I/sup 2/ technology for retinal prosthesis packaging effort are far-reaching. This technology obviates the need for a technician to create electrical and mechanical connections one by one. Instead, the technology is limited only by standard photolithography and standard microfabrication techniques, providing the capability for a reduction of an order of magnitude or more in the center-to-center pad distances that can be accommodated in the process. High-density electrode arrays are thereby feasible, because the fabrication process places no limit on the number of output pads that can reasonably be connected to the array.

Development of flexible arrays for in vivo neuronal recording and stimulation

Recent developments in low-power electronics and semiconductor fabrication techniques have found many applications in the life sciences. High-density electrode arrays are becoming well established as tools for the measurement of neuronal signals. The fabrication of arrays on flexible materials allows for 2D position sensitive recording of cellular activity in vivo and for the possibility of direct in vivo stimulus. Using flexible polymer materials, compliant with semiconductor fabrication techniques, we demonstrate a process allowing the fabrication of flexible multi-site microelectrode neuronal recording and stimulating arrays. We describe the development of both 8 and 61 electrode arrays on polyimide substrates with 50 and 5 μ m minimum linewidths respectively. Further studies have realised 8-electrode arrays using gold on Polydimethylsiloxane (PDMS), an alternative biocompatible material, with linewidths of 14 μ m . Implementing low noise amplification, 2.6 μ V rms (bandpass typically 80–2000 Hz), the polyimide 8-electrode arrays have been used to record electroretinogram and ganglion cell action potentials in situ from the frog retina ( Rana temporaria). Such arrays coupled to pixellated CMOS sensors, incorporating on-board neural networking should allow for the recovery of basic functionality in the human retina. More specifically, where retinal degeneration has affected only the photosensitive elements of the eye we can utilise the remaining neuronal pathways. Initial stimulation studies for electro-deposited platinum electrodes of 4 nA/ μ m 2 indicate upper breakdown limits for charge density approaching 40 μ C m - 2 . Investigations of lifetime stimulation of a 50 μ m diameter electrode, of typical impedance less than 20 k Ω at 1 kHz, suggest operational limits over lifetime in the order of 10 μ C m - 2 . These charge densities are adequate for neuronal cell stimulation.

Polytrodes: High-Density Silicon Electrode Arrays for Large-Scale Multiunit Recording

We developed a variety of 54-channel high-density silicon electrode arrays (polytrodes) designed to record from large numbers of neurons spanning millimeters of brain. In cat visual cortex, it was possible to make simultaneous recordings from >100 well-isolated neurons. Using standard clustering methods, polytrodes provide a quality of single-unit isolation that surpasses that attainable with tetrodes. Guidelines for successful in vivo recording and precise electrode positioning are described. We also describe a high-bandwidth continuous data-acquisition system designed specifically for polytrodes and an automated impedance meter for testing polytrode site integrity. Despite having smaller interconnect pitches than earlier silicon-based electrodes of this type, these polytrodes have negligible channel crosstalk, comparable reliability, and low site impedances and are capable of making high-fidelity multiunit recordings with minimal tissue damage. The relatively benign nature of planar electrode arrays is evident both histologically and in experiments where the polytrode was repeatedly advanced and retracted hundreds of microns over periods of many hours. It was possible to maintain stable recordings from active neurons adjacent to the polytrode without change in their absolute positions, neurophysiological or receptive field properties.

Active positioning device for a perimodiolar cochlear electrode array

The authors report on an actuated positioning device for a cochlear prosthesis insertion procedure. The device consists of multiple high-aspect ratio fluidic actuator chambers integrated with a high-density silicon cochlear electrode array and manufactured in a tapered-helix form. Actuation chambers with cross-sectional sizes as small as 40 × 200 μm and lengths of 30 mm have been fabricated using flexible polymers. The device will allow for low-resistance basilar insertion of a stimulating electrode array into the cochlea providing for deep, perimodiolar position considered most beneficial for auditory nerve stimulation, while minimizing intracochlear trauma. Experimental measurements, FEA analysis, and modeling demonstrate a viable and appropriate actuation method for a cochlear implant procedure.

High-density electrode array for imaging in vitro electrophysiological activity

The development of a high-density active microelectrode array for in vitro electrophysiology is reported. Based on the Active Pixel Sensor (APS) concept, the array integrates 4096 gold microelectrodes (electrode separation 20 microm) on a surface of 2.5 mmx2.5 mm as well as a high-speed random addressing logic allowing the sequential selection of the measuring pixels. Following the electrical characterization in a phosphate solution, the functional evaluation has been carried out by recording the spontaneous electrical activity of neonatal rat cardiomyocytes. Signals with amplitudes from 130 microVp-p to 300 microVp-p could be recorded from different pixels. The results demonstrate the suitability of the APS concept for developing a new generation of high-resolution extracellular recording devices for in vitro electrophysiology.

A comparison of chronic multi-channel cortical implantation techniques: manual versus mechanical insertion

High-density multi-channel intra-cortical electrode arrays allow researchers to record simultaneously from populations of neurons for the purpose of understanding neural coding and plasticity. These devices have tens to hundreds of electrodes spaced within a few square millimeters. During insertion, the high-density probes can compress the cortex several millimeters prior to breaking through the pia. Compression of cortical tissue has been demonstrated to result in traumatic brain injury (TBI) which may be a major contributor to low electrode yield and decreased recording longevity. Two insertion techniques for chronically implanting multi-wire electrode arrays in layer IV of primary auditory cortex were compared. A mechanical insertion device, capable of rapidly inserting the electrode array without visible compression of the brain, was constructed. The neural responses to broadband clicks and pure tones recorded from the arrays inserted with the mechanical device were compared to the results from a manual insertion method using a micromanipulator. Both techniques result in a similar number of active channels directly following surgery with a mean signal-to-noise ratio of approximately 4.5. Over 60% of the animals implanted with the mechanical insertion device had driven activity at week 6 whereas none of the animals with manually inserted arrays exhibited functional responses after 3 weeks. This report provides initial evidence that mechanical insertion devices, which prevent cortical compression, increase electrode recording longevity.

Attentional modulation of the human somatosensory evoked potential in a trial-by-trial spatial cueing and sustained spatial attention task measured with high density 128 channels EEG

We investigated the modulation of the somatosensory evoked potential (SEP) elicited by mechanical stimuli in a spatial sustained attention and a spatial trial-by-trial cueing design by means of high density electrode array EEG recordings. Subjects were instructed to detect rare tactile target stimuli at the to-be-attended hand while ignoring stimuli at the other hand. Analysis of the SEP revealed a highly complex pattern of results. The P50 component was significantly increased for attended stimuli in the sustained attention as opposed to the trial-by-trial cueing condition. However, no difference in amplitude was found for attended as opposed to unattended stimuli. High density electrode array recordings revealed a centero-frontal N140 component (N140c), which preceded the parietal N140 (N140p) by about 20 ms. The N140c exhibited an attention effect in particular in the trial-by-trial spatial cueing condition. The N140p was significantly enlarged with attention across both experimental conditions, but a closer inspection demonstrated that this was mainly due to the great attention effect in the trial-by-trial spatial cueing condition. The late positive component (190–380 ms after stimulus onset) exhibited a significant attention effect in both experimental conditions. The present experiment provides evidence that the attentional modulation of the SEP is different when tactile as opposed to electrical stimuli were used and when only somatosensory stimuli are presented with no further sensory stimulation in other modalities. Furthermore, transient as opposed to sustained spatial attention affected various components of the SEP in a different way.

Positioning living cells on a high-density electrode array by negative dielectrophoresis

We present in this paper a large-scale microelectrode array, which allows a dielectrophoretic positioning of cells in a matrix form. The electrode structure was chosen to produce regularly spaced field minima, toward which particles are directed and concentrated, under conditions of negative DEP. The need to power thousands of electrodes at the same time guided the choice of a multilayer structure. Cells can also be directed toward small wells formed in silicon, where they remain when the field is removed.

A Neurobiological Theory of Meaning in Perception Part II: Spatial Patterns of Phase in Gamma EEGs from Primary Sensory Cortices Reveal the Dynamics of Mesoscopic Wave Packets

Domains of cooperative neural activity called "wave packets" have been discovered in the visual, auditory, and somatomotor cortices of rabbits that were trained to discriminate conditioned stimuli in these modalities. Each domain forms by a first order state transition, which strongly resembles a phase transition from vapor to liquid. In this view, raw sense data injected into cortex by sensory axons drive cortical action potentials in swarms like water molecules in steam. The increased activity destabilizes the cortex. Within 3 to 7 milliseconds of transition onset, the activity binds together into a state resembling a scintillating rain drop, which lasts ~80 to 100 milliseconds, then dissolves. Wave packets form at rates of 2 to 7/second in all sensory areas, overlapping in space and time. Results of sensory information processing are seen in spatial patterns of amplitude modulation (AM) of wave packets with carrier waves in the gamma range (20 to 80 Hz in rabbits). The AM patterns correspond to categories of CSs that the rabbits can discriminate. The patterns are found in electroencephalographic (EEG) potentials generated by dendrites and recorded with high-density electrode arrays. The state transitions by which AM patterns form are manifested in the spatial pattern of phase modulation (PM), which have the radial symmetry of a cone. The apex of a PM cone marks the site of nucleation of an AM pattern. The phase gradient gives a soft boundary condition, where the axonal delay in spread gives sufficient phase dispersion to reach the half-power level. The size of the wave packets (10 to 30 mm in diameter in rabbits) is determined largely by the conduction velocities of intracortical axons through which the neural cooperation is maintained. The findings show that significant cortical activity takes the form of mesoscopic interactions of millions of neurons in broad areas of cortex, which are more clearly detected in graded dendritic potentials than in action potentials. The distinction is analogous to the difference between statistical mechanical and thermodynamic descriptions of particle behavior. Both types of neural activity show spatial and temporal discontinuities but at distinctive scales of microns and msec versus mm and tenths of a second. The aim of measurement here is to establish the wave packet as the information carrier at the mesoscopic level in brain dynamics, comparable to the role of the action potential as the information carrier at the microscopic level in neuron dynamics.

Delayed activation and retrograde propagation in cardiac muscle: implication of virtual electrode effects.

Point cathodal stimulation of cardiac tissue was shown previously to produce both a dog-bone shaped virtual cathode transverse to the muscle fibers and two longitudinal virtual anodes. We hypothesize that virtual anodes can cause a region of delayed activation, separating two regions of early activation caused by the virtual cathode. Using a high-density electrode array in 42 superfused epicardial slices from 14 canine left ventricles, we observed regions of early and delayed activation and different pathways of retrograde propagation corresponding to the earlier patterns. Retrograde propagation was seen from the transversely located early activation area through areas of delayed activation toward the cathode, and from the early activation area toward the cathode directly. These pathways caused a wide dispersion in the direction of retrograde propagation (2 degrees +/- 31 degrees, n = 179, relative to the fast axis of threshold activation; radial velocity: 0.5 +/- 0.2 m/s, n = 95, in 12 slices from 8 hearts with stimuli of 330 micros, 0.8-30 mA). Delayed activations were observed 0 degrees +/- 6 degrees (n = 32) from the axis in 23 maps (at differing stimulation strengths) recorded in 13 slices from 10 hearts. We conclude that point cathodal stimulation induce delayed activation along the fiber axis and retrograde propagation both along and transverse to the axis.

A simulation study evaluating the performance of high-density electrode arrays on myocardial tissue.

Multielectrode arrays used to detect cellular activation have become so dense (electrodes per square millimeter) as to jeopardize the basic assumptions of activation mapping; namely, that electrodes are points adequately separated as to not interfere with the tissue or each other. This paper directly tests these assumptions for high-density electrode arrays. Using a finite element model with modified Fitzhugh-Nagumo kinetics, we represent electrodes as isopotential surfaces of varying widths and spacing ratio (SR) (center-to-center spacing divided by electrode width). We examine the signal strength and ability of a single electrode to detect activation due to a passing wavefront. We find that high-density arrays do not cause significant wavefront curvature or alter activation timing in the underlying tissue. Relationships between signal strength, cross talk, and array design are explained by the interaction of the propagating wavefront and induced sources on the isopotential electrodes. Sensitivity analysis shows that these results may be generalized to a wide range of physiologically relevant designs and applications. We conclude that electrode array designs in which electrode spacing greatly exceeds electrode diameter are overly conservative and that arrays with a SR of less than 2.0 may perform successfully in electrophysiological studies.

Construction of a very high-density extracellular electrode array.

Cellular activation mapping (specifying in time and space the electrical activation sequence of cells) is a well-established basic research tool in cardiac, neural, and gastric physiology. Much recent research in cardiac mapping has focused on large arrays (>200 electrodes) with small electrodes (<500 microm). Construction of such arrays using standard techniques is tedious and yields irregular electrode spacing. We present a novel construction technique that rapidly produces large arrays with regularly spaced small electrodes. For methods, fine-pitch copper ribbon cables, insulated with either polyvinylchloride (PVC) or polyimide (flexible printed circuit; FPC), were assembled together such that the active surface was the cut end of the cable. The cut end was sanded and polished, then coated with silver and sometimes silver chloride. Once completed, the alternating current (AC) root-mean-square (rms) potential was measured between two adjacent, individual electrodes. Polarization testing was conducted according to a previously reported protocol (Witkowski FX and Penkoske PA. J Electrocardiol 21: 273-282, 1988). Activation mapping was conducted in the open-chest guinea pig with both pacing- and defibrillation- strength stimuli. In terms of results, four PVC and three FPC arrays were constructed, ranging from 4 to 400 electrodes. Two hours of labor were needed to create a complete electrode array, independent of the number of electrodes, including connectors and silver/silver chloride coating. As expected, the addition of a silver/silver chloride coating significantly reduced (0.76-0.42 mV, P < 0.001) the AC rms potential difference between two electrodes. A nearly immediate recovery of the potential difference between adjacent pairs of silver/silver chloride electrodes was observed after defibrillation stimuli.

Endocardial mapping of ventricular tachycardia in the intact human ventricle: Evidence for reentrant mechanisms

A balloon array of 112 electrodes was used to obtain simultaneous recordings of endocardial electrograms during intraoperative mapping studies of ventricular tachycardia. Introduction of the balloon through a left atriotomy and across the mitral valve allowed endocardial activation maps to be obtained in the intact left ventricie. Of 20 patients with coronary artery disease studied in this way, suggestive evidence of endocardial reentry was found in 6. Three separate reentrant mechanisms were observed. In two patients, a single broad wave front of continuous recirculating activation reminiscent of a vortex was initiated by the formation of a functional arc of block in response to premature stimuli. In five patients, premature stimuli again produced a functional arc of block, which was circumvented by two opposing wave fronts that united on the distal side. Retrograde penetration by a narrow isthmus of slow conduction through the block initiated the tachycardia, whose activation sequence was comblent with figure eight reentry.In one patient, premature stimuli produced a region of delayed potentials. Critical timing of these resulted in microreentry in an adjacent circumscribed site, which formed the site of origin of the ensuing tachycardia. The microreentrant signals were not detected by standard unipolar recordings, but were seen on simultaneously recorded high gain electrograms. In 14 patients, although mapping identified a site of origin, the activation patterns showed either radial spread or incomplete circles.Detection of reentrant mechanisms daring intraoperative mapping required high density electrode arrays and refined high gain recordings. An intact ventricle may facilitate intraoperative initiation of tachycardia.