Event-triggered weak biomedical electrical signal detection circuit and bioelectrical signal device

Abstract

The present invention provides an event-triggered weak biomedical electrical signal detection circuit and a bioelectrical signal device, which relate to the field of integrated circuits. A filtering module filters the bioelectrical signal to obtain a noise-reduced nerve signal; a voltage-to-current module converts the noise-reduced nerve signal into a corresponding current signal; an ADC module generates a pulse signal according to the current signal and combines the pulse signal and a Gray code clock to generate the time value of the Gray code clock corresponding to any pixel within one sampling period; a global processing module processes the historical time value of the Gray code clock corresponding to the previous sampling period and the real-time time value of the Gray code clock corresponding to the current sampling period in an event-triggered manner, and outputs the real-time time value or a comparison result. The structure of the present invention is simple, which reduces the power consumption and noise of the system, improves the signal-to-noise ratio, and improves the reliability. It realizes the compression of the data code stream, which is beneficial to the acquisition and research of large-scale bioelectrical signal data.

Description

Technical Field

[0001] The present invention relates to the field of integrated circuits, and particularly to an event-triggered weak biomedical electrical signal detection circuit and a bioelectrical signal device. Background Art

[0002] In the field of biomedical electricity, how to detect weak signals is a current research frontier hotspot. EEG (electroencephalogram) reveals the mechanism behind human behaviors such as thinking, feeling, and movement by recording the electrical activities of brain neurons; ECG (electrocardiogram) records the electrical signals during each contraction and relaxation of the heart by measuring the electrical activities of the heart to assist doctors in diagnosis and evaluation; EMG (electromyogram) helps diagnose and treat neuromuscular diseases by measuring muscle electrical activities. Through the research on weak signal detection, it not only helps medical workers have a deeper understanding of how the human body works and how nerve signals are transmitted, but also assists in the research of disease diagnosis, treatment, and prevention.

[0003] In the field of electronic technology, the detection of bioelectrical signals establishes a direct communication and control channel between the human body and electronic devices, thus better assisting scientists in developing wearable assistive medical devices and bringing new possibilities to the development of human-computer interaction in human society.

[0004] For currently existing related devices for bioelectrical signal detection, the readout chips for bioelectrical signals occupy a large area and consume a large amount of power, may damage tissues, and pose health risks; and the quality of the readout signals is poor, with low linearity, signal-to-noise ratio, and resolution; the amount of data that needs to be processed for bioelectrical signals is large, complex, and variable, which is not conducive to the acquisition and research of large-scale bioelectrical signal data. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide an event-triggered weak biomedical electrical signal detection circuit and a bioelectrical signal device that solve or partially solve the above problems.

[0006] In the first aspect of an embodiment of the present invention, an event-triggered weak biomedical electrical signal detection circuit is provided. The event-triggered weak biomedical electrical signal detection circuit includes: a filtering module, a voltage-to-current module, an ADC module, and a global processing module;

[0007] The filtering module is used to filter the bioelectrical signal from the neural probe to obtain a noise-reduced neural signal and transmit it to the voltage-to-current module;

[0008] The voltage-to-current module is used to convert the noise-reduced neural signal into a corresponding current signal and transmit it to the ADC module;

[0009] The ADC module is configured to generate a pulse signal according to the current signal, and combine the pulse signal and the Gray code clock from the global processing module to generate the time value of the Gray code clock corresponding to any pixel within one sampling period and transmit it to the global processing module;

[0010] The global processing module is configured to process the historical time value of the Gray code clock corresponding to the previous sampling period and the real-time time value of the Gray code clock corresponding to the current sampling period in an event-triggered manner, and output the difference or the comparison result between the historical time value and the real-time time value to achieve data stream compression;

[0011] Among them, the filtering module, the voltage-to-current module, and the ADC module are all arranged in the neural probe array, and the global processing module is arranged outside the chip.

[0012] Optionally, the ADC module includes: an oscillation unit, a frequency division unit, and a sampling and storage unit;

[0013] The oscillation unit is configured to convert the current signal into an oscillation signal and transmit it to the frequency division unit;

[0014] The frequency division unit is configured to divide the oscillation signal, and when the number of oscillations corresponding to the oscillation signal reaches a preset number, send the pulse signal to the sampling and storage unit;

[0015] The sampling and storage unit is configured to use the Gray code clock to count the duration from the start of sampling to receiving the pulse signal within one sampling period, obtain the time value and transmit it to the global processing module.

[0016] Optionally, the oscillation signal is a square wave oscillation signal whose frequency is proportional to the current signal;

[0017] The frequency division unit includes: a k-bit asynchronous counter;

[0018] The preset number is 2 k times.

[0019] Optionally, the sampling and storage unit uses a readout signal and a transmission signal to transmit the historical time value and the real-time time value to the global processing module;

[0020] Among them, the high-level duration of the transmission signal is less than the high-level duration of the readout signal, and the transmission signal changes from low level to high level at the middle position of the high-level duration of the readout signal;

[0021] Read the historical time value before the transmission signal changes from low level to high level, and transmit it to the global processing module;

[0022] Read the real-time time value after the transmission signal changes from high level to low level, and transmit it to the global processing module.

[0023] Optionally, the global processing module includes: a Gray code counter, a control unit, a Gray code to binary unit, a memory, a subtractor, and an event extraction unit;

[0024] The control unit is configured to send control signals to the Gray code counter, the Gray code to binary unit, the subtractor, and the event extraction unit respectively;

[0025] The Gray code counter is configured to send the Gray code clock to the ADC module through a bus according to the control signal sent by the control unit;

[0026] The Gray code to binary unit is configured to convert the historical time value into binary count and transmit it to the memory, and convert the real-time time value into binary count and transmit it to the subtractor;

[0027] The memory is configured to temporarily store the binary count converted from the historical time value, and after the Gray code to binary unit converts the real-time time value into binary count, transmit the binary count converted from the historical time value to the subtractor;

[0028] The subtractor is configured to calculate the difference between the binary count converted from the real-time time value and the binary count converted from the historical time value, and transmit it to the event extraction unit;

[0029] The event extraction unit is configured to output the difference or the comparison result between the historical time value and the real-time time value according to the situation of the difference.

[0030] Optionally, when the historical time value is exactly the same as the real-time time value, the difference is 0, and both the high signal and the low signal output by the event extraction unit are 0;

[0031] When the real-time time value is not exactly the same as the historical time value, if the difference is 1, the high signal output by the event extraction unit is 1 and the low signal is 0;

[0032] When the real-time time value is not exactly the same as the historical time value, if the difference is -1, the high signal output by the event extraction unit is 0 and the low signal is 1;

[0033] When the real-time time value is not exactly the same as the historical time value, if the difference is not 1 or -1, both the high signal and the low signal output by the event extraction unit are 1, and the difference is output.

[0034] Optionally, the event extraction unit includes: a NAND gate, a first AND gate, a second AND gate, a first OR gate, a second OR gate, and a third OR gate;

[0035] The NAND gate, the first OR gate, the second OR gate, and the third OR gate all receive the difference;

[0036] The output end of the NAND gate and the output end of the first OR gate are both connected to the input end of the first AND gate;

[0037] The output end of the first OR gate and the output end of the second OR gate are both connected to the input end of the second AND gate;

[0038] The output end of the first AND gate outputs a high signal;

[0039] The output end of the second AND gate outputs a low signal.

[0040] Optionally, taking a filter module composed of two capacitors and one resistor as an example: the filtered noise-reduced nerve signal V is output at both ends of the resistor S The expression is as follows:

[0041]

[0042] In the above formula, V IN represents the voltage value of the bioelectric signal, V CM represents the common-mode voltage value, s represents s in the frequency characteristic, s = iw, R represents the resistance value of the resistor, and C represents the capacitance value of the capacitor;

[0043] The noise-reduced nerve signal is converted into a corresponding current signal I S The expression is:

[0044] I S = Gm × V S

[0045] In the above formula, Gm represents the transconductance coefficient of the voltage-to-current module;

[0046] According to the magnitude of the current signal I S being different, the oscillation frequency corresponding to the oscillation unit is also different. According to the conservation of charge, there is:

[0047] I S × T S = 2 k × Q u

[0048] In the above formula, Q u represents the amount of charge transferred during one oscillation of the oscillation unit, which is a fixed value. T S represents the time required when the number of oscillations corresponding to the oscillation signal reaches a preset number of times.

[0049] Optionally, the oscillation unit includes: an OSC oscillator based on the time-to-digital principle.

[0050] In a second aspect of the embodiments of the present invention, a bioelectric signal device is provided. The bioelectric signal device includes the event-triggered weak biomedical electric signal detection circuit according to any one of the first aspect.

[0051] The event-triggered weak biomedical electric signal detection circuit provided by the present invention includes: a filtering module, a voltage-to-current module, an ADC module, and a global processing module. The filtering module, the voltage-to-current module, and the ADC module are all arranged in the neural probe array, and the global processing module is arranged outside the chip.

[0052] The filtering module filters the neural signals from the neural probe to obtain noise-reduced neural signals and transmits them to the voltage-to-current module; the voltage-to-current module converts the noise-reduced neural signals into corresponding current signals and transmits them to the ADC module. The ADC module generates pulse signals according to the current signals, and combines the pulse signals and the Gray code clock from the global processing module to generate the time value of the Gray code clock corresponding to any pixel within one sampling period and transmits it to the global processing module.

[0053] The global processing module processes the historical time value of the Gray code clock corresponding to the previous sampling period and the real-time time value of the Gray code clock corresponding to the current sampling period in an event-triggered manner, and outputs the difference or the comparison result between the historical time value and the real-time time value to achieve the compression of the data code stream.

[0054] The filtering module, the voltage-to-current module, and the ADC module on the pixel of the present invention realize the filtering and reading of bioelectric signals, ensure the quality of bioelectric signals while realizing the continuous monitoring of bioelectric data, have a simple structure, better realize the miniaturization of the neural probe array, ensure the minimum tissue damage and tissue displacement during the chip implantation process, and also reduce the power consumption and noise of the system as much as possible. Converting the current signal into a square wave oscillation improves the linearity and realizes accurate and efficient analog-to-digital signal conversion. At the same time, the power consumption and noise of the system are reduced as much as possible during parameter design, and the signal-to-noise ratio is improved.

[0055] A Gray code counter is adopted. In one count, only one bit of the Gray code value will flip, which is half less likely to flip compared to a binary counter. This not only reduces the error rate, improves reliability, but also reduces power consumption. At the same time, an event-triggered method is proposed for signal processing, further reducing the output data bit stream: during global signal processing, specific digital readings are output only when the real-time time value and the historical time value readings differ by more than 1 bit, and in other cases, only the comparison result of the two is output. Therefore, the present invention realizes the compression of the data bit stream, which is beneficial to the acquisition and research of large-scale bioelectrical signal data. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0057] Figure 1 It is a schematic diagram of the overall structure of the event-triggered weak biomedical electrical signal detection circuit according to an embodiment of the present invention;

[0058] Figure 2 It is a schematic diagram of a preferred structure of the event extraction unit in the event-triggered weak biomedical electrical signal detection circuit according to an embodiment of the present invention;

[0059] Figure 3 It is OUT in an embodiment of the present invention <n:1>Schematic structure diagram of an event extraction unit with 6 bits and OUT<6> being the two's complement;

[0060] Figure 4 It is the timing relationship diagram of three pixel points in two sampling periods in the embodiment of the present invention. Detailed implementation manners

[0061] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0062] The event-triggered weak biomedical electrical signal detection circuit proposed by the present invention includes: a filtering module, a voltage-to-current module, an ADC module, and a global processing module. Among them, the filtering module, the voltage-to-current module, and the ADC module are all arranged in the neural probe array. That is, the circuit structure composed of the filtering module, the voltage-to-current module, and the ADC module is the circuit structure within the pixel array. And the global processing module is arranged outside the chip, that is, outside the pixel array.

[0063] In the event-triggered weak biomedical electrical signal detection circuit proposed by the present invention, the filtering module is used to filter the neural signals from the neural probe to obtain noise-reduced neural signals and transmit them to the voltage-to-current module. The filtering module can be composed of any currently known circuit structure or components.

[0064] In the event-triggered weak biomedical electrical signal detection circuit proposed by the present invention, the voltage-to-current module is used to convert the noise-reduced neural signals into corresponding current signals and transmit them to the ADC module; the ADC module is used to generate pulse signals according to the current signals and, in combination with the pulse signals and the Gray code clock from the global processing module, generate the time value of the Gray code clock corresponding to any pixel within one sampling period and transmit it to the global processing module.

[0065] In the event-triggered weak biomedical electrical signal detection circuit proposed by the present invention, the global processing module is used to process the historical time value of the Gray code clock corresponding to the previous sampling period and the real-time time value of the Gray code clock corresponding to the current sampling period in an event-triggered manner, and output the difference or the comparison result between the historical time value and the real-time time value to achieve the compression of the data code stream.

[0066] In a possible embodiment, the ADC module includes: an oscillation unit, a frequency division unit, and a sampling and storage unit; the oscillation unit is configured to convert a current signal into an oscillation signal and transmit it to the frequency division unit; preferably, the oscillation signal is a square wave oscillation signal whose frequency is proportional to the current signal; and a preferred way to implement this square wave oscillation signal is to use an OSC oscillator based on the time-to-digital conversion (TDC) principle.

[0067] The frequency division unit is configured to divide the frequency of the oscillation signal and, when the number of oscillations corresponding to the oscillation signal reaches a preset number of times, send a pulse signal to the sampling and storage unit; a preferred choice for the frequency division unit is to use a k-bit asynchronous counter to implement the frequency division function, and the preset number of times is 2 k times. That is, when the number of oscillations corresponding to the oscillation signal reaches 2 k times, a pulse signal is sent to the sampling and storage unit.

[0068] The sampling and storage unit is configured to use a Gray code clock to count the duration from the start of sampling to the reception of the pulse signal within a sampling period, obtain a time value, and transmit it to the global processing module. In a specific implementation, a preferred method is: the sampling and storage unit uses a read signal and a transmission signal to transmit the historical time value and the real-time time value to the global processing module; where the high-level duration of the transmission signal is less than the high-level duration of the read signal, and the transmission signal changes from low level to high level at the middle position of the high-level duration of the read signal; read the historical time value before the transmission signal changes from low level to high level and transmit it to the global processing module; read the real-time time value after the transmission signal changes from high level to low level and transmit it to the global processing module.

[0069] Taking the example that the neural probe array is composed of a total of M pixel points: At each pixel point, the bioelectric signal acquisition electrode inputs the acquired bioelectric signal into the filtering module; the filtering module filters the bioelectric signal to obtain a noise-reduced neural signal; the voltage-to-current module linearly converts the noise-reduced bioelectric signal into a current signal.

[0070] The OSC oscillation unit converts the current into a square wave oscillation whose frequency is proportional to the input current, and then the frequency division unit outputs a pulse signal when the square wave oscillation reaches a fixed number of times, which is 2 k times; the sampling and storage unit records the duration on the bus when the pulse signal is received, and this duration is essentially a Gray code clock value, which can be represented by GRAY <n:1>It is represented by. When the readout signal of each pixel arrives, that is, within the high-level duration of the readout signal, the historical time value is read out and transmitted to the global processing module before transmitting the signal, and the real-time time value is read out and transmitted to the global processing module after transmitting the signal. The transmission signal is a pulse signal, and its high-level duration is much shorter than the high-level duration of the readout signal.

[0071] Bioelectric signals generally first pass through a high-pass filter (HPF) to filter out the unwanted low-frequency stimulus artifacts in the bioelectric signals, obtaining noise-reduced neural signals. Since signals less than 1 Hz need to be filtered out, the resistance value in the high-pass filter HPF needs to reach the order of hundreds of megohms. Generally, a pseudo-resistor is used to replace the resistor to reduce the area. Taking a filter module composed of two capacitors and one resistor as an example: The filtered noise-reduced neural signal V is output at both ends of the resistor. S The expression is as follows:

[0072]

[0073] In the above formula, V IN represents the voltage value of the bioelectric signal, V CM represents the common-mode voltage value, s represents s in the frequency characteristic, s = iw, R represents the resistance value of the resistor, and C represents the capacitance value of the capacitor.

[0074] The noise-reduced neural signal V S Then, it passes through a voltage-to-current module with a transconductance coefficient of Gm to linearly convert the voltage signal into a current signal I S The expression is:

[0075] I S = Gm × V S

[0076] The current signal I S is the current value transmitted to the oscillation unit. The current is converted into a square-wave oscillation whose frequency is proportional to the input current, improving the linearity and achieving accurate and efficient analog-to-digital signal conversion.

[0077] According to the different magnitudes of the current signal I S , the oscillation frequency of the corresponding oscillation unit is also different. According to the conservation of charge, there is:

[0078] I S × T S = 2 k × Q u

[0079] In the above formula, Q u represents the amount of charge transferred during one oscillation of the oscillation unit, which is a fixed value, and T S represents the time required when the number of oscillations corresponding to the oscillation signal reaches a preset number of times.

[0080] The frequency division unit performs frequency division and, when the number of oscillations corresponding to the oscillation signal reaches a preset number, sends a pulse signal to the sampling and storage unit. The sampling and storage unit is used to count the duration from the start of sampling to the reception of the pulse signal within one sampling period using the Gray code clock to obtain a time value, i.e., the Gray code value GRAY <n:1>, when the readout signal of each pixel arrives, the historical value of the recorded Gray code clock is read out before transmitting the signal, and the real-time value of the Gray code clock is read out after transmitting the signal. The reason for using a Gray code counter is that only one bit of the Gray code value flips during each count, which is half less likely to flip compared to a binary counter. This not only reduces the error rate, improves reliability, but also reduces power consumption.

[0081] In a possible embodiment, the global processing module includes: a Gray code counter, a control unit, a Gray code to binary conversion unit, a memory, a subtractor, and an event extraction unit.

[0082] The control unit is configured to send control signals to the Gray code counter, the Gray code to binary conversion unit, the subtractor, and the event extraction unit respectively. The Gray code counter is configured to send the Gray code clock to the ADC module via a bus according to the control signal sent by the control unit;

[0083] The Gray code to binary conversion unit is configured to convert the historical time value into a binary count and transmit it to the memory, and convert the real-time time value into a binary count and transmit it to the subtractor.

[0084] As described above, the transmission signal is a pulse signal. The historical time value is read out and transmitted to the global processing module before this transmission signal, and the real-time time value is read out and transmitted to the memory of the global processing module after the transmission signal. Therefore, the historical time value is converted into a binary count by the Gray code to binary conversion unit before the real-time time value. For the subsequent operation of the subtractor, a memory needs to be set up. This memory is used to temporarily store the binary count converted from the historical time value, and after the Gray code to binary conversion unit converts the real-time time value into a binary count, it transmits the binary count converted from the historical time value to the subtractor.

[0085] Naturally, it can be understood that the subtractor is used to obtain the difference between the binary count converted from the real-time time value and the binary count converted from the historical time value, that is, the binary count converted from the real-time time value - the binary count converted from the historical time value, and transmit the obtained difference to the event extraction unit.

[0086] The event extraction unit is configured to output the difference or the comparison result between the historical time value and the real-time time value according to the situation of the difference obtained by the subtractor. In a preferred setting, when the historical time value is exactly the same as the real-time time value, the difference between the two is naturally 0, and both the high signal and the low signal output by the event extraction unit are 0.

[0087] When the real-time time value and the historical time value are not exactly the same, if the difference between the two is 1, the high signal output by the event extraction unit is 1 and the low signal is 0; if the difference between the two is -1, the high signal output by the event extraction unit is 0 and the low signal is 1. In this case, regardless of whether the difference between the two is 1 or -1, it is essentially that the binary count corresponding to the real-time time value differs from the binary count corresponding to the historical time value by 1 bit. Generally, it is considered that for the three cases where the difference is 0, 1, or -1, the corresponding bioelectric signal remains basically unchanged or undergoes a small mutation.

[0088] When the real-time time value and the historical time value are not exactly the same, if the difference between the two is not 1 or -1, both the high signal and the low signal output by the event extraction unit are 1, and the difference is also output. In such cases, it is generally considered that the corresponding bioelectric signal undergoes a large mutation.

[0089] To better explain and illustrate the event-triggered weak biomedical electrical signal detection circuit proposed by the present invention, refer to Figure 1 , which exemplarily shows the overall structural schematic diagram of the event-triggered weak biomedical electrical signal detection circuit according to the embodiment of the present invention. Figure 1 In [the reference], the filtering module is exemplarily represented by two capacitors C1, C2 and a resistor R, which filters the bioelectric signal V IN and the common-mode voltage signal V CM to obtain the noise-reduced nerve signal V S .

[0090] Figure 1 In [the reference], the voltage-to-current module is exemplarily represented by Gm, and OSC and frequency division represent the oscillation unit and the frequency division unit. The current signal I S converted by the voltage-to-current module Gm is transmitted to OSC and frequency division. The oscillation unit converts the current signal I S into an oscillation signal and transmits it to the frequency division unit. The frequency division unit divides the oscillation signal, and when the number of oscillations corresponding to the oscillation signal reaches a preset number, it sends a pulse signal SH to the sampling and storage unit ( Figure 1 exemplarily represented by storage sampling in [the reference]). The sampling and storage unit is used to count the duration from the start of sampling to the reception of the pulse signal within one sampling period by using the Gray code clock (from the Gray code counter) to obtain a time value, that is, the Gray code value GRAY <n:1>, when the read signal of each pixel arrives, before transmitting the signal, the historical value of the recorded Gray code clock is read, and after transmitting the signal, the real-time value of the Gray code clock is read, and both are transmitted to the Gray code to binary unit ( Figure 1 exemplarily represented in Gray code to binary in

[0091] The control unit respectively sends control signals to the Gray code counter, the Gray code to binary, the subtractor, and the event extraction unit ( Figure 1 exemplarily represented by event extraction in

[0092] The Gray code to binary unit converts the historical time value into binary count and transmits it to the memory. The Gray code to binary unit converts the real-time time value into binary count and transmits it to the subtractor.

[0093] The subtractor calculates the difference between the binary count converted from the real-time time value and the binary count converted from the historical time value, that is, the binary count converted from the real-time time value - the binary count converted from the historical time value, and transmits the obtained difference to the event extraction unit.

[0094] The event extraction unit is used to output the real-time time value or the comparison result between the historical time value and the real-time time value according to the difference obtained by the subtractor. When the historical time value and the real-time time value are exactly the same, the difference between the two is naturally 0, and the high signal ( Figure 1 exemplarily represented by UP in Figure 1 exemplarily represented by DOWN in

[0095] are both 0.

[0095] If the difference between the real-time time value and the historical time value is 1, the event extraction unit outputs: UP = 1, DOWN = 0; if the difference between the real-time time value and the historical time value is -1, the event extraction unit outputs: UP = 0, DOWN = 1. In this case, regardless of whether the difference between the two is 1 or -1, essentially, the binary count corresponding to the real-time time value and the binary count corresponding to the historical time value differ by 1 bit. Generally, it is considered that there are three types of situations for the difference of 0, 1, and -1, and the corresponding bioelectric signals remain basically unchanged or mutate slightly.

[0096] If the difference between the real-time time value and the historical time value is not 1 or -1, the event extraction unit outputs: UP = DOWN = 1, and outputs the difference ( Figure 1 exemplarily represented by OUT in <n:1>(representation). In such cases, it is generally considered that the corresponding bioelectrical signal has a large mutation.

[0097] In the event-triggered weak biomedical electrical signal detection circuit proposed by the present invention, a preferred structure of the event extraction unit is referred to Figure 2 as shown. The event extraction unit includes: a NAND gate nand, a first AND gate and1, a second AND gate and2, a first OR gate or1, a second OR gate or2, and a third OR gate or3;

[0098] The NAND gate nand, the first OR gate or1, the second OR gate or2, and the third OR gate or3 all receive the difference OUT between the real-time time value and the historical time value <n:1>The output terminal of the NAND gate nand and the output terminal of the first OR gate or1 are both connected to the input terminal of the first AND gate and1.

[0099] The output terminal of the second OR gate or2 and the output terminal of the third OR gate or3 are both connected to the input terminal of the second AND gate and2; the output terminal of the first AND gate and1 outputs a high signal UP; the output terminal of the second AND gate and2 outputs a low signal DOWN.

[0100] It should be noted that since the difference is not 0 or 1 or -1, the event extraction unit outputs: UP = DOWN = 1, and also needs to output the difference OUT <n:1>, so the event extraction unit also has a direct output loop ( Figure 2 not shown in <n:1>In addition, when the difference OUT <n:1>When the number of bits is large, the NAND gate nand, the first OR gate or1, the second OR gate or2, and the third OR gate or3 that receive the difference may not necessarily meet the signal input requirements, so it is necessary to expand the logic gates at the input end only. For example Figure 3 shown with OUT <n:1>Taking 6 bits as an example, and OUT<6> being the two's complement, in the logic combination of the output high signal UP, two AND gates are expanded to receive 6 bits, and two OR gates are expanded to receive 6 bits. In the logic combination of the output low signal DOWN, four OR gates are expanded to receive 6 bits respectively. In other cases, those skilled in the art can obtain them through simple reasoning.

[0101] Refer to Figure 4 In the timing relationship diagram of two sampling periods of three pixel points in the embodiment of the present invention shown, it is assumed that different pixels are sampled with a period of T. M different pixels are sampled with a period of T, and the sampling period of each pixel is T / M. The sampling of different pixels is performed by the sampling signal RS to control. During the sampling period RS In the middle is the transmission signal TR , for realizing the reading of historical values and real-time values. When the sampling signal RS of each pixel arrives, at the transmission signal TR Read the previous recorded Gray code clock history value during the transmission signal TR Read out the real-time value later.

[0102] The reset period for each pixel is also T / M, and reset is performed immediately after the sampling window ends. The reset of different pixels is controlled by the reset signal EN. to control, the i-th pixel is at EN Reset when = 0, at EN Works properly when =1 until the next sampling signal RS Temporary sampling.

[0103] The reset sampling clock module in the global processing ( Figure 1 not shown in the figure) transmits the reset signal EN<0> and the sampling signal RS<0> to the first pixel point respectively, and then connects the timing control modules of each pixel point through the control clocks CK_EN and CK_RS on the bus. Both CK_EN and CK_RS have a period of T / M and are sequentially transmitted to the subsequent pixel points.

[0104] Combined with Figure 4 , in the first pixel point, the historical time value in the first period is Ts1(1), and the real-time time value in the second period is Ts2(1). If Ts2(1) - Ts1(1) > δT, then up = 1 and down = 0 are output.

[0105] In the second pixel point, the historical time value in the first period is Ts1(2), and the real-time time value in the second period is Ts2(2). If Ts2(2) – Ts1(2) < -δT, then up = 0 and down = 1 are output.

[0106] In the third pixel point, the historical time value in the first period is Ts1(3), and the real-time time value in the second period is Ts2(3). If |Ts2(3) - Ts1(3)| < δT, then up = 0 and down = 0 are output.

[0107] Taking the sampling frequency of the chip as 40KHz, the neural probe array has 256 pixel points, and each pixel point samples and records 12bit once as an example. If the existing related equipment for detecting bioelectric signals is used, the data transmission rate required by its data stream processing method is:

[0108] 256 × 40K × 12bit = 122.9Mbps

[0109] Since bioelectric signals remain unchanged most of the time, mutate slightly for a small part of the time, and mutate greatly for an extremely small part of the time. Assuming that the bioelectric signal remains unchanged in 90% of the cases, changes by 1bit in 9% of the cases, then using the method of only outputting the changed values of the event-triggered weak bio-medical electrical signal detection circuit proposed in the present invention, only 2bit data streams of UP and DOWN are output in the above two cases, and only 1% of the signals mutate greatly, and an additional 12bit data stream is output for this mutation. Then the data transmission rate is:

[0110] 256 × 40K × (2bit + 1% × 12bit) = 21.7Mbps

[0111] Compared with the transmission rate of 122.9 Mbps, the event-triggered weak biomedical electrical signal detection circuit proposed by the present invention is equivalent to compressing the data bit stream to 18% of the original. It is extremely beneficial for the acquisition and research of large-scale biomedical electrical signal data.

[0112] Based on the above event-triggered weak biomedical electrical signal detection circuit, an embodiment of the present invention further proposes a biomedical electrical signal device, and the biomedical electrical signal device includes the event-triggered weak biomedical electrical signal detection circuit described in any one of the above.

[0113] Through the above examples, the event-triggered weak biomedical electrical signal detection circuit provided by the present invention includes: a filtering module, a voltage-to-current module, an ADC module, and a global processing module. The filtering module, the voltage-to-current module, and the ADC module are all arranged in the neural probe array, and the global processing module is arranged outside the chip.

[0114] The filtering module filters the neural signals from the neural probe to obtain noise-reduced neural signals and transmits them to the voltage-to-current module; the voltage-to-current module converts the noise-reduced neural signals into corresponding current signals and transmits them to the ADC module. The ADC module generates pulse signals according to the current signals, and combines the pulse signals and the Gray code clock from the global processing module to generate the time value of the Gray code clock corresponding to any pixel in one sampling period and transmits it to the global processing module.

[0115] The global processing module processes the historical time value of the Gray code clock corresponding to the previous sampling period and the real-time time value of the Gray code clock corresponding to the current sampling period in an event-triggered manner, and outputs the difference or the comparison result of the historical time value and the real-time time value to achieve the compression of the data bit stream.

[0116] The filtering module, voltage-to-current module, and ADC module on the pixel of the present invention realize the filtering and reading of biomedical electrical signals, ensure the quality of biomedical electrical signals while realizing the continuous monitoring of biomedical data, have a simple structure, better realize the miniaturization of the neural probe array, ensure the minimum tissue damage and tissue displacement during the chip implantation process, and at the same time reduce the power consumption and noise of the system as much as possible. Converting the current signal into a square wave oscillation improves the linearity and realizes accurate and efficient analog-to-digital signal conversion. At the same time, the power consumption and noise of the system are reduced as much as possible during parameter design, and the signal-to-noise ratio is improved.

[0117] A Gray code counter is adopted. In one count, only one bit of the Gray code value will flip, which is half less likely to flip than a binary counter. This not only reduces the error rate, improves reliability, but also reduces power consumption. At the same time, an event-triggered method is proposed for signal processing, further reducing the output data bit stream: during global signal processing, specific digital readings are only output when the difference between the current real-time value and the historical value readings is more than 1 bit, and only the comparison result of the two is output in other cases. Therefore, the present invention realizes the compression of the data bit stream, which is beneficial to the acquisition and research of large-scale bioelectrical signal data.

[0118] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, article or device including the element.

[0119] The embodiments of the present invention have been described above in conjunction with the accompanying drawings, but the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive.

[0120]

[0121] Under the circumstances of the protected scope, many forms can still be made, and these all fall within the protection scope of the present invention.

Claims

1. An event-triggered weak biomedical electrical signal detection circuit, characterized in that The event-triggered weak biomedical electrical signal detection circuit includes: a filtering module, a voltage-to-current module, an ADC module, and a global processing module; The filtering module is used to filter the biomedical electrical signal from the neural probe to obtain a noise-reduced neural signal and transmit it to the voltage-to-current module; The voltage-to-current module is used to convert the noise-reduced neural signal into a corresponding current signal and transmit it to the ADC module; The ADC module is used to generate a pulse signal according to the current signal, and combine the pulse signal and the Gray code clock from the global processing module to generate the time value of the Gray code clock corresponding to any pixel in one sampling period and transmit it to the global processing module; The global processing module is used to process the historical time value and the real-time time value of the Gray code clock corresponding to the previous sampling period and the current sampling period in an event-triggered manner, and output the difference between the binary counts converted from the real-time time value and the historical time value or output the comparison result between the historical time value and the real-time time value to achieve data stream compression; Among them, the filtering module, the voltage-to-current module, and the ADC module are all arranged in the neural probe array, and the global processing module is arranged outside the chip; The global processing module includes: a Gray code counter, a control unit, a Gray code-to-binary unit, a memory, a subtractor, and an event extraction unit; The control unit is used to send control signals to the Gray code counter, the Gray code-to-binary unit, the subtractor, and the event extraction unit respectively; The Gray code counter is used to send the Gray code clock to the ADC module through the bus according to the control signal sent by the control unit; The Gray code-to-binary unit is used to convert the historical time value into a binary count and transmit it to the memory, and convert the real-time time value into a binary count and transmit it to the subtractor; The memory is used to temporarily store the binary count converted from the historical time value, and after the Gray code-to-binary unit converts the real-time time value into a binary count, transmit the binary count converted from the historical time value to the subtractor; The subtractor is used to calculate the difference between the binary count converted from the real-time time value and the binary count converted from the historical time value and transmit it to the event extraction unit; The event extraction unit is used to output the difference or the comparison result between the historical time value and the real-time time value according to the situation of the difference.

2. The event-triggered weak biomedical electrical signal detection circuit according to claim 1, wherein The ADC module includes: an oscillation unit, a frequency division unit, and a sampling and storage unit; The oscillation unit is used to convert the current signal into an oscillation signal and transmit it to the frequency division unit; The frequency division unit is used to divide the frequency of the oscillation signal, and when the number of oscillations corresponding to the oscillation signal reaches a preset number, send the pulse signal to the sampling and storage unit; The sampling and storage unit is configured to use the Gray code clock to count the duration from the start of sampling to the reception of the pulse signal within one sampling period, obtain the time value, and transmit it to the global processing module.

3. The event-triggered weak biomedical electrical signal detection circuit according to claim 2, characterized in that, The oscillation signal is a square wave oscillation signal whose frequency is proportional to the current signal; The frequency division unit includes: a k-bit asynchronous counter; The preset number of times is 2 k times 4. The event-triggered weak biomedical electrical signal detection circuit according to claim 2, characterized in that, The sampling and storage unit uses the read signal and the transmission signal to transmit the historical time value and the real-time time value to the global processing module; Wherein, the high-level duration of the transmission signal is less than the high-level duration of the read signal, and the transmission signal changes from low level to high level at the middle position of the high-level duration of the read signal; Read the historical time value before the transmission signal changes from low level to high level, and transmit it to the global processing module; Read the real-time time value after the transmission signal changes from high level to low level, and transmit it to the global processing module.

5. The event-triggered weak biomedical electrical signal detection circuit according to claim 1, wherein When the historical time value is exactly the same as the real-time time value, the difference is 0, and both the high signal and the low signal output by the event extraction unit are 0; When the real-time time value is not exactly the same as the historical time value, if the difference is 1, the high signal output by the event extraction unit is 1 and the low signal is 0; When the real-time time value is not exactly the same as the historical time value, if the difference is -1, the high signal output by the event extraction unit is 0 and the low signal is 1; When the real-time time value is not exactly the same as the historical time value, if the difference is not 1 or -1, both the high signal and the low signal output by the event extraction unit are 1, and the difference is output.

6. The event-triggered weak biomedical electrical signal detection circuit according to claim 1, characterized in that, The event extraction unit includes: a NAND gate, a first AND gate, a second AND gate, a first OR gate, a second OR gate, and a third OR gate; The NAND gate, the first OR gate, the second OR gate, and the third OR gate all receive the difference; The output terminal of the NAND gate and the output terminal of the first OR gate are both connected to the input terminal of the first AND gate; The output terminal of the second OR gate and the output terminal of the third OR gate are both connected to the input terminal of the second AND gate; The output terminal of the first AND gate outputs a high signal; The output terminal of the second AND gate outputs a low signal.

7. The event-triggered weak biomedical electrical signal detection circuit according to claim 2, wherein When a filtering module is composed of two capacitors and a resistor: the filtered noise-reduced neural signal V is output across the resistor S The expression of which is as follows: In the above formula, V IN represents the voltage value of the bioelectric signal, V CM represents the common-mode voltage value, s represents s in the frequency characteristic, s = iw, R represents the resistance value of the resistor, and C represents the capacitance value of the capacitor; The noise-reduced neural signal is converted into a corresponding current signal I S The expression is as follows: I S =Gm×V S In the above formula, Gm represents the transconductance coefficient of the voltage-to-current module; According to the current signal I S With different magnitudes, the oscillation frequencies corresponding to the oscillation unit are also different. According to the law of conservation of charge, we have: I S × T S =2 k × Q u In the above formula, Q u represents the amount of charge transferred during one oscillation of the oscillation unit, which is a fixed value. T S represents the time required when the number of oscillations corresponding to the oscillation signal reaches a preset number of times.

8. The event-triggered weak biomedical electrical signal detection circuit according to claim 2, characterized in that, The oscillation unit includes: an OSC oscillator based on the time-to-digital principle.

9. A bioelectric signal device, characterized in that, The bioelectric signal device includes the event-triggered weak bio-medical electric signal detection circuit according to any one of claims 1-8.

Images