Charge injection error in the presence of subthreshold effects has been analyzed. It is confirmed that the subthreshold effect is significant at low voltage falling rates. A simplified model is derived using an appropriate approximation. Predictions are compared to the results of a SPICE simulation, a nonquasi-static (NQS) model simulation and experimental results. Excellent agreement between the modified and NQS model and recently published experimental results was obtained. This analytical model is computationally efficient compared to the SPICE and NQS models and provides physical insight into the switching errors

Keywords: MOS analogue integrated circuits SPICE circuit analysis computing integrated circuit modelling switching circuits MOS analog switch SPICE simulation charge injection error nonquasi-static model simulation subthreshold analysis switching errors voltage falling rates

The commonly used two-stage CMOS amplifier suffers from two basic performance limitations due to the RC compensation network around the second gain stage. Frisr, this frequency compensation technique provides stable operation for limited range of capacitive loads, and second, the power supply rejection shows severe degradation above the open-loop pole frequency. The technique described here provides stable operation for a much larger range of capacitive loads, as well as much improved V_BB power supply rejection over very wide bandwidths for the same basic op amp circuit. This paper presnets mathematical analysis of this new technique in terms of its frequency and noise characteristics followed by its implementation in all n-well CMOS process. Experimental results show 70 dB negative power supply rejection at 100 kHz and an input noise density of 58 nV/sqrt(Hz) at 1 kHz.

This paper reports the development of an implantable, fully integrated, multichannel peripheral neural recording system, which is powered and controlled using an RF telemetry link. The system allows recording of ±500 μ V neural signals from axons regenerated through a micromachined silicon sieve electrode. These signals are amplified using on-chip 100 Hz to 3.1 kHz bandlimited amplifiers, multiplexed, and digitized with a low-power (<2 mW), moderate speed (8 μ s/b) current-mode 8-b analog-to-digital converter (ADC). The digitized signal is transmitted to the outside world using a passive RF telemetry link. The circuit is implemented using a bipolar CMOS process. The signal processing CMOS circuitry dissipates only 10 mW of power from a 5-V supply while operating at 2 MHz and consumes 4× 4 mm² of area. The overall circuit including the RF interface circuitry contains over 5000 transistors, dissipates 90 mW of power, and consumes 4× 6 mm² of area

Keywords: BiCMOS analogue integrated circuits biomedical electronics biomedical telemetry medical signal processing micromachining neurophysiology radiofrequency amplifiers radiotelemetry -500 to 500 muV 10 mW 100 Hz to 3.1 kHz 2 MHz 5 V 90 mW RF interface circuitry RF telemetry link axons bandlimited amplifiers biomedical sensors bipolar CMOS process micromachined sieve electrode wireless implantable multichannel digital neural recording system

Discusses the single-unit recording characteristics of microelectrode arrays containing on-chip signal processing circuitry. Probes buffered using on-chip unity-gain operational amplifiers provide an output resistance of 200 Ω with an input-referred noise of 11 μ V root-mean-square (rms) (100 Hz-10 kHz). Simultaneous in vivo recordings from single neurons using buffered and unbuffered (passive) iridium recording sites separated by less than 20 um have shown that the use of on-chip circuitry does not significantly degrade system noise. Single-unit neural activity has also been studied using probes containing closed-loop preamplifiers having a voltage gain of 40 dB and a bandwidth of 13 kHz, and several input dc-baseline stabilization techniques have been evaluated. Low-noise in vivo recordings with a multiplexed probe have been demonstrated for the first time using an external asymmetrical clock running at 200 kHz. The multiplexed system adds less than 8 μ V_rms of noise to the recorded signals, suppressing the 5-V clock transitions to less than 2 ppm.

Keywords: bioelectric potentials biomedical electrodes biomedical electronics microelectrodes micromechanical devices neurophysiology operational amplifiers preamplifiers probes 100 Hz to 10 kHz 11 muV 13 kHz 20 mum 200 kHz 200 ohm 5 V 8 muV Ir active microelectrode arrays clock transitions suppression input debaseline stabilization techniques input-referred noise multiplexed probe on-chip signal processing circuitry on-chip unity-gain operational amplifiers recorded signals simultaneous in vivo recordings single-unit neural recording

In many brain regions, high levels of background activity can obscure action potentials of neurons close to a recording electrode, particularly in the presence of a stimulus. As recorded on a multi-channel electrode array, this neural noise appears to be highly correlated among the channels. We have developed an array processing technique to remove the correlated component of noise, improving the signal-to-noise ratio. In addition, we show that the weighting vector of the array algorithm can be manipulated to facilitate sorting of the action potentials. An example of the technique applied to the guinea pig cochlear nucleus is shown.

Keywords: Spike sorting; Signal processing; Silicon micro-electrode; Tetrode; Electrophysiology

Multielectrode arrays (MEAs) have emerged as a leading technology for extracellular neural recording and stimulation. Their large number of recording sites promises to yield important insight into neural systems. As the density of recording sites increases, interfacing to each electrode becomes increasingly difficult. Introducing electronics onto the MEA substrate provides a technique for preliminary signal conditioning to take place at the MEA itself, reducing the complexity of offpackage electronics. This paper introduces a custom MEA system with integrated preamplifiers. MEA fabrication, cell-culturing, and electrical performance are discussed.

Keywords: electrode, multielectrode arrrays, MEA, MEMS, extracellular recording, low-noise

The surface-to-volume ratio increases with decreasing scale, thus, controlling and changing the surface properties of microstructures can be a powerful tool in the design, fabrication, and use of microsystems. This paper overviews several recent projects that utilize the modulation of surfaces from hydrophobic to hydrophilic and vice versa, or from protein adsorbing to non-fouling, with applications in biomedical microdevices and self-assembling microelectromechanical systems (MEMS).

Keywords: surface modification, self-assembled monolayer, hydrophobic, hydrophilic, protein adsorption, biofouling, self-assembly, MEMS.

A planar 6×6 array of iridium electrodes with four reference electrodes has been developed for use with neural tissue preparations. Precise knowledge of the relative locations of the array elements allows for spatial neurophysiological analyses. The 10 μ m diameter platinized iridium electrodes on a 100 μ m pitch have been used to stimulate acutely prepared slices of spinal cord from free-ranging rodents. An intracellular recording from a single neuron in the substantia gelatinosa (SG) using the whole-cell, tight-seal technique allowed low noise, high resolution studies of excitatory or inhibitory electrical responses of a given neuron to inputs from the primary afferent fibers or from stimulation by individual electrodes of the array. The resulting maps of responses provide an indication of the interconnectivity of neural processes. The pattern emerging is that of limited interconnectivity in the SG from areas surrounding a recorded neuron but with strong excitatory or inhibitory effects from those oriented in a longitudinal (rostral-caudal) direction relative to the neuron. The observations to date suggest the neurons of the SG are arranged in sets of independent networks, possibly related to sensory modality and input from particular body regions.

Keywords: Spinal cord slice; Microelectrode array; Substantia gelatinosa; Focal stimulation; Fabrication

Systems designed to significantly reduce equipment cost and size for neurophysiological studies and hybrid biosensor applications were developed. Custom integrated circuits, each providing 18 channels of amplification and filtering were designed, fabricated and tested. Planar arrays of iridium microelectrodes were fabricated and packaged in a standard 40 pin dual-in-line package for cultured cell and neural slice preparation studies. An impedance imaging system was developed to monitor the impedance of the cell/electrode interface across the array, thereby expanding the possible biosensor applications to non-electrically active cell types. Thermal regulation was achieved via a Peltier effect thermoelectric device allowing temperature control both above and below ambient temperature. While designed to work together the system components presented may be easily applied to existing systems for enhancement of capabilities while reducing size and cost.

Keywords: arrays bioelectric potentials biological techniques biosensors cellular biophysics electric impedance measurement electrodes neurophysiology Ir Peltier effect thermoelectric device ambient temperature cell/electrode interface impedance monitoring cultured cell studies custom integrated circuits electrophysiological research instrumentation hybrid biosensor applications impedance measurements iridium microelectrodes neural recording neural slice preparation studies neurophysiological studies nonelectrically active cell types planar electrode array systems standard 40 pin dual-in-line package

This paper describes an ad hoc modified version of the electrical circuit analysis program SPICE, which has been optimized for detailed simulations of the behaviour of neurons. An equivalent-circuit description of the simulation building blocks is provided, and the SPICE modifications are specified. These modifications, in contrast to previous uses of SPICE, allows one to simulate the behaviour of neurons of Hodgkin-Huxley type (excitable membrane) and of postsynaptic membranes without any approximations. Simulation results are reported and compared, both with data previously analysed in the literature by other authors and with experimental data recently obtained by coupling neurons to planar extracellular microelectrodes. Details of the circuit elements used in the simulations are reported. The improvements of our proposed model are discussed in comparison with a previous SPICE-based model described in the literature.

Keywords: NETWORK ANALYSIS PROGRAM, ELECTRICAL BEHAVIOR, PROPAGATION, MEMBRANE

Known elementary wide-band amplifiers suffer from a fundamental tradeoff between noise figure (NF) and source impedance matching, which limits the NF to values typically above 3 dB. Global negative feedback can be used to break this tradeoff, however, at the price of potential instability. In contrast, this paper presents a feedforward noise-canceling technique, which allows for simultaneous noise and impedance matching, while canceling the noise and distortion contributions of the matching device. This allows for designing wide-band impedance-matching amplifiers with NF well below 3 dB, without suffering from instability issues. An amplifier realized in 0.25 μ m standard CMOS shows NF values below 2.4 dB over more than one decade of bandwidth (i.e., 150-2000 MHz) and below 2 dB over more than two octaves (i.e., 250-1100 MHz). Furthermore, the total voltage gain is 13.7 dB, the -3 dB bandwidth is from 2 MHz to 1.6 GHz, the IIP2 is +12 dBm, and the IIP3 is 0 dBm. The LNA drains 14 mA from a 2.5 V supply and the die area is 0.3× 0.25 mm².

Keywords: CMOS analogue integrated circuits impedance matching integrated circuit noise thermal noise wideband amplifiers 0.25 micron 13.7 dB 14 mA 150 to 2000 MHz 2 MHz to 1.6 GHz 2.5 V LNA drains distortion canceling feedforward noise-canceling technique global negative feedback noise distortion noise figure source impedance matching thermal noise canceling voltage gain wide-band CMOS low-noise amplifier

The electrical double-layer at a solid electrode does not in general behave as a pure capacitance but rather as an impedance displaying a frequency-independent phase angle different from 90^o. Ways are indicated how to analyse the interfacial impedance if such a complication arises in the presence of a faradaic process, both on the supposition that the double-layer behaviour is due to surface inhomogeneity and on the supposition that it is a double-layer property per se. As examples, the equations derived are successfully applied to a totally irreversible and an ac quasi-reversible electrode process at a gold electrode.

This paper presents an in-depth analysis of transconductances in CMOS for advanced analog IC design. Transconductances in a 0.25 μ m CMOS technology have been measured over a large range of geometries and bias conditions. Gate (g_mg), source (g_ms), drain (g_md) and bulk (g_mb) transconductances are consistently normalized and represented vs. inversion coefficient (IC) from very weak to moderate and strong inversion. The ideal transconductance behavior in particular in weak inversion is analyzed via the analytical structure of the EKV MOSFET model. The new EKV 3.0 MOSFET model shows excellent abilities to correctly represent transconductances at all levels of inversion and channel lengths.

Keywords: CMOS analogue integrated circuits MOSFET circuit CAD integrated circuit design integrated circuit modelling semiconductor device measurement semiconductor device models 0.25 micron CMOS analog IC design EKV MOSFET model MOSFET geometry/bias conditions MOSFET inversion transconductance analysis channel lengths ideal transconductance behavior inversion coefficient inversion levels moderate inversion normalized gate/source/drain/bulk transconductances strong inversion transconductance measurement very weak inversion

Polycrystalline silicon thin films were explored with respect to their application as low-impedance microelectrodes for extracellular stimulation and recording of cells. Microelectrode arrays (MEAs) comprising polysilicon microelectrodes were fabricated using CMOS-compatible processes. Overall capacitance of an electrode with a diameter of 20 μ m is on the order of 200-300 pF. Chemical and morphological stability in physiological saline solution was excellent over a period of at least 5 months. This finding renders applications in neuronal implants or bio-chips. Nanoporous polysilicon electrodes were created by anodic oxidation in hydrofluoric acid (HF). However, no considerable decrease of electrode impedance was observed although pore formation was clearly confirmed by transmission electron microscopy (TEM).

Keywords: Impedance spectroscopy; Micro electrode arrays; Polycrystalline silicon; Porous silicon; Thin-film microelectrodes

Although measurement of sealing resistance is an important tool in the assessment of the electrical contacts between cultured cells and substrate embedded microelectrodes, it does not offer information about the type of cell, i.e. neuron or non-neuronal cell. Also, rules for translation of a measured sealing resistance into parameters for successful stimulation, i.e. eliciting an action potential, are not available yet. Therefore, a method is proposed for the detection of active membrane currents, elicited by extracellular current stimulation. The method is based on the prediction of the linear part of the response to an applied stimulus current pulse using an impedance model of the neuron-electrode contact. Active membrane currents are detected in the nonlinear response, which is obtained by subtraction of the predicted linear response from the measured response. The required impedance model parameters are extracted from impedance spectroscopy or directly from the measured responses.

Keywords: Neuron-electrode contact; Extracellular stimulation; Impedance spectroscopy; Sealing resistance; Cultured neurons; Microelectrode arrays

The electrical contact between a substrate embedded microelectrode and a cultured neuron depends on the geometry of the neuron-electrode interface. Interpretation and improvement of these contacts requires proper modeling of all coupling mechanisms. In literature, it is common practice to model the neuron-electrode contact using lumped circuits in which large simplifications are made in the representation of the interface geometry. In this paper, the finite-element method is used to model the neuron-electrode interface, which permits numerical solutions for a variety of interface geometries. The simulation results offer detailed spatial and temporal information about the combined electrical behavior of extracellular volume, electrode-electrolyte interface and neuronal membrane.

Keywords: bioelectric phenomena   biological techniques   finite element analysis   microelectrodes   neurophysiology   physiological models   combined electrical behavior   coupling mechanisms   cultured neuron   electrical contact   electrode-electrolyte interface   extracellular volume   geometry-based finite-element modeling   interface geometry representation   lumped circuits   neuron-electrode interface   neuronal membrane   neuroscience method   simulation results

An improved frequency compensation technique is presented in this paper. It is based on a cascade of a voltage amplifier and a transconductor to form a composite gain-enhanced feedforward stage in a two-stage amplifier so as to broaden the gain bandwidth via low-frequency pole-zero cancellation at heavy capacitive loads, but yet without increasing substantial power consumption. The technique has been confirmed by the experimental results. An operational amplifier has been designed to drive a capacitive load of 300 pF. The amplifier exhibits a dc gain of 87 dB, a gain bandwidth of 10.4 MHz at 63.7^o phase margin, an average slew rate of 3.5 V/μ s, a compensation capacitor of only 6 pF while consuming 2.45 mW at a 3 V supply in a standard 0.6 μ m CMOS technology.

Keywords: Feedforward transconductance amplifier, frequency compensation, pole zero cancellation, two-stage CMOS amplifier

Multichannel recording provides integral information about electrical brain activities at one instant in time. In this study, multielectrode probes were fabricated to record the thalamic field potentials (FPs) responding to the electrical stimulation of nerve at the rat tail. At first, the number of sweeps used to form the evoked FP average and the spatial sampling density were determined by using cross-correlation functions, which were then statistically analyzed. The difference was significant at P<0.05, if the number of sweeps for averaging was more than 50 and the spatial interval between two consecutive recording sites was less than 50 μ m in the anteroposterior, mediolateral and ventrodorsal directions. The responsive area was distributed vertically in the thalamus (ventral posterior lateral (VPL) nucleus); therefore, the recording sites were arranged in one linear array. Sixteen recording sites, which were 50 μ m apart from each other, were distributed in the ventrodorsal direction. A 16-channel silicon probe was fabricated by using a standard photolithography process and laser micromachining techniques. The probe provides capabilities to record multiple thalamic evoked FPs and multiunit activities simultaneously.

Keywords: Field potential; Thalamus; Cross-correlation; Multielectrode probe; Laser micromachining

The detection of extracellular potentials by means of multi-electrode arrays (MEA) is a useful technique for multi-site long-term monitoring of cultured neuronal activity with single-cell resolution. To optimize the geometry of the MEA it is advantageous to localize the cellular compartments that constitute the generators of these signals. For this purpose, an in vitro technique for the detection of extracellular signals with subcellular resolution has been developed. It makes use of easy-to-manufacture large-tip pipettes, monitoring of electrode-cell gap resistance for precise electrode positioning and low-density (100 cells/mm²) dissociated hippocampal cultures. Negative monophasic extracellular spikes, typically 60 μ V, were measured over putative axonal processes and monophasic, biphasic and triphasic signals were recorded over the soma. A compartmental simulation suggests that different somatic conductance densities of Na⁺ (1-10 mS/cm²) and K⁺ (5-10 mS/cm²) channels can produce characteristic somatic extracellular potentials, with a variety of shapes similar to those observed experimentally.

Keywords: Extracellular potentials; Cultured hippocampal neurons; Multi-electrode arrays; Axon-hillock; Compartmental simulation; Neuron model

Operation of MOS devices in the strong, moderate, and weak inversion regions is considered. The advantages of designing the input differential stage of a CMOS op amp to operate in the weak or moderate inversion region are presented. These advantages include higher voltage gain, less distortion, and ease of compensation. Specific design guidelines are presented to optimize amplifier performance. Simulations that demonstrate the expected improvements are given.

Keywords: MOS analogue integrated circuits amplifiers CMOS op amps MOS devices amplifier performance optimization compensation input stage optimization low-distortion voltage gain weak inversion region

This paper describes a planar array of microelectrodes developed for monitoring the electrical activity of cells in culture. The device allows the incorporation of surface topographical features in an insulating layer above the electrodes. Semiconductor technology is employed for the fabrication of the gold electrodes and for the deposition and patterning of an insulating layer of silicon nitride. The electrodes have been tested using a cardiac cell culture of chick embryo myocytes, and the physical beating of the cultured cells correlated with the simultaneous extracellular voltage measurements obtained. It was found that extracellular stimulation of the cells was possible via the same electrodes used for recording.

Keywords: extracellular recording; microelectrodes; cell guidance; cardiac cells

Parallel recording of neuronal activity in the behaving animal is a prerequisite for our understanding of neuronal representation and storage of information. Here we describe the development of micro-machined silicon microelectrode arrays for unit and local field recordings. The two-dimensional probes with 96 or 64 recording sites provided high-density recording of unit and field activity with minimal tissue displacement or damage. The on-chip active circuit eliminated movement and other artifacts and greatly reduced the weight of the headgear. The precise geometry of the recording tips allowed for the estimation of the spatial location of the recorded neurons and for high-resolution estimation of extracellular current source density. Action potentials could be simultaneously recorded from the soma and dendrites of the same neurons. Silicon technology is a promising approach for high-density, high-resolution sampling of neuronal activity in both basic research and prosthetic devices.

This paper reports on the development of a fully integrated 32-channel integrated circuit (IC) for recording neuronal signals in neurophysiological experiments using microelectrode arrays. The IC consists of 32 channels of low-noise preamplifiers and bandpass filters, and an output analog multiplexer. The continuous-time RC active filters have a typical passband of 20-2000 Hz; the low and the high cut-off frequencies can be separately controlled by external reference currents. This chip provides a satisfactory signal-to-noise ratio for neuronal signals with amplitudes greater than 50 V. For the nominal passband setting, an equivalent input noise of 3 V_rms has been achieved. A single channel occupies 0.35 mm² of silicon area and dissipates 1.7 mW of power. The chip was fabricated in a 0.7 μ m CMOS process.

The preamplifier is a fairly standard design: differential pair, CD, CS. The authors make frequent use of MOSFETs in the linear region as resistors. Picofarad capacitors are necessary in each amplifier. Compare to Harrison's design [62] and Ji's design [72].

Keywords: Low-noise amplifier; Bandpass filters; Extracellular recording; Microelectrode arrays; VLSI electronics

The influence of surface roughness of solid electrodes on electrochemical measurements is critically examined. A model and its mathematical consequences are presented which describe the effects in at least a semi-quantitative way. The conclusion is drawn that the neglect of surface roughness has led to apparent misinterpretations in literature, and that great caution should be exercised in this respect.

The use of cell-based biosensors outside of the laboratory has been limited due to many issues including preparation of the sample, maintenance of the biological environment, and integration of the electronics for data collection and analysis. This paper describes a system that addresses several of these issues with the development of an integrated silicon-polydimethylsiloxane cell-cartridge. The cell-cartridge contains a CMOS silicon chip that incorporates a digital interface, temperature control system, microelectrode electrophysiology sensors, and analog signal buffering. Additionally, the cell-cartridge supports two separate cell populations in two 10 μ l sealed chambers that have independent fluidic channels for sample injection. A portable, microcontroller-based electronics system capable of monitoring the action potential (AP) activity within the cell-cartridges was also developed. The AP activities of cardiomyocyte syncytia in the two chambers differentially responded to the flow of a control medium versus the flow of a biochemical agent. The cell-cartridges and portable electronics system were used to successfully record AP activity from cardiomyocytes outside of the laboratory under realistic application conditions.

Keywords: Cell-based biosensor; Portable biosensor; CMOS cell-cartridge

We describe a photoreceptor circuit that can be used in massively parallel analog VLSI silicon chips, in conjunction with other local circuits, to perform initial analog visual information processing. The receptor provides a continuous-time output that has low gain for static signals (including circuit mismatches), and high gain for transient signals that are centered around the adaptation point. The response is logarithmic, which makes the response to a fixed image contrast invariant to absolute light intensity. The 5-transistor receptorcan be fabricated in an area of about 70 μ m by 70 μ m in a 2 μ m single-poly CMOS technology. It has a dynamic range of 1-2 decades at a single adaptation level, and a total dynamic range of more than 6 decades. Several technical improvements in the circuit yield an additional 1-2 decades dynamic range over previous designs without sacrificing signal quality. The lower limit of the dynamic range, defined arbitrarily as the illuminance at which the bandwidth of the receptor is 60 Hz, is at approximately 1 lux, which is the border between rod and cone vision and also the limit of current consumer video cameras. The receptor uses an adaptive element that is resistant to excess minority carrier diffusion. The continuous and logarithmic transduction process makes the bandwidth scal eiwht intensity. As a result, the total A.C. RMS receptor noise is constatn, independent of intensity. The spectral density of the noise is within a factor or two of pure photon shot noise and varies inversely with intensity. The connection between shot and thermal noise in a system goberned by Boltzman statistics is beautifilly illustrated.

Attempts are underway to construct a retinal prosthesis to recover limited vision for blind patients with retinitis pigmentosa using implantable electronic devices. These microchips provide electrical stimulation to damaged retinal tissues using an array of stimulus circuits. This paper describes improvements to conventional circuit designs with significantly decreased implementation area and the ability to support arbitrary stimulus waveforms where an array of such stimulus circuits is required. This yields greater spatial resolution in stimulation owing to more stimulus circuits per chip area. Also introduced are digital-to-analog converter gain prescalar and dc-offset circuits which tune the stimulus circuits to an optimally effective range due to variation in retinal degradation. The prototype chip was fabricated by MOSIS in 1.2 μ m CMOS technology.

Keywords: CMOS integrated circuits biomedical electronics digital-analogue conversion prescalers sensory aids 1.2 micron CMOS microchip DC offset circuit age-related macular degeneration blind patient circuit design electrical stimulation implantable electronic device multibias digital-to-analog converter gain prescalar retinal tissue retinitis pigmentosa stimulus circuit array visual prosthesis

BiCMOS and CMOS versions of an operational transconductance amplifier (OTA) with an adjustable linear range that is independent of tail current are presented. The circuit is analysed based on the operaion of transistors in weak and strong inversion and the analytical results are compared with the experimental data.

Keywords: VLSI, Operational transconductance amplifiers

A micropower signal-processor IC is the key component of an implantable telemetry system for neurophysiology. The bipolar/JFET/I²L chip uses digital and low-noise analog circuitry to amplify, filter, and multiplex eight channels of neutral, electrogram, and temperature data from unanesthetized and freely moving animals. Fully integrated continuous-time bandpass amplifiers incorporate a frequency-sensitive feedback network to prevent the amplification of input offset voltage. The system can telemeter data for over 500 h, permitting long-term neurophysiological investigations.

Keywords: Biological techniques and instruments Biomedical electronics Monolithic integrated circuits Multiplexing equipment Neurophysiology Signal processing equipment Telemetering equipment Telemetering systems biological techniques and instruments biomedical electronics monolithic integrated circuits multiplexing equipment neurophysiology signal processing equipment telemetering equipment telemetering systems

In this paper the properties of a low-level nonlinear continuous-time circuit element-termed a Bernoulli Cell (or Operator)-are described in a systematic way. This cell is composed of an n-p-n BJT and an emitter-connected grounded capacitor, and is governed by a differential equation of the Bernoulli form. Although this cell has the potential for application in both linear and nonlinear analog signal processing, this paper focuses on the field of input-output linear log-domain Filtering. The Bernoulli Cell can be utilized in both the analysis and synthesis of log-domain circuits. The Bernoulli Cell approach leads to the creation of a system of linear differential equations with time dependent coefficients and state variables nonlinearly related to currents internal to the circuit; this set of equations is termed ``log-domain state-space'', and can be used for the synthesis of linear log-domain filters. Four design examples-including a bandpass biquad-are presented.

Keywords: band-pass filters bipolar analogue integrated circuits biquadratic filters continuous time filters integrated circuit design integrating circuits linear differential equations linear network synthesis low-pass filters state-space methods Bernoulli cell Bernoulli form Bernoulli operator analog signal processing bandpass biquad emitter-connected grounded capacitor linear differential equations linear log-domain filter synthesis log-domain circuits Log-Domain_Filtering log-domain state-space low-level continuous-time circuit element n-p-n BJT nonlinear continuous-time circuit element state variables time dependent coefficients transistor-level approach

Despite the significance of matched devices in analog circuit design, mismatch modeling for design application has been lacking. This paper addresses misconceptions about MOSFET mismatch for analog design. V_t mismatch does not follow a simplistic 1/( sqrt(area)) law, especially for wide/short and narrow/long devices, which are common geometries in analog circuits. Further, V_t and gain factor are not appropriate parameters for modeling mismatch. A physically based mismatch model can be used to obtain dramatic improvements in prediction of mismatch. This model is applied to MOSFET current mirrors to show some nonobvious effects over bias, geometry, and multiple-unit devices.

Keywords: MOSFET SPICE analogue integrated circuits current mirrors integrated circuit design semiconductor device models MOSFET current mirrors MOSFET mismatch SPICE analog IC design analog circuit design gain factor physically based mismatch model semiconductor device modeling threshold voltage mismatch

This paper describes the principle and design of a CMOS rail-to-rail input operational amplifier with THD performance of -90 dB which is suited for high-quality audio systems. A new output stage has been used featuring an output suing that extends to either supply rail and is capable of driving a low ohmic load (32 Ω). The opamp, which is realized in a 0.5 μ m 3.3-V digital CMOS process, uses a standard two-stage Miller configuration. The rail-to-rail input functionality is achieved with a new area-efficient on-chip charge pump which provides the local supply voltage for the input differential pair. THD levels below -90 dB have not yet been shown with existing rail-to-rail techniques. This rail-to-rail input configuration also behaves independently of the common mode level with respect to transconductance and slewing characteristics

Keywords: CMOS analogue integrated circuits audio-frequency amplifiers harmonic distortion operational amplifiers 0.5 micron 3.3 V CMOS rail-to-rail input operational amplifier THD audio system local charge pump slewing characteristics transconductance two-stage Miller configuration

This paper describes a flexible, metallic multielectrode array, made on kapton to fit in a recording chamber for interface-type organotypic cultures. This multisite recording system is designed for continuous multisite monitoring of electrophysiological activity in rat brain organotypic slice cultures. The system is composed of a signal conditioning set-up, which also masters electrical stimulation paradigms and a card containing the microelectrode array. The card comprises a perfusion chamber closed by a rigid and permeable membrane on which the pierced microelectrode array supporting the slice culture is placed. Once closed with a gaseous chamber, the inside of the card remained sterile and free of contamination and could be maintained inside or outside the incubator for electrophysiological analyses. Dimensions of each 28-plated gold microelectrode recording site are 50 μ m × 100 μ m. The design of the chambers and the card makes it possible to modify both the perfusion medium and the gaseous atmosphere in sterile conditions, allowing thus analyses of long-term effects of pharmacological compounds. Using this array one can perform stimulation and recordings of the electrical activity of the slice. Signals obtained with this reusable system exhibit a good signal-to-noise ratio. This device was tested to follow the evolution and modifications of the evoked and/or spontaneous electrical activity of the same groups of neurones during several days.

Keywords: Microelectrode array; Field potentials; In vitro; Organotypic culture

A 64-channel amplifier system for the recording of extracellular signals with planar metal microelectrodes is presented. Gold metal microelectrodes on glass wafers were fabricated using standard photolithographic techniques. The measurement system was divided into a headstage preamplifier and a main amplifier. The inherent noise of the extracellular recording system was minimized by using an independent battery supply. The metal electrodes were directly connected to the gates of low noise junction field effect transistors (JFETs) using a specially designed electronic circuit. With this set-up, it was possible to record extracellular signals with planar metal microelectrodes without any surface modification for impedance reduction. A feedback circuit in the first amplification stage compensated slow drifts of the gold microelectrodes, which made online sampling of all 64 channels with a sampling rate of 10 kHz possible. Recordings were taken from rat cardiac myocytes cultured on fibronectin coated sensor chips. The system exhibited a good signal-to-noise ratio. It was able to detect the signal propagation within the cardiac cell layer and it could be used for pharmacological investigations involving the heart.

Keywords: Extended gate electrodes; Extracellular recording; Sensor array; Action potential; Rat cardiac myocytes

Spatiotemporally coordinated activity of neural networks is crucial for brain functioning. To understand the basis of physiological information processing and pathological states, simultaneous multisite long-term recording is a prerequisite. In a multidisciplinary approach we developed a novel system of organotypically cultured rat hippocampal slices on a planar 60-microelectrode array (MEA). This biohybrid system allowed cultivation for 4 weeks. Methods known from semiconductor production were employed to fabricate and characterize the MEA. Simultaneous extracellular recording of local field potentials (LFPs) and spike activity at 60 sites under sterile conditions allowed the analysis of network activity with high spatiotemporal resolution. To our knowledge this is the first realization of hippocampus cultured organotypically on multi-microelectrode arrays for simultaneous recording and electrical stimulation. This biohybrid system promises to become a powerful tool for drug discovery and for the analysis of neural networks, of synaptic plasticity, and of pathophysiological conditions such as ischemia and epilepsy.

Keywords: Correlation analysis; Hippocampus; Multielectrode array; Organotypic long-term culture; Paired pulse facilitation; Photolithography; Rat; Plasticity

Tetrodes allow isolation of multiple neurons at a single recording site by clustering spikes. Due to electrode drift and perhaps due to time-varying neuronal properties, positions and shapes of clusters change in time. As data is typically collected in sequential files, to track neurons across files one has to decide which clusters from different files belong to the same neuron. We report on a semi-automated neuron tracking procedure that uses computed similarities between the mean spike waveforms of the clusters. The clusters with the most similar waveforms are assigned to the same neuron, provided their similarity exceeds a threshold. To set this threshold, we calculate two distributions: of within-file similarities, and of best matches in the across adjacent file similarities. The threshold is set to the value that optimally separates the two distributions. We compare different measures of similarity (metrics) by their ability to separate these distributions. We find that these metrics do not differ drastically in their performance, but that taking into account the cross-channel noise correlation significantly improves performance of all metrics. We also demonstrate the method on an independent dataset and show that neurons, as assigned by the procedure, have consistent physiological properties across files.

Keywords: Tracking neurons; Tetrode; Electrode drift; Extracellular recording

A fully analytical MOS transistor model dedicated to the design and analysis of low-voltage, low-current analog circuits is presented. All the large and small-signal variables, namely the currents, the transconductances, the intrinsic capacitances, the non-quasi-static transadmittances and the thermal noise are continuous in all regions of operation, including weak inversion, moderate inversion, strong inversion, conduction, and saturation. The same approach is used to derive all the equations of the model: the weak and strong inversion asymptotes are first derived, then the variables of interest are normalized and linked using an appropriate interpolation function. The model exploits the inherent symmetry of the device by referring all the voltages to the local substrate. It is shown that the inversion Q'_inv is controlled by the voltage differencd V_p - V_ch, where V_ch is the channel voltage, defined as the difference between the quasi-Fermi potentials of the carriers. The pinch-off voltage V_p is defined as the particular value of V_ch such that the inversion charge is zero for a given gate voltage. It depends only on the gate voltage and can be interpreted as the equivalent effect of the gate voltage referred to the channel. The various modes of operation of the transistor are then presented in terms of voltages V_p - V_s and V_p - V_d. Using the charge sheet model with the assumption of constant doping in the channel, the drain current I_d is derived and expressed as the difference between a forward component I_f and a reverse component I_r. Each of these is proportional to a function of V_p - V_s, respectively V_p - V_d, through a specific current I_s. This function is exponential in weak inversion and quadratic in strong inversion. The current in the moderate inversion is then modelled by using an appropriate interpolation function resulting in a continuous expression valid from weak to strong inversion. A quasi-static small-signal model including the transconductances and the intrinsic capacitances are modelled in moderate inversion using the same interpolation function and without any additional parameters. This small-signal model is then extended to higher frequencies by replacing the transconductances by first order transadmittances obtained from a non-quasi-static calculation. All these transadmittances have the same characteristic time constant which depends on the bias conditions in a continuous manner. To complete the model, a general expression for the thermal noise valid in all regions of operation is derived. This model has been successfully implemented in several computer simulation programs and has only 9 physical parameters, 3 fine tuning fitting coefficients and 2 additional temperature parameters.

Keywords: MOS transistor, device modeling, low-voltage, low-current

Sensor arrays are a key tool in the field of neuroscience for noninvasive recording of the activity of biological networks, such as dissociated neurons or neural tissue. A high-density sensor array complementary metal-oxide-semiconductor chip is presented with 16 Kpixels, a frame rate of 2 kiloframes per second, and a pitch of 7.8 μ m× 7.8 μ m for imaging of neural activity. The related circuit and system issues as well as process aspects are discussed. A mismatch-canceling calibration circuitry with current mode signal representation is used. Results from first biological experiments are presented, which prove full functionality of the chip.

Keywords: Bioelectric potentials, biological cells, biological tissues, biomedical transducers, image sensors, multi-electrode array (MEA), nervous system.

Plasma deposition and treatment processes, of interest for different applications in the field of biomaterial surfaces, are briefly reviewed in this paper. In particular, details are given on processes developed in our laboratory, namely: the deposition of non fouling (PEO-like) and bacterial resistant (Ag/PEO-like) coatings, that of -COOH functional coatings, and combined processes utilized for obtaining micro-patterned polymer surfaces.

The platinized platinum electrode is the most widely used type of electrode. It makes a regular appearance in conductance cells, forms the basis of the hydrogen reference electrode, and is unsurpassed as an electrocatalyst in fuel cells. Yet the mechanism of the formation of platinum deposit has been investigated only recently, and information on the properties of the electrode (its appearance, structure, area, and reproducibility), and how these are affected by its method of prefaration, is widely scattered and not well knows. The main factors are critically assessed in this review.

Keywords: electrode, platinum

A high-performance CMOS power amplifier consisting of a new input stager especially suited to power amplifier applications and a variation on a class AB output stage is presented. The amplifier has been fabricated using a conventional silicon gate p-well process. The configuration results in several performance improvements over previously reported high-output current amplifiers without requiring process enhancements. Design details and experimental results are described.

Keywords: CMOS integrated circuits Linear integrated circuits Power amplifiers linear integrated circuits power amplifiers

A CMOS buffer amplifier which achieves significant improvements in linearity and drive capability over previously reported ``high-swing'' amplifiers is described. The buffer operates from a 5-V supply, is capable of rail-to-rail operation at both the input and output, an exhibits a remarkably high linearity of 0.05% THD while driving 3 V_p-p into 100 Ω at 20 kHz.

Keywords: Amplifiers Buffer circuits CMOS integrated circuits Linear integrated circuits amplifiers buffer circuits linear integrated circuits

A simple change in the form of the state variables clarifies previously published analyses of log-domain circuits. The change highlights the role of bias currents and leads to a simple design procedure based an self-contained building blocks. A new nonlinear equivalent circuit is presented fur the building blocks with a topology that emphasizes the applicability of small-signal analysis for log-domain circuits. The nonlinear equivalent circuit suggests a novel single-ended inverting cell that works at very low supply voltage VDD and dues not require p-n-p transistors.

Keywords: active filters equivalent circuits nonlinear network synthesis bias current building blocks design-oriented analysis log-domain circuit nonlinear equivalent circuit single-ended inverting cell small-signal analysis state variables topology

Companding is discussed as it concerns log domain filters. Here, companding is assumed to be implemented via modulation of additive biasing currents. From this perspective, only class AB log domain filters are considered ``instantaneously'' companding filters. Furthermore, class A and class AB filters are seen to be limiting cases of syllabically companding filter.

Keywords: compandors filtering theory nonlinear filters additive biasing currents class-A filters class-AB filters instantaneous companding log domain filters modulation syllabic companding

A novel approach to filter design, based on Adams' `log-domain' filters, is proposed that yields a truly current-mode circuit realisation. Adams' idea, which was introduced in a limited context, is generalised to permit a complete distortionless synthesis procedure, which results in circuit implementations readily realisable using complementary bipolar processes. It is shown that, by introducing an exponential map on the state-space description of the desired linear system, a log-domain filter can be fully realised with transistors configured in current mirror-type groupings, current sources and capacitors. Owing to the mapping, the state variables are intrinsically related to current, and not voltage, in the resulting circuits, a fact that emphasises the current-mode nature of the design. A general biquadratic filter section is designed, and, following discussion of cascading sections, a seventh-order Chebychev lowpass filter is designed. All designed circuits are shown to be tunable over a two-decade range in frequency while their characteristics are accurately preserved, even for biquad sections whose f₀ Q product is greater than f_T / 10. The Chebychev filter is shown in simulation to possess nearly 60 dB dynamic range relative to 0.9% THD, with a cutoff frequency of nearly 5 MHz, using transistor models from AT&T's CBIC-R 300 Hz complementary bipolar process

Keywords: Chebyshev approximation active filters bipolar integrated circuits low-pass filters 5 MHz active filters biquadratic filter section complementary bipolar processes current mirror-type groupings current sources current-mode filtering cutoff frequency distortionless synthesis procedure dynamic range filter design Log-Domain_Filtering seventh-order Chebychev lowpass filter state-space description transistor models two-decade range

Three schemes for linearising the transconductance of the basic differential pair in subthreshold CMOS are examined: (i) multiple asymmetric differential pairs, (ii) source degeneration via symmetric diffusers, and (iii) source degeneration via a single diffuser. Using a maximally flat optimising criterion, the linear range of the basic differential pair can be increased by 4-8 times

Keywords: CMOS analogue integrated circuits continuous time filters differential amplifiers linear range linearised differential transconductors maximally flat optimising criterion multiple asymmetric differential pairs source degeneration subthreshold CMOS symmetric diffusers

Metal-filled microelectrodes are best for high-frequency work; fluid-filled ones are best for low frequencies and dc. Both have advantages and drawbacks. This paper gives the results of experience with both sorts of probe. Practical hints and recipes are included because these seldom appear in detail.

There are several groups of researchers developing cell-based biosensors for chemical and biological warfare agents based on electrophysiologic monitoring of cells. In order to transition such sensors from the laboratory to the field, a general-purpose hardware and software platform is required. This paper describes the design, implementation, and field-testing of such a system, consisting of cell-transport and data acquisition instruments. The cell-transport module is a self-contained, battery-powered instrument that allows various types of cell-based modules to be maintained at a preset temperature and ambient CO₂ level while in transit or in the field. The data acquisition module provides 32 channels of action potential amplification, filtering, and real-time data streaming to a laptop computer. At present, detailed analysis of the data acquired is carried out off-line, but sufficient computing power is available in the data acquisition module to enable the most useful algorithms to eventually be run real-time in the field. Both modules have sufficient internal power to permit realistic field-testing, such as the example presented in this paper.

Keywords: Extracellular recording; Portable; Biosensor

A fundamental technical hurdle in systems neuro- physiology has been to record the activity of individual neurons in situ while using microstimulation to activate inputs or outputs. Stimulation artifact at the recording electrode has largely limited the usefulness of combined stimulating and recording to using single stimulation pulses (e.g., orthodromic and antidromic activation) or to presenting brief trains of pulses to look for transient responses (e.g., paired-pulse stimulation). Using an adaptive filter, we have developed an on-line method that allows continuous extracellular isolation of individual neuron spikes during sustained experimental microstimulation. We show that the technique accurately and robustly recovers neural spikes from stimulation-corrupted records. Moreover, we demonstrate that the method should generalize to any recording situation where a stereotyped, triggered transient might obscure a neural event.

Keywords: Action potential, adaptive filter, microstimulation, spike detection, stimulation artifact

A detailed characterization of the neuron-to-microtransducer junction, based on the equivalent electric-circuit approach, is provided. The recording of action potentials is then simulated with the general-purpose network-analysis program SPICE. Both noble-metal microelectrodes and insulated-gate FETs are considered. The responses of such devices are characterized as functions of several parameters, e.g. sealing impedance, density of ionic currents in the cell membrane, and spatial discontinuities of the adhesion process. It is shown that the various signal shapes reported in the literature can be reproduced and interpreted in terms of time derivatives of the action potential. In this way, the shape of any experimental signal can be interpreted on the basis of a specific sealing condition. Possible future improvements in microtransducer design, based on the proposed approach, are also suggested.

Keywords: bioelectric potentials biological techniques and instruments cellular biophysics equivalent circuits neurophysiology physiological models SPICE action potential time derivatives adhesion process cell membrane equivalent electric-circuit approach experimental signal shape extracellular recording general-purpose network-analysis program insulated-gate FETs ionic currents density neural modeling neuron-to-microtransducer junction noble-metal microelectrodes patch recording sealing impedance spatial discontinuities

The majority of techniques for separating multiple single-unit spike trains from a multi-unit recording rely on the assumption that different cells exhibit action potentials having unique amplitudes and waveforms. When this assumption fails, due to the similarity of spike shape among different cels or to the presence of complex spikes with declining intra-burst amplitude, these methods lead to errors in classification. In an effort to avoid these errors, the stereotrode (McNaugton et al., 1983) and later the tetrode (O'Keefe and Reece, 1993l Wilson and McNaughton, 1993) recording techniques were developed. Because the altter technique has been applied primarily to the hippocampus, we sought to evaluate its performance in the neocortex. Multi-unit recordings, wising single tetrodes, were made at 28 sites in area 17 of 3 anesthetized cats. Neurons were activated with moving bars and square wave gratings. Single units were separated by components of the waveforms recorded on each channel. Using tetrodes, we recorded a total of 154 single cells (mean = 5.4, max = 9). By cross-checking the performance of the tetrode with the stereotrode and electrode, we found that the best of the 6 possible stereotrode pairs and the best of 4 possible electrodes from each tetrode yielded 102 (mean = 3.6, max = 7) and 95 (mean =3.4, max = 6) cells, respectively. Moreover, we found that the number of cells isolated at each site by the tetrode was greater than the stereoftrode or electrode in 16/28 and 28/28 cases, respectively. Thus, both stereotrodes, and particularly electrodes, often lumped 2 or more cells in a single cluster that could be easily separated by the tetrode. We conclude that the tetrode recording currently provides the best and most reliable method for the isolation of multiple single units in the neocortex using a single probe.

Keywords: Multi-unit recording; Visual cortex, area 17; Spike detection; Principal components analysis

Presents an overview of current design techniques for operational amplifiers implemented in CMOS and NMOS technology at a tutorial level. Primary emphasis is placed on CMOS amplifiers because of their more widespread use. Factors affecting voltage gain, input noise, offsets, common mode and power supply rejection, power dissipation, and transient response are considered for the traditional bipolar-derived two-stage architecture. Alternative circuit approaches for optimization of particular performance aspects are summarized, and examples are given.

Keywords: Electron device noise Field effect integrated circuits Linear integrated circuits Linear network synthesis Operational amplifiers electron device noise field effect integrated circuits linear integrated circuits linear network synthesis operational amplifiers

Mammalian spinal neuronal networks growing on arrays of photo-etched electrodes in culture provide a highly stable system for the long-term monitoring of multichannel, spontaneous or evoked electrophysiological activity. In the absence of the homeostatic control mechanisms of the central nervous system, these networks show remarkable sensitivities to minute chemical changes and mimic some of the properties of sensory tissue. These sensitivities could be enhanced by receptor up-regulation and altered by the expression of unique receptors. The fault-tolerant spontaneous network activity is used as a dynamic platform on which large changes in activity signify detection of chemical substances. We present strategies for the expression of novel supersensitivities to foreign molecules via genetic engineering that involves the grafting of ligand binding cDNA onto truncated native receptor DNA and the subsequent expression of such chimeric receptors.

Keywords: sensomimes; nerve cell biosensors; chimeric receptors; transfection; extracellular recording; liposomes

As part of an exploration of the feasibility of an epi-retinal prosthesis, we developed an experimental method to electrically stimulate and record from retinal neurons using a micro-fabricated multi-electrode array. An isolated retina is placed on an array of 10 um diameter disk electrodes with the ganglion cell side of the retina facing the electrode surfaces. The retina is bathed in oxygenated Ames' medium and warmed in order to sustain it in vitro for the duration of an experiment, typically 4-9 h. To reduce stimulus artifacts, the electrodes are grouped into two clusters-one used for stimulation and the other for recording-spaced several hundred microns apart, and electrodes are insulated with both silicon nitride and a 10 μ m thick layer of polyimide. Stimuli are delivered to the array using an optically isolated current source stimulator, and the resulting responses recorded with an eight channel nerve response amplifier. Stimulation and recording are performed under computer control. A variety of physiologic measurements is described in order to illustrate the strengths and drawbacks of this method.

Keywords: Electrode array; Extracellular stimulation; Electric stimulation; Retina; Rabbit; Neural recording

This paper proposes a new model concerning the channel charges in weak inversion injected from a turn-off MOSFET into a holding capacitor. This portion of charge injection has recently been newly observed, showing a significant contribution to the switch-induced error voltage on the switched capacitor. Our model is derived at the critical point where the device is operated in the transition region between strong inversion and weak inversion. This point has been expressed explicitly as a function of the DC input voltage, the threshold voltage, and the fall time of the gate voltage. The ability of the model in accurately determining quantitatively the impact of the weak inversion charge injection on the error voltage has been extensively judged experimentally and by two-dimensional mixed-mode simulation for a wide variety of design parameters such as the channel width and length, the holding capacitance, the fall time of the gate voltage, and the DC input voltage The assumptions utilized in the model development have also been validated.

Keywords: MOSFET capacitance field effect transistor switches semiconductor device models 2D mixed-mode simulation DC input voltage MOSFET analog switches channel charges channel length channel width design parameters gate voltage fall time holding capacitance holding capacitor quantitative model switch-induced error voltage switched capacitor threshold voltage turnoff MOSFET weak inversion charge injection

As extracellular electrode arrays with 100 or more active recording sites become more widely used for simultaneous recording of neural ensembles, practical data acquisition systems that can efficiently accommodate high electrode counts are needed. To reduce the high data rates associated with extracellular recordings from these arrays, various algorithms and systems have been designed to provide complete online detection and classification of extracellular spike waveforms. However, many of these algorithms require significant user supervision to ensure accurate performance. In this paper, we discuss the design and validation of a 100-channel PC-based system that can be used with arrays of extracellular electrodes such as the Utah Electrode Array. Instead of comprehensive online spike analysis, the system performs online detection and storage of the spike waveforms for offline classification. This strategy preserves the data of interest, reduces system complexity, and requires less user supervision during experiments.

Keywords: Multi-channel; Extracellular; Electrode; Spikes; Acquisition; Real-time

Standard photolithography is used to pattern a poly (ethylene glycol) (PEG)-like polymer onto silicon substrates. The coating has excellent non-fouling properties and good adhesion to various substrate materials, such as silicon, oxide, nitride, gold, and platinum. This method allows precise control of the shape, size, and alignment of the polymer, thus providing a reliable tool to pattern protein sheets as well a cell cultures. This method also enables the incorporation of patternened cell cultures with various predefined elements such as electrodes, channels, and sensors. To demonstrate the properties of our technique, we apply it to build cell cultures and to protect metallic electrodes from protein and cell adhesion. We show that the thin coatings provide excellent protection without compromising the conductivity of the electrodes.

Keywords: Bio-MEMS; Bio-fouling; Proteins; Cell cultures

There is a need among scientists and clinicians for low-noise, low-power biosignal amplifiers capable of amplifying signals in the mHz to kHz range while rejecting large dc offsets generated at the electrode-tissue interface. The advent of fully-implantable multielectrode arrays has created the need for fully-integrated micropower amplifiers. We designed and tested a novel bioamplifier that uses a MOS-bipolar pseudo-resistor to amplify signals down to the mHz range while rejecting large dc offsets. We derive the theoretical noise-power tradeoff limit-the noise efficiency factor-for this amplifier and demonstrate that our VLSI implementation approaches that limit. The resulting amplifier, built in a standard 1.5 μ m CMOS process, passes signals from 0.1 mHz to 7.2 kHz with an input-referred noise of 2.2 μ V_rms and a power dissipation of 80 μ W while consuming 0.16 mm² of chip area.

There is a need among scientists and clinicians for low-noise low-power biosignal amplifiers capable of amplifying signals in the millihertz-to-kilohertz range while rejecting large dc offsets generated at the electrode tissue interface. The advent of fully implantable multielectrode arrays has created the need for fully integrated micropower amplifiers. We designed and tested a novel bioamplifier that uses a MOS-bipolar pseudoresistor element to amplify low-frequency signals down to the millihertz range while rejecting large dc offsets. We derive the theoretical noise power tradeoff limit the noise efficiency factor for this amplifier and demonstrate that our VLSI implementation approaches this limit by selectively operating MOS transistors in either weak or strong inversion. The resulting amplifier, built in a standard 1.5 μ m CMOS process, passes signals from 0.025 Hz to 7.2 kHz with an input-referred noise of 2.2 μ V_rms and a power dissipation of 80 W while consuming 0.16 mm² of chip area. Our design technique was also used to develop an electroencephalogram amplifier having a bandwidth of 30 Hz and a power dissipation of 0.9 μ W while maintaining a similar noise power tradeoff.

Keywords: Analog integrated circuits, biosignal amplifier, low noise, low-power circuit design, neural amplifier, noise efficiency factor, subthreshold circuit design, weak inversion.

We developed an transistor-only version of our autozeroing floating-gate amplifier (AFGA). We use a subthreshold transistor to model the behavior of an electron-tunneling device, and we use another subthreshold transistor to model the behavior of pFET hot-electron injection. We have derived analytical models that completely characterize the amplifier and that are in good agreement with experimental data. This circuit is a bandpass filter, and behaves similarly to the AFGA with different operating parameters. Both the low-frequency and high-frequency cutoffs are controlled electronically, as is done in continuous-time filters. This circuit has a low-frequency cutoff at frequencies above 1 Hz, and therefore complements the operating regimes of the AFGA.

Keywords: amplifiers band-pass filters field effect transistor circuits hot carriers tunnelling autozeroing floating-gate amplifier bandpass filter electron tunneling device pFET hot electron injection subthreshold transistor transistor circuit model

We have developed a bandpass floating-gate amplifier that uses tunneling and pFET hot-electron injection to set its dc operating point adaptively. Because the hot-electron injection is an inherent part of the pFET's behavior, we obtain this adaptation with no additional circuitry. Because the gate currents are small, the circuit exhibits a high-pass characteristic with a cutoff frequency less than 1 Hz. The high-frequency cutoff is controlled electronically, as is done in continuous-time filters. We have derived analytical models that completely characterize the amplifier and that are in good agreement with experimental data for a wide range of operating conditions and input waveforms. This autozeroing floating-gate amplifier demonstrates how to use continuous-time floating-gate adaptation in amplifier design.

Keywords: MOS analogue integrated circuits continuous time filters high-pass filters hot carriers tunnelling autozeroing floating-gate amplifier bandpass floating-gate amplifier cutoff frequency dc operating point gate currents high-frequency cutoff high-pass characteristic pFET hot-electron injection tunneling

Signal degradation and an array size dictated by the number of available interconnects are the two main limitations inherent to standalone microelectrode arrays (MEAs). A new biochip consisting of an array of microelectrodes with fully-integrated analog and digital circuitry realized in an industrial CMOS process addresses these issues. The device is capable of on-chip signal filtering for improved signal-to-noise ratio (SNR), on-chip analog and digital conversion, and multiplexing, thereby facilitating simultaneous stimulation and recording of electrogenic cell activity. The designed electrode pitch of 250 μ m significantly limits the space available for circuitry: a repeated unit of circuitry associated with each electrode comprises a stimulation buffer and a bandpass filter for readout. The bandpass filter has corner frequencies of 100 Hz and 50 kHz, and a gain of 1000. Stimulation voltages are generated from an 8-bit digital signal and converted to an analog signal at a frequency of 120 kHz. Functionality of the read-out circuitry is demonstrated by the measurement of cardiomyocyte activity. The microelectrode is realized in a shifted design for flexibility and biocompatibility. Several microelectrode materials (platinum, platinum black and titanium nitride) have been electrically characterized. An equivalent circuit model, where each parameter represents a macroscopic physical quantity contributing to the interface impedance, has been successfully fitted to experimental results.

Keywords: Author Keywords: Microelectrode array; CMOS; Electrogenic cells; Cardiomyocites; Cell-based biosensor; Impedance

This paper describes how an extracellular microelectrode may be used to stimulate neurons with brief, rectangular pulses and afterwards directly record the resultant activity. Two obstacles are the stimulus artifact lingering in the electrical circuitry and transient tip potentials (TTPs) arising from ion depletion at the electrode-tissue interface. Electronic switching between the stimulus source and the recording amplifier eliminates direct stimulus artifact from the electrical circuitry, although high but acceptable switching artifact remains. TTPs revert with time constants that are prominent in the desired recording (0.1-1 ms) and can reach 50 mV when more than 1 μ A passes through a typical electrolyte-filled micropipette (for example 2-4 MΩ, filled with 3 M NaCl, and placed in 0.1 M NaCl). They are always negative when cations flow into the tip, they are accompanied by a rise in microelectrode impedance, and they increase as a function of the resting electrode impedance, the duration and amplitude of applied current, and the dilution of the external electrolyte. TTPs were subtracted by differential recording and stimulation through matched micropipettes (one in the brain and one in contiguous electrolyte) and in addition were reduced by pressure ejection of electrolyte. Directly elicited spikes (single or multiple) were detected about 0.5 ms after delivery of a rectangular stimulus pulse in the cerebellar cortex of pentobarbital-anesthetized rats. Typically, 3-4 units could be excited by less than 3 μ A cathodal currents at any recording site. All-or-nothing properties, thresholds, and refractoriness to a second pulse within 2-4 ms verified the neuronal nature of the recorded signals. Complex wave forms, probably generated synaptically, were also seen. The technique of coincident extracellular recording and stimulation can be used as a universal search stimulus during microelectrode penetrations through the brain and in determining threshold-distance relations for extracellular stimulation. Where cell penetrations are unstable, it might be usefully substituted for intracellular technique in testing a neuron's behavioral or physiological influences or in exploring a cell membrane's response to drugs (in terms of excitability rather than voltage and impedance).

Keywords: Extracellular recording; Stimulation method; Single neurons

Several multi-electrode array devices integrating planar metal electrodes were designed in the past 30 years for extracellular stimulation and recording from cultured neuronal cells and organotypic brain slices. However, these devices are not well suited for recordings from acute brain slice preparations due to a dead cell layer at the tissue slice border that appears during the cutting procedure. To overcome this problem, we propose the use of protruding 3D electrodes, i.e. tip-shaped electrodes, allowing tissue penetration in order to get closer to living neurons in the tissue slice. In this paper, we describe the design and fabrication of planar and 3D protruding multi-electrode arrays. The electrical differences between planar and 3D protruding electrode configuration were simulated and verified experimentally. Finally, a comparison between the planar and 3D protruding electrode configuration was realized by stimulation and recording from acute rat hippocampus slices. The results show that larger signal amplitudes in the millivolt range can be obtained with the 3D electrode devices. Spikes corresponding to single cell activity could be monitored in the hippocampus CA3 and CA1 region using 3D electrodes.

Keywords: Multi-electrode array; Protruding electrodes; Tip-shaped electrodes; Simulation; Acute slice; Rat hippocampus; Electrical stimulation; Extracellular recording

This paper considers compact, large singal linear, log-domain circuits that exploit the translinear characteristics of MOS transistors in weak inversion for low power applications in the audio frequency range. We synthesize a single pole log-domain lowpass filter from its state space description and extend the circuit topology to implement a two pole bandpass. Experimental results from a custom fabricated 2 μ m CMOS chip are presented that confirm circuit funcionality and tunability.

This paper presents a new reversed nested Miller compensation technique for multistage operational amplifier (opamp) design. The new compensation technique inverts the sign of the right half complex plane zero and shifts the frequency of the complex conjugate poles to a higher frequency. Simulation results indicate that the gain-bandwidth product and settling time are improved by factors of two and three, respectively, without degrading stability and power consumption. To verify the proposed technique, a three-stage opamp is fabricated with 0.6 μ m CMOS technology. The measured results of the test circuit agree with the results that are obtained from theoretical analysis and circuit simulation.

Keywords: CMOS analogue integrated circuits, buffer circuits, compensation, operational amplifiers, poles and zeros, 0.6 micron, CMOS circuit, gain-bandwidth product, nulling resistor, multistage operational amplifier, poles and zeros, reversed nested Miller compensation, settling time, voltage buffer

This paper describes a 3-D microelectrode array for the chronic recording of single-unit activity in the central nervous system. The array is formed by a microassembly of planar silicon multishank microporbes, which are precisely positioned in a micromachined platform that resides on the surface of the cortex. Interconnects between the probes and the platform are formed using electroplated nickel lead transfers, implemented using automated computer control. All dimensions are controlled to ±1 μ m and shank/probe separations as small as 100 μ m are possible. Four-probe 16-shank prototype arrays have been tested chronically in guinea pig cortex. After three months in vivo, no significant tissue reaction has been observed surrounding these structures when they remain free to move with the brain, with normal apperaring tissue between shanks spaced at 150 μ m to 200 μ m intervals. The array structure is compatible with the use of signal processing circuitry both on the probes and on the platform. A platform-based signal processing system has been designed to interface with several active probes, providing direct analog access to the recording sites, performing on-chip analog-to-digital conversion of neural activity, and providing simple binary-output recognition of a single-unit spike events using a user-input threshold voltage.

Future hybrid neuron-semiconductor chips will consist of complex neural networks that are directly interfaced to electronic integrated circuits. They will help us to understand the dynamics of neuronal networks and may lead to novel computational facilities. Here we report on an elementary step towards such neurochips. We designed and fabricated a silicon chip for multiple two-way interfacing, and cultured on it pairs of neurons from the pedal ganglia of the snail Lymnaea stagnalis. These neurons were joined to each other by an electrical synapse, and to the chip by a capacitive stimulator and a recording transistor. We obtained a set of neuroelectronic units with sequential and parallel signal transmission through the neuron-silicon interface and the synapse, with a bidirectionally interfaced neuron-pair and with a signal path from the chip through a synaptically connected neuron pair back to the chip. The prospects for assembling more involved hybrid networks on the basis of these neuroelectronic units are considered.

A second-generation multichannel probe designed for measuring single-unit activity in neural structures is described. The probe includes CMOS circuitry for electronically positioning the recording sites with respect to the active neurons and for amplifying and multiplexing the recorded signals. The probe selects eight active recording sites from among 32 on the probe shank using a static input channel selector. The neural signals on the selected channels are then amplified and multiplexed to the outside world. The probe offers a typical AC gain of 300 (15 Hz to 7 kHz), a DC gain of 0.3, and an equivalent input noise of 15 μ V_rms. Operating from a single 5-V supply, the probe dissipates 2.5 mW of power and implements channel selection, self-testing, data output, and initialization using three external leads. The probe is realized using 12 masks in a high-yield single-sided dissolved wafer process with a 3 μ m feature size for the circuitry and a 3 μ m pitch on the electrode shanks

The spatial propagation of synchronized activity in cortical slice cultures was characterized by multi-site extracellular recording. Spontaneous activity was studied in normal culture medium, and in bicuculline- or kainic acid-containing media. A common feature in all these conditions was that activity was generated first in superficial layers (i.e., layer I/II) before spreading over the whole area of the slice. In culture medium or bicuculline-containing medium, the initiation site of the activity was not constant and showed a large variety of patterns of horizontal propagation. Kainic acid induced epileptiform activity, consisting of intense initial bursts followed by repetitive after-discharges. Though the patterns of spatial propagation of the bursts were variable as in the other conditions, the after-discharges followed a constant path. Cross-correlation analysis indicated that the network moved in a graded fashion to a steady state during the sequence of after-discharges.

Keywords: Planar electrode arrays; Synchronized activity; Cortical slice cultures

The capability for multisite stimulation is one of the biggest potential advantages of microelectrode arrays (MEAs). There remain, however, several technical problems which have hindered the development of a practical stimulation system. An important design goal is to allow programmable multisite stimulation, which produces minimal interference with simultaneous extracellular and patch or whole cell clamp recording. Here, we describe a multisite stimulation and recording system with novel interface circuit modules, in which preamplifiers and transistor transistor logic-driven solid-state switching devices are integrated. This integration permits PC-controlled remote switching of each substrate electrode. This allows not only flexible selection of stimulation sites, but also rapid switching of the selected sites between stimulation and recording, within 1.2 ms. This allowed almost continuous monitoring of extracellular signals at all the substrate-embedded electrodes, including those used for stimulation. In addition, the vibration-free solid-state switching made it possible to record whole-cell synaptic currents in one neuron, evoked from multiple sites in the network. We have used this system to visualize spatial propagation patterns of evoked responses in cultured networks of cortical neurons. This MEA-based stimulation system is a useful tool for studying neuronal signal processing in biological neuronal networks, as well as the process of synaptic integration within single neurons.

The authors attribute the artifact to the dc offset of the electrode and charge storage during stimulation. They construct a stimulation and recording system, using discrete components, to account for these sources. A sample and hold circuit stores the dc offset, which is added to the stimulation voltage. This also makes the effect of stimulation on cells independent of the dc offset. Connecting the electrode to a low impedance discharge path after stimulation reduces the effect of charge storage. The authors demonstrate reproducibility of evoked responses and elimination of the stimulation artifact.

Keywords: Electrical stimulation, MEA, neuron, spike.

Activity-dependent modification of synaptic efficacy is widely recognized as a cellular basis of learning, memory, and developmental plasticity. Little is known, however, of the consequences of such modification on network activity. Using electrode arrays, we examined how a single, localized tetanic stimulus affects the firing of up to 72 neurons recorded simultaneously in cultured networks of cortical neurons, in response to activation through 64 different test stimulus pathways. The same tetanus produced potentiated transmission in some stimulus pathways and depressed transmission in others. Unexpectedly, responses were homogeneous: for any one stimulus pathway, neuronal responses were either all enhanced or all depressed. Cross-correlation of responses with the responses elicited through the tetanized site revealed that both enhanced and depressed responses followed a common principle: activity that was closely correlated before tetanus with spikes elicited through the tetanized pathway was enhanced, whereas activity outside a 40-ms time window of correlation to tetanic pathway spikes was depressed. Response homogeneity could result from pathway-specific recurrently excitatory circuits, whose gain is increased or decreased by the tetanus, according to its cross-correlation with the tetanized pathway response. The results show how spatial responses following localized tetanic stimuli, although complex, can be accounted for by a simple rule for activity-dependent modification.

A biosensor based on the extracellular recording of action potentials from excitable cells via a planar microelectrode array is being developed. The function of the sensor is a function of both the cellular response to environmental toxins and the coupling of the cellular potentials to the microelectrodes in the array. Because good electronic coupling between cell and amplifier requires that the microelectrodes have low impedance, the microelectrodes are electroplated with platinum black and the arrays are characterized by impedance measurements and by imaging using scanning electron microscopy (DEM) and x-ray photoelectron spectroscopy (XPS). SEM provides highly detailed images of the shape and structure of well-formed deposits of thickness on the order of 1 μ m or more. XPS reveals the presence of platinum deposits that are too thin to be detected by SEM. Impedance measurements show reductions in the electrical resistance at 100 Hz from roughly 60 MΩ or more 1 MΩ. The overall electronic coupling of biopotentials to the microelectrodes was demonstrated by recordings obtained from beating rat myocytes and from rat spinal cord cells.

Keywords: micorelectrodes, arrays, biosensors, cell culture, X ray photoelectron spectroscopy, scanning electron microscopy, electric impedance measurement, bioelectric potentials, platinum

The effects of pole-zero pairs (doublets) on the frequency response and settling time of operational amplifiers are explored using analytical techniques and computer simulation. It is shown that doublets which produce only minor changes in circuit frequency response can produce major changes in settling time. The importance of doublet spacing and frequency are examined. It is shown that settling time always improves as doublet spacing is reduced whereas the effect of doublet frequency is different for 0.1 and 0.01 percent error bands. Finally it is shown that simple analytical formulas can be used to estimate the influence of frequency doublets on amplifier settling time.

Keywords: Computer-aided circuit analysis Frequency response Operational amplifiers Poles and zeros Simulation computer-aided circuit analysis frequency response operational amplifiers poles and zeros simulation

Cultured spinal cord networks grown on microelectrode arrays display complex patterns of spontaneous burst and spike activity. During disinhibition with bicuculline and strychnine, synchronized burst patterns routinely emerge. However, the variability of both intra- and interculture burst periods and durations are typically large under these conditions. As a further step in simplification of synaptic interactions, we blocked excitatory AMPA synapses with 2,3-dioxo-6-nitro-1,2,3,4-tetrahydrobenzoquinoxaline-7-sulphonamide (NBQX), resulting in network activity mediated through the N-methyl-D-aspartate (NMDA) receptor (NMDAONLY). This activity was APV sensitive. The oscillation under NMDAONLY conditions at 37 ^oC was characterized by a period of 2.9±0.3 s (16 separate cultures). More than 98% of all neurons recorded participated in this highly rhythmic activity. The temporal coefficients of variation, reflecting the rhythmic nature of the oscillation, were 3.7, 4.7, and 4.9% for burst rate, burst duration, and interburst interval, respectively [mean coefficients of variation (CVs) for 16 cultures]. The oscillation persisted for at least 12 h without change (maximum observation time). Once established, it was not perturbed by agents that block mGlu receptors, GABAB receptors, cholinergic receptors, purinergic receptors, tachykinin receptors, serotonin (5-HT) receptors, dopamine receptors, electrical synapses, burst afterhyperpolarization, NMDA receptor desensitization, or the hyperpolarization-activated current. However, the oscillation was destroyed by bath application of NMDA (20-50 μ M). These results suggest a presynaptic mechanism underlying this periodic rhythm that is solely dependent on the NMDA synapse. When the AMPA/kainate synapse was the sole driving force (n = 6), the resulting burst patterns showed much higher variability and did not develop the highly periodic, synchronized nature of the NMDAONLY activity. Network size or age did not appear to influence the reliability of expression of the NMDAONLY activity pattern. For this reason, we suggest that the NMDAONLY condition unmasks a fundamental rhythmogenic mechanism of possible functional importance during periods of NMDA receptor-dominated activity, such as embryonic and early postnatal development.

We have utilized cultured neuronal networks grown on microelectrode arrays to demonstrate rapid, reliable detection of a toxic compound, trimethylolpropane phosphate (TMPP). Initial experiments, which were performed blind, demonstrated rapid classification of the compound as a convulsant, a finding consistent with previous whole animal neurobehavioral studies. TMPP (2-200 μ M) reorganized network spike activity into synchronous, quasi-periodic burst episodes. Integrated burst amplitudes invariably increased, reflecting higher spike frequencies within each burst. The variability of network burst parameters, quantified as coefficients of variation (CVs), was decreased. Mean CVs for burst duration, interburst interval, and burst rate were lowered by 42±13, 58±5.5, and 62±1.8%, respectively (mean±SEM, n=8 cultures, 197 channels). These changes in network activity paralleled the effects induced by bicuculline, a known disinhibitory and seizure-inducing drug, and confirmed classification of TMPP as a potential epileptogenic compound. Simple pharmacological tests permit exploration of mechanisms underlying observed activity shifts. The EC50 for GABA inhibition of network activity was increased from 2.8 to 7.0 μ M by 20 μ M TMPP and to 20.5 μ M by 200 μ M TMPP. Parallel dose-response curves suggest that TMPP acts by a competitive antagonism of GABA inhibition, and are consistent with reported patch-clamp analysis of TMPP-induced reduction of inhibitory postsynaptic current amplitudes. The potency of TMPP in generating epileptiform activity in vitro was comparable to concentrations reported for in vivo studies. TMPP and bicuculline produced both increases and decreases in burst rate depending on native spontaneous bursting levels. These results demonstrate a need for multivariate analysis of network activity changes to yield accurate predictions of compound effects.

Keywords: Trimethylolpropane phosphate; Neuronal network; Neurotoxicants

This paper describes low-voltage neural stimulating circuitry developed using fully complementary BiCMOS (FC-BiCMOS) process technology for providing charge-balanced bipolar stimulating currents to tissue in the central nervous system. The electronics features an FC-BiCMOS buffer, a 7-b biphasic current-output digital-to-analog converter, a 14-b frequency divider, a nonoverlapping two-phase clock generator, and an auto timeout safety scheme while driving any two of eight selected sites from 0 to ±126 μ A with ±2 μ A resolution. The circuit area is 1.6 mm² in 3 μ m features. Micropower circuit techniques allow the probe to dissipate < 10 μ W in standby and operate at 10 MHz from ±2.5 V supplies.

Keywords: BiCMOS integrated circuits bioelectric phenomena biomedical electronics neural nets 10 MHz 10 muW 2.5 V biological neural network central nervous system charge-balanced bipolar stimulating current fully complementary BiCMOS circuit low-voltage electronics micropower circuit stimulation tissue

We present a method for the detection of action potentials, an essential first step in the analysis of extracellular neural signals. The low signal-to-noise ratio (SNR) and similarity of spectral characteristic between the target signal and background noise are obstacles to solving this problem and, thus, in previous studies on experimental neurophysiology, only action potentials with sufficiently large amplitude have been detected and analyzed. In order to lower the level of SNR required for successful detection, we propose an action potential detector based on a prudent combination of wavelet coefficients of multiple scales and demonstrate its performance for neural signal recording with varying degrees of similarity between signal and noise. The experimental data include recordings from the rat somatosensory cortex, the giant medial nerve of crayfish, and the cutaneous nerve of bullfrog. The proposed method was tested for various SNR values and degrees of spectral similarity. The method was superior to the Teager energy operator and even comparable to or better than the optimal linear detector. A detection ratio higher than 80% at a false alarm ratio lower than 10% was achieved, under an SNR of 2.35 for the rat cortex data where the spectral similarity was very high.

Keywords: bioelectric potentials   biological techniques   cellular biophysics   neurophysiology   signal detection   signal processing   somatosensory phenomena   spectral analysis   wavelet transforms   action potential detection   bullfrog   crayfish   cutaneous nerve   detection ratio   extracellular neural signal recording   extracellular neural signals analysis   giant medial nerve   low signal-to-noise ratio   multiple scales   neural signal recording   rat cortex data   rat somatosensory cortex   spectral similarity   spectral similarity degree

For more than three decades, it has been possible to use microlithographically fabricated extracellular electrodes to record action potential (AP) signals from electrically active cells such as neurons of intact organisms. There has also been a steady evolution of techniques to interface between electronic circuits and neural or cardiac cells cultured on arrays of such electrodes, mainly directed toward basic science goals. More recently, such combinations of living cells and electronics have been harnessed as tools for the detection of chemical and biological toxins, and for screening of pharmacologically active compounds. This paper presents a survey of the relevant technologies, cell types available, specific requirements for applications, and discussion of opportunities for future development.

Keywords: biosensors cellular biophysics microelectrodes action potential signal cardiac culture cell-based biosensor electronic sensor living cellular component microelectrode array neural culture pharmaceutical screening toxin detection

A technique that enables the variation of bias currents in a filter without causing disturbances at the output is presented. Thus, the bias current can be kept at the minimum value necessary for the total input signal being processed, reducing the noise and power consumption. To demonstrate this approach, a dynamically biased log-domain filter has been designed in a 0.25 μ m BiCMOS technology. The chip occupies 0.52 mm². In its quiescent condition, the filter consumes 575 μ W and has an output noise of 4.4 nA_rms. Signal-to-noise ratio greater than 50 dB over 3 decades of input and total harmonic distortion less than 1% for inputs less than 2.5 mA peak are achieved. The bias can be varied to minimize noise and power consumption without disturbing the output.

Keywords: BiCMOS analogue integrated circuits active filters harmonic distortion integrated circuit noise low-power electronics 0.25 micron 2.5 mA 575 muW BiCMOS technology analog active filter dynamic bias current log-domain filter output noise power consumption signal-to-noise ratio total harmonic distortion

In this paper, we show and validate a reliable circuit design technique based on source voltage shifting for current-mode signal processing down to femtoamperes. The technique involves specific-current extractors and logarithmic current splitters for obtaining on-chip subpicoampere currents. It also uses a special on-chip sawtooth oscillator to monitor and measure currents down to a few femtoamperes. This way, subpicoampere currents are characterized without driving them off chip and requiring expensive instrumentation with complicated low leakage setups. A special current mirror is also introduced for reliably replicating such low currents. As an example, a simple log-domain first-order low-pass filter is implemented that uses a 100-fF capacitor and a 3.5-fA bias current to achieve a cutoff frequency of 0.5 Hz. A technique for characterizing noise at these currents is also described and verified. Finally, transistor mismatch measurements are provided and discussed. Experimental measurements are shown throughout the paper, obtained from prototypes fabricated in the AMS 0.35 μ m three-metal two-poly standard CMOS process.

Keywords: CMOS analogue integrated circuits   current mirrors current-mode circuits   integrated circuit design leakage currents   low-pass filters   0.35 micron   0.5 Hz   100 fF   3.5 fA   current mirror   current-mode signal processing   cutoff frequency   femtoampere current-mode circuits   leakage setups   log-domain first-order low-pass filter   logarithmic current splitters   on-chip sawtooth oscillator   on-chip subpicoampere currents   source voltage shifting specific-current extractors   three-metal two-poly standard CMOS   transistor mismatch

A description is given of a functional silicon micromachined device that permits non-invasive, bidirectional, highly specific communication with cultured mammalian neurons. The heart of the system is a well structure that holds the cell in close proximity to a metal extracellular electrode while permitting normal outgrowth of axons and dendrites. An iterative approach is used to create a design that allows normal growth of the neurons while preventing their escape. An array of 16 such neurowells makes it possible to perform studies of biological neural network development and function with unprecedented detail.

Keywords: HIPPOCAMPAL-NEURONS, MICROELECTRODE ARRAY, ACTION-POTENTIALS, RAT HIPPOCAMPUS, PROPAGATION, DENDRITES, BEHAVIOR, DESIGN, AXONS, SLICE

The neurochip is a silicon micromachined device upon which cultured mammalian neurons can be continuously and individually monitored and stimulated. The neurochip is based upon a 4×4 array of metal electrodes, each of which has a caged well structure designed to hold a single mature cell body while permitting normal outgrowth of neural processes. We demonstrate that this device is capable of maintaining cell survival, and that the electrodes can both record and stimulate electrical activity in individual cells with no crosstalk between channels.

A pair of symmetrical and versatile equivalent circuits suitable for two pole single and two stage operational amplifiers is presented. The improved accuracy of the new equivalent circuits is necessary when designing the compensation network of high speed amplifiers to minimise their very strongly phase-dependent settling time, which is important for sampled data analogue signal processing circuits. These models also identify a critical effective capacitance, which is shown to govern the settling behaviour of the operational amplifier.

Keywords: equivalent circuits operational amplifiers poles and zeros sampled data systems signal processing equipment GaAs operational amplifier compensation network critical effective capacitance high speed sampled data operational amplifier phase-dependent settling time sampled data analogue signal processing circuits symmetrical equivalent circuits two pole single operational amplifier two stage operational amplifiers versatile equivalent circuits

The phenomena of learning and memory are inherent to neural systems that differ from each other markedly. The differences, at the molecular, cellular and anatomical levels, reflect the wealth of possible instantiations of two neural learning and memory universals: (i) an extensive functional connectivity that enables a large repertoire of possible responses to stimuli; and (ii) sensitivity of the functional connectivity to activity, allowing for selection of adaptive responses. These universals can now be fully realized in ex-vivo developing neuronal networks due to advances in multi-electrode recording techniques and desktop computing. Applied to the study of ex-vivo networks of neurons, these approaches provide a unique view into learning and memory in networks, over a wide range of spatio-temporal scales. In this review, we summarize experimental data obtained from large random developing ex-vivo cortical networks. We describe how these networks are prepared, their structure, stages of functional development, and the forms of spontaneous activity they exhibit (Sections 2-4). In Section 5 we describe studies that seek to characterize the rules of activity-dependent changes in neural ensembles and their relation to monosynaptic rules. In Section 6, we demonstrate that it is possible to embed functionality into ex-vivo networks, that is, to teach them to perform desired firing patterns in both time and space. This requires `closing a loop' between the network and the environment. Section 7 emphasizes the potential of ex-vivo developing cortical networks in the study of neural learning and memory universals. This may be achieved by combining closed loop experiments and ensemble-defined rules of activity-dependent change.

The purpose of this paper is to characterize the neuron-microelectrode junction, based on the equivalent electric-circuit approach. As a result, recording of action potentials can be simulated with a general-purpose circuit simulation program such as HSPICE. The response of the microelectrode was analyzed as a function of parameters such as sealing resistance and adhesion conditions. The models of the neuron and microelectrode implemented in HSPICE were first described. These models were used to simulate the behavior of the junction between a patch of neuronal membrane (described by the compartmental model) and a microelectrode.

Keywords: Circuit model Hspice simulation extracellular recording microelectrode neuro-electronic junction neurons

Neural processes display rhythmic oscillations in local field potentials; identification of their characteristic frequencies is complicated due to their highly non-stationary nature. A simple technique, combining Fourier transforms and correlation coefficients yields unambiguous determinations of the frequencies without a priori filtering. This procedure also provides quantitative information concerning interactions between frequencies. Fundamental frequencies in local field potential data acquired from the hippocampus, cortex, and striatum from awake, behaving rats were calculated using this technique. Characteristic frequencies identified using this technique from hippocampus and cortex agreed with known oscillations. Application to dorsal striatal local field potentials identified a low-frequency theta component as well as a narrow gamma band oscillation at 50-55 Hz.

Keywords: Fundamental frequency; Oscillation; Gamma rhythm; Theta rhythm; 1/f Noise; Hippocampus; Postsubiculum; Striatum

A review of various aspects of electrode-electrolyte interface impedance is presented. The effect of electrode topography on the form and magnitude of the interface impedance is discussed. The work of Schwan and his colleagues on the non-linearity of the interface impedance is presented and interpreted. The electrical properties of silver-silver chloride electrodes (much used in a wide range of biomedical applications) are also briefly reviewed.

Keywords: electrode-electrolyte interface; A.C. impedance; nonlinearity; silver-silver chloride

Most pioneering biosensors employ a biological molecule-an enzyme, antibody, nucleic acid, or other element-to recognize sample molecules of interest. Recognition takes place via biochemical binding through hydrogen bonding, charge-charge interactions, and so forth. A secondary process, such as a colorimetric indicator reaction or the amplification of a weak bioelectric signal, informs the user of the primary molecular recognition event. A few such molecular recognition biosensors are familiar as consumer products, including glucose monitors (enzyme-based), pregnancy test strips (antibody-based), and paternity test kits (nucleic acid-based). This article describes the use and design features of modern biosensors.

This paper presents a low-power low-noise fully integrated bandpass operational amplifier for a variety of biomedical neural recording applications. A standard two-stage CMOS amplifier in a closed-loop resistive feedback configuration provides a stable ac gain of 39.3 dB at 1 kHz. A subthreshold PMOS input transistor is utilized to clamp the large and random dc open circuit potentials that normally exist at the electrode-electrolyte interface. The low cutoff frequency of the amplifier is programmable up to 50 Hz, while its high cutoff frequency is measured to be 9.1 kHz. The tolerable dc input range is measured to be at least ±0.25 V with a dc rejection factor of at least 29 dB. The amplifier occupies 0.107 mm² in die area, and dissipates 115 μ W from a 3 V power supply. The total measured input-referred noise voltage in the frequency range of 0.1-10 kHz is 7.8 μ V_rms. It is fabricated using AMI 1.5 μ m double-poly double-metal n-well CMOS process. This paper presents full characterization of the dc, ac, and noise performance of this amplifier through in vitro measurements in saline using two different neural recording electrodes.

Keywords: Bandpass amplifier dc baseline stabilization fully integrated low-noise low-power neural recording

The realization of a commercially viable, general-purpose quad CMOS amplifier is presented, along with discussions of the tradeoffs involved in such a design. The amplifier features an output swing that extends to either supply rail, together with an input common-mode range that includes ground. The device is especially well suited for single-supply operation and is fully specified for operation from 5 to 15 V over a temperature range of -55 to +125 ^oC. In the areas of input offset voltage, offset voltage drift, input noise voltage, voltage gain, and load driving capability, this implementation offers performance that equals or exceeds that of popular general-purpose quads or bipolar of Bi-FET construction. On a 5-V supply the typical V_os is 1 mv, V_os drift is 1.3 μ V/ ^oC, 1-kHz noise is 36 nV Hz^1/2, and gain is one million into a 600 Ω load. This device achieves its performance through circuit design and layout techniques as opposed to special analog CMOS processing, thus lending itself to use on system chips built with digital CMOS technology.

Keywords: CMOS integrated circuits Linear integrated circuits Operational amplifiers linear integrated circuits operational amplifiers

This active probe can be used for the long-term recording of extracellular neural biopotentials and as a basis for closed-loop neural prostheses. The probe incorporates on-chip circuitry for amplifying, multiplexing, and buffering neural signals recorded from ten recording electrodes spaced 100 μ m apart. It requires only three leads and operates from a single 5-V supply. On-chip self-test circuitry for testing electrode impedance levels is provided. The on-chip circuitry is fabricated in a die area of 1.3 mm² using 6 μ m LOCOS enhancement-depletion (E-D) NMOS technology, and dissipates 5 mW of power. The probe is 4.7 mm long and 15 um thick, and has a shank which tapers from 160 μ m near the base to less than 15 μ m near the tip.

Keywords: Bioelectric potentials Biomedical electronics Biomedical equipment Field effect integrated circuits Probes Prosthetics bioelectric potentials biomedical electronics biomedical equipment field effect integrated circuits probes prosthetics

We report the use of a gold coating on microelectrode arrays (MEAs) to enable the use of the relatively reliable surface modification chemistry afforded by alkanethiol self-assembled monolayers (SAMs). The concept is simple and begins with planar MEAs, which are commercially available for neuronal cell culture and for brain slice studies. A gold film, with an intermediate adhesive layer of titanium, is deposited over the insulation of an existing MEA in a manner so as to be thin enough for transmission light microscopy as well as to avoid electrical contact to the electrodes. The alkanethiol-based linking chemistry is then applied for the desired experimental purpose. Here we show that polylysine linked to alkanethiol SAM can control the geometry of an in vitro hippocampal neuronal network grown on the MEA. Furthermore, recordings of neuronal action potentials from random and patterned networks suggest that the gold coating does not significantly alter the electrode properties. This design scheme may be useful for increasing the number of neurons located in close proximity to the electrodes. Realization of in vitro neuronal circuits on MEAs may significantly benefit basic neuroscience studies, as well as provide the insight relevant to applications such as neural prostheses or cell-based biosensors. The gold coating technique makes it possible to use the rich set of thiol-based surface modification techniques in combination with MEA recording.

Keywords: Cell patterning, hippocampal pyramidal cells, MEA, neural recording, neuronal network, self-assembled monolayer (SAM)

We have developed a three-dimensional silicon electrode array which provides 100 separate channels for neural recording in cortex. The device is manufactured using silicon micromachining techniques, and we have conducted acute recording experiments in cat striate cortex to evaluate the recording capabilities of the array. In a series of five acute experiments, 58.6% of the electrodes in the array were found to be capable of recording visually evoked responses. In the most recent acute study, the average signal-to-noise ratio for recordings obtained from 56 of the electrodes in the array was calculated to be 5.5:1. Using standard window discrimination techniques, an average of 3.4 separable spikes were identified for each of these electrodes. In order to compare the two-dimensional mapping capabilities of the array with those derived from other technologies, orientation preference and ocular dominance maps were generated for each of the evoked responses. Histological evaluation of the implant site indicates some localized tissue insult, but this is likely due to the perfusion procedure since high signal-to-noise ratio neural responses were recorded. The recording capabilities of the Utah Intracortical Electrode Array in combination with the large number of electrodes available for recording make the array a tool well suited for investigations into the parallel processing mechanisms in cortex.

Keywords: Multielectrode; Recording; Electrode array; Electrophysiology; Parallel processing; Visual cortex; visual cortex; single unit activity

A technique has been developed in which a planar array of 32 microelectrodes, arranged in a 4 by 8 pattern with 200 μ m separation, is used to record from and stimulate the hippocampal slice preparation at multiple sites. Control of media flow past the tissue is critical to observe signals and preserve viability. Active supression circuitry is used to prevent device saturation due to large stimulation artifacts. The field potentials recorded are spatially unique and provide a 2-dimensional description of the underlying population activity in the various pyramidal strata and subpopulations. Multisite stimulation is also possible with the array, permitting the experimenter to quickly stimulate and record from brain slices in many spatial patterns.

Keywords: Multichannel recordings; Multisite stimulation; Electrode array; Hippocampal slice; Brain slice

A general treatment of the effect of surface roughness on the impedance of ideally polarizable (blocking) electrodes is proposed. In terms of fractal geometry, surface irregularities are characterized solely by the effective fractional dimension, D. The advantage of this approach is that the structure of the irregularities is irrelevant if the surface is self-similar. The admittance, Y, of self-similar blocking electrodes is shown to depend on the frequency ω as Y = σ(i^α), ie, any blocking electrode with fractal surface behaves as a constant phase element (CPE) observed experimentally in many and diverse systems. The fractional exponent α is directly related to D as α = 1/(D - 1), hence α can be regarded as a measure of surface roughness. The coefficient σ is shown to be a simple explicit function of electrolyte conductivity and double-layer capacitance thus enabling one to study the latter even when the interface behaves as a CPE instead of being an ideal capacitance.

We have developed, manufactured, and tested two analog CMOS integrated circuit ``neurochips'' for recording from arrays of densely packed neural electrodes. Device A is a 16-channel buffer consisting of parallel noninverting amplifiers with a gain of 2 V/V. Device B is a 16-channel two-stage analog signal processor with differential amplification and high-pass filtering. It features selectable gains of 250 and 500 V/V as well as reference channel selection. The resulting amplifiers on Device A had a mean gain of 1.99 V/V with an equivalent input noise of 10 μ V_rms. Those on Device B had mean gains of 53.4 and 47.4 dB with a high-pass filter pole at 211 Hz and an equivalent input noise of 4.4 μ V_rms. Both devices were tested in vivo with electrode arrays implanted in the somatosensory cortex.

Keywords: CMOS analogue integrated circuits   amplifiers   arrays   biomedical electrodes   biomedical electronics   high-pass filters   medical signal processing   neurophysiology   prosthetics   somatosensory phenomena   1.99 V   10 muV   16-channel two-stage analog signal processor   211 Hz   4.4 muV   47.4 dB   53.4 dB   artificial devices interfacing with brain   differential amplification   high-pass filtering   multichannel integrated circuits   neural amplifier   neural recording   neural signal processing   neurochip   neuroprosthetics

We present the design and testing of a 16-channel analog amplifier for processing neural signals. Each channel has the following features: (1) variable gain (70-94 dB), (2) four high pass Bessel filter poles (f_-3dB=445 Hz), (3) five low pass Bessel filter poles (f_-3dB=6.6 kHz), and (4) differential amplification with a user selectable reference channel to reject common mode background biological noise. Processed signals are time division multiplexed and sampled by an on-board 12-bit analog to digital converter at up to 62.5k samples/s per channel. The board is powered by two low dropout voltage regulators which may be supplied by a single battery. The board measures 8.1 cm × 9.9 cm, weighs 50 g, and consumes up to 130 mW. Its low input-referred noise (1.0 μ V_RMS) makes it possible to process low amplitude neural signals; the board was successfully tested in vivo to process cortically derived extracellular action potentials in primates. Signals processed by this board were compared to those generated by a commercially available system and were found to be nearly identical. Background noise generated by mastication was substantially attenuated by the selectable reference circuit. The described circuit is light weight and low power and is used as a component of a wearable multichannel neural telemetry system.

Keywords: Single unit recording; Neural signal processing; Low-power

We present the design, testing, and evaluation of a 16 channel wearable telemetry system to facilitate multichannel single unit recordings from freely moving test subjects. Our design is comprised of (1) a 16-channel analog front end board to condition and sample signals derived from implanted neural electrodes, (2) a digital board for processing and buffering the digitized waveforms, and (3) an index-card sized 486 PC equipped with an IEEE 802.11b wireless ethernet card. Digitized data (up to 12 bits of resolution at 31.25 k samples/s per channel) is transferred to the PC and sent to a nearby host computer on a wireless local area network. Up to 12 of the 16 channels were transmitted simultaneously for sustained periods at a range of 9 m. The device measures 5.1 cm × 8.1 cm × 12.4 cm, weighs 235 g, and is powered from rechargeable lithium ion batteries with a lifespan of 45 min at maximum transmission power. The device was successfully used to record signals from awake, chronically implanted macaque and owl monkeys.

Keywords: Neural telemetry; Single unit recording; Wireless; Neural data acquisition; Portable

This paper describes a fully-integrated bandpass amplifier for neural recording applications. Diode-connected sub-threshold-biased NMOS transistors in the feedback loop of the amplifier realize the high on-chip impedance necessary to eliminate the dc baseline potential of the electrode while amplifying the neural field and action potentials. The amplifier has an in-band gain of 38.2 dB, a dc gain of 0, an upper cutoff frequency of 24 kHz and a low frequency cutoff of 66 mHz. It consumes 92 μ W from ±1.5 V supplies and has an input-referred noise of 16.6 μ V_rms integrated from 100 Hz-10k Hz. The amplifier occupies 0.082 mm² in 3um features and is being used on a 64-site neural recording probe.

Keywords: Action potentials, dc baseline potential, field potentials, microelectrode, neural recording amplifier

This paper describes a 64 site, 8 channel silicon microelectrode for single-unit neural recording. The probe features integrated CMOS circuitry for electronic positioning of the active recording sites with respect to active neurons. On-chip capacitively coupled pre-amplifiers eliminate the DC baseline polarization of the electrode while providing a per channel gain of 1000. Time-division multiplexing circuitry is provided for sampling the 8 active channels onto one data lead. The on-chip circuitry consumes 834 μ W of power from ±1.5 V supplies and occupies 4.34 mm² of die area. The probe is fabricated using an 18 mask, single-sided, micromachined CMOS process with a 3 μ m minimum feature size

Keywords: CMOS analogue integrated circuits biocontrol biomedical electrodes biomedical electronics biomedical telemetry data acquisition microelectrodes micromachining neurophysiology preamplifiers prosthetics silicon time division multiplexing 1.5 V 834 muW CMOS circuitry DC baseline potential DC feedback path Si active recording arrays active recording sites capacitively coupled preamplifiers closed-loop fully-implanted prosthetic system control circuitry electronic positioning in-vivo data acquisition microelectrode micromachined CMOS process neural prosthesis neural recording probe on-chip electronics on-chip front-end selection silicon neural recording arrays single-unit neural recording telemetry chip time-division multiplexing circuitry

A folding architecture for a subthreshold CMOS transconductance amplifier is described. Good linearity is obtained for an extremely large differential input voltage, without loss in the common-mode voltage range. Theoretical noise analysis indicates a 6 dB improvement in the dynamic range compared to a simple single-pair MOS implementation. A prototype has been fabricated in a 2 μ m CMOS process, and experimental results are presented.

Keywords: CMOS analogue integrated circuits feedback amplifiers integrated circuit noise 2 micron CMOS transconductance amplifier common-mode voltage range folding architecture large-signal subthreshold amplifier linearity noise analysis

In this work, we describe a novel framework aimed at enhancing the communication and signal processing technology of microimplanted devices used for recording and stimulating neural cells. The power of the proposed framework stems from providing simple algorithms, yet efficient signal processing power that is suitable for on-chip microprobe design. The framework unifies our previous work on multiresolution analysis and array processing that was aimed at performing typical neural signal processing tasks such as noise suppression, source detection and separation, and information coding. Strategies for optimizing the information transfer have shown to greatly benefit from the optimal array processing mechanisms used and the compression achieved by expressing the data in the multiresolution domain. We demonstrate through simulated and experimental results that the framework provides the basis for simple and practical implementation for today's biosensor array technology requirements without compromising issues of bandwidth, detection and classification.

Keywords: array signal processing bioelectric potentials biomedical electrodes discrete wavelet transforms medical signal processing microelectrodes neurophysiology signal classification signal reconstruction MASSIT algorithm Poisson processes SVD module additive noise array model array signal processing technology biosensor array technology cascade of modules data compression information coding microimplanted devices microprobe neural recording devices multichannel neural recording multiresolution analysis neural signal processing noise suppression on-chip microprobe design optimal array processing mechanisms orthonormal set of eigenvectors rank ordered set signal reconstruction source detection source separation spike detection module unified framework user interface wavelet transform

We investigate a new technique for noise reduction in multichannel neural recordings based on the discrete wavelet transform. Starting with the denoising technique proposed by Donoho et al. (IEEE Trans. Inform. Theory 41 (1995) 613-627), we suggest a new thresholding method for the multiresolution decomposition of the multichannel data. The potential of this technique lies in the fact that thresholds at different resolution levels of the wavelet transform are estimated spatially to account for significant correlation of the wavelet coefficients across channels. The method is applied to a simulated multichannel data as well as real silicon microprobe recordings obtained in our laboratory. Preliminary results show the ability of the technique to reduce both spatially correlated and uncorrelated noise components in the neural recordings. Results are compared to existing techniques and the overall performance is evaluated.

Keywords: Multichannel recording; Silicon probes; Array processing; Wavelet denoising

This paper deals with the spike classification problem encountered in multi-unit recordings of neural activity in the brain. We recently developed a new methodology for estimating and classifying multi-units recorded by means of multichannel silicon probes from the observed spike trains (Proceedings of the ICASSP'01, May 2001, pp. 2813-2816; Proceedings of the IEEE 35th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, November 2001). In this work, we demonstrate the robustness of the technique to single unit spike amplitude variation often encountered in burst activity or long term chronic recordings. In low signal-to-noise ratio scenarios where variability in spike threshold crossings during classical detection is always a problem, we show that the technique is extremely robust to shifts in spike event times. Results showing the efficiency of the algorithm from simulated and experimental data are presented.

Keywords: Multichannel recording; Spike sorting; Array processing; Wavelet transform

The impedance of solid electrodes in the absence of faradaic reactions usually deviates from purely capacitive behaviour. The widely accepted explanations of this ``capacitance dispersion'' are based on the assumption that owing to surface roughness, or porosity, or spatially inhomogeneous capacitance density, the current density along the surface is not homogeneous, and thus capacitance dispersion is purely of geometric origin. We show that this view is not correct in the case of rough electrodes because capacitance dispersion due to irregular geometry appears at much higher frequencies than is usual in electrochemical methodologies. We present impedance spectra measured on platinum electrodes of various roughnesses in aqueous solutions to demonstrate that capacitance dispersion on rough electrodes is of interfacial origin and is due to adsorption effects. The old finding that the rougher the surface the larger the capacitance dispersion, can be rationalized alternatively in such a way that increasing roughness may broaden the time constant distribution of adsorption kinetics and may therefore increase the capacitance dispersion.

This paper presents a technique for implementing analog filters with wide dynamic range and low power dissipation and chip area. The desired dynamic range of the filter is divided into subranges, each covered by a different filtering path optimized specifically for this subrange. This results in small admittance levels for the individual filtering paths and correspondingly small power dissipation and chip area. The system provides undisturbed output during range switching, contrary to conventional automatic gain control (AGC)/filter arrangements that generate disturbances every time the gain of the AGC changes. We also report on a low-noise highly linear CMOS transconductor useful for high-frequency applications. A chip implementing the ideas of this paper was fabricated in a 0.25 μ m digital CMOS process. The intended application of the filter is channel selection in an 802.11a/Hiperlan2 Wireless Ethernet receiver. The chip dissipates 9 mA, occupies an area of 0.7 mm², and maintains a signal/(noise + IM3 distortion) ratio of at least 33 dB over a 48 dB signal range, with good blocker immunity. This performance represents at least an order of magnitude improvement over existing channel selection filters, even those that do not achieve disturbance-free operation.

Keywords: CMOS digital integrated circuits automatic gain control continuous time filters divide and conquer methods 0.25 micron 9 mA Hiperlan2 Wireless Ethernet receiver admittance levels analog filters automatic gain control channel selection filters chip area companding continuous-time filters digital CMOS filtering paths high-frequency applications low power dissipation low-noise highly linear CMOS transconductor range switching signal distortion ratio signal noise ratio wide dynamic range

This paper compares the main performance parameters of the current feedback opamp (CFOA) with those of a conventional voltage opamp (VOA). To make the comparison effective, a folded cascode VOA is considered (which is characterized by similar features and topology) and the same power consumption was assumed for both amplifiers. The work confirms that the CFOA can provide higher bandwidth, albeit at the expense of reduced loop gain. Noise performance is also analyzed. Input-referred noise generators are determined and some peculiar CFOA features, having no equivalence in conventional opamps, have been highlighted. It is shown that the CFOA has slightly lower noise voltage, but a larger noise current. Simulations are given which are in very good agreement with expected results.

Keywords: feedback amplifiers integrated circuit noise low-power electronics noise generators operational amplifiers bandwidth current-feedback amplifiers folded cascode circuit input-referred noise generators loop gain noise current noise voltage power consumption voltage operational amplifiers

A surface coating technique is investigated to enhance device biocompatibility by eliminating bio-fouling, the strong but nonspecific affinity of proteins and cells to attach to surfaces. This coating is a conformal, thin poly(ethylene glycol)-like film deposited in a glow discharge of tetraglyme. Substrates with different chemistries are successfully modified, and exhibit ultralow protein adsorption and cell attachment with the coating. This ``stealth'' or ``non-fouling'' coating can also be faithfully patterned using standard photolithography processes. The interaction of proteins and cells with patterned surfaces is limited only to the protein-adhesive domains, thus creating heterogeneous patterns of proteins and cell cultures on the surface. The potential benefits of our technique to applications such as cell-based assays and micro-electrodes are discussed.

Biosensors incorporate a biological sensing element that converts a change in an immediate environment to signals conducive for processing. Biosensors have been implemented for a number of applications ranging from environmental pollutant detection to defense monitoring. Biosensors have two intriguing characteristics: (1) they have a naturally evolved selectivity to biological or biologically active analytes; and (2) biosensors have the capacity to respond to analytes in a physiologically relevant manner. In this paper, molecular biosensors, based on antibodies, enzymes, ion channels, or nucleic acids, are briefly reviewed. Moreover, cell-based biosensors are reviewed and discussed. Cell-based biosensors have been implemented using microorganisms, particularly for environmental monitoring of pollutants. Biosensors incorporating mammalian cells have a distinct advantage of responding in a manner that can offer insight into the physiological effect of an analyte. Several approaches for transduction of cellular signals are discussed; these approaches include measures of cell metabolism, impedance, intracellular potentials, and extracellular potentials. Among these approaches, networks of excitable cells cultured on microelectrode arrays are uniquely poised to provide rapid, functional classification of an analyte and ultimately constitute a potentially effective cell-based biosensor technology. Three challenges that constitute barriers to increased cell-based biosensor applications are presented: analytical methods, reproducibility, and cell sources. Possible future solutions to these challenges are discussed.

Keywords: Antibody, Environmental monitoring, Functional assay, Chemical warfare, Extracellular potential, Impedance, Microelectrode, Patterning, Stem cells

A portable cell-based biosensor has been developed and characterized. The prototype system relies on extracellular recording from excitable cells cultured over an array of platinized gold microelectrodes. Extracellular potentials were bandpass filtered between 80 Hz to 2.8 kHz and amplified with a selectable gain of either 1000 or 5000. The input-referred noise level of the system was only 8.7 μ V_RMS in the laboratory setting, reaching only 10.6 μ V_RMS in an outdoor environment, more than sufficient for measurement of extracellular potentials from excitable cells. The system also incorporates a feedback control system for temperature regulation and a 36-channel multiplexer for selection of up to four output channels for simultaneous display. Wherever possible, low-cost `off-the-shelf' components were utilized in this prototype biosensor design. Using this system, extracellular recordings from chick myocardiocytes were performed under both laboratory and outdoor conditions.

Keywords: Amplifier; Biosensor; Cardiac myocytes; Extracellular recording; Portable; Biosensors; Toxicity; Portable equipment; Cell culture; Electrochemical electrodes; Feedback control; Antibodies; Portable cell based biosensors; Toxin detection; Cardiac myocytes; Extracellular recording

An extracellular recording system incorporating an electrode array and an amplifier/stimulator CMOS chip is described and characterized. Important features of this custom VLSI chip include 16 instrumentation amplifiers with a gain of 50 and the incorporation of a cross-point array allowing designation of an extracellular microelectrode as either a stimulator or sensor. The planar array consisted of 32 microelectrodes, 14 mm in diameter, and four larger reference electrodes. Microelectrodes, interconnecting traces, and bond pads were patterned with a 500-nm layer of gold. The interconnecting traces were passivated with a 1-mm thick layer of silicon nitride to provide chemical and electrical insulation and microelectrode impedance was lowered utilizing electrode position of platinum black. The amplifier exhibited a nearly flat frequency response with high pass and low pass corner frequencies of 0.7 Hz and 50 kHz, respectively. The input referred noise over the 50 kHz bandwidth was 12-16 μ V_rms, well below the magnitude of previously reported extracellular potentials. Crosstalk between neighboring channels resulted in an output signal below the amplifier noise level, even for relatively large extracellular potentials. Using this system, extracellular recordings were demonstrated yielding typical peak-to-peak biopotentials of magnitude 0.9-2.1 mV and 100-400 mV for chick cardiac myocytes and rat spinal cord neurons, respectively. The key components of this extracellular recording system can be manufactured using industry standard thin film photolithographic techniques.

Keywords: Amplifier; Cardiac myocytes; CMOS; Extracellular recording; Spinal cord neurons

A simple four-transistor, linear, tunable, high-frequency transconductance element is described. By using a pair of composite n-channel-p-channel devices, the circuit achieves its linearity by current differencing without undue matching requirements. It is shown that linearity and frequency response can be optimized simultaneously by appropriate choice of device dimensions. The performance is verified by SPICE simulations, and an operational transconductance amplifier (OTA) is used as one example for the many applications of the proposed element.

We have designed, fabricated, and characterized a microminiaturized ``neuroport'' for brain implantable neuroprosthesis applications, using an analog CMOS integrated circuit and a silicon based microelectrode array. An ultra-low power, low-noise CMOS preamplifier array with integral multiplexing was designed to accommodate stringent thermal and electrophysiological requirements for implantation in the brain, and a hybrid integration approach was developed to fabricate a functional microminiaturized neuroprobe device. Measurements showed that our fully scalable 16-channel CMOS amplifier chip had an average gain of 44 dB, bandwidth from 10 Hz to 7.3 kHz, and an equivalent input noise of approximately 9 μ V_rms with an average power consumption per preamplifier of 52 μW, which is consistent with simulation results. As a proof-of-concept demonstration, we have measured local field potentials from thalamocortical brain slices of rats, showing oscillatory behavior with an amplitude about 0.5 mV and a period ranging 80-120 ms. The results suggest that the hybrid integrated neuroport can form a prime platform for the development of a next level microminiaturized neural interface to the brain in a single implantable unit.

Keywords: Brain computer interface integrated neural probe array low-noise preamplifier neuroprosthesis

Dissociated cell cultures of neurons from neonatal rat superior cervical ganglia have been grown in specially prepared dishes, the bottoms of which consist of glass coverslips on which thin-film microcircuits have been deposited. The microcircuit provides 32 microelectrodes per dish, each approximately 8× 10 μ m in area. Extracellular recordings of action potentials from individual neurons have been made, with good signal-to-noise ratios, for cells within 40 μ m of the electrode centers. The microelectrodes are also suitable for passing the current required for extracellular stimulation and action potentials have been evoked by stimulating cell bodies and processes through the electrodes.

Keywords: MEA

The basic principles for recording and stimulation with extracellular electrodes are described and how they relate to the use of multielectrode arrays, MEAs, for studies of neural networks in culture. The main engineering issues for building arrays are described, and some examples are given of their use. The ``neurochip'' which has evolved from standard arrays is described, as well as other possible future developments.

Keywords: Neurons, networks, cultures, multielectrode

We have developed a new method for culturing cells that maintains their health and sterility for many months. Using conventional techniques, primary neuron cultures seldom survive more than 2 months. Increases in the osmotic strength of media due to evaporation are a large and underappreciated contributor to the gradual decline in the health of these cultures. Because of this and the ever-present likelihood of contamination by airborne pathogens, repeated or extended experiments on any given culture have until now been difficult, if not impossible. We surmounted survival problems by using culture dish lids that form a gas-tight seal, and incorporate a transparent hydrophobic membrane (fluorinated ethylene propylene) that is selectively permeable to oxygen (O₂) and carbon dioxide (CO₂), and relatively impermeable to water vapor. This prevents contamination and greatly reduces evaporation, allowing the use of a non-humidified incubator. We have employed this technique to grow dissociated cortical cultures from rat embryos on multi-electrode arrays. After more than a year in culture, the neurons still exhibit robust spontaneous electrical activity. The combination of sealed culture dishes with extracellular multi-electrode recording and stimulation enables study of development, adaptation, and very long-term plasticity, across months, in cultured neuronal networks. Membrane-sealed dishes will also be useful for the culture of many other cell types susceptible to evaporation and contamination.

Keywords: Cultured mammalian neurons; Multi-electrode arrays; Contamination; Hyperosmolality; Osmolarity; pH; Teflon membrane; Sealed culture chambers; Mold infection

We have developed a simple and expandable procedure for classification and validation of extracellular data based on a probabilistic model of data generation. This approach relies on an empirical characterization of the recording noise. We first use this noise characterization to optimize the clustering of recorded events into putative neurons. As a second step, we use the noise model again to assess the quality of each cluster by comparing the within-cluster variability to that of the noise. This second step can be performed independently of the clustering algorithm used, and it provides the user with quantitative as well as visual tests of the quality of the classification.

Keywords: Data clustering; Expectation-maximization; Antennal lobe; Locust; Sampling jitter; Multi-electrode; Tetrode

This paper reports a study of the adsorption of four proteins-fibrinogen, lysozyme, pyruvate kinase, and RNAse A-to self-assembled monolayers (SAMs) on gold. The SAMs examined were derived from thiols of the structure HS(CH₂)₁₀R, where R was CH_3, CH_2OH, and oligo(ethy1eneoxide). Monolayers that contained a sufficiently large mole fraction of alkanethiolate groups terminated in oligo(ethy1eneoxide) chains resisted the kinetically irreversible, nonspecific adsorption of all four proteins. Longer chains of oligo(ethy1ene oxide) were resistant at lower mole fractions in the monolayer. Resistance to the adsorption of proteins increased with the length of the oligo(ethy1ene oxide) chain: the smallest mole fraction of chains that prevented adsorption was proportional n^-0.4 to where n represents the number of ethylene oxide units per chain. Termination of the oligo(ethy1ene oxide) chains with a methoxy group instead of a hydroxyl group had little or no effect on the amount of protein adsorbed. The amount of pyruvate kinase that adsorbed to mixed SAMs containing hexa(ethy1ene oxide)-terminated chains depended upon the temperature. When the mole fraction of oligo(ethy1ene oxide) groups in the monolayer was below the level needed to prevent adsorption, more pyruvate kinase adsorbed to the monolayer at 37 ^oC than at 25 ^oC. No difference was observed between adsorption at 25 and 4 ^oC.

The ability to assemble neuronal networks with designed topology would allow uniquely defined experiments on neurocomputing. We describe a fundamental step, the controlled formation of synapses by guided outgrowth, in vitro for the first time combining simple neuritic geometry with predefined connectivity. We used neurons from the A-clusters in the pedal ganglia of the snail Lymnaea stagnalis. They were cultured on a substrate with linear patterns made by adsorption of brain-derived conditioning factors and photolithography. We induced and observed the frontal collision of two growth cones on narrow lanes. Following such encounters, individual electrical synapses formed that were sometimes strong enough for prolonged presynaptic stimulation to reach the threshold of postsynaptic firing.

This paper discusses certain important issues involved in the design of a nerve signal preamplifier for implantable neuroprostheses. Since the electroneurogram signal measured from cuff electrodes is typically on the order of 1 μ V, a very low-noise interface is essential. We present the argument for the use of BiCMOS technology in this application and then describe the design and evaluation of a complete preamplifier fabricated in a 0.8-um double-metal double-poly process. The preamplifier has a nominal voltage gain of 100, a bandwidth of 15 kHz, and a measured equivalent input-referred noise voltage spectral density of 3.3 (nV)/(sqrt(Hz)) at 1 kHz. The total input-referred rms noise voltage in a bandwidth 1 Hz-10 kHz is 290 nV, the power consumption is 1.3 mW from ±2.5-V power supplies, and the active area is 0.3 mm^2.

Keywords: BiCMOS analogue integrated circuits   bioelectric potentials   biomedical electronics   instrumentation amplifiers   integrated circuit noise   medical signal processing   neurophysiology   -2.5 V   1.3 mW   15 kHz   2.5 V   BiCMOS technology   ENG signals   LNA design   double-metal double-poly process electroneurogram signal   implantable ENG recording system   implantable neuroprostheses   low-noise interface   low-noise preamplifier   nerve cuff electrode recording   nerve signal preamplifier

This paper presents a solution for the output signal-to-noise ratio of a bandpass nonlinearity which is attractive for numerical evaluations of certain nonlinearities of interest. The solution is obtained by solving a differential equation which results in an expression for thenth order Chebyshev transform of an odd-order nonlinearity in terms of the (n - 1)^st-order Chebyshev transform of the derivative of the nonlinearity. The particular cases treated included a linear-logarithmic amplifier, an arctangent limiter, and a piecewise-linear limiter. The approach and results are related to previous efforts.

Keywords: Nonlinear distortions

The electrical properties of metal microelectrodes for single-unit recording are reviewed. An equivalent circuit is presented, the elements of which are discusses. The most important element is the electrolytic capacitor formed by the metal-electrolyte interface. Its value is about 0.2 pF/μ^2 at 1 kHz. the effects of exposed metallic area at the tip and platinziation are described, and some consideration is given to problems peculiar to the operation of these electrodes in neural tissue.

This paper describes the implementation of a low-power floating-gate MOS (FGMOS)-based log-domain integrator that reduces the minimum required voltage supply and the risk of instabilities. The performance of the block is illustrated with the experimental results of a second-order low-pass/bandpass filter working in the audio range with a 1-V voltage supply and a maximum power consumption of 2 μ W. The experimental results show that the FGMOS transistor is a powerful device that enables the design of low-voltage-supply low-power-consumption filters which have very simple topologies.

Keywords: Floating-gate MOS (FGMOS) log-domain filters low power low voltage

A fundamental, technical hurdle for systems neurophysiologists has been to extracellularly record from individual neurons while simultaneously applying micro-stimulation. Unfortunately, stimulation pulses corrupt neural recordings and obscure action potentials that occur during the time course of the artifact. In recent literature, there are several techniques proposed to remove stimulation artifacts; these, among others, include spectral cancellation, rapid polynomial fitting, and sample-hold and discharge schemes. Despite the growing number of proposed artifact removal schemes, discussions on the actual cause of stimulation artifacts are notably absent. Ultimately, the best prevention of stimulation artifacts will come from an understanding of their source. In this paper, a comprehensive model of the electrode-electrolyte interface is presented. Preliminary results from this model are used to suggest that the stimulation artifact is caused by the switching circuitry, itself, coupled with the non-linearities of the electrode-electrolyte interface.

Keywords: stimulation artifact

Efficient and selective electrical stimulation and recording of neural activity in peripheral, spinal, or central pathways requires multielectrode arrays at micrometer scale. ``Cultured probe'' devices are being developed, i.e., cell-cultured planar multielectrode arrays (MEAs). They may enhance efficiency and selectivity because neural cells have been grown over and around each electrode site as electrode-specific local networks. If, after implantation, collateral sprouts branch from a motor fiber (ventral horn area) and if they can be guided and contacted to each ``host'' network, a very selective and efficient interface will result. Four basic aspects of the design and development of a cultured probe, coated with rat cortical or dorsal root ganglion neurons, are described. First, the importance of optimization of the cell-electrode contact is presented. It turns out that impedance spectroscopy, and detailed modeling of the electrode-cell interface, is a very helpful technique, which shows whether a cell is covering an electrode and how strong the sealing is. Second, the dielectrophoretic trapping method directs cells efficiently to desired spots on the substrate, and cells remain viable after the treatment. The number of cells trapped is dependent on the electric field parameters and the occurrence of a secondary force, a fluid flow (as a result of field-induced heating). It was found that the viability of trapped cortical cells was not influenced by the electric field. Third, cells must adhere to the surface of the substrate and form networks, which are locally confined, to one electrode site. For that, chemical modification of the substrate and electrode areas with various coatings, such as polyethyleneimine (PEI) and fluorocarbon monolayers promotes or inhibits adhesion of cells. Finally, it is shown how PEI patterning, by a stamping technique, successfully guides outgrowth of collaterals from a neonatal rat lumbar spinal cord explant, after six days in culture

Keywords: biomedical electrodes biomedical electronics cellular biophysics electrophoresis neuromuscular stimulation prosthetics FES cellular engineering central pathways chemical modification collateral sprouts cultured multielectrode arrays cultured probe dielectrophoretic trapping method dorsal root ganglion neurons electrode-specific local networks impedance spectroscopy neural activity recording neuroelectronic interfacing peripheral pathways planar multielectrode arrays polyethyleneimine patterning rat cortical neurons secondary force selective electrical stimulation spinal pathways

A simple cascode with the gate voltage of the cascode transistor being controlled by a feedback amplifier called a regulated cascode is presented. In comparison to the standard cascode circuit, the minimum output voltage is lower by about 30 to 60% while the output conductance and the feedback capacitance are lower by about 100 times. An analytical large-signal, small-signal, and noise analysis is carried out. Some applications like current mirrors and voltage amplifiers are discussed. Experimental results confirming the theory are presented

Keywords: MOS integrated circuits amplifiers feedback linear integrated circuits network analysis noise transient response MOS cascode circuit analogue circuit applications current mirrors feedback amplifier gate voltage control high swing type high-impedance large signal analysis noise analysis regulated cascode small signal analysis voltage amplifiers

An updated version of a tutorial paper (see IEEE Circuits Devices Mag., vol. 2, no. 1, p. 20-32, 1985) on active filters using operational transconductance amplifiers (OTAs) is presented. The integrated circuit issues involved in active filters (using CMOS transconductance amplifiers) and the progress in this field in the last 15 years is addressed. CMOS transconductance amplifiers, nonlinearized and linearized, as well as frequency limitations and dynamic range considerations are reviewed. OTA-C filter architectures, current-mode filters, and other potential applications of transconductance amplifiers are discussed.

Keywords: CMOS analogue integrated circuits active filters current-mode circuits linearisation techniques operational amplifiers CMOS OTA CMOS transconductance amplifiers OTA-C filter architectures active filters amplifier architectures current-mode filters dynamic range frequency limitations integrated circuit linearized type nonlinearized type operational transconductance amplifiers

The theoretical and experimental results for white noise in the low-power subthreshold region of operation of an MOS transistor are discussed. It is shown that the measurements are consistent with the theoretical predictions. Measurements of noise in photoreceptors-circuits containing a photodiode and an MOS transistor-that are consistent with theory are reported. The photoreceptor noise measurements illustrate the intimate connection of the equipartition theorem of statistical mechanics with noise calculations.

Keywords: insulated gate field effect transistors metal-insulator-semiconductor devices random noise resistors semiconductor device noise thermal noise white noise MOS transistors low-power subthreshold region photodiode photoreceptor noise measurements photoreceptors shot noise thermal noise white noise

Extensive measurements of drain current thermal noise are presented for 3 different CMOS technologies and for gate lengths ranging from 2 μ m down to 0.17 μ m. Using a surface-potential-based compact MOS model with improved descriptions of carrier mobility and velocity saturation, all the experimental results can be described accurately without invoking carrier heating effects or introducing additional parameters

Keywords: MOSFET carrier mobility semiconductor device models semiconductor device noise surface potential thermal noise 0.17 to 2 micron MOSFET carrier mobility deep-submicron CMOS technology drain current surface potential thermal noise model velocity saturation

Cultures of neurons can be grown on microelectrode arrays (MEAs), so that their spike and burst activity can be monitored. These activity patterns are quite sensitive to changes in the environment, such as chemical exposure, and hence the cultures can be used as biosensors. One key issue in analyzing the data from neuronal networks is how to quantify the level of synchronization among different units, which represent different neurons in the network. In this paper, we propose a synchronization metric, based on the statistical distribution of unit-to-unit correlation coefficients. We show that this synchronization metric changes significantly when the networks are exposed to bicuculline, strychnine, or 2,3-dioxo-6-nitro-l,2,3,4-tetrahydrobenzoquinoxaline-7-sulphonamide (NBQX). For that reason, this metric can be used to characterize pharmacologically induced changes in a network, either for research or for biosensor applications.

Keywords: Spinal-cord cultures; Microelectrode arrays; Correlation; Burst; Seizure

The results presented here demonstrate selective learning in a network of real cortical neurons. We focally stimulate the network at a low frequency (0.3-1 Hz) until a desired predefined response is observed 50 ± 10 msec after a stimulus, at which point the stimulus is stopped for 5 min. Repeated cycles of this procedure ultimately lead to the desired response being directly elicited by the stimulus. By plotting the number of stimuli required to achieve the target response in each cycle, we are able to generate learning curves. Presumably, the repetitive stimulation is driving changes in the circuit, and we are selecting for changes consistent with the predefined desired response. To the best of our knowledge, this is the first time learning of arbitrarily chosen tasks, in networks composed of real cortical neurons, is demonstrated outside of the body.

Keywords: learning; multielectrode array; cultured neurons; neural network; reward; drive reduction

An analytical model for switch-induced error voltage on a switched capacitor is derived. A compact expression contains the effects of gate voltage falling rate, threshold voltage, and storage capacitance. It can be used to quickly predict the error voltage. The model is in good agreement with computer simulations using SPICE program and experiment.

Keywords: MOS integrated circuits, analog Switched-capacitor circuits

Charge injection in MOS switches has been analyzed. The analysis has been extended to the general case of including source resistance and source capacitance. Universal plots of percentage channel charge injected are presented. Normalized variables are used to facilitate usage of the plots. A small-geometry switch, slow switching rate, and small source resistance can help reduce the charge injection effect.

Keywords: Charge injection MOSFET switches Sample/hold circuits Switched-capacitor circuits

The analysis has been extended to the general case including signal-source resistance and capacitance. Universal plots of percentage channel charge injected are presented. Normalized variables are used to facilitate usage of the plots. The effects of gate voltage falling rate, signal-source level, substrate doping, substrate bias, switch dimensions, as well as the source and holding capacitances are included in the plots. A small-geometry switch, slow switching rate, and small source resistance can reduce the charge injection effect. On-chip test circuitry with a unity-gain operational amplifier, which reduces the disturbance imposed by measurement equipment to a minimum, is found to be an excellent monitor of the switch charge injection. The theoretical results agree with the experimental data.

Keywords: Field effect integrated circuits Sample and hold circuits Semiconductor switches Switched capacitor networks field effect integrated circuits sample and hold circuits semiconductor switches switched capacitor networks

A number of recent methods developed for automatic classification of multiunit neural activity rely on a Gaussian model of the variability of individual waveforms and the statistical methods of Gaussian mixture decomposition. Recent evidence has shown that the Gaussian model does not accurately capture the multivariate statistics of the waveform samples' distribution. We present further data demonstrating non-Gaussian statistics, and show that the multivariate t-distribution, a wide-tailed family of distributions, provides a significantly better fit to the true statistics. We introduce an adaptation of a new expectation-maximization based competitive mixture decomposition algorithm and show that it efficiently and reliably performs mixture decomposition of t-distributions. Our algorithm determines the number of units in multiunit neural recordings, even in the presence of significant noise contamination resulting from random threshold crossings and overlapping spikes.

Keywords: Spike sorting; Multi-unit recording; Electrode array; Unsupervised classification; Mixture models; Expectation-maximization; Multivariate t-distribution

Fully differential amplifiers yield large differential gains and also high common mode rejection ratio (CMRR), provided they do not include any unmatched grounded component. In biopotential measurements, however, the admissible gain of amplification stages located before dc suppression is usually limited by electrode offset voltage, which can saturate amplifier outputs. The standard solution is to first convert the differential input voltage to a single-ended voltage and then implement any other required functions, such as dc suppression and dc level restoring. This approach, however, yields a limited CMRR and may result in a relatively large equivalent input noise. This paper describes a novel fully differential biopotential amplifier based on a fully differential dc-suppression circuit that does not rely on any matched passive components, yet provides large CMRR and fast recovery from dc level transients. The proposed solution is particularly convenient for low supply voltage systems. An example implementation, based on standard low-power op amps and a single 5-V power supply, accepts input offset voltages up to ±500 mV, yields a CMRR of 102 dB at 50 Hz, and provides, in accordance with the AAMI EC38 standard, a reset behavior for recovering from overloads or artifacts.

Keywords: AC coupling biopotential amplifiers electrode offset potential

AC coupling is essential in biopotential measurements. Electrode offset potentials can be several orders of magnitude larger than the amplitudes of the biological signals of interest, thus limiting the admissible gain of a dc-coupled front end to prevent amplifier saturation. A high-gain input stage needs ac input coupling. This can be achieved by series capacitors, but in order to provide a bias path, grounded resistors are usually included, which degrade the common mode rejection ratio (CMRR). This paper proposes a novel balanced input ac-coupling network that provides a bias path without any connection to ground, thus resulting in a high CMRR. The circuit being passive, it does not limit the differential dc input voltage. Furthermore, differential signals are ac coupled, whereas common-mode voltages are dc coupled, thus allowing the closed-loop control of the dc common mode voltage by means of a driven-right-leg circuit. This makes the circuit compatible with common-mode dc shifting strategies intended for single-supply biopotential amplifiers. The proposed circuit allows the implementation of high-gain biopotential amplifiers with a reduced number of parts, thus resulting in low power consumption. An electrocardiogram amplifier built according to the proposed design achieves a CMRR of 123 dB at 50 Hz.

Keywords: bioelectric potentials   biomedical electrodes   biomedical electronics   closed loop systems   electrocardiography   instrumentation amplifiers   50 Hz   AC-coupled front-end   balanced input ac-coupling network   bias path   biological signals   biopotential measurements   closed-loop control   common mode rejection ratio   common-mode dc shifting strategies   dc common mode voltage   dc coupled common-mode voltages   design   differential dc input voltage   differential signals   driven-right-leg circuit   electrocardiogram amplifier   electrode offset potentials   grounded resistors   high-gain biopotential amplifiers   high-gain input stage   low power consumption   series capacitors   single-supply biopotential amplifiers

When recording from multi-electrode arrays, only a short period around the time of a threshold crossing is generally saved for later analysis. Then, waveforms are often sorted automatically to identify templates of spikes from individual neurons near an electrode. As spikes sum from different neurons and noise is present, some spikes may be missed and others erroneously accepted. This paper describes methods for identifying and correcting errors in recorded spike trains to recover the pattern of spikes from each neuron as faithfully as possible. These methods are complementary to, but distinct from methods to reconstruct waveforms that arise from summation of individual templates that overlap one another. Our methods are based on the local statistics of the firing rates or inter-spike intervals and the methods work best for neurons that fire regularly (small standard deviation relative to the mean interval). First, we test whether accepting more spikes, whose waveforms are close to the templates that have been identified, will increase the regularity or smoothness of the firing rates. Then, after accepting spikes that increase regularity, we test whether individual intervals are sufficiently longer (or shorter) than their neighbors to identify spikes that have been omitted (or accepted) erroneously. The methods are tested on simulated spike trains, where spikes have been inserted or deleted at random, and on spike trains recorded from multi-electrode arrays in dorsal root ganglia of cats walking on a treadmill.

Keywords: Neurons; Spikes; Electrode arrays; Intervals

A CMOS low-power low-noise monolithic instrumentation amplifier (IA) is described. The power drain is reduced by use of current feedback and by use of only single-stage operational transconductance amplifiers in the low-frequency loop. The bandwidth of the IA is designed for medical purposes (0.5-500 Hz) and it can produce variable gains of 14/20/26/40 dB, which are set by control software.

Keywords: Biomedical electronics CMOS integrated circuits Feedback Instrumentation amplifiers Linear integrated circuits biomedical electronics feedback instrumentation amplifiers linear integrated circuits

Locomotion in vertebrates is controlled by central pattern generators in the spinal cord. The roles of specific network architecture and neuronal properties in rhythm generation by such spinal networks are not fully understood. We have used multisite recording from dissociated cultures of embryonic rat spinal cord grown on multielectrode arrays to investigate the patterns of spontaneous activity in randomised spinal networks. We were able to induce similar patterns of rhythmic activity in dissociated cultures as in slice cultures, although not with the same reliability and not always with the same protocols. The most reliable rhythmic activity was induced when a partial disinhibition of the network was combined with an increase in neuronal excitability, suggesting that both recurrent synaptic excitation and neuronal excitability contribute to rhythmogenesis. During rhythmic activity, bursts started at several sites and propagated in variable ways. However, the predominant propagation patterns were independent of the protocol used to induce rhythmic activity. When synaptic transmission was blocked by CNQX, APV, strychnine and bicuculline, asynchronous low-rate activity persisted at approx. 50% of the electrodes and approx. 70% of the sites of burst initiation. Following the bursts, the activity in the interval was transiently suppressed below the level of intrinsic activity. The degree of suppression was proportional to the amount of activity in the preceding burst. From these findings we conclude that rhythmic activity in spinal cultures is controlled by the interplay of intrinsic neuronal activity and recurrent excitation in neuronal networks without the need for a specific architecture.

A microelectrode array (MEA) consisting of 34 silicon nitride passivated Pt-tip microelectrodes embedded on a perforated silicon substrate (porosity 35%) has been realized. The electrodes are 47 μ m high, of which only the top 15 um are exposed Pt-tips having a curvature of 0.5 mu m. The MEA is intended for extracellular recordings of brain slices in vitro. Here we report the fabrication, characterization and initial electrophysiological evaluation of the first generation of Pt-tip MEAs.

Keywords: extracellular recording; impedance; microelectrode arrays; Pt-tip microelectrode; microelectrode; brain function; silicon nitride; silicon

Electrical activity can be recorded extracellularly from contracting heart cells in vitro with the electrodes of 30-element microelectrode arrays built into the culture chambers. The arrays are fabricated in the laboratory by etching thin metal films deposited on glass coverslips; the fabrication employs techniques developed by the microelectronics industry.

Rat spinal networks generate a spontaneous rhythmic output directed to motoneurons under conditions of increased excitation or of disinhibition. It is not known whether these differently induced rhythms are produced by a common rhythm generator. To investigate the generation and the propagation of rhythmic activity in spinal networks, recordings need to be made from many neurons simultaneously. Therefore extracellular multisite recording was performed in slice cultures of embryonic rat spinal cords grown on multielectrode arrays. In these organotypic cultures most of the spontaneous neural activity was nearly synchronized. Waves of activity spread from a source to most of the network within 35-85 ms and died out after a further 30-400 ms. Such activity waves induced the contraction of cocultured muscle fibres. Several activity waves could be grouped into aperiodic bursts. Disinhibition with bicuculline and strychnine or increased excitability with high K^+ or low Mg^2+ solutions could induce periodic bursting with bursts consisting of one or several activity waves. Whilst the duration and period of activity waves were similar for all protocols, the duration and period of bursts were longer during disinhibition than during increased excitation. The sources of bursting activity were mainly situated ventrally on both sides of the central fissure. The pathways of network recruitment from one source were variable between bursts, but they showed on average no systematic differences between the protocols. These spatiotemporal similarities under conditions of increased excitation and of disinhibition suggest a common spinal network for both types of rhythmic activity.

The authors consider the use in signal processing of noise reduction techniques such as syllabic companding (compressing and expanding), developed for transmission and storage. Applications discussed include increasing selectivity in single sinusoid detection. Whether or not companding will be useful in a given filtering application depends on the type and response of the filter and the type of signals present at the input.

Keywords: compandors filtering and prediction theory interference suppression signal processing speech analysis and processing companding filtering noise reduction techniques selectivity signal processing single sinusoid detection speech signals syllabic companding

The state of the art of continuous-time filter design is reviewed. Several techniques are discussed and compared in terms of performance and implementation feasibility in different fabrication technologies. This review does not aim at historical completeness, but rather emphasizes techniques that have proven their worth in commercial applications. Brief mention is also made of experimental work which, in the opinion of the author, shows promise for the future.

Keywords: active filters linear integrated circuits monolithic integrated circuits operational amplifiers reviews tuning continuous-time filter design fabrication technologies implementation feasibility integrated filters performance review

The development and in vivo test of a fully integrated differential CMOS amplifier, implemented with standard 0.7-mu mCMOS technology (one poly, two metals, self aligned twin-well CMOS process) intended to record extracellular neural signals is described. In order to minimize the flicker noise generated by the CMOS circuitry, a chopper technique has been chosen. The fabricated amplifier has a gain of 74 dB, a bandwidth of 3 kHz, an input noise of 6.6 nV/( Hz )^0.5, a power dissipation of 1.3 mW, and the active area is 2.7 mm^2. An ac coupling has been used to adapt the electrode to the amplifier circuitry for the in vivo testing. Compound muscle action potentials, motor unit action potentials, and compound nerve action potentials have been recorded in acute experiments with rats, in order to validate the amplifier.

Keywords: CMOS Chopper technique neural recording

Discusses the use of sentinel and surrogate animal species data for evaluating the potential effects of chemicals and pollutants to humans. Data as an additional weight of evidence in risk assessment; Factors impending the application of sentinel species approaches.

The wavelet transform is compared with the more classical short-time Fourier transform approach to signal analysis. Then the relations between wavelets, filter banks, and multiresolution signal processing are explored. A brief review is given of perfect reconstruction filter banks, which can be used both for computing the discrete wavelet transform, and for deriving continuous wavelet bases, provided that the filters meet a constraint known as regularity. Given a low-pass filter, necessary and sufficient conditions for the existence of a complementary high-pass filter that will permit perfect reconstruction are derived. The perfect reconstruction condition is posed as a Bezout identity, and it is shown how it is possible to find all higher-degree complementary filters based on an analogy with the theory of Diophantine equations. An alternative approach based on the theory of continued fractions is also given. These results are used to design highly regular filter banks, which generate biorthogonal continuous wavelet bases with symmetries.

Keywords: band-pass filters digital filters filtering and prediction theory signal processing transforms Bezout identity FIR filters bandpass filters complementary high-pass filter continuous wavelet bases design discrete wavelet transform filter design higher-degree complementary filters low-pass filter multiresolution signal processing perfect reconstruction filter banks regularity signal analysis theory of continued fractions

We describe an algorithm for suppression of stimulation artifacts in extracellular micro-electrode array (MEA) recordings. A model of the artifact based on locally fitted cubic polynomials is subtracted from the recording, yielding a flat baseline amenable to spike detection by voltage thresholding. The algorithm, SALPA, reduces the period after stimulation during which action potentials cannot be detected by an order of magnitude, to less than 2 ms. Our implementation is fast enough to process 60-channel data sampled at 25 kHz in real-time on an inexpensive desktop PC. It performs well on a wide range of artifact shapes without re-tuning any parameters, because it accounts for amplifier saturation explicitly and uses a statistic to verify successful artifact suppression immediately after the amplifiers become operational. We demonstrate the algorithm's effectiveness on recordings from dense monolayer cultures of cortical neurons obtained from rat embryos. SALPA opens up a previously inaccessible window for studying transient neural oscillations and precisely timed dynamics in short-latency responses to electric stimulation.

Keywords: Artifact suppression; Micro-electrode array; Stimulation; Real-time; Local regression; Multi-electrode array; MEA; Spikes

Electrical stimulation through multi-electrode arrays is used to evoke activity in dissociated cultures of cortical neurons. We study the efficacies of a variety of pulse shapes under voltage-as well as current-control, and determine useful parameter ranges that optimize efficacy while preventing damage through electrochemistry. For any pulse shape, stimulation is found to be mediated by negative currents. We find that positive-then-negative biphasic voltage-controlled pulses are more effective than any of the other pulse shapes tested, when compared at the same peak voltage. These results suggest that voltage-control, with its inherent control over limiting electrochemistry, may be advantageous in a wide variety of stimulation scenarios, possibly extending to in-vivo experiments.

Keywords: Electrical stimulation; dissociated culture; multi-electrode array; MEA; stimulation parameters

Over the last few decades, technology to record through ever increasing numbers of electrodes has become available to electrophysiologists. For the study of distributed neural processing, however, the ability to stimulate through equal numbers of electrodes, and thus to attain bidirectional communication, is of paramount importance. Here, we present a stimulation system for multi-electrode arrays that interfaces with existing commercial recording hardware, and allows stimulation through any electrode in the array, with rapid switching between channels, without impairing recording performance. The system is controlled through real-time Linux, making it extremely flexible: Stimulation sequences can be constructed on-the-fly, and arbitrary stimulus waveforms can be used if desired. A key feature of this design is that it can readily and inexpensively be reproduced in other labs, since it interfaces to standard PC parallel ports and uses only off-the-shelf components. Moreover, adaptation for use with in-vivo multi-electrode probes would be straightforward. In combination with our freely available data acquisition software, MeaBench, this system can provide feedback stimulation in response to recorded action potentials within 15 ms.