System and method for noise cancellation in a recording system

Abstract

A system and method for obtaining electrical recording data is disclosed, comprising: providing an object of interest at least partially located in a recording environment; providing at least one signal logging electrode or transducer adapted to conduct or convert signal data of interest from the object of interest; obtaining signal data of interest; obtaining noise data from a location within the recording environment; modulating the signal to remove noise; and obtaining noise cancelled data, the obtaining of the noise cancelled data comprising subtracting the noise data from the signal data of interest.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to Australian Provisional Patent Application No. 2023902596, filed on August 15, 2023, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] This disclosure relates to methods for noise cancellation and signal calibration, such as in multichannel and / or electro-optic acquisition of bioelectricity. Background Technology

[0004] Interference from ambient electromagnetic noise, internal Johnson-Nyquist noise, and motion artifacts pose significant challenges to recording processing, particularly from excitable biological tissues, including those involved in neuroscience and cardiac electrophysiology recording systems. These systems are used in isolated tissues and organs in laboratories (i.e., ex vivo or in vitro), but also in in vivo formulations implanted in animals and humans.

[0005] This interference leads to a decrease in the signal-to-noise ratio (SNR), which can be significant if the absolute value of the signal of interest is very small or very small relative to the background noise. For example, extracellular bioelectrical signals are typically less than 1 millivolt (mV) and usually less than 100 microvolts (μV).

[0006] In traditional acquisition systems based on multi-electrode arrays (MEAs), it is the user's responsibility to attempt to reduce interference from external electromagnetic noise through equipment grounding and electrical isolation of the recording device. Some examples include the use of shielded cables, Faraday cages, or Faraday chambers. Another approach is to use biological instruments placed near the recording source.

[0007] Traditionally, different aspects of noise processing lie in signal post-processing. For example, some systems employ noise modeling and reconstruction, and then remove the reconstructed noise signal from the recorded signal as their noise reduction strategy.

[0008] The inventors were among the first to develop a photoelectrode (optical pole) for use in electrophysiological recording. Noise interference problems also arise with photoelectrodes (optical poles). Because in devices based on these electro-optic transducers, the output signal (i.e., the change in optical reflectivity) is proportional to the input voltage, but the scaling factor is unknown, noise reduction becomes complex.

[0009] Any discussion of documents, actions, materials, devices, articles, etc., already included in this specification is for the purpose of providing context for the invention only. This should not be construed as an admission that any or all of these matters constitute part of the prior art or common general knowledge in the field related to this invention simply because they existed prior to the priority date of each claim of this application. Summary of the Invention

[0010] In a first aspect, this disclosure provides a recording device. The device includes a reference electrode and a plurality of conductive devices. The plurality of conductive devices includes: at least one signal transduction device adapted to transduce a signal from an object of interest (e.g., biological tissue, cell, cell culture, or another biological or non-biological sample of interest) at least partially located in the recording environment; and at least one noise transduction device adapted to transduce a noise signal.

[0011] At least one noise conduction device may include one or more first noise conduction devices, which are adapted to be exposed to a recording environment during use. Furthermore, the first noise conduction devices and the signal conduction devices may have substantially the same material and dimensions. In some cases, this means that the half-cell potential or electrode potential of the first noise conduction device is substantially the same as the half-cell (or electrode) potential of the signal conduction device.

[0012] At least one noise conduction device may include one or more second noise conduction devices adapted to be isolated from the recording environment during use. In another embodiment, at least some of the plurality of conductive devices may be disposed in the substrate. One or more second noise conduction devices may also be embedded in the substrate. A reference electrode may be disposed in the substrate.

[0013] Multiple conductive devices can be a multi-electrode array.

[0014] Alternatively, at least some of the conductive devices are transducers. In some embodiments, one or more first noise conducting devices include at least one noise electrode and at least one noise sensing transducer.

[0015] The track used for conduction with at least one noise electrode and the track used for conduction with a reference electrode can have substantially the same material.

[0016] Each transducer may be adapted to convert optical signals into electrical signals, sense returned electrical signals, and transmit output optical signals converted from returned electrical signals. The substrate may include liquid crystal.

[0017] The device may also have a timer or controller adapted to time the switching between the input of the signal conduction device and the input of the noise conduction device.

[0018] The device may also have a signal holding circuit system for periodically holding the output of the signal conduction device or noise conduction device.

[0019] In a second aspect, this disclosure provides a measurement multichannel recording apparatus. The apparatus includes a reference electrode and a plurality of photoelectric transducers embedded in a light-transmitting substrate. The plurality of photoelectric transducers includes at least one signal recording transducer adapted to convert an optical input into an electrical input to an object of interest. The plurality of photoelectric transducers also includes at least one noise recording transducer adapted to receive a response signal from the object of interest and convert the response signal into an output optical signal. The apparatus further includes a detector adapted to detect the output optical signal.

[0020] In a third aspect, this disclosure provides an electrical measuring device including a reference electrode and at least one signal recording device adapted to record signals from an object of interest. The device is capable of providing input to the object of interest periodically, intermittently, or at times determined by a user.

[0021] At least one signal measuring device may be at least one photoelectric transducer embedded in a light-transmitting substrate, said at least one photoelectric transducer being adapted to convert optical input into electrical input to an object of interest.

[0022] At least one photoelectric transducer can be adapted to convert an electrical signal from an object in response to an electrical input into an output optical signal.

[0023] Optical input can be provided at user-defined times in the following ways: turning the light source on and off; interrupting light transmission from the light source; moving at least one photoelectric transducer away from the light receiving position.

[0024] In a fourth aspect, this disclosure provides a recording system that includes the electrophysiological recording devices mentioned in any of the foregoing aspects.

[0025] In a fifth aspect, this disclosure provides a signal recording method comprising: providing an input signal to an object of interest at least partially located in a recording environment; obtaining a response signal from the object of interest in response to the input signal, the response signal being transmitted or converted by at least one signal recording electrode or transducer; obtaining noise data from a location within a measurement chamber; and obtaining noise cancellation data, the obtaining noise cancellation data comprising subtracting a noise signal from the response signal.

[0026] The method may include providing at least one noise recording electrode or transducer.

[0027] Obtaining a noise signal may include obtaining a first noise signal from at least one first noise electrode or transducer, said first noise electrode or transducer being positioned isolated from the object of interest but exposed to the recording environment.

[0028] Obtaining a noise signal may include obtaining a second noise signal from at least one second noise electrode or transducer, said at least one second noise electrode or transducer being fully embedded within a substrate located in the recording environment such that it is not exposed to the recording environment.

[0029] At least one first noise electrode or transducer may be at least one first noise electrode, and the noise signal includes a third noise signal from at least one noise transducer, which is positioned to be isolated from the object of interest but exposed to the recording environment.

[0030] This method may include obtaining a reference signal from a reference electrode.

[0031] The method may include obtaining a voltage offset voltage value, wherein obtaining the voltage offset voltage value includes calculating the difference between a reference signal and a first noise signal.

[0032] The method may include applying a compensation voltage to a substrate located within a recording environment, the compensation voltage being a voltage offset voltage with opposite polarity.

[0033] Obtaining noise cancellation data can include subtracting the voltage offset value from the object signal data.

[0034] This method may include providing input signals with multiple input values.

[0035] The method may include: obtaining multiple voltage offset values, each for a corresponding input value, and calculating a baseline based on the multiple voltage offset values.

[0036] This method may include calibrating the response signal to remove the baseline and determine the gain.

[0037] This method may include adjusting the response signal and / or the noise signal.

[0038] Adjustment may include smoothing or filtering the response signal and / or noise signal.

[0039] Input signals can be provided intermittently.

[0040] Obtaining a noise signal may include obtaining a signal from at least one signal electrode or transducer when the input signal is not provided to the object of interest.

[0041] The input signal can be an optical input. The optical input can be provided at user-defined times in the following ways: turning the light source on and off; periodically interrupting the light transmission from the light source; moving at least one signal electrode or transducer to the light receiving position and moving at least one signal electrode or transducer away from the light receiving position.

[0042] In a sixth aspect, this disclosure provides a computer program including instructions for controlling a computer or embedded processor to implement the methods mentioned in the fifth aspect above.

[0043] In a seventh aspect, this disclosure provides an apparatus for obtaining functional measurements of an object of interest, including a computer having a processor and memory supporting computer processing, the computer implementing the method mentioned in the fifth aspect above.

[0044] In an eighth aspect, this disclosure provides an electro-optic detection apparatus. The apparatus includes: a first optical electrode configured to receive a first optical signal and generate a first optical electrode signal, the first optical electrode being configured to contact a sample of interest; and a second optical electrode configured to receive a second optical signal and generate a second optical electrode signal. The second optical electrode is configured to be isolated from the sample of interest. The apparatus also includes a receiver configured to determine an output signal by applying an active noise cancellation algorithm to the first and second optical electrode signals.

[0045] In some embodiments, the device further includes: a light source configured to generate an optical signal; and a light source splitter configured to split the optical signal into a first optical signal and a second optical signal.

[0046] In some embodiments, generating the first optical signal includes: receiving a first optical signal by a first optical electrode and reflecting the first optical signal into a photodiode, generating an analog electrical signal by the photodiode, and converting the analog electrical signal by an analog-to-digital converter to generate the first optical signal.

[0047] In some embodiments, applying the active noise cancellation algorithm includes subtracting the second optical pole signal from the first optical pole signal. In some embodiments, applying the active noise cancellation algorithm includes applying a Wiener filter to the first and second optical pole signals.

[0048] In some embodiments, applying the active noise cancellation algorithm includes applying a single-tap adaptive filter to the first optical pole signal. In some embodiments, applying the active noise cancellation algorithm includes determining the cross-correlation between the first and second optical pole signals. In some embodiments, applying the active noise cancellation algorithm includes determining the autocorrelation of the second optical pole signal. In some embodiments, applying the active noise cancellation algorithm includes determining filter coefficients based on the cross-correlation and autocorrelation. In some embodiments, applying the active noise cancellation algorithm includes applying a high-pass filter to the first optical pole signal.

[0049] In some embodiments, the first optical pole includes one or more of a polarization-maintaining fiber, a liquid crystal layer, and a reflector.

[0050] In a ninth aspect, this disclosure provides a method for detecting neural signals. The method includes generating an optical signal from a light source and dividing the optical signal into a first optical signal and a second optical signal. The method further includes generating a first optical pole signal based on the first optical signal from a first optical pole configured to contact a sample of interest, and generating a second optical pole signal based on the second optical signal from a second optical pole configured to be isolated from the sample of interest. The method also includes determining an output signal by applying an active noise cancellation algorithm to the first and second optical pole signals.

[0051] In some embodiments, generating the first optical signal includes: receiving a first optical signal by a first optical electrode and reflecting the first optical signal into a photodiode, generating an analog electrical signal by the photodiode, and converting the analog electrical signal by an analog-to-digital converter to generate the first optical signal.

[0052] In some implementations, applying an active noise cancellation algorithm includes subtracting the second optical pole signal from the first optical pole signal.

[0053] In some implementations, applying the active noise cancellation algorithm includes applying a single-tap active filter to the first and second optical pole signals.

[0054] In some implementations, applying the active noise cancellation algorithm includes applying a Wiener filter to the first and second optical pole signals. Attached Figure Description

[0055] The implementation will now be described by way of example only with reference to the accompanying drawings, in which:

[0056] Figure 1 This is a cross-sectional view of a multi-electrode array with added exposed and concealed photoelectric poles and added exposed electrodes for noise acquisition and signal calibration.

[0057] Figure 2 This is a top view of a multi-electrode array, which has added exposed and concealed photoelectric poles as well as added exposed electrodes for noise acquisition and signal calibration.

[0058] Figure 3 This is a cross-sectional view of a conventional MEA, which has added exposed and concealed electrodes for noise acquisition and baseline calibration;

[0059] Figure 4 This is a top view of a traditional MEA, which has added exposed and concealed electrodes for noise acquisition and baseline calibration;

[0060] Figure 5It is a conceptual diagram illustrating the arrangement used for oversampling and sample-and-hold to isolate signal and noise data sequences;

[0061] Figure 6-1 is a schematic diagram illustrating a time-based implementation, wherein the same electrodes or photoelectric poles are used for signal sensing and for noise sensing;

[0062] Figure 6-2 is a representation of the oversampled data sequence obtained from the time-based implementation method;

[0063] Figure 6-3 shows the signal and noise sequences separated from the oversampled data sequence shown in Figure 6-2;

[0064] Figure 7-1 is a schematic diagram illustrating another time-based implementation, wherein input is alternatively provided to a signal sensing electrode or optical electrode and a noise sensing electrode or optical electrode;

[0065] Figure 7-2 is an example of an oversampled data sequence obtained from the time-based implementation shown in Figure 7-1;

[0066] Figure 7-3 shows the signal and noise sequences separated from the oversampled data sequence shown in Figure 7-2;

[0067] Figure 8-1 is a schematic diagram illustrating another time-based implementation, wherein the substrate is positioned such that incident light is directed toward the signal sensing photoelectrode;

[0068] Figure 8-2 depicts the embodiment shown in Figure 8-1, wherein the substrate is positioned such that incident light is directed toward the noise sensing photoelectrode;

[0069] Figure 8-3 is an example of an oversampled data sequence obtained from the time-based implementation shown in Figure 8-1;

[0070] Figure 9-1 is a schematic diagram illustrating another time-based implementation, wherein the substrate is positioned such that incident light is directed toward the signal sensing photoelectrode.

[0071] Figure 9-2 depicts the embodiment shown in Figure 9-1, wherein the substrate is positioned such that incident light is directed toward the noise sensing photoelectrode;

[0072] Figure 9-3 is an example of an oversampled data sequence obtained from the time-based implementation shown in Figure 9-1;

[0073] Figure 10 An example of a calibration process for generating a noise-cancelled signal is schematically depicted, in which an array of optical poles is used to record the signal;

[0074] Figure 11An example of a calibration process for generating a noise-cancelled signal is schematically depicted, in which a multi-electrode array is used to record the signal;

[0075] Figure 12 An example of a general process for obtaining the final signal of interest is conceptually described;

[0076] Figure 13 The structure of the optical pole according to an embodiment is shown.

[0077] Figure 14 A design for an electro-optic detection system according to an embodiment is shown.

[0078] Figure 15 This is a block diagram of an active noise cancellation (ANC) system 1500 according to an embodiment.

[0079] Figure 16 The data flow in the system according to the embodiment is shown.

[0080] Figure 17 Equations (a) to (e) according to the implementation are shown.

[0081] Figure 18 This is a schematic diagram of a light receiver board according to an embodiment.

[0082] Figure 19 A digital signal processing structure according to an embodiment is shown.

[0083] Figure 20 This is a graph showing the cross-correlation between the active and passive light source channels at different light source powers when there is no input signal at the active light source according to the embodiment.

[0084] Figure 21 Two pairs of graphs according to an embodiment are shown, illustrating the signals before (active optical channel) and after (noise cancellation output channel) the noise cancellation process.

[0085] Figure 22 This is a graph showing a comparison between the RMS values ​​of the channel 1 input and the DSP output under different light source currents when there is no input signal at the photoelectrode, according to an embodiment.

[0086] Figure 23 The diagram illustrates the signal-to-noise ratio (SNR) and SNR improvement for a 200 mV sinusoidal input under different source currents, according to an embodiment, before and after the active noise cancellation process.

[0087] Figure 24 The diagram illustrates the signal-to-noise ratio (SNR) and SNR improvement for a 10 mV sinusoidal input under different light source currents, before and after the active noise cancellation process, according to an embodiment.

[0088] Figure 25 A comparison of the active photoelectric channel (a) and noise-cancelled output (b) with a 10 mV 1 kHz sinusoidal input signal at the active photoelectric electrode and a photocurrent of 500 mA is shown according to an embodiment.

[0089] Figure 26 Experimental results of an active optical electrode with motion artifacts according to an embodiment are shown.

[0090] Figure 27 Experimental results of noise cancellation output with reduced motion artifacts according to the embodiment are shown. Detailed Implementation

[0091] In the following detailed description, reference is made to the accompanying drawings, which form a part of the detailed description. The illustrative embodiments described in the detailed description, depicted in the drawings, and defined in the claims are not intended to be limiting. Other embodiments may be utilized and other changes may be made without departing from the spirit or scope of the subject matter presented. It will be readily understood that aspects of this disclosure, as illustrated in the general description and drawings herein, can be arranged, substituted, combined, separated, and designed in a variety of different configurations, all of which are contemplated in this disclosure.

[0092] Photoelectrodes, or "photo poles," are devices used to detect nerve signals. Figure 13 The structure of the optical pole according to an embodiment is shown. Light is incident from the polarization-maintaining fiber 1302 and enters the liquid crystal layer. The light is then reflected by a mirror and returns to the fiber through the liquid crystal layer. The rotation angle of the liquid crystal is controlled by a voltage across the two ends of the layer (an example of which is shown in reference [2]), which is also the voltage across the two metal pins. As the light passes through and returns within the liquid crystal, the optical power decreases based on the rotation of the liquid crystal. Therefore, the higher the voltage across the metal pins, the higher the optical power.

[0093] Figure 14 A design for an electro-optic detection system according to an embodiment is shown. An optical transducer is used to convert neural signals into optical signals. A light source transmits light to the optical transducer via a circulator. Then, based on the neural signal voltage, a liquid crystal layer in the optical transducer changes the light intensity.

[0094] The light emitted from the photodiode is guided to the receiver board. Photodiodes on the receiver board convert the light into an analog signal, which is then sampled by the data acquisition device.

[0095] Compared with conventional electrode detection systems, Figure 14The illustrated design can achieve lower electrical interference, lower signal attenuation, and a compact size. However, the output noise level of the current electro-optic system is significantly higher than that of conventional electrode systems. This means that neural activity may be more difficult to observe, especially at smaller amplitudes.

[0096] In some designs, most of the noise originates from the light source. Therefore, a straightforward approach to reducing noise from the system is to design a low-noise current source to power the light source. The example implementations provided in this paper include another method for noise reduction: active noise cancellation.

[0097] Active noise cancellation technology is used in a variety of applications, such as noise-canceling headphones, automotive noise cancellation, and RF signal noise cancellation. The basic principle of active acoustic noise cancellation is that when two waves with essentially the same frequency and amplitude are out of phase, they cancel each other out.

[0098] For optical implementations, digital signal noise is generated. Therefore, it is preferable to apply filtering techniques rather than simply subtracting the two signals. Wiener filters (as described in reference

[10] ) can be used in the signal processing portion of active noise cancellation applications. Since the neural signals being detected by the optical system have a similar frequency range to audible sound frequencies, active noise cancellation and Wiener filters can be effective for the optical system.

[0099] This paper presents a system for actively reducing noise in existing electro-optical detection systems.

[0100] In one embodiment of the noise cancellation method provided in this disclosure, one or more additional channels are added to the data acquisition system to record noise characteristics simultaneously with the signal of interest. The location of the additional recording channels will determine the noise characteristics (e.g., but not limited to, inherent noise of the device, noise of the electrode-electrolyte bilayer, mechanical noise).

[0101] In traditional multichannel electrophysiological systems, additional channels can be embedded in the holding substrate to capture noise features. In electro-optic (optical pole) based electrophysiological recording systems, one of the electrodes can be insulated, and this reflective optical pole channel is used to capture noise features. Alternatively or additionally, additional electrodes or optical poles that are in contact with the external environment but not with the object of interest can be added to capture noise features from different sources.

[0102] In one aspect, this disclosure provides a system for noise cancellation and signal calibration. It relates to acquiring noise characteristics separately from the same channel used to acquire the signal of interest (time-based implementation) or from an additional channel substantially juxtaposed with or positioned adjacent to the channel used to acquire the signal of interest (additional channel-based implementation).

[0103] In implementations based on providing additional channels (e.g., see...) Figures 1 to 4 The recording system includes a sensing array that, in addition to channels for recording electrophysiological signals of interest (“signal channels” or “signal probes”), includes one or more channels for recording noise signals (“noise channels”). One or more noise channels are adapted to record noise features simultaneously with the signals of interest. The position of additional electrodes or photodiodes is adjustable and will determine the components / sources of the noise features (e.g., but not limited to, inherent noise of the device, noise from the electrode-electrolyte bilayer, external electromagnetic noise, motion artifacts).

[0104] In traditional MEA recording systems, some of the additional channels are electrodes embedded in the holding substrate and insulated from the external environment by the substrate material to capture noise characteristics.

[0105] The inventors previously developed an optical-based voltage sensing device and method for acquiring electrophysiological signals. These devices are referred to as “optical poles.” They are described in PCT Publication WO2013 / 110141, the contents of which are incorporated herein by reference. In electro-optic (i.e., optical pole) based electrophysiological recording systems, additional optical poles are added, whose metal vias are preferably insulated from the external environment by the same material as the substrate, to capture noise characteristics.

[0106] These noise characteristics mainly originate from internal device noise.

[0107] When this implementation is carried out as a modification of a system involving a multi-electrode array or an optical array, it involves adding one or more electrodes or optical poles to the system, such as a conventional multi-electrode array or optical array.

[0108] In some examples involving optically based systems, electrodes (e.g., Figure 1 Electrode 15 can be embedded in the holding substrate, exposing it to the external environment and connecting it to the data acquisition system 40 via wires / tracks. The electrode can be metallic or a non-metallic material, such as a conductive polymer. This electrode provides an additional recording channel for capturing additional noise characteristics. In doing so, the embedded electrode allows the system to provide protection for the liquid crystal layer of the photoelectrode, which is sensitive to direct current (DC) voltage, by measuring and compensating for any DC offset on the photoelectrode.

[0109] In one example, the additional electrodes or photoelectrodes are in contact with the external environment but not with the sample of interest. Depending on the location of this or these additional electrodes, the feature can include electrolyte-electrode double-layer noise as well as external electromagnetic noise. Therefore, they are adapted to capture noise features different from those of the noise electrodes provided in alternative implementations (e.g., noise electrodes positioned corresponding to the location of the object of interest).

[0110] Therefore, additional channels (probes, in this example electrodes or optical poles) can be used, depending on their placement and material selection, in conjunction with the material selection of the signal recording channels (probes, in this example electrodes or optical poles), to generate noise characteristics representing different noise sources. These noise characteristics can be used to characterize the noise components in the captured features of the signal of interest, as well as the characteristics of the reference signal that captured the measurement baseline. By comparing these different characteristics rather than by applying more complex data processing techniques, the true signal of interest, or its representation, can be generated. The characteristics of various noises and the baseline voltage level also change over time. This also increases the complexity of characterizing the noise using post-data processing of the entire recorded signal (rather than using individual channel signals as taught in this disclosure).

[0111] The following describes a channel-based implementation in two example systems, one with a recording arrangement using optical electrode transducers (optical electrodes) and the other with a recording arrangement using a multi-electrode array.

[0112] 1. Light pole

[0113] Figure 1 and Figure 2 A schematic representation of a multichannel recording system 100 using optical electrodes or light poles is depicted.

[0114] In some embodiments, the multichannel recording system 100 is used to record electrical signals. For in vivo recording, the system 100 is adapted to be embedded in the recording region or object of interest 1, such as tissue, cells, cell cultures, or another sample of interest 1. Of course, the system 100 is also suitable for in vitro or ex vivo measurements of biopotentials.

[0115] System 100 includes recording devices, here an array of conductive devices 10, 13, 14 positioned separately from each other. The conductive devices 10, 13, 14 may be transducers. In this example, they are optical electrode transducers or "optical electrodes".

[0116] Conductive electrodes 10, 13, and 14 are each adapted to transmit a signal voltage from a measured region proximal to the electrode. Conductive electrodes 10, 13, and 14 are supported in substrate 2. Electrodes 10, 13, and 14 are arranged, for example, in a matrix array or any other configuration suitable for a particular application. As will be explained later, photoelectrode 10 is adapted to contact the sample or object 1. Therefore, it is photoelectrode 10 that provides the signal. The other photoelectrodes 13 and 14 each provide noise signals attributable to different sources. Since photoelectrode 13 does not contact object 1 or the recording environment, it will measure noise and / or baseline shift caused by the system arrangement of optics or arrays. Since photoelectrode 14 is in contact with the external environment (measurement chamber) but not with the tissue sample, it is adapted to capture noise characteristics from sources outside the sample (outside the array or included within the recording system itself). As will be mentioned below, they can also be used to determine the presence of a non-zero baseline in the potential within the measurement chamber.

[0117] System 100 also includes a first reference electrode 11, which is arranged in use not to contact the sample 1 to provide a baseline voltage for the signal generated by the sensing electrode. A first conductive track 18 is provided for the first reference electrode 18. In this example, the first conductive track 18 is shown as embedded in the substrate 2. The first conductive track 18 may be grounded.

[0118] However, in different embodiments, the first reference electrode 11 may be placed away from the photoelectric arrays 10, 13, and 14. In electrophysiological recording, if the first reference electrode 11 is placed away from the measurement chamber, it is preferable to place it at a location on the object of measurement where movement is minimal, or / and where electrical continuity between the reference electrode 11 and the measurement chamber (i.e., the photoelectric arrays 10, 13, and 14) can be maintained. It is also preferable that, in this case, the electrode 11 is made of an inert or substantially inert material to further minimize baseline voltage sources outside the measurement chamber.

[0119] Return to reference Figure 1 and Figure 2 In the embodiment shown, system 100 may optionally further include a noise sensing electrode 15, which is arranged such that it does not come into contact with the sample during use. Figure 1 In the example shown, only one noise sensing electrode 15 is included. However, more noise electrodes 15 may be included at various locations in the system 100. A second conductive track 20 is provided for each noise sensing electrode 15. In this example, the noise electrode conductive track 20 is also embedded in the substrate 2.

[0120] Depending on the placement of the noise sensing electrode 15, it will have a baseline voltage contributed by different sources, such as ambient noise, instrument noise, and motion noise (e.g., from the breathing subject). Figure 1In the example shown, the noise sensing electrode 15 is placed in the measurement chamber but does not contact the sample of interest. Therefore, the noise sensing electrode 15 will capture the baseline voltage attributable to the junction potential of the electrode, inherent instrument noise, and, where applicable, movement noise (e.g., if the substrate is embedded in a moving sample).

[0121] 1.1. Noise Cancellation

[0122] exist Figure 1 and Figure 2 In the example shown, photoelectrodes 10, 13, and 14 are disposed in photoelectrode array 30. Each photoelectrode is a layered structure comprising a sensing electrode embedded in substrate layer 2, a liquid crystal layer for converting photoelectric signals, and an output conductive layer 9 made of a transparent conductive material. Photoelectrode array 30 is arranged such that photoelectrodes 10, 13, and 14 share a layered structure but have individual sensing electrodes.

[0123] The photoelectric array includes one or more signal recording photoelectric electrodes 10 for recording signals from the sample of interest. The array also includes one or more noise sensing photoelectric electrodes 13 embedded in a substrate 2. These are embedded within the substrate 2 and insulated from the external environment and adjacent photoelectric electrodes 10, 14 by the substrate 2. The substrate 2 can be a rigid or flexible structure. Using a flexible substrate will provide conformal object / sample contact and better signal transmission from the sample to the exposed electrodes, and improve the comfort of the subject acquiring the recording.

[0124] A single noise sensing electrode 13 can be used for the entire multi-electrode array. That is, the entire array includes only one noise sensing electrode 13, with the rest being signal sensing electrodes 10. Another approach is to provide one noise sensing electrode 13 paired with each signal recording electrode 10. An intermediate arrangement provides multiple noise sensing electrodes 13, but fewer than the number of signal recording electrodes 10.

[0125] In a preferred embodiment, the noise sensing photoelectrode 13 is located between the signal recording photoelectrodes. However, other placements are also possible.

[0126] Light 12 is projected toward each of the noise sensing electrodes 13 so that a noise feature is generated by each noise sensing electrode 13. The noise feature may be attributed to noise within the light source or photoelectric device, such as that caused by refraction, wiring noise, motion artifacts, inherent noise, or random noise. The electrical signal is converted into an optical signal by the photoelectric electrodes 10, 13, and 14, and then detected and processed.

[0127] exist Figure 1In the example shown, the second noise sensing photodiode 14 is embedded in the substrate 2. Unlike the insulated noise sensing photodiode 13, these photodiodes 14 are exposed to the recording environment, i.e., the measurement chamber. They are also arranged such that they will be located at a distance from the sample 1 of interest during use. The exposed noise sensing photodiodes 14 have the same or substantially the same construction as the signal recording photodiode 10, although they do not contact the tissue or sample 1 of interest. Because the noise sensing photodiodes 14 are exposed to the external environment, they will provide different noise characteristics compared to those captured by the insulated noise sensing photodiodes 13. As will be mentioned, having these noise channels makes it possible to determine the amount of noise attributable to a specific source by manipulating the noise levels recorded from different noise channels.

[0128] In this example, the photodiode array includes a liquid crystal layer 8 to polarize incident light to help optimize the operation of the photodiode array. The liquid crystal layer 8 is located between the substrate layer 2 that holds the sensing electrodes 11, 13, and 14 and a transparent conductive material layer 9 (e.g., indium tin oxide). The transparent conductive material layer 9 allows the output of the photodiode array to be transmitted to the data acquisition system 40.

[0129] The liquid crystal layer 8 will have inherent liquid crystal noise. The system 100 in this example includes an additional noise sensing electrode 15. However, more noise sensing electrodes 15 can be provided. The noise sensing electrodes 15 are exposed to the external environment. For example, the noise sensing electrodes 15 will be in direct contact with a solution (e.g., a saline solution) added to the tissue or sample being measured. The noise sensing electrodes 15 have the same or substantially the same material and dimensions as the through-holes (vertical interconnect pathways) of the signal recording photodiodes 10, but are insulated from the liquid crystal and photodiode array via the substrate 2. The noise sensing electrodes 15 are connected to the data acquisition system (conceptually shown by reference numeral 40) via tracks in the substrate 2. If two or more noise sensing electrodes 15 are provided, each noise sensing electrode 15 will be connected to the data acquisition system 40 via a separate track. Due to their placement, the noise sensing electrodes 15 provide additional noise characteristics, which include electrolyte-electrode double-layer noise, but not the inherent liquid crystal noise. The additional noise characteristics can be processed and used for noise cancellation. The data acquisition system 40 can be located in the same or separate processing system as the control system for the optical instruments (i.e., the light source for the photodiode array).

[0130] Therefore, given the above arrangement, system 100 is adapted to record noise characteristics from noise sensing photodiodes 13, 14 and / or electrode 15. These characteristics, associated with different noise sources, are processed and subtracted from the signal of interest recorded by signal recording photodiode 10. Thus, they are useful in improving signal quality.

[0131] The improved signal quality resulting from the implementation of the techniques described herein can increase the information elucidated from electrophysiological experiments and recordings, and also increase the sensitivity of clinical diagnostic electrophysiological systems. For example, when applied to electrophysiological measurements involving the brain, the improved signal quality can improve controllers used in brain-computer interfaces, as well as controllers used in feedback circuits in neural prostheses.

[0132] In one embodiment of a system for electro-optical recording, the techniques described herein can improve the sensitivity of the transducer. This increased sensitivity allows for the detection and recording of ultra-low voltage signals.

[0133] Therefore, in another aspect, the currently disclosed techniques include methods for recording noise characteristics, performing real-time and offline analysis of the captured noise characteristics, and calibrating the device. Software implementing the methods described herein is another aspect of this disclosure.

[0134] 1.2. Signal Calibration

[0135] The noise characteristics from the insulating noise sensing electrode 13 can be used to establish a “baseline” for the optical signal acquired using the recording electrode 10. This allows for calibration of the reflected signal from the signal recording electrode 10, enabling the removal of reflected noise components from measurements obtained using the signal recording electrode 10.

[0136] In an electro-optic transducer (i.e., an optical pole), the change in reflectivity is proportional to the sensed potential. However, the offset of the output signal relative to the input (e.g., caused by a non-zero baseline voltage) and the scaling (e.g., caused by the optical gain in the optical pole) are unknown. By applying a series of known potentials across the device (to the optical pole 10) and measuring the characteristics of noise sensing 13 and 14 as well as the signal of the signal recording optical pole 10, a calibration curve can be constructed for each of the recording optical poles 10.

[0137] 1.3. Protection of LCD

[0138] The liquid crystal in the photoelectric device may be damaged by a DC potential applied across the device. (Return to reference) Figure 1 The noise sensing electrode 15 provides a mechanism to protect the liquid crystal 8 by detecting and canceling any DC voltage generated at both ends of the device.

[0139] Since the noise sensing electrode 15 will be in direct contact with the solution, or, in the case of an in vivo formulation, with the surrounding fluid, a half-cell potential (junction potential) is generated. This electrode 15 has the same or substantially the same material and dimensions as the through-hole of the recording electrode 10, which is also in contact with the same external solution. Therefore, it is reasonable to assume that the half-cell potential of the noise sensing electrode 15 is close to the half-cell potential of the through-hole of the recording electrode 10. Therefore, in Figure 1In the illustrated embodiment, the half-cell potential difference between the reference electrode 11 and the noise sensing electrode 15 is comparable to the half-cell potential difference between the ground electrode 11 and the recording optical electrode 10. This could generate a potential difference across the liquid crystal layer, which, as mentioned, can damage the liquid crystal layer 8. Therefore, if the two half-cell potentials are not the same, a compensation potential of the desired amplitude and polarity is applied to the ground / reference electrode to compensate for the difference between the half-cell potentials.

[0140] Figure 10 An example calibration process 300 for generating noise cancellation signal 316 is shown. Data features 302 are recorded from signal recording photoelectrode 10. A first noise feature 304 is generated by substrate-insulated photoelectrode 13. The first noise feature 304 can be attributed to, for example, inherent instrument noise (e.g., due to optical gain), motion artifacts, random noise, wiring noise, reflectivity, electromagnetic noise, etc. A second noise feature 306 is generated by exposed noise recording photoelectrode 14. The second noise feature 306 can be attributed to external noise and, for example, a junction potential that is theoretically matched to the junction potential of signal recording photoelectrode 10. Removing noise signals 304, 306 from the signal 302 from the recording photoelectrode will produce a calibrated tissue signal 314.

[0141] The half-cell potential 308 of the noise sensing electrode 15 is compared with the half-cell potential 310 of the ground / reference electrode 11 to generate an offset measurement 312. The offset measurement 312 can be used to generate a voltage to compensate for the DC offset caused by the unequal half-cell potentials between the ground electrode 11 and the recording electrode 10. Applying the compensation voltage helps protect the liquid crystal layer.

[0142] In addition, this measurement provides a baseline for electrode measurements of the recording electrode 4 compared to the reference electrode 11.

[0143] The calibration signal data 314 and the offset measurement 312 can therefore be used in the comparison process to generate the calibration electrical measurement 316.

[0144] 2. Multi-electrode array (MEA)

[0145] The implementation of the technology described herein is also applicable to the recording system 200 using MEA. Each electrode in system 200 will have a separate track 3 connected to the data acquisition system 40.

[0146] 2.1. Noise Cancellation

[0147] exist Figure 3 and Figure 4In the illustrated embodiment, the recording arrangement 200 includes a multi-electrode array (MEA) 50 comprising a plurality of electrodes 4, 5, 6, and 7 embedded in a substrate 2. Electrodes 4, 7, and 6 include one or more signal recording electrodes 4 for recording signals from the tissue of interest or sample 1. Electrodes 4, 7, and 6 include one or more first noise sensing electrodes 6 fully embedded in the substrate 2. Each of the first noise sensing electrodes 6 is insulated from the external environment and adjacent electrodes 4, 7 by the substrate 2. As in embodiments involving optical electrodes (i.e., optical electrode-based embodiments), the substrate 2 may be a rigid structure or a more flexible structure conforming to the subject.

[0148] A first noise sensing electrode 6 can be used for the entire MEA 50. Alternatively, a noise sensing electrode 6 can be paired with each signal recording electrode 4. Alternatively, a compromise between these two methods can be adopted, resulting in more than one noise sensing electrode but fewer than the number of standard recording electrodes. In a preferred embodiment, the noise sensing electrodes 6 are located between the standard signal recording electrodes 4, but other placements are also permitted.

[0149] Additionally, in this example, a second noise sensing electrode 7 is also embedded in the substrate 2. However, unlike the insulated noise sensing electrode 6, the second noise sensing electrode 7 is exposed to the external environment. Each of the second noise sensing electrodes 7 is located at a certain distance from the tissue or sample 1 of interest. The second, or "exposed," noise sensing electrode 7 has the same or substantially the same construction (material, size) as the signal recording electrode 4. When exposed to the external environment, the exposed noise sensing electrode 4 will provide a different noise characteristic compared to the insulated noise sensing electrode 6.

[0150] Each electrode will have a separate track 3 to conduct signals to the data acquisition system 40 (see Figure 3 The noise characteristics from the noise sensing electrodes 6 and 7 are processed and subtracted from the signal of interest recorded by the standard recording electrode 4.

[0151] 2.2. Signal Calibration

[0152] The signal from the exposed noise sensing electrode 7, which is far from the sample of interest 1, can be used to calibrate the baseline of the recording arrangement 200.

[0153] The baseline level of the signal recorded by each recording electrode 4 is at least partially attributed to the difference in half-cell potential between the recording electrode 4 and the reference / ground electrode 5. It can also be attributed to any background bioelectrical signals from tissue distal to the region of interest that are directly detected by the recording electrode 4.

[0154] Since the exposed noise sensing electrode 7 does not come into contact with the biological material or the sample of interest 1, the baseline of the signal captured by the exposed noise sensing electrode 7 is instrument-based and is primarily attributed to the half-cell potential difference between the noise sensing electrode 7 and the ground / reference electrode 5.

[0155] The exposed noise sensing electrode 7 has the same or substantially the same material and dimensions as the signal recording electrode 4. Therefore, its half-cell potential (i.e., junction potential) is approximately equal to the half-cell potential (i.e., junction potential) of the standard recording electrode 4. Any difference between the two recorded half-cell potentials will be primarily attributed to the instrument-based DC offset (DC). This offset will affect the baseline recorded by the signal recording electrode 4.

[0156] Figure 11 An example of the calibration process is depicted, which utilizes the... Figure 3 The signal provided in the arrangement shown is measured using a channel (i.e., a measurement using a multi-electrode array). Subtracting the baseline 404 of the signal from the distal noise sensing electrode 7 from the baseline 402 of the signal acquired by each standard recording electrode 4 will produce an estimate 406 of the background noise (e.g., due to electrophysiological activity). Subtracting the baseline 404 of the distal noise sensing electrode 7 from the baseline 408 of the reference electrode 5 will produce any instrument-based DC offset 410 in the recording electrode 4. Therefore, in Figure 3 The arrangement shown includes electrode channels that allow for the separation of the instrument DC offset 410 from the baseline electrophysiological signal level 406. The DC offset 410 and the baseline electrophysiological signal 406 can be used to construct a baseline 412 for the signal of interest recorded by the recording electrode 4. Therefore, the baseline 412 can be subtracted from the physiological recording to generate a noise-modulated signal.

[0157] 3. Common characteristics

[0158] In the examples discussed above, both the optically based and the conventional multi-electrode-based recording configurations require reference or ground electrodes 5, 11. Reference electrodes 5, 11 record signals from external ambient noise. The electrodes also have junction (i.e., half-cell) potentials.

[0159] Real-time or offline signal processing can be applied to subtract noise features from the signal of interest. This process involves minimizing the distortion of the signal of interest. The processing can involve time-domain and / or Fourier-domain subtraction of the noise features.

[0160] Real-time or offline processing can also be used for optical poles that record signals (such as...) Figure 1 The signal output from the arrangement 100 shown is calibrated. It also allows for the measurement and adjustment of DC offset in MEA-based and multi-polar array-based systems.

[0161] In some implementations, signal processing is implemented in hardware. For example, such as... Figure 5 As shown, the system will include a differential amplifier, a sample and hold circuit system, and other hardware processing units for processing the master signal and reference signal, or a software equivalent of such a system.

[0162] Figure 5 An example is also shown in which the sample and hold circuitry 508 holds the sample points acquired during the oversampling period (or during the recording period). A subtraction circuit 510 (e.g., a subtraction circuit including a differential amplifier) ​​can then be used to output a noise cancellation output 512, which is the difference between the optical output during the oversampling period 514 and the optical output during the recording period 516 of the object signal of interest.

[0163] Alternatively, a comparison between the noise acquired during the oversampling period and the signal of interest collected during the "real sampling" or "recording period" can be performed using signal processing in software.

[0164] The following provides several examples of how time-based implementations can be achieved. However, it will be understood that these are not exhaustive. For example, in a system using an optical pole positioned to sense data of interest, the time-based approach involves oversampling and sample-and-hold. The system's data acquisition and digital-to-analog converter will oversample, meaning the system will acquire data samples at a rate higher than the actual sampling rate of the signal of interest. During the oversampling period, for each "true" sample point (i.e., the sample point from the signal of interest), at least one additional "noisy" sample point will be acquired where optical transmission is interrupted and no light reaches the object of interest.

[0165] For example, as shown in Figure 6-1, shutter 506 can be positioned between light source 502 and optical components 532 (e.g., mirrors, pinholes, collimators, optical fibers, circulators, etc.) that guide light to photoelectrode 536 to interrupt light incident on photoelectrode 536. Alternatively or additionally, shutter 506 can be positioned between optical components 532 that guide the output from photoelectrode 536 to photodetector 534 that detects the output to interrupt detection of the output from photoelectrode 536. Interruption of light transmission can be accomplished using other mechanisms, such as turning the light source on and off, or having a resonant mirror that intermittently (e.g., periodically) deflects the incident light from its predetermined path to the photoelectrode. As an alternative to shutter, the voltage supply to the photoelectrode unit used in the arrangement can be turned on and off.

[0166] When no light is incident on the photoelectrode, the samples collected during the oversampling interval will contain noise information but not the signal of interest. The noise collected during the oversampling interval will contain information about the system.

[0167] Depending on the placement of the shutter or resonator in the optical path, or whether the light at the light source is periodically switched off, the noise signal will contain information about noise originating from the system up to the shutter or resonator insertion point. If light transmission between the light source and photoelectrode 536 is interrupted, the noise data will include noise contributed by system components downstream of the light source 502. If light transmission between photoelectrode 536 and photodetector 534 is interrupted, the noise data will include noise contributions from the photodetector and downstream instruments.

[0168] Oversampling (i.e., sampling at a rate higher than the required sampling rate for the actual data) generates a data sequence 538 (see Figure 6-2), where black circles represent data from the signal photoelectric pole and gray circles represent data from the noise sensing photoelectric pole. The data sequence 538 can then be split into a signal data sequence 542 and a noise data sequence 540 (see Figure 6-3) using software or hardware such as a shutter timer or counter. The signal data sequence 542 and the noise data sequence 540 are then processed for various functions discussed herein, such as noise cancellation to obtain noise-cancelled data, and the establishment of various baselines. The data sequences 538, 542, and 540 shown are for illustrative purposes only and do not limit the scope of the invention in any way.

[0169] Referring to Figure 7, an alternative to the optical interruption shown in Figure 6 is an operation switch 520. Switch 520 toggles between signal line 3 from signal electrode 522 and noise sensing electrode 524. This switching results in an oversampled data sequence 526, where black circles represent data from the signal electrode and patterned (dashed) circles represent data from the noise sensing electrode. The data sequence 526 can then be split into a signal data sequence 528 and a noise data sequence 530 using hardware or software processing. For example, a timer or counter can be used. The signal data sequence 528 and the noise data sequence 530 are then processed for various functions discussed herein, such as noise cancellation and the establishment of various baselines. The data sequences 526, 528, and 530 shown are for illustrative purposes only and do not limit the scope of the invention in any way.

[0170] Figures 8-1 and 8-2 depict an alternative mechanical implementation. In this implementation, a substrate 2, in which photodiodes 522 and 524 are embedded, moves periodically. In Figure 8-1, the substrate 2 is positioned such that light incident from the light source 502 and redirected by the optical component 532 reaches the signal photodiode 522. In Figure 8-2, the substrate 2 is positioned such that the redirected light reaches a noise sensing photodiode 524 at another point on the substrate 2. The position changes at a rate specified by the oversampling rate to obtain an oversampled data sequence 550, which can be divided into a noise data sequence 551 and a signal data sequence of interest 552 based on the position of the substrate 2 at each data point acquired (Figure 8-3). The movement of the substrate will introduce noise into the signal; however, this noise can be recorded and removed from the signal. A variation of this implementation is in which the substrate is fixed in place, but the optical component guiding the light input moves periodically.

[0171] Figures 9-1 and 9-2 illustrate another alternative time-based implementation. In this implementation, the optical components include a redirection section 552. The redirection section is a component movable between multiple locations. At one location, it directs light from the light source 502 toward the signal electrode 522, as shown in Figure 9-1. At another location, it directs light from the light source 502 toward the noise sensing electrode 524, as shown in Figure 9-2. The movable section 552 moves at an “oversampling rate” (as shown in Figure 9-3) to obtain an oversampled data sequence 560 comprising both signal data and noise data. The data sequence 560 can be split into a signal sequence 562 of interest and a noise sequence 564 by software or hardware processing. In a variation of this implementation, the redirection section can be a resonator (e.g., a resonant crystal) that oscillates between different locations when excited by certain frequencies or multiple frequencies. This can be accomplished, for example, using standard components in a scanning microscope system.

[0172] Figure 12 A general conceptualization of a process 600 for obtaining a final signal 602 of interest, according to one embodiment, is depicted. Measurements or recordings 604 from signal electrodes or optical electrodes provide a total signal containing the signal of interest and noise signals. Measurements or recordings 606 from noise-sensing electrodes or optical electrodes that are not in contact with the recording chamber or the object of interest capture noise from the measurement system (and, in the case of optical electrodes, optical noise). These measurements 606 can be used to determine noise and / or baseline shift due to the system itself.

[0173] Measurements or recordings 608 from noise sensing electrodes that are in contact with the recording chamber but not with the object of interest will capture at least some of the noise attributable to the electronics (e.g., noise caused by junction potential as discussed above). These measurements 608 can be used to determine noise and / or baseline shift due to the electronics. For example, a baseline component may exist if there is a non-zero difference in the junction potential in the electrodes or photoelectrodes used.

[0174] Therefore, in the offset determination step 612, measurements 606 and / or 608 from the noise sensing electrode will be used to determine the signal offset.

[0175] Measurements 604, 606, and 608 may optionally be processed or regulated using software or hardware processing in adjustment step 610. This includes, for example, signal smoothing or filtering. Next, in noise subtraction step 614, the regulated signal is processed to separate and subtract the noise component from the signal components to obtain the noise-subtracted signal of interest. The order in which adjustment step 610 and noise subtraction step 614 are performed can be reversed.

[0176] From the perspective of the overall signal, in the offset compensation step 616, the baseline calculated using measurements 606 and 608 using the noise electrode (or optical electrode) is subtracted from the noise subtraction signal of interest to obtain the final signal of interest 602.

[0177] The implementation of the techniques described herein involves using the same channel as the acquisition channel for the signal of interest or using a separate channel positioned close to the acquisition channel to acquire noise features individually from one or more sources. The advantage is that in vivo electrophysiological recording can be performed by embedding the recording arrangement into any excitable tissue of interest.

[0178] For example, when recording electrophysiological signals from the heart to illustrate cardiac motion, existing methods involve injecting compounds into heart tissue to induce interruptions in the heartbeat. This method is not suitable for in vivo recording. However, when using a signal channel or a channel substantially juxtaposed with the signal channel for noise acquisition, motion artifacts affect both noise and signal acquisition in essentially the same way. That is, the noise characteristics will be affected by motion artifacts to a similar or identical degree to the signal of interest. Therefore, by processing the noise and signal characteristics, the effects of motion artifacts can be substantially eliminated. Of course, the advantage of eliminating or minimizing the effects of motion artifacts also exists when the sample is non-biological or non-physiological.

[0179] The embodiments described herein achieve noise cancellation, output signal calibration, or both. The embodiments of this disclosure can be applied not only to electrophysiological recording systems, but also to any application / field where single-channel or multi-channel recording of ultra-low voltage signals is desired.

[0180] Please understand that the accompanying drawings are not drawn to scale. The arrangements depicted in the drawings are for illustrative purposes only.

[0181] Changes and modifications may be made to the previously described parts without departing from the spirit or scope of this disclosure.

[0182] For example, the electrodes / photons can be arranged in any configuration, not just the layout shown in the example depicted in the accompanying drawings.

[0183] It will also be understood that the optical or electrode materials can be selected to suit the arrangement of a specific application. However, as mentioned above, for baseline calibration purposes, in an optically based system, the additional noise sensing optical electrode is preferably made of the same or substantially the same material as the standard recording optical electrode. The additional noise sensing electrode is preferably made of the same or substantially the same material as the ground electrode via, such that the half-cell potentials of the additional noise sensing electrode and the ground electrode are substantially the same or similar. Ideally, the connections and conductive tracks to the data acquisition system 40 should be made of the same or substantially the same material as the material used for the reference electrode.

[0184] In the MEA system, the additional noise sensing electrodes are preferably made of the same or substantially the same material as the standard recording electrodes. The connections and conductive tracks to the data acquisition system 40 used for these additional electrodes are also preferably made of the same or substantially the same material as the standard recording electrodes.

[0185] While embodiments relating to examples involving optical or multi-electrode recording have been disclosed, they are applicable to various recording systems in which physical signals (reflections, sound, or other signals) from a tissue or sample of interest are converted into electrical signals. The sample may be biological or non-biological.

[0186] In the preceding text, the term "measuring chamber" refers to the location or space in which a recording device is situated in order to obtain both a signal and one or more noise signals from the recording area (i.e., the sample or tissue of interest). It can be more generally referred to by the term "recording environment," which should be understood to be unrestricted by a specific "chamber," as in cases such as ex vivo or in vitro recording.

[0187] Therefore, as summarized above, the system may include one or more transducer units that, instead of being insulated within the substrate, sense the signal of interest. The transducer units are provided together with one or more first noise transducer units, which are insulated from the signal of interest and may be disposed within the substrate. An additional channel provided by the first noise transducer units captures a first noise feature. Furthermore, in some embodiments, the system includes one or more second transducer units that capture a second noise feature. The sum of the first and second noise features is subtracted from the recorded signal of interest to determine a calibration signal, wherein at least some of the noise associated with the recorded signal has been removed.

[0188] 4. Example Implementation

[0189] Described with Figures 13 to 26 Related example implementations. Figure 17 Equations (a) to (e) according to the implementation are shown.

[0190] 4.1 Active Noise Cancellation (ANC) System

[0191] In some implementations, the light source contributes a significant portion of the noise to the output signal. By placing a beam splitter at the output of the light source, two optical output channels with nearly identical noise levels can be obtained. These two optical channels can then be connected to an active optical electrode, a passive optical electrode, and two channels from a photodiode receiver. With this setup, one message channel contains neural signals and system noise, while the other noise channel contains only system noise. By transferring the signals from both channels to an active noise cancellation algorithm, the noise in the message channel can be significantly reduced based on the information contained in the noise channel.

[0192] Figure 15 This is a block diagram of an Active Noise Cancellation (ANC) system 1500 according to an embodiment. The ANC system consists of two boards: a receiver board and an FPGA development board. The optical signals from both channels are received by the receiver board, amplified by a maximum gain of 140 dBΩ, and converted into digital signals. Then, the digital signals from both channels are processed by a noise cancellation algorithm on the FPGA development board, which outputs clean signals.

[0193] 4.2 Receiver Board

[0194] Figure 18This is a schematic diagram of an optical receiver board according to an embodiment. The optical receiver board has two power supply stages. The first stage low-dropout (LDO) regulator (LT3045EMSE) converts an external voltage supply ranging from 7 V to 20 V to 6.2 V. To reduce interference and noise on the board, there are three second-stage LDOs (LT3045EMSE), two for analog power and one for digital power, all of which convert 6.2 V to 3.3 V.

[0195] Light is reflected at different reflectivities at the photoelectrodes, the reflectivities being determined by the angle at which the liquid crystal is rotated by the neural signal voltage. The increased AC neural signal portion is less than 10% of the total light intensity. Since only the neural signal portion of the light is of interest, and the signal needs to be amplified by at least 120 dB, the DC portion of the light must be removed before amplification. Therefore, the amplifier is AC coupled through a large capacitor immediately following the photodiode. When the photodiode is configured to reverse bias, the system suffers from 1 / f noise from the LDO. Therefore, photodiode D1 is configured in “zero mode” (an example of which is described in reference

[13] ). In this mode, the sensitivity is reduced due to the absence of DC bias, but the noise from the DC bias power supply is also removed, which greatly improves the overall signal-to-noise ratio.

[0196] Refer again Figure 18 Operational amplifiers U1 and U2 are two-stage amplifiers. A large capacitor C1 ensures that only AC current is sent to operational amplifier U1, which is configured as a transimpedance amplifier. The gain of the transimpedance amplifier is set by a variable resistor R1. Operational amplifier U2, together with R2 and R3, forms an inverting amplifier, and its gain is controlled by a variable resistor R3. The resistors R1 and R3 are both derived from digital potentiometers, with a minimum resistance of 390 Ω and a maximum resistance of 10,000 KΩ, which allows the total gain of the two-stage amplifiers to range from 43.6 dBΩ to 140 dBΩ. Operational amplifiers U1 and U2 (ADA4896) were selected for very low noise. Operational amplifiers U1 and U2 have a gain of 2.3 at 10 Hz. Input reference voltage noise and 11 The input reference current noise is a factor. However, the operational amplifier also introduces an input bias current of -11 uA, which is undesirable even with the large gain. With R1 set to 10,000 kΩ, the -11 uA input bias current will result in an output voltage of -1.1 V. Therefore, Vref is set to 2.1 V to have a peak-to-peak signal range of approximately 2 V and a bottom gap space of 1.1 V. The maximum input current from the photodiode is 2V / 140dBΩ = 200nA.

[0197] The output voltage Vx from the two-stage amplifier is then passed through an active fourth-order anti-aliasing low-pass filter consisting of U3 and U4 (LT6233). The cutoff frequency is set to 12 kHz to balance signal hold up to 10 kHz and anti-aliasing at the 64 kHz ADC sampling frequency.

[0198] Finally, a 24-bit 64 kHz analog-to-digital converter (MAX11254) converts the amplified analog signal into a digital signal, which is then sent to the FPGA board (Arty Z7) via the SPI port.

[0199] 4.2 Digital Signal Processor (DSP)

[0200] Brain-computer interfaces can utilize arrays of hundreds of optical electrodes working together. The expectation is that the processing of signals generated by the optical electrodes occurs in real time.

[0201] The Wiener filter (described in reference

[10] ) has been tested to effectively eliminate noise using signals recorded from existing setups. However, the Wiener filter processes only a portion of the signal in the box at a time and requires significant computational resources when the box size becomes very large. Experimental data show that the optical system exhibits almost no correlation between data streams of more than three samples with zero hysteresis. In contrast, the single-tap filter shows significant noise reduction, but increasing the number of taps shows little improvement. Furthermore, to accommodate algorithms capable of processing hundreds of channels simultaneously, the algorithm for each channel must be small in size and consume very little power. Therefore, a single-tap adaptive filtering algorithm based on the Wiener filter was designed and implemented on an FPGA board.

[0202] Figure 19 A digital signal processing architecture according to an embodiment is shown. It is assumed that transfer function H1 is removed from the system, and transfer function H2 is an integrator. The system has two inputs. and And the output y signal. It is a noisy message signal, and It's a noise signal. The system is trying to remove it. Zhongyu The signal contains some noise related to the noise itself. The value of can be determined according to equation 17(a), where n is and The noise in both, s is The signal in the signal plus other noise, m is Other noise in the signal, and k is a scaling factor. s, n, and m are random signals and are uncorrelated with each other.

[0203] Let N be and The total number of samples in the sample. Take... for and The cross-correlation; and for Autocorrelation. It can be determined according to Equation 17(b).

[0204] According to Equation 17(c), the filter coefficients can be determined by dividing the cross-correlation by the autocorrelation. Since s, n, and m are uncorrelated, according to Equation 17(d), their cross-correlation should be equal to zero.

[0205] In some implementations, experimental results show that kn is The main noise in the equation is shown in Equation 17(e). Therefore, Equation 17(c) can be simplified according to Equation 17(f).

[0206] The output signal y can be determined according to Equation 17(g).

[0207] Due to kn m, therefore with In comparison, the noise level in y is greatly improved.

[0208] To make the system more practical, a high-pass filter H1 is implemented to remove DC offset from the signal, improving cross-correlation and autocorrelation results. A leakage integrator H2 is used, so the system has a moving average to track the most recent signal for approximately half a second. This leakage integrator compensates for losses caused by temperature changes or physical motion. and The variation of different noise gains in the data.

[0209] Figure 16 The data flow in the system according to the embodiment is shown. Light is emitted from the light source and split into two identical beams at a beam splitter. One beam then enters an active photodiode, where it carries neural signals, while the other beam enters a passive photodiode and does not carry additional information. Both beams are converted into current by a photodiode, then into voltage by an amplifier, and finally into digital signals by an ADC, all occurring on the light receiver board.

[0210] The digital signal is then transmitted to the FPGA for digital signal processing. The light source is a superluminescent diode with a wavelength of 1550 nm and a variable current of up to 800 mA, operating at a constant temperature of 20°C. The frequency range of the neural signal is 10 Hz to 10 kHz.

[0211] The USB serial port embedded in the FPGA board can only stream data to a PC at 300 Hz, which is insufficient. However, it can stream real-time data at 64 kHz via suitable peripherals.

[0212] 5. Experimental Results

[0213] 5.2 Cross-correlation

[0214] Preferably, for the digital signal processing algorithm to function well, the noise in the two optical channels must be very similar. The cross-correlation between the two receiver channels can be calculated by measuring directly from the two receiver channels without any processing algorithm, with the light source on and no input to the light source. Figure 20 This is a graph showing the cross-correlation between the active and passive light source channels at different light source powers when there is no input signal at the active light source according to the embodiment.

[0215] At lower light source power, the noise on the photoelectrode and photoreceiver board is considerably larger than the noise from the light source. Neural signals and light source noise increase proportionally with increasing light power, while the noise on the photoreceiver board remains constant. When the light source current exceeds 500 mA, the light source noise dominates the overall noise in the system, and the normalized cross-correlation remains stable at 0.8. This suggests the possibility of active noise cancellation in this photoelectrode system.

[0216] 5.2 Noise Comparison

[0217] Figure 21 Two pairs of graphs are shown, according to an embodiment, illustrating the signals before (active optical channel) and after (noise cancellation output channel) the noise cancellation process.

[0218] Graphs (a) and (b) show the noise signals from a light source powered by a 500mA current. The active optical channel signal shown in graph (a) has a much larger amplitude range than the noise-cancelled output signal shown in graph (b), meaning that the noise in the noise-cancelled output channel has been reduced. However, in both graphs (a) and (b), the two signals still have a normalized cross-correlation of 0.4, indicating that there is still room for improvement in performance.

[0219] Figure 21 Graph (c) shows the active photoelectric signal with a current of 500 mA at the light source. Graph (d) shows the corresponding noise-cancelled output channel, where a 200 mV sinusoidal signal input has been applied to the active photoelectric signal shown in graph (c). Advantageously, the signal in the noise-cancelled output channel (graph (d)) is smoother and more standard than the signal in the active photoelectric channel (graph (c)).

[0220] Figure 22This is a graph comparing the RMS values ​​of Channel 1 input and DSP output under different light source currents when there is no input signal at the light pole, according to an embodiment. The blue bar shows the Channel 1 input before the signal processing algorithm. As the light source current increases, the RMS value of the blue bar increases, confirming that the noise in the light source is proportional to the light source current. The red bar shows the RMS value of the digital signal processing output. At lower light source currents, the noise reduction from the blue bar to the red bar is relatively small, but it improves with increasing light source current and remains at around 50% when the light source current is above 500 mA, which matches the cross-correlation results.

[0221] 5.3 Optical Input Test Using Sine Wave Signals

[0222] Figure 23 The diagram illustrates the signal-to-noise ratio (SNR) and SNR improvement for a 200 mV sinusoidal input under different source currents, according to an embodiment, before and after the active noise cancellation process.

[0223] Figure 24 The improvement in signal-to-noise ratio (SNR) and signal-to-noise ratio (SNR) before and after the active noise cancellation process is shown under different light source currents for a 10 mV sinusoidal input, according to the embodiment.

[0224] There are standout data points at the 300 mA input, but overall it matches previous results, namely that the SNR improvement is rather small at lower source currents, but increases with increasing source currents and then stabilizes at higher source currents.

[0225] Figure 25 A comparison of the active photoelectric channel (a) and noise-cancelled output (b) at a photocurrent of 500 mA with a 10 mV 1 kHz sinusoidal input signal at the active photoelectric electrode is shown according to an embodiment. The 1 kHz signal level remains constant in both graphs, while the amplitude elsewhere decreases by 5 dB to 10 dB, which matches the results of the readings above.

[0226] 5.4 Eliminating physical motion

[0227] When a neural signal detection device is installed on the human body, the system produces some physical motion artifacts. Figure 26 and Figure 27 The corresponding experimental results graphs according to the implementation method are shown. Figure 26 Experimental results of an active optical electrode with motion artifacts according to an embodiment are shown. Figure 27 The experimental results of noise cancellation output according to the embodiment are shown, which correspond to Figure 26 The experimental results showed a reduction in motion artifacts.

[0228] Although some low-frequency noise remains in the DSP channel, the motion-induced spikes have been largely removed, indicating that the system implementation is effective in removing physical motion artifacts.

[0229] Those skilled in the art will understand that many variations and / or modifications can be made to the above embodiments without departing from the broad general scope of this disclosure. Furthermore, those skilled in the art will understand that the embodiments disclosed herein can be combined with one or more other embodiments disclosed herein without departing from the broad general scope of this disclosure. Therefore, these embodiments are to be considered illustrative rather than restrictive in all respects.

[0230] Those skilled in the art will understand that any suitable functional distribution among the different functional units can be used without diminishing the invention. For example, a function shown to be performed by a single device may be performed by the same device. Similarly, a function shown to be performed by a single device may be distributed among several devices. Therefore, references to specific functional units are to be regarded only as references to suitable means for providing the described function, and not as indications of a strict logical or physical structure or organization.

[0231] Those skilled in the art will understand that the operations performed in the processes and methods disclosed herein may be implemented in different orders. Furthermore, the steps and operations outlined are provided by way of example only, and some steps and operations may be optional, may be combined into fewer steps and operations, or may be extended to other steps and operations without diminishing the essence of the disclosed embodiments.

[0232] Throughout this specification, the word “comprise” or variations thereof, such as “comprises” or “comprising”, shall be understood to mean including the stated element, whole or step, or group of elements, whole or steps, but not excluding any other element, whole or step, or group of elements, whole or steps.

[0233] As used herein, any reference to “one embodiment” or “implementation” means that a particular element, feature, structure, or characteristic described in connection with that embodiment is included in at least one embodiment. The phrase “in one embodiment” appearing in various places throughout the specification does not necessarily refer to the same embodiment. Similarly, the use of “a” or “an” before an element or component is merely for convenience. This description should be understood to mean that there are one or more elements or components, unless it is obvious otherwise.

[0234] Unless explicitly stated otherwise, “or” refers to an inclusive “or”, not an exclusive “or”. For example, conditions A or B are satisfied by any of the following: A is true (or exists) and B is false (or does not exist), A is false (or does not exist) and B is true (or exists), and both A and B are true (or exist).

[0235] References

[0236] The following references are included in this article:

[0237] [2] E. Scherschener, CD Perciante, EA Dalchiele, EM Frins, M. Korn and JA Ferrari, “Polymer-dispersed liquid-crystal voltage sensor”, Applied Optics, Vol. 45, No. 15, p. 3482, 2006.

[0238]

[10] N. Benvenuto, S. Tomasin and G. Cherubini’s Algorithms for communications systems and their applications. Wiley, 2021.

[0239]

[13] Y. Wei, T. Lehmann, L. Silvestri, H. Wang and F. Ladouceur, “Photodiode working in zero-mode: Detecting light power change with dcrejection and ac amplification,” Optics Express, Vol. 29, No. 12, p. 18915, 2021.

Claims

1. A recording device comprising: a reference electrode; a plurality of electrically conductive devices; the plurality of electrically conductive devices comprising at least one signal conducting device adapted to conduct a signal from a subject of interest located at least partially in a recording environment; and at least one noise conducting device adapted to conduct a noise signal.

2. The apparatus of claim 1, wherein, the at least one noise conducting device comprising one or more first noise conducting devices adapted in use to be exposed to the recording environment.

3. The apparatus of claim 2, wherein, the first noise conducting devices and the signal conducting device have substantially the same material and size.

4. The apparatus of any one of claims 1 to 3, wherein, the at least one noise conducting device comprising one or more second noise conducting devices adapted in use to be isolated from the recording environment.

5. The apparatus of claim 4, wherein, at least some of the plurality of electrically conductive devices are disposed in a substrate.

6. The apparatus of claim 5, wherein, the one or more second noise conducting devices are embedded within the substrate.

7. The apparatus of claim 5 or claim 6, wherein, the reference electrode is disposed in the substrate.

8. The apparatus of any one of claims 1 to 7, wherein, the plurality of electrically conductive devices is a multi-electrode array.

9. The apparatus of any one of claims 1 to 7, wherein, at least some of the electrically conductive devices are transducers.

10. The apparatus of claim 9, when dependent directly or indirectly from claim 2, wherein, the one or more first noise conducting devices comprise at least one noise electrode and at least one noise sensing transducer.

11. The apparatus of claim 10, wherein, the tracks for conducting with the at least one noise electrode and the tracks for conducting with the reference electrode have substantially the same material.

12. The apparatus of any one of claims 9-11, wherein, the transducers are each adapted to convert an optical signal into an electrical signal, receive a return electrical signal, and transmit an output optical signal converted from the return electrical signal.

13. The device of any one of claims 1 to 12, further comprising a timer or controller adapted to time a switch between an input of the signal conducting device and an input of the noise conducting device.

14. The device of claim 13, further comprising signal holding circuitry for periodically holding an output of the signal conducting device or the noise conducting device.

15. A multi-channel recording device comprising: a reference electrode; a plurality of opto-electric transducers embedded in a light-transmissive substrate; the plurality of opto-electric transducers comprising at least one signal recording transducer adapted to convert an optical input into an electrical input to a subject of interest; the plurality of opto-electric transducers comprising at least one noise recording transducer adapted to receive a response signal from the subject of interest and convert the response signal into an output optical signal; a detector adapted to detect the output optical signal.

16. An electrical measurement device comprising: a reference electrode; at least one signal recording device adapted to record a signal from a subject of interest; an input device adapted to analog a subject at a user-defined time.

17. The apparatus of claim 16, wherein, The at least one signal recording device is at least one opto-electric transducer embedded in a light-transmissive substrate, the at least one opto-electric transducer being adapted to convert an optical input into an electrical input to the object of interest.

18. The apparatus of claim 17, the at least one opto-electric transducer being adapted to convert an electrical signal from the object of interest in response to the electrical input into an output optical signal.

19. The apparatus of claim 18, wherein, The optical input is provided at a user-defined time by: switching on and off a light source; interrupting the transmission of light from a light source at a user-defined time; moving the at least one opto-electric transducer away from a light-receiving position.

20. A recording system comprising a recording apparatus according to any one of claims 1 to 19.

21. A signal recording method comprising: providing an input signal to an object of interest located at least partially in a recording environment; obtaining a response signal from the object of interest in response to the input signal; the response signal being transmitted or converted by at least one signal recording electrode or transducer; obtaining noise data from a location within a measurement environment; obtaining noise cancellation data, the obtaining noise cancellation data comprising subtracting the noise signal from the response signal.

22. The method of claim 21, comprising: providing at least one noise recording electrode or transducer.

23. The method of claim 22, wherein, The obtaining the noise signal comprises obtaining a first noise signal from at least one first noise electrode or transducer, the at least one first noise electrode or transducer being positioned in isolation from the object of interest but exposed within the measurement environment.

24. The method of claim 22 or claim 23, wherein, The obtaining the noise signal comprises obtaining a second noise signal from at least one second noise electrode or transducer, the at least one second noise electrode or transducer being fully embedded within a substrate located in the recording environment such that it is not exposed in the recording environment.

25. The method of any one of claims 21 to 24, wherein, The at least one first noise electrode or transducer is at least one first noise electrode and the noise signal comprises a third noise signal from at least one noise transducer, the at least one noise transducer being positioned in isolation from the object of interest but exposed within the recording environment.

26. The method of any of claims 21-25, further comprising: obtaining a reference signal from a reference electrode.

27. The method of claim 26, when directly or indirectly dependent on claim 23, further comprising obtaining a value of a voltage offset voltage, wherein, The obtaining a voltage offset voltage value comprises calculating a difference between the reference signal and the first noise signal.

28. The method of claim 27, further comprising: applying a compensation voltage to a substrate located within the measurement environment, the compensation voltage being a voltage offset voltage having an opposite polarity.

29. The method of claim 27 or claim 28, wherein, The obtaining noise cancellation data comprises subtracting the voltage offset value from the response signal.

30. The method of any of claims 27-29, comprising: The input signal is provided at a plurality of input values.

31. The method of claim 30, comprising: A plurality of voltage offset values are obtained, each at a corresponding input value, and a baseline is calculated from the plurality of voltage offset values.

32. The method of claim 31, comprising: The response signal data is calibrated to remove the baseline.

33. The method of 21-32, further comprising: The response signal and / or the noise signal is conditioned.

34. The method of claim 33, wherein, The conditioning comprises smoothing or filtering the response signal and / or the noise signal.

35. The method of any one of claims 21 to 34, wherein, The input signal is provided intermittently.

36. The method of claim 35, wherein, The obtaining the noise signal comprises obtaining a signal from the at least one signal electrode or transducer when the input signal is not provided.

37. The method of claim 36, wherein, The input signal is an optical input.

38. The method of claim 37, wherein, The optical input is provided at a user-defined time by: turning on and off a light source; interrupting the transmission of light from the light source; moving the at least one signal electrode or transducer to a light receiving position and moving the at least one signal electrode or transducer away from the light receiving position.

39. A computer program comprising instructions for controlling a computer to implement the method of any one of claims 21 to 38.

40. A device for obtaining a functional measurement of a subject of interest, comprising a computer having a processor and a memory supporting computer processing, the computer implementing the method of any one of claims 21 to 38.

41. An electro-optical detection device, comprising: a first optrode configured to receive a first optical signal and generate a first optrode signal, the first optrode configured to be in contact with a sample of interest; a second optrode configured to receive a second optical signal and generate a second optrode signal, the second optrode configured to be isolated from the sample of interest; a receiver configured to: determine an output signal by applying an active noise cancellation algorithm to the first optrode signal and the second optrode signal.

42. The device of claim 41, further comprising: a light source configured to generate an optical signal; and a light source splitter configured to split the optical signal into the first optical signal and the second optical signal. Generating a first optrode signal comprises:

43. The apparatus of any one of claims 41-42, wherein, receiving, by the first optrode, the first optical signal and reflecting the first optical signal into a photodiode; generating, by the photodiode, an analog electrical signal, converting, by an analog-to-digital converter, the analog electrical signal to generate the first optrode signal. Applying an active noise cancellation algorithm comprises:

44. The arrangement of any one of claims 41 to 43, wherein, subtracting the second optrode signal from the first optrode signal.

45. The apparatus of any one of claims 41-44, wherein, Applying an active noise cancellation algorithm comprises:

46. The apparatus of any one of claims 41-45, wherein, applying a Wiener filter to the first optrode signal and the second optrode signal.

47. The apparatus of any one of claims 41-46, wherein, Applying an active noise cancellation algorithm comprises:

48. The apparatus of claim 47, further comprising: applying a single-tap adaptive filter to the first optrode signal.

49. The apparatus of claim 48, further comprising: Applying an active noise cancellation algorithm comprises:

50. The apparatus of any one of claims 41-49, wherein, determining a cross-correlation of the first optrode signal and the second optrode signal.

51. The apparatus of any one of claims 41-50, wherein, Determining an autocorrelation of the second optrode signal. Determining filter coefficients based on the cross-correlation and the autocorrelation. Applying an active noise cancellation algorithm comprises: applying a high-pass filter to the first optrode signal. The first optrode comprises: a polarization-maintaining optical fiber; a liquid crystal layer; and a mirror.

52. A method of detecting a neural signal, comprising: generating, by a light source, an optical signal; 53. The method of claim 53, wherein, splitting the optical signal into a first optical signal and a second optical signal; generating, by a first optrode, a first optrode signal based on the first optical signal, the first optrode configured to be in contact with a sample of interest; generating, by a second optrode, a second optrode signal based on the second optical signal, the second optrode configured to be isolated from the sample of interest; determining an output signal by applying an active noise cancellation algorithm to the first optrode signal and the second optrode signal. Generating the first optrode signal comprises: receiving, by the first photodiode, the first optical signal and reflecting the first optical signal into a photodiode; generating, by the photodiode, an analog electrical signal, converting, by an analog-to-digital converter, the analog electrical signal to generate the first photodiode signal.

54. The method of any one of claims 52-53, wherein, applying an active noise cancellation algorithm includes subtracting the second photodiode signal from the first photodiode signal.

55. The method of any one of claims 52-54, wherein, applying an active noise cancellation algorithm includes applying a single-tap active filter to the first photodiode signal and the second photodiode signal.

56. The method of any one of claims 52-55, wherein, applying an active noise cancellation algorithm includes applying a Wiener filter to the first photodiode signal and the second photodiode signal.

Images