Multi-channel electroencephalogram acquisition and electrical stimulation multiplexing electrode circuit and control method thereof

Abstract

This invention discloses a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit and its control method. The circuit comprises a Bluetooth module, a multi-channel switch multiplexing module, an EEG acquisition module, a programmable constant current electrical stimulation module, and a timing logic control module that controls the operation of each module. The Bluetooth module receives instructions and coordinates communication with the multi-channel switch multiplexing module and the EEG acquisition module, which acquire signals through electrodes and transmit the data back to the Bluetooth module for analysis or transmission. The Bluetooth module is also connected to the programmable constant current electrical stimulation module to provide precise electrical stimulation. The timing logic control module ensures that EEG acquisition and electrical stimulation do not occur simultaneously, avoiding mutual interference. The multi-channel switch multiplexing module switches its operating mode according to Bluetooth instructions. By implementing the circuit of this invention, the overall integration and channel density of the system are improved, thereby reducing device size and energy consumption, meeting the needs of wearable technology and high-density array applications.

Description

Technical Field

[0001] This invention relates to the field of biomedical signal detection technology, and more specifically to a multi-channel electroencephalogram (EEG) acquisition and electrical stimulation multiplexing electrode circuit and its control method. Background Technology

[0002] Currently, EEG acquisition and transcranial or neurostimulation devices are mainly divided into two categories. The first category employs a completely separate acquisition and stimulation channel design, i.e., two independent electrode and circuit systems. This design leads to problems such as large device size, complex wiring, and bulky head-mounted devices, greatly affecting user comfort and portability. The second category is a simple multiplexing scheme, using a common switch to switch between acquisition and stimulation functions. However, this method has significant technical drawbacks, such as high-voltage crosstalk causing damage to the acquisition front end, severe stimulation artifacts affecting signal quality, poor security, and low integration.

[0003] Within the existing technological framework, the same electrode cannot be efficiently reused simultaneously or in a time-division manner for EEG signal acquisition and electrical stimulation output, which greatly limits channel utilization. Furthermore, due to the lack of effective safety isolation measures, the high voltage during stimulation can easily damage the acquisition chip. Moreover, stimulation artifacts severely interfere with the authenticity and reliability of the signal, making it extremely difficult to achieve closed-loop real-time control. In addition, the large number of discrete components used in the system not only increases the overall size and power consumption of the device but also hinders the development of compact wearable devices required for high-density array applications.

[0004] Therefore, it is necessary to design a new circuit that enables efficient multiplexing of a single electrode to support EEG acquisition and electrical stimulation output; provides a high-voltage safety isolation protection mechanism to ensure the safety of the front-end acquisition circuit; significantly reduces the impact of stimulation artifacts, promoting the realization of a closed-loop acquisition-stimulation system; and improves the overall integration and channel density of the system, thereby reducing device size and energy consumption to meet the needs of wearable technology and high-density array applications. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit and its control method.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, comprising: a Bluetooth module, a multiplexed switch module, an EEG acquisition module, a programmable constant current electrical stimulation module, and a timing logic control module; the Bluetooth module and the programmable constant current electrical stimulation module are respectively connected to the multiplexed switch module; the multiplexed switch module is connected to the electrodes; the EEG acquisition module is connected to the multiplexed switch module and the electrodes; the EEG acquisition module is connected to the Bluetooth module; and the timing logic control module is connected to the Bluetooth module.

[0007] The system receives instructions via Bluetooth, and the timing logic control module coordinates the operation of each module to ensure that the EEG acquisition module and the programmable constant current stimulation module do not operate simultaneously. The multiplexer switches the working mode connected to the electrodes according to the instructions from the Bluetooth module to perform data transmission and electrical stimulation applications. The EEG acquisition module processes the EEG signals and transmits the data back to the Bluetooth module for analysis or transmission.

[0008] The further technical solution is as follows: the EEG acquisition module is connected to the Bluetooth module via an SPI bus.

[0009] The further technical solution is as follows: the Bluetooth module includes a Bluetooth communication chip.

[0010] The further technical solution is as follows: the multiplexer module includes a TI analog switching chip; the TI analog switching chip integrates several independently controlled single-pole double-throw switches.

[0011] The further technical solution is as follows: the EEG acquisition module includes an EEG acquisition chip, which integrates a multi-channel electrophysiological amplifier array, a 16-bit ADC, and an SPI interface; the SPI interface is connected to the SPI bus.

[0012] The further technical solution is as follows: the programmable constant current stimulation module includes a programmable constant current stimulation chip, which integrates multiple constant current sources and controls the current of each channel through an 8-bit DAC. The further technical solution is as follows: the programmable constant current electrical stimulation chip adopts a four-wire SPI interface, which is configured by the Bluetooth module or the timing logic control module by reading and writing multiple registers corresponding to each channel.

[0013] The further technical solution is as follows: In the acquisition mode, the electrode of the timing logic control module is connected to the multiplexer module through the multiplexer module, and the stimulation circuit is turned off and isolated; in the stimulation mode, the electrode is connected to the programmable constant current stimulation module through the multiplexer module, and the multiplexer module is set to a high-impedance state; in the closed-loop mode, stimulation is performed during the stimulation pulse output period, and the acquisition mode is switched back during the pulse interval.

[0014] This invention also provides a control method for a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, the method using the aforementioned multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, comprising: The Bluetooth module receives instructions, and the timing logic control module coordinates the work of each module to ensure that the EEG acquisition module and the programmable constant current stimulation module do not operate simultaneously. The multiplexer switches the working mode connected to the electrodes according to the instructions from the Bluetooth module to perform data transmission and electrical stimulation applications. The EEG acquisition module processes the EEG signals and transmits the data back to the Bluetooth module for analysis or transmission.

[0015] Its further technical solutions include: In acquisition mode, weak EEG signals are processed, amplified, and filtered using the highly integrated RHD2132 chip; in stimulation mode, the SX-32 chip provides bipolar constant current stimulation, and high-voltage isolation technology is used for signal isolation.

[0016] The advantages of this invention compared to existing technologies are as follows: By integrating a Bluetooth module, a multiplexed switch module, an EEG acquisition module, a programmable constant current electrical stimulation module, and a timing logic control module, this invention achieves efficient multiplexing of a single electrode to support EEG signal acquisition and electrical stimulation output. Specifically, the Bluetooth module receives external commands and transmits them to the timing logic control module, which coordinates the work of each module to ensure that EEG acquisition and electrical stimulation do not occur simultaneously, avoiding the risks of short circuits and crosstalk. The multiplexed switch module switches the operating mode connected to the electrode according to the received commands, thereby seamlessly switching between data transmission and electrical stimulation applications. The high-voltage safety isolation protection mechanism effectively prevents damage to the front-end acquisition circuit from the stimulation voltage and reduces the impact of stimulation artifacts. The entire system is compactly designed, improving overall integration and channel density, reducing device size and energy consumption, and enhancing the system's adaptability and flexibility. It is particularly suitable for wearable technology and high-density arrays, promoting the realization of closed-loop acquisition-stimulation systems.

[0017] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic block diagram of a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit provided in an embodiment of the present invention; Figure 2 A specific circuit diagram of the Bluetooth module provided in the embodiments of the present invention; Figure 3 The specific circuit diagram of the multi-channel switch multiplexing module provided in the embodiments of the present invention; Figure 4 This is a detailed circuit diagram of the EEG acquisition module provided in an embodiment of the present invention; Figure 5 A detailed circuit diagram of the programmable constant current electrical stimulation module provided in this embodiment of the invention; Figure 6 A detailed circuit diagram of the interface chip provided in the embodiments of the present invention; Figure 7 This is a flowchart illustrating the control method for a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit provided in an embodiment of the present invention. Figure 8 This is a schematic diagram of a single-channel multiplexing circuit provided in an embodiment of the present invention; Figure 9 This is a timing waveform diagram of the stimulation and acquisition switching provided in an embodiment of the present invention; Explanation of the markings in the image: 10. Bluetooth module; 20. Multiplexer module; 30. EEG acquisition module; 40. Programmable constant current stimulation module; 50. Timing logic control module. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0022] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0024] Current EEG acquisition and transcranial or neurostimulation devices are mainly divided into two categories: the first category uses a completely separate acquisition and stimulation channel design, resulting in large device size, complex wiring, and poor comfort; the second category uses ordinary switching functions, but suffers from high-voltage crosstalk, severe stimulation artifacts, and low safety. Neither of these methods can effectively achieve efficient simultaneous or time-division multiplexing of the same electrode for signal acquisition and electrical stimulation output, limiting channel utilization. Furthermore, the lack of safety isolation measures makes the acquisition chip susceptible to damage, affecting signal accuracy and system reliability. Additionally, the numerous discrete components increase size and power consumption, hindering the development of compact wearable devices.

[0025] To this end, embodiments of the present invention provide a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, enabling efficient multiplexing of a single electrode to support EEG acquisition and electrical stimulation output; providing a high-voltage safety isolation protection mechanism to ensure the safety of the front-end acquisition circuit; significantly reducing the impact of stimulation artifacts and promoting the realization of a closed-loop acquisition-stimulation system; and improving the overall integration and channel density of the system, thereby reducing the size and energy consumption of the device and meeting the needs of wearable technology and high-density array applications.

[0026] Specifically, this circuit integrates a Bluetooth module 10, a multiplexed switch module 20, an EEG acquisition module 30, a programmable constant current electrical stimulation module 40, and a timing logic control module 50, achieving efficient multiplexing of a single electrode to support both EEG signal acquisition and electrical stimulation output. The timing logic control module 50 coordinates the operation of each module, ensuring that acquisition and stimulation operations do not occur simultaneously, thus avoiding damage to the front-end acquisition circuitry from high-voltage crosstalk. The multiplexed switch module 20 switches the electrode's operating mode according to instructions, and combined with high-voltage isolation technology, effectively reduces the impact of stimulation artifacts, ensuring signal authenticity and system security. Furthermore, by employing highly integrated chips, the overall integration and channel density of the system are improved, reducing device size and power consumption, meeting the needs of compact wearable technology and high-density array applications, and promoting the realization of a closed-loop acquisition-stimulation system.

[0027] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0028] Please see Figure 1 A multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit includes: a Bluetooth module 10, a multiplexed switch module 20, an EEG acquisition module 30, a programmable constant current electrical stimulation module 40, and a timing logic control module 50; the Bluetooth module 10 and the programmable constant current electrical stimulation module 40 are respectively connected to the multiplexed switch module 20; the multiplexed switch module 20 is connected to the electrodes; the EEG acquisition module 30 is connected to the multiplexed switch module 20 and the electrodes; the EEG acquisition module 30 is connected to the Bluetooth module 10; and the timing logic control module 50 is connected to the Bluetooth module 10. The timing logic control module 50 coordinates the operation of each module by receiving instructions through the Bluetooth module 10, ensuring that the EEG acquisition module 30 and the programmable constant current stimulation module 40 do not operate simultaneously. The multiplexer module 20 switches the working mode connected to the electrodes according to the instructions of the Bluetooth module 10 to perform data transmission and electrical stimulation application. The EEG acquisition module 30 processes the EEG signals and transmits the data back to the Bluetooth module 10 for analysis or transmission.

[0029] In this embodiment, the core of the entire circuit lies in the efficient reuse of a single electrode through the collaborative work of a series of modular components, enabling both the acquisition of EEG signals and the output of electrical stimulation. The following is a detailed description of each component and its collaborative mechanism: Bluetooth Module 10: As the core communication hub of the system, it supports wireless data transmission using the BLE 5.0 protocol. It is responsible not only for receiving external commands and sending processed EEG signals, but also for managing the system's power domain enable, sensor data acquisition, and other functions.

[0030] Multiplexing Module 20: This module is key to achieving single-electrode multiplexing. It can dynamically switch between acquisition and stimulation modes, ensuring that only one operation is activated at any given time, thus avoiding potential damage to the front-end acquisition circuitry from high-voltage stimulation. Furthermore, this module effectively isolates interference between different channels, improving signal quality.

[0031] EEG Acquisition Module 30: Used to amplify, filter, and digitize weak EEG signals. This module is equipped with a high-performance electrophysiological amplifier array and a 16-bit ADC, ensuring signal accuracy and stability. It also supports real-time impedance measurement, contributing to improved safety during use.

[0032] Programmable Constant Current Stimulation Module 40: Designed to generate safe and precise bipolar constant current stimulation pulses. Each current source can be independently configured via an 8-bit DAC to meet different experimental needs. Manufactured using high-voltage technology, this module exhibits excellent compliance and adaptability, suitable for various types of stimulation electrodes.

[0033] The timing logic control module 50 coordinates the workflows of the various modules mentioned above, ensuring that the acquisition and stimulation processes do not overlap. Specifically, during the acquisition phase, it connects electrodes to the EEG acquisition module 30; while during the stimulation phase, it switches to the electrical stimulation module and places the acquisition front end in a high-resistance state to prevent artifacts.

[0034] In summary, the entire circuit significantly improves the overall integration and channel density of the system, reduces device size and power consumption, and also enhances security and signal quality, making it very suitable for application in cutting-edge fields such as wearable brain-computer interfaces and neurorehabilitation.

[0035] In one embodiment, the EEG acquisition module 30 described above is connected to the Bluetooth module 10 via an SPI bus.

[0036] In one embodiment, please refer to Figure 2 The aforementioned Bluetooth module 10 includes a Bluetooth communication chip.

[0037] In this embodiment, the Bluetooth communication chip is the nRF52832, a highly integrated Bluetooth Low Energy (BLE) master control chip. It integrates a 32-bit ARM Cortex-M4F processor, a BLE RF transceiver, various peripheral interfaces, and rich power management functions, serving as the core control and wireless communication unit of the entire system. Figure 2 As shown, U1 connects to multiple peripheral circuits through its pins to realize complete wireless communication and system control functions.

[0038] For power supply, the chip's VDD and VSS pins are connected to a 3.3V power supply (VDD_battery) and ground, respectively. A stable clock source is formed by capacitors C6 (12pF) and C9 (12pF) and a crystal oscillator X1 (32MHz) to ensure normal MCU operation. Simultaneously, the chip's VDD_aRF pin is filtered by capacitors C4 (0.1μF) and C5 (4.7μF) to support stable RF signal transmission. An external antenna is connected to the RF pin via a matching network of L2 (3nH) and C10 (1pF) to achieve efficient wireless signal transmission and reception.

[0039] To meet the power requirements of different operating modes, the system is designed with multiple voltage control circuits. The 3.3V_VDD-3.3V_VCC power control section uses an MD5133 (D1) as a voltage regulator, along with resistor R33, to achieve 3.7V lamp power control, ensuring stable system operation under different loads. Furthermore, devices such as the SIA319ED-T1-GE3 (Q1) and MAX17048G+ (D3) are used to implement battery voltage detection, power switching, and overvoltage protection functions, ensuring the system's safety and reliability.

[0040] In terms of data communication, the nRF52832 connects to external memory (such as BM270) via SPI interfaces (SCL, SDA, CS) for program burning and data storage; and connects to the debugging interface via SWDCLK and SWDIO pins for easy development and debugging. Simultaneously, the chip integrates peripherals such as ADC, GPIO, and PWM to support sensor data acquisition and device control. For example, the magnetic element sensor MT8831DF-I616(U5) is connected to the chip via an analog input channel to detect changes in the ambient magnetic field, enhancing the system's intelligence level.

[0041] In one embodiment, please refer to Figure 3 The aforementioned multiplexed switch module 20 includes a TI analog switching chip; the TI analog switching chip integrates several independently controlled single-pole double-throw switches.

[0042] This module is built around multiple TMUX6234 chips. Each TMUX6234 is a high-precision, low-on-resistance multi-channel CMOS analog switch, which integrates four independently controlled single-pole double-throw (SPDT) switches. It supports a voltage range of ±18V, has excellent high-voltage withstand capability and low crosstalk characteristics, and its inter-channel crosstalk can be as low as -105dB, effectively suppressing the interference of electrical stimulation on EEG signal acquisition. Figure 3 The image shows multiple TMUX6234 chips (D9 to D20) connected in parallel, each used to control different groups of electrode channels, enabling efficient multiplexing management of 32 or more electrodes.

[0043] Each TMUX6234 chip is connected to its corresponding electrode input (Electrode) and output (Iout) via its S1A / S1B and S2A / S2B pins. The SEL1–SEL4 control signals select the path state of each SPDT switch, thus switching between "acquisition" and "stimulation" modes. For example, when SEL1 is high, the electrode signal is accessed to the EEG acquisition front-end via the S1A path; when SEL1 is low, it is accessed to the electrical stimulation circuit via the S1B path. Simultaneously, the EN pin is an active-low enable signal, internally pulled down to ensure all channels are closed when not enabled, preventing short circuits or interference caused by accidental operation.

[0044] In addition, the circuit incorporates multiple N-channel MOSFETs (such as Q2 to Q43) as grounding or isolating switches on the electrode side, working in conjunction with the TMUX6234 to achieve three-state control: acquisition, stimulation, and high-impedance floating. These MOSFETs are driven by external control signals (such as CTL_XX), allowing the electrodes to be grounded during stimulation to reduce artifacts, or placed in a high-impedance state during non-operation, improving system safety and signal integrity. The power supply uses ±15.6V_VCC to meet the high-voltage stimulation requirements, while decoupling capacitors such as C67 to C71 stabilize the power supply voltage and prevent noise coupling.

[0045] The entire multiplexed switching module 20 coordinates the switching timing of each chip through control logic (provided by the Bluetooth module 10 or the MCU) to ensure that acquisition and stimulation do not occur simultaneously, achieving secure isolation and low-artifact transmission. This design not only achieves efficient multiplexing of a single electrode but also significantly improves the system's integration and reliability, making it suitable for high-density, wearable brain-computer interface systems.

[0046] In one embodiment, please refer to Figure 4 The aforementioned EEG acquisition module 30 includes an EEG acquisition chip, which integrates a multi-channel electrophysiological amplifier array, a 16-bit ADC, and an SPI interface; the SPI interface is connected to the SPI bus.

[0047] In this embodiment, the EEG acquisition chip is the RHD2132, a highly integrated electrophysiological signal acquisition chip that integrates a 32-channel high-precision instrumentation amplifier array, a 16-bit analog-to-digital converter (ADC), and a standard four-wire SPI interface, specifically designed for acquiring weak bioelectrical signals. Figure 4As shown, the chip (D5) receives EEG signal input from the multiplexed module 20 through its in0 to in31 pins. Each input channel is connected to an external electrode signal path, enabling high-fidelity acquisition of scalp or body surface EEG signals. The ref_elec pin is connected to a reference electrode and is used to construct a differential amplification structure, improving the common-mode rejection ratio (CMRR) and effectively suppressing power frequency interference and environmental noise.

[0048] The EEG acquisition chip is powered by a 3.3V_RH power supply, with the VDD and GND pins connected to the positive and negative power supplies and ground respectively to ensure stable operation. The chip's integrated ADC operates at a speed of up to 1.05 million samples per second, supporting simultaneous sampling of all 32 channels at a rate of 30,000 samples per second, meeting the needs of high-density EEG acquisition. Its input reference noise is as low as a typical 2.4μV RMS, exhibiting excellent signal-to-noise ratio performance within a bandwidth of 0.5–100Hz. Furthermore, the chip supports configuring the upper and lower cutoff frequencies of each channel via registers; the upper limit can be adjusted between 100Hz and 20kHz, and the lower limit between 0.1Hz and 500Hz, suitable for acquisition needs across different neural signal frequency bands.

[0049] The EEG acquisition chip also integrates multi-frequency on-site electrode impedance measurement, which can be performed in real-time via the elec_test pin in conjunction with external circuitry to assess electrode-skin contact quality. Simultaneously, the chip incorporates an optional DSP high-pass filter to remove DC offset and improve signal stability. An auxiliary ADC input (auxin1) can be used to input data from other sensors, expanding system functionality. In terms of power management, each amplifier channel can be independently turned on or off, enabling on-demand activation and significantly reducing overall power consumption to meet the low-power requirements of wearable devices.

[0050] In terms of communication, the EEG acquisition chip connects to the Bluetooth module 10 via an SPI interface (RH_SCLK, RH_MOSI, RH_MISO, RH_CS) to enable configuration register writing and data acquisition reading. The CS pin is used for chip select, SCLK for the clock signal, MOSI for data transmission, and MISO for data reception, supporting high-speed, low-power data transmission. Additionally, the LVDS_en pin enables LVDS output mode, improving anti-interference capabilities for long-distance transmission. The power supply section uses decoupling capacitors (C27-C30) for filtering to ensure power stability. R13 (OR) and R14 (100kΩ) are the reference electrode feedback resistor and the right leg drive resistor, respectively, working with external circuitry to achieve common-mode rejection and noise suppression, further improving signal quality. The overall design achieves highly integrated, low-noise, and highly reliable EEG acquisition functionality, providing high-quality raw signal support for closed-loop neuromodulation systems.

[0051] In one embodiment, please refer to Figure 5 The aforementioned programmable constant current stimulation module 40 includes a programmable constant current stimulation chip. The programmable constant current stimulation chip integrates multiple constant current sources and controls the current of each channel through an 8-bit DAC. In one embodiment, please refer to Figure 5 The aforementioned programmable constant current electrical stimulation chip uses a four-wire SPI interface and is configured by the Bluetooth module 10 or the timing logic control module 50 by reading and writing multiple registers corresponding to each channel.

[0052] In this embodiment, the programmable constant current stimulation chip is the SX-32, a multi-channel, high-degree-of-freedom electrical stimulation integrated circuit designed specifically for neuromodulation applications. It integrates 32 independent constant current sources, each supporting dual-range output (±22.5V). Within a set range, it can achieve precise control of the stimulation current via an 8-bit DAC, with a typical resolution of 0.1μA, meeting various neuromodulation needs. Figure 5 As shown, the Iout1 to Iout32 pins of the programmable constant current electrical stimulation chip (D8) are connected to the external electrode output path, respectively, for applying precise bipolar constant current stimulation pulses to scalp or body surface electrodes. Its power input terminals VstimP and VstimN are connected to +1.65V_VCC and -1.65V_VCC respectively, forming an excitation voltage range of up to ±33V, supporting a stimulation compliance power supply of up to 45V, and suitable for various commercially available electrodes and complex stimulation scenarios.

[0053] As can be seen, this chip contains 32 independent current sources, each of which can be set to two current output ranges. Within the set range, precise stimulation current control is achieved through an 8-bit DAC. Simultaneously, the chip is fabricated using a high-voltage process, providing ±22.5V and supporting stimulation compliance power supplies up to 45V, making it compatible with most commercially available stimulation electrodes. The chip also features a stimulation controller based on a standard four-wire SPI bus. This controller is equipped with eight independent 40-bit registers for each stimulation channel, allowing independent setting of stimulation current, stimulation pulse time parameters, etc., for each channel by modifying and reading these registers. By adjusting the stimulation pulse parameters, a stimulation pulse repetition frequency of 0.1Hz to 125kHz can be achieved, and the number of cycles per stimulation repetition or a continuous loop mode can also be set. Properly configuring the eight registers for each channel enables complex stimulation control of multiple stimulation channel combinations and multi-electrode charge flow.

[0054] The programmable constant current electrical stimulation chip is manufactured using high-voltage technology, possessing excellent electrical isolation performance and effectively preventing high-voltage stimulation signals from flowing back to other parts of the system. Its integrated constant current source has good load adaptability, and the output current is unaffected by electrode impedance fluctuations, ensuring stimulation stability. Simultaneously, the chip communicates with external controllers (such as Bluetooth module 10 or timing logic control module 50) via a four-wire SPI interface (mSCLK, mMOSI, mMISO, mCSB) to achieve independent configuration and status monitoring of each channel. Each stimulation channel is equipped with eight 40-bit registers, which can be read and written to set parameters such as stimulation current, pulse width, frequency (0.1Hz~125kHz), duty cycle, and number of repetition cycles, supporting single stimulation or continuous loop modes, thereby achieving complex multi-channel combination timing and charge flow control.

[0055] In terms of circuit design, the power management section of the programmable constant current electrical stimulation chip consists of multiple voltage regulation and filtering circuits. For example, U3 is an LDO regulator that stabilizes the input voltage to the required value; U4 is a high-voltage switching regulator that provides efficient power conversion; and U5 is a MOSFET driver used to control the on and off of the high-voltage path. In addition, D6 and D7 are bidirectional voltage sensing chips (VIA) used to monitor the voltage states of VDD1 and VDD2 to ensure safe system operation. Ceramic capacitors such as C31, C33, and C35 are used for power supply decoupling to reduce noise interference. All analog grounds (GNDs) and digital grounds (GNDs) are connected through a low-impedance path to avoid ground loop interference. The overall design achieves high-precision, high-reliability, and high-safety electrical stimulation functions, providing a powerful and flexible stimulation output capability for closed-loop neuromodulation systems.

[0056] In one embodiment, the circuit also includes an interface chip, which is mainly used to achieve precise control of peripheral components such as the multiplexer module 20 and the electrical stimulation chip. The core function of this circuit is to serve as the output driver of the timing logic control module 50, ensuring that high-voltage crosstalk and artifact interference can be effectively avoided during the acquisition and stimulation process.

[0057] First, from a power supply perspective, the circuit uses a 3.3V power supply and is connected to ground via a decoupling capacitor C60 (nominal value 0.1μF) to reduce power supply noise and improve system stability. This design is crucial for ensuring the purity of digital signals, especially in applications processing weak electrical signals.

[0058] Next, observing the I / O port configuration, we can see that 32 pins, from IO0_0 to IO5_7, are assigned to different control tasks. These pins correspond to CTL_0 to CTL_31, and each CTL_x signal is directly associated with the control input of an external device, such as the SEL pin of the TMUX6234 multiplexer module 20. This allows the main control unit to flexibly select which one or more signal paths to activate, thereby achieving effective management of complex circuit topologies.

[0059] In addition, the EX_GPIO_SDA and EX_GPIO_SCL pins support the I²C communication protocol, allowing the interface chip to exchange data with other peripherals. This means that in addition to basic control functions, it can also play a role in data transmission, enhancing the interconnectivity and scalability of the entire system.

[0060] It is worth noting that resistor R32 (10kΩ) is connected between EX_GPIO_SDA and EX_GPIO_SCL, providing a simple way to enhance signal integrity and prevent data loss or errors due to excessive line length or environmental interference.

[0061] In summary, the design cleverly integrates functions such as level conversion, signal buffering, and logic isolation, which not only enables precise control of multiple peripheral components but also improves the overall system's anti-interference capability and reliability. Such a design scheme can provide users with stable and reliable support.

[0062] In one embodiment, in the acquisition mode, the electrodes of the aforementioned timing logic control module 50 are connected to the multiplexer module 20 via the multiplexer module 20, and the stimulation circuit is turned off and isolated; in the stimulation mode, the electrodes are switched to be connected to the programmable constant current stimulation module 40 via the multiplexer module 20, and the multiplexer module 20 is set to a high-impedance state; in the closed-loop mode, stimulation is performed during the stimulation pulse output period, and the acquisition mode is switched back during the pulse interval.

[0063] In one embodiment, the timing logic control module 50 is designed to support three operating modes: acquisition mode, stimulation mode, and closed-loop mode. This design aims to ensure that the system can perform its functions efficiently and accurately in different application scenarios, while guaranteeing signal quality and user safety.

[0064] Acquisition Mode: In this mode, the electrodes are connected to the acquisition front end via a multiplexer module 20 to acquire bioelectrical signals. Simultaneously, the stimulation circuit is shut off and isolated to prevent any potential interference or damage. Specifically, when the system is in acquisition mode, the multiplexer module 20 selects a specific channel for signal transmission based on a control signal, ensuring that only the selected channel is active, while other unused channels remain in a high-impedance state to avoid unnecessary crosstalk and signal contamination. Furthermore, the acquisition front end employs a high-input-impedance instrumentation amplifier and integrates a right-leg drive circuit to improve the common-mode rejection ratio, further enhancing signal quality.

[0065] Stimulation Mode: When switching to stimulation mode, the electrodes are connected to the programmable constant current electrical stimulation module 40 via the multiplexer module 20 to provide precise electrical stimulation to the target tissue. At this time, the acquisition front end is set to a high-resistance state to avoid damage during stimulation. The key component in this mode is the bipolar programmable constant current source, which allows users to adjust parameters such as current intensity, pulse width, frequency, and duty cycle to meet different treatment or research needs. To ensure safety, a high-voltage safety isolation module is activated, using a high-voltage diode array and current limiting measures to prevent stimulation voltage backflow to the acquisition front end, while also limiting leakage current to meet safety standards.

[0066] Closed-loop mode: This is one of the most complex modes, designed to minimize artifacts. In this mode, the system performs stimulation during the stimulation pulse output and rapidly switches back to acquisition mode during the intervals between pulses to capture the purest possible bioelectrical signals. This process requires the timing logic control module 50 to have highly precise timing control capabilities to ensure seamless switching between stimulation and acquisition. Furthermore, a real-time impedance detection mechanism checks the electrode-skin contact impedance before and after each switch. If any abnormality is detected (such as excessively high or low impedance), stimulation will be automatically shut off immediately and an alarm will be issued to protect the user from potential risks.

[0067] In summary, by cleverly combining acquisition, stimulation, and closed-loop control strategies, the timing logic control module 50 in this embodiment not only achieves efficient signal processing and electrical stimulation functions, but also greatly improves the system's integration, channel utilization, and security. It is suitable for various brain-computer interface application fields, including but not limited to wearable brain-computer interfaces, transcranial electrical stimulation, and neurorehabilitation devices.

[0068] Please see Figure 7After the 32 electrodes are connected to their respective protection circuits, they are connected to the EEG acquisition module 30 and the programmable constant current stimulation module 40 via a two-to-one switch. Both the EEG acquisition module 30 and the programmable constant current stimulation module 40 communicate with the timing logic control module 50 via bidirectional SPI or I²C interfaces to achieve data transmission and parameter configuration. The timing logic control module 50 is responsible for receiving wireless commands (such as Bluetooth communication), generating control signals, and driving the multiplexer module 20 to complete channel switching. The power supply module supplies power to the entire system and supports multiple voltage outputs to meet the needs of different chips. This architecture achieves physical separation and logical coordination between acquisition and stimulation, ensuring that only one functional path is active at any given time, thereby avoiding short circuits and crosstalk.

[0069] In one embodiment, please refer to Figure 8 Electrode 1 first passes through an ESD protection circuit to prevent electrostatic discharge; then it enters a high-voltage isolation two-way switch, driven by a control signal, which switches between "acquisition" and "stimulation" modes. In acquisition mode, the signal enters the EEG acquisition module 30 via a "signal filtering" path, where it is amplified, filtered, and converted by an ADC. When switched to stimulation mode, the electrode connects to the stimulation module, outputting a constant current pulse. Simultaneously, the system also features a clamping protection mechanism to prevent damage to sensitive circuits from overvoltage or reverse voltage. This design achieves high-voltage isolation and non-overlapping switching, effectively preventing high-voltage stimulation from flowing back to the acquisition front end, significantly reducing artifacts and improving system safety.

[0070] In one embodiment, please refer to Figure 9 It supports a closed-loop stimulation mode with independent configuration of multiple channels. Each channel can perform personalized stimulation tasks according to preset parameters (such as stimulation waveform, frequency, and current intensity). For example, channel 1 can be set to "square wave - 100Hz - 2mA" for 10 seconds; channel 2 to "square wave - 100Hz - 1mA" for 20 seconds; and channel n can be set to "square wave - 1kHz - 1mA" for 30 seconds. This flexible multi-channel timing control capability enables the system to implement complex neuromodulation strategies, making it suitable for scenarios such as neurorehabilitation and cognitive enhancement. The stimulation parameters of all channels are uniformly scheduled by the main control module and coordinated by the timing logic control module 50 to ensure that each channel switches in an orderly manner within different time windows, avoiding mutual interference.

[0071] This embodiment features a highly integrated, multi-channel, securely isolated, and low-artifact EEG acquisition and electrical stimulation multiplexing electrode circuit. Its core lies in achieving efficient multiplexing of a single electrode channel between acquisition and stimulation functions through an innovative system architecture and the collaborative work of key modules. Furthermore, the system boasts an input impedance of ≥100 MΩ, input noise of ≤1 μVrms (0.5–100 Hz), a common-mode rejection ratio of ≥110 dB, and an isolation voltage of ±22 V, ensuring high-quality acquisition of weak bioelectrical signals. It supports up to 32 channels and employs a timing-controlled non-overlapping switching method, significantly reducing the number of electrodes and wiring complexity, and substantially improving channel utilization. The entire device adopts a single-chip / small layout highly integrated design, resulting in a small size and low power consumption. It is suitable for cutting-edge fields such as wearable brain-computer interfaces, transcranial electrical stimulation, neurorehabilitation devices, closed-loop neuromodulation systems, and high-density flexible arrays, demonstrating excellent scalability and promising clinical application prospects.

[0072] The method described in this embodiment is intended for use in wearable brain-computer interfaces, neurorehabilitation, transcranial electrical stimulation, and closed-loop neuromodulation systems. The circuit includes a Bluetooth module 10 supporting the BLE 5.0 protocol for wireless communication; a multi-channel analog switch and multiplexed electrode interface module that provides high-voltage safety isolation protection while enabling channel selection and mode switching; an EEG acquisition module 30 responsible for amplifying, filtering, and suppressing weak EEG signals; a programmable constant-current electrical stimulation module 40 that outputs safe and precise bipolar constant-current stimulation pulses; and a timing logic control module 50 to coordinate the switching timing between EEG acquisition and electrical stimulation to reduce artifact interference. This design not only achieves efficient multiplexing of a single electrode for EEG data acquisition and electrical stimulation output but also improves the overall system integration and channel density, reduces device size and power consumption, while ensuring the safety and signal quality of the front-end circuitry, meeting the advanced application requirements of modern medical and health fields.

[0073] The aforementioned multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit integrates a Bluetooth module 10, a multiplexer module 20, an EEG acquisition module 30, a programmable constant current electrical stimulation module 40, and a timing logic control module 50. This achieves efficient multiplexing of a single electrode to support both EEG signal acquisition and electrical stimulation output. Specifically, the Bluetooth module 10 receives external commands and transmits them to the timing logic control module 50, which coordinates the operation of each module to ensure that EEG acquisition and electrical stimulation do not occur simultaneously, avoiding the risks of short circuits and crosstalk. The multiplexer module 20 switches the operating mode connected to the electrode according to the received commands, thus seamlessly switching between data transmission and electrical stimulation applications. A high-voltage safety isolation protection mechanism effectively prevents damage to the front-end acquisition circuit from the stimulation voltage and reduces the impact of stimulation artifacts. The entire system is compact, improving overall integration and channel density. This not only reduces device size and energy consumption but also enhances the system's adaptability and flexibility, making it particularly suitable for wearable technology and high-density arrays, thus promoting the realization of closed-loop acquisition-stimulation systems.

[0074] In one embodiment, a control method for a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit is also provided. This method utilizes the aforementioned multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, including: The timing logic control module 50 coordinates the operation of each module by receiving instructions through the Bluetooth module 10, ensuring that the EEG acquisition module 30 and the programmable constant current stimulation module 40 do not operate simultaneously. The multiplexer module 20 switches the working mode connected to the electrodes according to the instructions of the Bluetooth module 10 to perform data transmission and electrical stimulation application. The EEG acquisition module 30 processes the EEG signals and transmits the data back to the Bluetooth module 10 for analysis or transmission.

[0075] In addition, it also includes: In acquisition mode, weak EEG signals are processed, amplified, and filtered using the highly integrated RHD2132 chip; in stimulation mode, the SX-32 chip provides bipolar constant current stimulation, and high-voltage isolation technology is used for signal isolation.

[0076] In one embodiment, the circuit further includes a spatiotemporal frequency domain coupling analysis module. This module is integrated as a system-on-a-chip (SoC) in the signal link between the EEG acquisition module 30 and the programmable constant current stimulation module 40. It establishes a bidirectional data channel with the Bluetooth module 10 and the multiplexer module 20 via a high-speed SPI bus, and is directly interconnected with the hardware state machine of the timing logic control module 50 to form a closed-loop sensing network of stimulation-acquisition-analysis-regulation. The module first initiates a spatial reference calibration procedure: by driving a miniature structured light projector integrated into the electrode cap frame to project an coded grating onto the skull surface, or by reading the acceleration and angular velocity data of the built-in nine-axis inertial measurement unit (IMU), it identifies and precisely locates the nasal root point, the external occipital protuberance point, the left and right pre-auricular points, and the geometric center point of the ground electrode (usually the earlobe or mastoid electrode) in the electrode array, which serves as an electrical reference. These anatomical landmarks and physical electrodes are defined together as spatial reference points. Based on these reference points, the module runs a three-dimensional coordinate calculation algorithm. Through a coordinate transformation matrix, the absolute coordinates of each reference point are mapped to a local scalp coordinate system with the root of the nose as the origin, the line connecting the two preauricular points as the X-axis, and the line connecting the root of the nose to the external occipital protuberance as the Y-axis. The module then calculates the spatial coordinates (x, y, z) of each electrode channel in this relative coordinate system, as well as the Euclidean distance (reference distance) between adjacent electrodes. This constructs a three-dimensional spatial positioning reference frame covering the entire electrode array. This frame not only records the geometric position of each electrode but also associates and records the corresponding brain functional regions (such as the frontal lobe, parietal lobe, temporal lobe, and occipital lobe) for each channel, providing anatomical semantic mapping for subsequent fault localization.

[0077] Based on the constructed spatial positioning reference framework, the spatiotemporal frequency domain coupling analysis module executes a spatial segmentation strategy for the electrode array: according to the correlation between scalp curvature changes and function, adjacent 2-4 electrode channels (e.g., F3, Fz, and F4 combined according to a 10-20 system to form a central frontal lobe monitoring segment, or C3, Cz, and C4 combined to form a central region monitoring segment) are logically combined into independent monitoring segments. Each monitoring segment is considered as a signal propagation unit with local homogeneous medium characteristics. For each monitoring segment, the module initiates a dual-modal signal propagation characteristic analysis: under the precise timing control of the timing logic control module 50, the programmable constant current electrical stimulation module 40 injects a low-frequency probe pulse train into the stimulation electrode of the target monitoring segment. This pulse train uses linear frequency modulation (LFM) or pseudo-random code modulation, with a frequency range covering 1Hz to 100Hz (covering the main EEG frequency bands and tissue impedance sensitive frequency bands), an adjustable pulse width of 20μs to 500μs, and the current intensity maintained below the sensory threshold (usually below 100μA) to ensure safety. The probe pulse penetrates the stratum corneum, epidermis, dermis, and skull, generating reflection, scattering, and transmission phenomena at the electrode-cortical contact interface. During the pulse interval (with microsecond-level precise control), the EEG acquisition module 30 switches to acquisition mode to capture voltage fluctuation data and current intensity waveforms between electrodes within the monitoring segment at a high sampling rate (≥1kHz), forming a multi-channel echo response signal matrix.

[0078] The spatiotemporal frequency domain coupling analysis module performs deep signal processing on the received echo response signal: First, it separates the response components of different frequency bands through a bandpass filter bank. Then, it performs cross-correlation operation on the time domain waveform to extract the signal propagation time delay (accurate to the microsecond level). This delay reflects the propagation speed of the pulse signal in the scalp tissue and indirectly characterizes the tightness of the electrode's fit with the skin. At the same time, the module performs Fast Fourier Transform (FFT) or Wavelet Transform on the echo signal to generate the bioimpedance spectrum of the monitoring segment, identify the energy peak of the characteristic frequency band, and calculate the resonant frequency offset (relative to the reference resonant frequency under no-load conditions). Combining the structural quantification parameters of each monitoring segment—including the electrode-scalp contact impedance (kΩ level) measured in real time by the electrode impedance detection circuit, the axial position coordinates of the electrode in three-dimensional space (reflecting whether the electrode is perpendicular to the scalp surface), and the hair density characteristics around the electrode estimated by the image sensor (hair increases contact impedance and causes signal attenuation)—the module runs a bioelectric signal propagation model. Based on the three-layer spherical head model, it calculates the theoretical propagation path of the signal in the multi-layer medium of scalp-skull-brain tissue and predicts the expected signal attenuation distribution and noise coupling distribution caused by interface impedance mismatch.

[0079] To further enhance environmental adaptability, the module introduces a multi-physics coupling correction mechanism: real-time data collection of scalp surface temperature and relative humidity is achieved through integrated temperature and humidity sensors. Based on a preset impedance-temperature-humidity coupling coefficient matrix, the calculated bioimpedance characteristic values ​​are corrected for humidity expansion (increased humidity leads to enhanced hydration of the stratum corneum, reducing contact impedance) and temperature drift (temperature changes affect skin ionic conductivity and the noise level of electronic devices), generating a corrected spatiotemporal frequency domain coupling characteristic parameter set. Based on this parameter set, the module's built-in parameter optimization algorithm (which can employ genetic algorithms or particle swarm optimization) calculates the dual-modal collaborative optimization parameters for each monitoring segment: For the EEG acquisition side, it calculates the gain compensation value of each channel's programmable gain amplifier (PGA) (e.g., automatically increasing the gain by 20-40dB when there is poor contact), the optimal band selection of the bandpass filter (e.g., dynamically adjusting the high-pass cutoff frequency to address EMG interference), the sampling time window setting (avoiding the stabilization period after stimulation artifacts), and the real-time adjustment value of the common-mode inhibition ratio (CMRR); for the electrical stimulation side, it calculates the compensation value of the stimulation current (e.g., moderately increasing the current to ensure effective stimulation when impedance increases) and the pulse width adjustment value (narrow pulses are used for fine-tuning, and wide pulses are used for deep activation). These parameters are integrated spatiotemporally with spatiotemporal consistency through an adaptive collaborative control strategy, ensuring that the stimulation parameters and acquisition parameters within the same monitoring segment achieve the optimal balance between physiological safety and signal quality, and are configured in real-time to the EEG acquisition chip and the constant-current stimulation chip via the SPI bus.

[0080] During the continuous monitoring phase, the spatiotemporal frequency domain coupling analysis module configures the operating mode of the EEG acquisition module 30 according to the adaptive collaborative control strategy parameters, acquiring EEG signals from each monitoring segment at high resolution and simultaneously recording time-domain waveforms and spectral response data. The module performs multi-dimensional feature extraction on the optimized EEG signal dataset: in the time domain, it extracts the dynamic changes in signal propagation time delay through peak detection and zero-crossing rate analysis; in the frequency domain, it extracts the characteristic frequency band energy shifts of alpha waves (8-13Hz), beta waves (13-30Hz), theta waves (4-8Hz), and delta waves (0.5-4Hz) through power spectral density (PSD) analysis, identifying abnormal energy attenuation or enhancement in specific frequency bands due to poor electrode contact; in the energy domain, it calculates the signal attenuation rate by comparing the energy difference between the received signal and the reference electrode signal in each monitoring segment. The module normalizes the extracted multi-dimensional feature parameters, converting them into standardized feature vectors with uniform dimensions.

[0081] Subsequently, the module performs intelligent diagnosis of the electrode-cortical coupling state: the standardized feature vector is input into a pre-set threshold library of electrode contact abnormalities and stimulation circuit integrity features. This threshold library is pre-trained by machine learning algorithms and contains feature fingerprints of various abnormal modes, including the complete electrode detachment mode (manifested as an extremely high signal attenuation rate, energy across the entire frequency band tending to the noise floor, and infinite time-domain delay), the progressive poor contact mode (manifested as a moderate attenuation rate, relatively preserved low-frequency energy while significant attenuation in the high-frequency band, and resonant frequency shifting to low frequencies), the high impedance mode caused by electrode-gel dehydration (manifested as an abnormally high proportion of 50Hz power frequency interference energy and a decrease in common-mode rejection ratio), and the motion artifact mode caused by the patient's head movement (manifested as abrupt changes in time-domain waveform, a surge in high-frequency energy, and jitter in delay). The module calculates the Euclidean distance or cosine similarity between the current feature vector and each abnormal pattern to obtain the abnormal similarity index of each monitoring segment. It then applies multi-parameter joint judgment rules (such as three-parameter weighted fusion judgment based on Bayesian inference or support vector machine) to comprehensively determine whether the electrode-cortex coupling state is normal by considering the joint deviation of signal attenuation rate, resonant frequency offset, and time domain delay. Finally, it generates a joint identification signal containing an abnormal monitoring segment identifier, precise spatial coordinates (corresponding to the anatomical location of the skull), and defect level (mild: signal quality degraded but still usable; moderate: requires immediate adjustment; severe: signal completely lost or risk of stimulation circuit breakage).

[0082] Based on the generated joint identification signal, the spatiotemporal frequency domain coupling analysis module initiates a hierarchical closed-loop feedback and dynamic calibration mechanism: The module first maps the identifier of the abnormal monitoring segment to the spatial positioning reference frame, obtains the anatomical location description corresponding to the monitoring segment (such as "left frontal lobe F7 electrode" or "right parietal lobe P4 electrode"), automatically generates a structured status diagnosis report, and pushes it to the user's smart terminal via Bluetooth module 10. The report content covers abnormality type diagnosis, spatial location visualization annotation, and targeted treatment recommendations (such as "Please press the F7 electrode cap to ensure contact" or "It is recommended to clean the skin in the Cz area and wear it again"). Simultaneously, the module sends a graded maintenance management instruction to the timing logic control module 50: if a severe abnormality is detected (such as simultaneous signal interruption in multiple channels within the monitoring segment, indicating a risk of breakage in the stimulation circuit or complete electrode detachment), the connection between the monitoring segment and the programmable constant current electrical stimulation module 40 is immediately cut off, and the electrical stimulation output is suspended to ensure patient safety; if the system is equipped with a redundant electrode array, the channel switching logic is automatically activated to seamlessly transfer the stimulation and acquisition tasks to the adjacent normal monitoring segment (such as automatically switching to F1 or F5 when F3 fails), maintaining the continuity of treatment; if it is a mild contact defect, the weight coefficient of the channel in the multi-channel signal fusion algorithm is dynamically adjusted (reducing its contribution), and a slight tactile feedback (such as a micro-vibrator built into the electrode cap) is triggered through the Bluetooth module 10 to prompt the user to adjust the specific position. In addition, the module establishes a long-term adaptive calibration mechanism: the spatiotemporal frequency domain coupling characteristic parameters of the current monitoring segment are statistically compared with the benchmark characteristics of the same anatomical location and similar temperature and humidity conditions in the historical fault database. If the deviation between the current feature and the historical mean exceeds the preset fault tolerance range (such as 3 times the standard deviation), it is determined that the electrode is aging or the scalp condition has changed over a long period of time. The algorithm recalibration process is automatically triggered, the weight coefficients of the frequency domain energy integration algorithm are corrected in reverse, the benchmark template of the bioimpedance spectrum is updated, and the adaptive collaborative control strategy parameters are recalculated. This enables the system to continuously adapt to long-term evolution processes such as the drying of electrode conductive paste, electrode oxidation, and changes in the diurnal rhythm of skin impedance, ensuring the long-term stability and safety of EEG acquisition and electrical stimulation intervention.

[0083] In one embodiment, the circuit described above is adapted to a hand nerve signal monitoring and functional electrical stimulation (FES) rehabilitation system. The spatiotemporal frequency domain coupling analysis module is integrated as a system-on-a-chip in the signal link between the interphalangeal electromyography acquisition module and the programmable constant current electrical stimulation module. This module first initiates a hand spatial reference calibration procedure: by driving a miniature flexible strain sensor array integrated into the finger joints of a wearable glove, or by reading the fingertip kinematic data from a built-in six-axis inertial measurement unit (IMU), it identifies and precisely locates the anatomical landmarks of the metacarpophalangeal joint (MCP), proximal interphalangeal joint (PIP), distal interphalangeal joint (DIP), and the radial and ulnar styloid processes of the wrist. These bony landmarks and the flexible electrode array are collectively defined as spatial reference points.

[0084] Based on these reference points, the module runs a 3D surface calculation algorithm. Through a non-rigid transformation matrix, the absolute coordinates of each reference point are mapped to a local hand coordinate system with the midpoint of the wrist crease as the origin, the line connecting the radial and ulnar styloid processes as the X-axis, and the longitudinal axis of the middle finger as the Y-axis. The module calculates the surface coordinates (u, v, w) of each electrode channel in the relative coordinate system and the geodesic distance along the finger length, constructing a 3D surface positioning reference frame covering the palm to the fingertip. This frame is associated with and records the corresponding nerve innervation areas of each channel (such as the thumb to middle finger area innervated by the median nerve, and the little finger and ulnar half of the ring finger innervated by the ulnar nerve), providing anatomical semantic mapping for subsequent motor intent decoding and stimulus targeting.

[0085] Based on the constructed spatial positioning reference framework, the module executes a functional segmentation strategy for the hand electrode array: according to the synergistic contraction characteristics of hand muscles and the nerve innervation topology, adjacent 2-3 electrode channels (e.g., electrodes at the thenar eminence of the thenar muscle group are combined into a thumb abduction monitoring segment, or electrodes at the intermembranous spaces of the interosseous muscle group are combined into a finger adduction monitoring segment) are logically combined into independent monitoring segments. For each monitoring segment, the module initiates a dual-modal signal propagation characteristic analysis: the programmable constant current stimulation module injects a high-frequency probe pulse train (frequency range covering 500Hz to 10kHz to match the excitation threshold of motor nerve fibers) into the stimulation electrodes of the target monitoring segment, with a pulse width adjustable from 50μs to 200μs, and the current intensity maintained below the pain threshold (usually below 5mA).

[0086] The electromyography (EMG) acquisition module switches to acquisition mode during the pulse interval, capturing the compound muscle action potentials (CMAP) and sensory nerve action potentials (SNAP) between electrodes within the monitoring segment at an ultra-high sampling rate (≥5kHz), forming a multi-channel neural conduction response signal matrix. The spatiotemporal frequency domain coupling analysis module performs in-depth processing on the received response signals: separating motor unit action potentials (MUAP) and sensory nerve signals through a bandpass filter bank, performing multi-template matching operations on the time-domain waveform to extract the motor unit recruitment order and nerve conduction velocity (accurate to 0.1m / s); simultaneously, performing a short-time Fourier transform (STFT) on the signal to generate a neuromuscular spectrogram of the monitoring segment, identifying the median frequency (MF) and mean power frequency (MPF) offset of the EMG signal, indirectly characterizing the functional state of the peripheral neuromuscular junction.

[0087] Combining the structural quantitative parameters of each monitoring segment—including the transcutaneous impedance measured in real time through the electrode-skin contact impedance detection circuit (affected by the thickness of the stratum corneum and sweat secretion), the contact angle of the electrode in the curved coordinate system (reflecting the degree of conformity between the flexible electrode and the skin surface), and the wearing tension distribution estimated through the pressure sensor array—the module runs a neural electrical signal propagation model. Based on a multi-layer cylindrical limb model (considering the multi-layer dielectric properties of skin-subcutaneous tissue-muscle-bone), it calculates the theoretical propagation path of the signal in the limb tissue and predicts the expected signal attenuation distribution caused by electrode displacement or skin dehydration.

[0088] To further enhance the adaptability of rehabilitation training, the module introduces a multi-physics coupling correction mechanism: Through an integrated geoskin response (GSR) sensor and temperature sensor, it collects real-time data on palm sweating and skin temperature. Based on a preset impedance-sweat-temperature coupling coefficient matrix, it performs sweat conductivity correction (increased sweating significantly reduces skin impedance) and temperature drift correction (temperature changes affect ion channel activity and electrode polarization noise) on the calculated neuromuscular impedance characteristic values, generating a corrected spatiotemporal frequency domain coupling characteristic parameter set. Based on this parameter set, the module's built-in parameter optimization algorithm calculates the dual-modal collaborative optimization parameters for each monitoring segment: For the electromyography (EMG) acquisition side, it calculates the gain compensation value of each channel's instrumentation amplifier (e.g., automatically adjusting gain when grasping movements cause micro-displacement of electrodes) and the optimal frequency band selection of the bandpass filter (e.g., dynamically adjusting the cutoff frequency to address residual EMG interference in the forearm); For the functional electrical stimulation side, it calculates the stimulation current compensation value (e.g., maintaining constant charge stimulation to ensure stable motor unit recruitment when impedance changes) and the stimulation phase adjustment value (biphasic pulses to balance charge and avoid tissue damage).

[0089] During the continuous rehabilitation monitoring phase, the module performs multi-dimensional feature extraction on the optimized electromyography (EMG) signal dataset: extracting motor unit firing rate and co-contraction index in the time domain; extracting the median frequency decline slope of muscle fatigue features in the frequency domain; and calculating the matching error between induced and expected muscle strength in each monitoring segment in the energy domain. The module normalizes the extracted multi-dimensional feature parameters and inputs them into a pre-set neuromuscular coupling state feature threshold library, which contains feature fingerprints of various abnormal patterns—including electrode slippage pattern (manifested as a sudden drop in CMAP amplitude and pseudo-increase in conduction velocity), muscle fatigue pattern (manifested as a continuous decrease in MPF ​​and widening of MUAP waveform), and nerve conduction block pattern (manifested as prolonged CMAP latency when distal stimulation is recorded proximally). The module determines whether the neuromuscular interface state is normal by calculating the similarity between the current feature vector and each abnormal pattern, and generates a joint identification signal containing abnormal phalanx identifiers, precise surface coordinates (corresponding to the anatomical position of the hand), and defect level.

[0090] Based on the generated joint identification signal, the module initiates a hierarchical closed-loop feedback and dynamic calibration mechanism: the module maps the identifier of the abnormal monitoring segment to the spatial positioning reference frame, obtains the description of the hand functional area corresponding to the monitoring segment (such as "thumb opposition muscle group" or "little finger abductor muscle group"), automatically generates a structured rehabilitation status diagnosis report, and pushes it to the rehabilitation therapist's terminal via Bluetooth module. The report content includes abnormal type diagnosis, visual annotation of hand functional areas, and targeted treatment suggestions (such as "please adjust the electrode fit at the proximal interphalangeal joint of the index finger" or "it is recommended to clean the sweaty area of ​​the palm and put on gloves again"). Simultaneously, the module sends a hierarchical maintenance management instruction to the timing logic control module: if a severe abnormality is detected (such as complete loss of nerve signals within the monitoring segment, indicating electrode breakage or nerve compression), the connection between the monitoring segment and the stimulation module is immediately cut off, and functional electrical stimulation output is suspended to ensure patient safety; if the system is equipped with a redundant electrode array, the channel switching logic is automatically activated to transfer the rehabilitation stimulation task to an adjacent normal monitoring segment (such as switching to the ulnar nerve compensation area when the median nerve innervation area fails), maintaining the continuity of rehabilitation training; if it is a mild contact malfunction, the weight coefficient of the channel in the multi-channel electromyography fusion algorithm is dynamically adjusted, and a slight vibration feedback is triggered via the Bluetooth module to prompt the patient to adjust their hand posture. In addition, the module establishes a long-term adaptive calibration mechanism: it statistically compares the spatiotemporal frequency domain coupling feature parameters of the current monitoring segment with the benchmark features of the same hand functional area and similar rehabilitation stage in the historical rehabilitation database. If the deviation between the current feature and the historical mean exceeds the preset fault tolerance range, it is determined that the electrode is aging or the patient's neuromuscular function is remodeled. The algorithm recalibration process is automatically triggered, the template library of the motor unit recognition algorithm is corrected in reverse, the benchmark template of the neuromuscular spectrogram is updated, and the adaptive collaborative control strategy parameters are recalculated. This enables the system to continuously adapt to the physiological evolution processes such as muscle atrophy reversal and nerve reinnervation that occur during long-term rehabilitation, ensuring the long-term stability and rehabilitation effect of hand function reconstruction.

[0091] In one embodiment, the circuit described above is adapted to an implantable spinal cord stimulation (SCS) pain management system. The spatiotemporal frequency domain coupling analysis module is integrated as a system-on-a-chip in the signal link between the epidural electrode acquisition module and the programmable constant current stimulation module. It establishes a bidirectional data channel with the external programmer and multi-channel electrode selector via a high-speed SPI bus. The module first initiates the spinal spatial reference calibration program: by driving the miniature X-ray imaging marker integrated into the proximal end of the electrode implantation guidewire or reading the positional change data of the built-in triaxial accelerometer, it identifies and accurately locates the spinous processes of the T8-T12 thoracic vertebrae, the spinous processes of the L1-L5 lumbar vertebrae, the sacral promontory, and the geometric center point of the metal shell of the implantable pulse generator (IPG) as an electrical reference. These anatomical landmarks and the epidural electrode array are collectively defined as spatial reference points.

[0092] Based on these reference points, the module runs a three-dimensional spine calculation algorithm. Through a rigid transformation matrix, the absolute coordinates of each reference point are mapped to a local trunk coordinate system with the sacral promontory as the origin, the line connecting the highest points of the iliac crests on both sides as the X-axis, and the longitudinal axis of the spine as the Z-axis. The module calculates the spatial coordinates (x, y, z) of each electrode contact point in this relative coordinate system, as well as the three-dimensional Euclidean distance between adjacent contacts and the geodesic distance along the spinal curve. This constructs a three-dimensional spatial positioning reference frame covering the epidural space of the thoracolumbar region. This frame not only records the geometric position of each electrode contact point but also associates and records the innervation areas of the corresponding spinal nerve segments (e.g., T10-L1 corresponding to the intercostal nerves and upper abdominal wall nerves, L2-S1 corresponding to the lumbar plexus and sacral plexus nerve roots), the anatomical projection areas of the dorsal column conduction tracts (fasciculus gracilis and fasciculus cuneatus), and the relative positional relationship between the epidural venous plexus and the cerebrospinal fluid space. This provides anatomical semantic mapping for subsequent stimulation field optimization and complication prevention.

[0093] Based on the constructed spatial positioning reference framework, the spatiotemporal frequency domain coupling analysis module executes a segmental division strategy for the epidural electrode array: according to the anatomical segments of the spinal cord and the distribution of pain dermatomes, the longitudinally arranged 4-8 electrode contacts (for example, in the 4-column × 8-row array of surgical electrodes, the contact corresponding to the T10-T12 segment is combined into a lower thoracic pain modulation segment, or the contact corresponding to the L3-S1 segment is combined into a lower limb pain modulation segment) are logically combined into independent monitoring segments. Each monitoring segment is regarded as a bioelectric propagation unit with local cerebrospinal fluid-dura mater-spinal cord tissue characteristics. For each monitoring segment, the module initiates a dual-modal signal propagation characteristic analysis: Under the precise timing control of the timing logic control module, the programmable constant current electrical stimulation module injects a low-frequency probe pulse train into the stimulation contact of the target monitoring segment. This pulse train adopts a charge-balanced biphasic waveform with a frequency range covering 2Hz to 120Hz (covering the 40-60Hz sensory abnormality substitution frequency band commonly used for pain suppression and the 10kHz high-frequency non-sensory abnormality frequency band). The pulse width is adjustable from 100μs to 1000μs, and the current intensity is maintained near the sensory threshold (usually adjustable from 0.1mA to 10mA).

[0094] The probe pulse penetrates the dura mater, the cerebrospinal fluid layer of the subarachnoid space, and the pia mater, generating an electric field distribution and nerve fiber depolarization at the electrode-spinal cord interface. During the pulse interval (precisely controlled at the millisecond level to avoid the post-stimulation refractory period), the epidural electrode acquisition module switches to acquisition mode, capturing evoked potentials (EP) and local field potentials (LFP) between each contact point within the monitoring segment at a high sampling rate (≥2kHz), forming a multi-channel neurophysiological response signal matrix.

[0095] The spatiotemporal frequency domain coupling analysis module performs deep signal processing on the received response signal: First, it eliminates stimulation artifacts and 50Hz power frequency interference through a band-stop filter. Then, it performs an average superposition operation on the time-domain waveform to extract the latency and amplitude of spinal cord evoked potentials (accurate to 0.1ms). This latency reflects the conduction speed of the electrical stimulation signal along the dorsal column conduction bundle, indirectly characterizing the physical distance between the electrode and the spinal cord dorsal column and the thickness of the cerebrospinal fluid layer. At the same time, the module performs Hilbert-Huang transform (HHT) on the field potential signal to generate a neural oscillation spectrum of the monitoring segment, identify the rhythmic activity of the theta band (4-8Hz) and the alpha band (8-13Hz), and calculate the energy envelope of the high-frequency oscillation (600-800Hz) evoked by the stimulation, indirectly reflecting the activation state of spinal cord interneurons and the degree of inhibition of the pain conduction pathway. Combining the structural quantitative parameters of each monitoring segment—including the electrode-tissue interface impedance measured in real time by the electrode impedance spectrum detection circuit (affected by fibrous tissue proliferation and electrode surface corrosion), the axial offset of the electrode contact in three-dimensional space (reflecting whether the electrode has shifted due to changes in body position), and the epidural space pressure change estimated by the implanted pressure sensor (increased pressure indicates changes in cerebrospinal fluid dynamics or fibrous scar formation)—the module runs the spinal cord electrical stimulation propagation model. Based on the three-dimensional bioelectrical model of the spine constructed by the finite element method (considering the multilayer non-uniform dielectric properties of vertebrae, intervertebral discs, dura mater, cerebrospinal fluid, and spinal cord gray and white matter), it calculates the three-dimensional distribution of the electric field in the spinal cord tissue and predicts the expected stimulation field distortion and pain coverage area shift caused by electrode displacement, tissue fibrosis, or cerebrospinal fluid loss.

[0096] To further enhance the long-term adaptability of chronic pain management, the module introduces a multi-physics coupling correction mechanism: by integrating a bioimpedance spectroscopy sensor and a temperature sensor, the tissue impedance characteristics and local temperature of the epidural space are collected in real time (the temperature rise is caused by the inflammatory response). Based on the preset impedance-fibrosis-inflammation coupling coefficient matrix, the calculated neurophysiological feature values ​​are corrected for fibrosis (fiber encapsulation leads to increased electrode impedance and increased stimulation threshold) and inflammation drift (inflammatory mediators change the tissue dielectric properties and nerve excitability), generating a corrected spatiotemporal frequency domain coupling feature parameter set. Based on this parameter set, the module's built-in parameter optimization algorithm (which can use simulated annealing or Bayesian optimization) calculates the dual-modal co-optimization parameters for each monitoring segment: For the neural signal acquisition side, it calculates the gain compensation value of the low-noise amplifier for each channel (e.g., automatically increasing the gain when signal attenuation is caused by fiber wrapping), the optimal frequency band selection of the bandpass filter (e.g., dynamically adjusting the high-pass cutoff frequency to preserve spinal cord signals in response to electromyographic interference), and the reference electrode configuration strategy (monopolar, bipolar, or tripolar recording to optimize common-mode suppression); For the electrical stimulation side, it calculates the compensation value of the stimulation current (e.g., appropriately increasing the current to maintain the target electric field strength when impedance increases), the contact polarity configuration (cathode / anode arrangement to focus the electric field or expand the coverage area), and the optimal combination of stimulation frequency and pulse width (for different mechanisms of neuropathic pain and ischemic pain).

[0097] During the continuous pain management phase, the spatiotemporal frequency domain coupling analysis module configures the working mode of the epidural electrode acquisition module according to the adaptive collaborative control strategy parameters. It acquires high-resolution spinal cord field potentials from each monitoring segment and correlates them with the patient's subjective pain reports (via VAS scores input from an external remote control), simultaneously recording time-domain evoked potentials and spectral response data. The module performs multi-dimensional feature extraction on the optimized neural signal dataset: in the time domain, it extracts functional integrity indicators of the spinal cord conduction tract through dynamic changes in evoked potential amplitude and latency; in the frequency domain, it extracts changes in the θ-α band coupling strength related to the pain state through power spectral density analysis, identifying specific frequency band energy abnormalities caused by electrode displacement or tissue fibrosis; in the energy domain, it calculates the activation efficiency index by comparing the energy consumption of stimulation and the energy conversion efficiency of evoked potential responses in each monitoring segment. The module normalizes the extracted multi-dimensional feature parameters, converting them into standardized feature vectors with uniform dimensions.

[0098] Subsequently, the module performs intelligent diagnosis of the electrode-spinal cord coupling state: the standardized feature vector is input into a pre-set threshold library of epidural electrode abnormalities and spinal cord responses. This threshold library is pre-trained by a machine learning algorithm and contains feature fingerprints of various abnormal patterns, including electrode displacement pattern (manifested as asymmetric evoked potential amplitude and increased difference in bilateral response latency), fiber encapsulation pattern (manifested as progressive increase in electrode impedance, rightward shift of the stimulation threshold curve, and better effect of high-frequency stimulation than low-frequency stimulation), cerebrospinal fluid loss pattern (manifested as abnormal expansion of the stimulation field diffusion range and unexpected sensory abnormalities in adjacent segments), and electrode breakage pattern (manifested as open-circuit impedance in multiple channels and failure of stimulation output). The module calculates the Mahalanobis distance or neural network classification probability between the current feature vector and each abnormal pattern to obtain the abnormal similarity index of each monitoring segment. It then applies multi-parameter joint judgment rules (such as multi-feature fusion judgment based on random forest or gradient booster) to comprehensively consider the degree of joint deviation of neural activation efficiency index, evoked potential symmetry, and impedance spectrum feature parameters to determine whether the electrode-spinal cord coupling state is normal. Finally, it generates a joint identification signal containing an abnormal monitoring segment identifier, precise spatial coordinates (corresponding spinal anatomical segment), and defect level (mild: stimulation parameters need fine-tuning; moderate: electrode position needs image-guided adjustment; severe: indication for surgical revision or electrode replacement).

[0099] Based on the generated joint identification signal, the spatiotemporal frequency domain coupling analysis module initiates a hierarchical closed-loop feedback and dynamic calibration mechanism: The module first maps the identifier of the abnormal monitoring segment to a spatial positioning reference frame, obtains the spinal cord segment and pain dermatome description corresponding to the monitoring segment (such as "T10 segment corresponds to the periumbilical pain area on the left" or "L4 segment corresponds to the pain area from below the knee to the medial side of the ankle"), automatically generates a structured programmed diagnostic report, and pushes it to the programming terminal of the pain management physician via Bluetooth module. The report content includes abnormality type diagnosis, visual annotation of spinal segments (overlaid on the patient's preoperative CT / MRI images), and targeted programming suggestions (such as "It is recommended to increase the stimulation current of T10 segment by 0.5mA to compensate for fiber encapsulation" or "The L4 segment electrode may have shifted 2mm towards the head, it is recommended to confirm with spinal X-ray"). Simultaneously, the module sends hierarchical maintenance management instructions to the timing logic control module: if a severe abnormality is detected (such as multiple channels in the monitoring segment simultaneously showing high impedance indicating electrode breakage, or the complete disappearance of evoked potentials indicating severe displacement), the connection between the monitoring segment and the programmable constant current electrical stimulation module is immediately cut off, the electrical stimulation output is suspended to ensure patient safety, and the patient is alerted via an external alarm; if the system is configured with a redundant electrode array (such as a multi-row design of surgical paddle electrodes), the contact switching logic is automatically activated to seamlessly transfer the stimulation and acquisition tasks to an adjacent normal monitoring segment (such as automatically switching to the side row contacts when the middle row fails), maintaining the continuity of pain treatment; if it is mild tissue fibrosis or electrode micro-displacement, the stimulation parameters of the monitoring segment are dynamically adjusted (current compensation, frequency adjustment, or contact recombination) to reconstruct the optimal stimulation field, and a slight tactile feedback is triggered by the external programmer to prompt the patient to adjust their posture (such as avoiding excessive forward flexion that could lead to electrode displacement).

[0100] In addition, the module establishes a long-term adaptive calibration mechanism: it statistically compares the spatiotemporal frequency domain coupling characteristic parameters of the current monitoring segment with the baseline characteristics of the same spinal cord segment and similar implantation duration (such as the fibrosis progression stage at 1 month, 3 months, 6 months, and 12 months post-operation) in the historical programmable database. If the deviation between the current characteristics and the historical average exceeds the preset fault tolerance range (such as the impedance increase rate exceeding 10% per month or the evoked potential amplitude decrease exceeding 30%), it is determined to be a progression of fibrosis in the tissue surrounding the electrode, electrode corrosion, or changes in spinal cord plasticity. The algorithm recalibration process is automatically triggered, which reversely corrects the dielectric parameters of the stimulation field calculation model (updating the conductivity and dielectric constant of the fibrotic tissue), updates the baseline template of the neural oscillation spectrum (adapting to changes in neural plasticity caused by chronic stimulation), and recalculates the adaptive collaborative control strategy parameters (such as gradually transitioning from a sensory abnormality replacement mode to a high-frequency non-sensory abnormality mode). This enables the system to continuously adapt to the processes of tissue healing, fibrous encapsulation formation, electrode aging, and long-term evolution of patient pain characteristics after implantation, ensuring the long-term stability, therapeutic effect, and patient safety of spinal cord electrical stimulation pain management.

[0101] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, characterized in that, include: Bluetooth module, multiplexer module, EEG acquisition module, programmable constant current stimulation module, and timing logic control module; The Bluetooth module and the programmable constant current electrical stimulation module are respectively connected to the multiplexed switch module; the multiplexed switch module is connected to the electrodes; the EEG acquisition module is connected to the multiplexed switch module and the electrodes; the EEG acquisition module is connected to the Bluetooth module; the timing logic control module is connected to the Bluetooth module. The system receives instructions via Bluetooth, and the timing logic control module coordinates the operation of each module to ensure that the EEG acquisition module and the programmable constant current stimulation module do not operate simultaneously. The multiplexer switches the working mode connected to the electrodes according to the instructions from the Bluetooth module to perform data transmission and electrical stimulation applications. The EEG acquisition module processes the EEG signals and transmits the data back to the Bluetooth module for analysis or transmission.

2. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 1, characterized in that, The EEG acquisition module is connected to the Bluetooth module via an SPI bus.

3. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 2, characterized in that, The Bluetooth module includes a Bluetooth communication chip.

4. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 3, characterized in that, The multiplexed switch module includes a TI analog switching chip; the TI analog switching chip integrates several independently controlled single-pole double-throw switches.

5. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 4, characterized in that, The EEG acquisition module includes an EEG acquisition chip, which integrates a multi-channel electrophysiological amplifier array, a 16-bit ADC, and an SPI interface; the SPI interface is connected to the SPI bus.

6. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 5, characterized in that, The programmable constant current stimulation module includes a programmable constant current stimulation chip, which integrates multiple constant current sources and controls the current of each channel through an 8-bit DAC.

7. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 6, characterized in that, The programmable constant current electrical stimulation chip uses a four-wire SPI interface and is configured by the Bluetooth module or the timing logic control module by reading and writing multiple registers corresponding to each channel.

8. The multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 7, characterized in that, In the acquisition mode, the electrodes are connected to the multiplexed ...

9. A control method for a multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit, characterized in that, The method uses the multichannel EEG acquisition and electrical stimulation multiplexing electrode circuit as described in any one of claims 1 to 8, including: The Bluetooth module receives instructions, and the timing logic control module coordinates the work of each module to ensure that the EEG acquisition module and the programmable constant current stimulation module do not operate simultaneously. The multiplexer switches the working mode connected to the electrodes according to the instructions from the Bluetooth module to perform data transmission and electrical stimulation applications. The EEG acquisition module processes the EEG signals and transmits the data back to the Bluetooth module for analysis or transmission.

10. The control method for the multi-channel EEG acquisition and electrical stimulation multiplexing electrode circuit according to claim 9, characterized in that, Also includes: In acquisition mode, weak EEG signals are processed, amplified, and filtered using the highly integrated RHD2132 chip; in stimulation mode, the SX-32 chip provides bipolar constant current stimulation, and high-voltage isolation technology is used for signal isolation.

Images