Method and device for fatigue real-time monitoring and negative feedback regulation and control, medium and program product

Abstract

The invention provides a fatigue real-time monitoring and negative feedback regulation method and device, a medium and a program product, and relates to the field of intelligent medical treatment. The method comprises the following steps: acquiring electroencephalogram and electro-oculogram signals and blood oxygen saturation data of a subject; the electroencephalogram and electro-oculogram signals comprise signals of FP1 and FP2 sites; electroencephalogram signals and electro-oculogram signals are separated from the electroencephalogram electro-oculogram signals; calculating fatigue based on the electroencephalogram signal, the electro-oculogram signal and the blood oxygen saturation data; and when the fatigue degree reaches a set threshold value, the transcranial electrical stimulation is performed on the subject by triggering an electrode positioned on the prefrontal cortex. According to the method, the brain cognition efficacy is improved through the multi-mode stimulation normal form, and a key technical system for brain function enhancement is perfected.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical treatment, and more specifically, to a method, device, medium and program product for real-time fatigue monitoring and negative feedback regulation. Background Art

[0002] Fatigue refers to the state in which the body's labor efficiency tends to decline due to long-term or excessive physical or mental labor under certain environmental conditions. In medicine, fatigue can be divided into physiological fatigue and psychological fatigue according to the nature of fatigue. The evaluation of fatigue status can be carried out through subjective and objective methods. The subjective evaluation method mainly uses subjective questionnaires, self-recording forms, sleep habit questionnaires and Stanford sleep scales to evaluate the fatigue level of the subjects. The objective evaluation method mainly starts from a medical perspective, using auxiliary tools such as medical instruments and equipment to test the changes in certain indicators of the subjects' human behavior, physiology, and biochemistry, so as to determine their fatigue level. Although the subjective evaluation method is simple, direct, low-cost, and has the advantages of no interference with task completion and easy acceptance, it is a widely used method for evaluating fatigue. However, this method is difficult to quantify the level and degree of fatigue, and because everyone's understanding is significantly different, the results are often unsatisfactory.

[0003] In complex working environments, personnel who perform high-load tasks or high-alertness positions, such as astronauts and pilots, need to maintain long-term, efficient brain information processing and cognitive abilities to ensure stable and efficient work performance. However, due to problems such as mental overload, tension, and ultra-intensive work, it is difficult to maintain high mental activity efficiency, and mental cognition is prone to fatigue, resulting in decreased brain cognitive functions such as alertness, memory, and analytical decision-making abilities. At present, the effective targets for brain function enhancement intervention are unclear, and the enhancement devices are not portable, and there is a lack of brain cognitive efficiency maintenance and enhancement measures that are highly specific, compliant, and effective. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention provides a method, device, medium and program product for real-time fatigue monitoring and negative feedback regulation; the method of the present invention improves brain cognitive efficiency by using a multimodal stimulation paradigm and improves the key technical system for brain function enhancement.

[0005] In a first aspect, the present application discloses a method for real-time fatigue monitoring and negative feedback control, the method comprising:

[0006] 101. Obtaining the EEG signal and blood oxygen saturation data of the subject; the EEG signal includes signals of FP1 and FP2 sites; separating the EEG signal and the EO signal from the EEG signal; 102. Calculating the fatigue level based on the EEG signal, the EO signal and the blood oxygen saturation data; 103. When the fatigue level reaches a set threshold, performing transcranial electrical stimulation on the subject by triggering electrodes located in the prefrontal cortex.

[0007] In some embodiments, the prefrontal cortex includes stimulation sites located at any two or more of the following: FP1, FP2, F3, Fz;

[0008] Optionally, the stimulation sites of the transcranial electrical stimulation also include: the vagus nerves of the left auricle and the right auricle;

[0009] Optionally, the blood oxygen saturation data is collected by a blood oxygen saturation collection sensor.

[0010] In some embodiments, the method for separating the EEG signal and the EOG signal from the EEG signal includes: calculating the long time difference of the EEG signal to measure the amplitude of the EOG fluctuation; obtaining the amplitude envelope of the EEG signal by low-pass filtering, and detecting the starting and ending points of the EOG signal by a double threshold method; decomposing the EEG signal, and introducing the Birgé-Massart strategy to adaptively determine the threshold of the wavelet coefficient; accurately estimating the EOG signal by wavelet reconstruction, and decoupling the estimated EOG signal from the original EEG signal to separate the EEG signal from the EOG signal.

[0011] In some embodiments, the method for calculating the fatigue level includes:

[0012] Inputting the EEG signal, EOG signal and blood oxygen saturation data into a real-time fatigue state assessment model to calculate fatigue; the real-time fatigue state assessment model is constructed based on a few-channel fatigue decoding algorithm of Transformer and attention mechanism;

[0013] Optionally, the method for constructing the real-time fatigue state assessment model includes:

[0014] Obtain a training set to execute the PVT paradigm dataset; input the dataset into the CNN structure to extract the local features of the EEG signal, use Transformer to extract the long-distance correlation of the EEG signal, and pay attention to the local and global features of the data; introduce channel attention mechanism and spatial attention mechanism to measure the importance weights of different channels and different spatial positions of features for the decoding task; adopt a feature fusion strategy to extract feature maps containing multiple layers of information; input the feature map into the classifier for continuous optimization to obtain a constructed real-time fatigue status assessment model.

[0015] In some embodiments, the transcranial electrical stimulation scheme includes any one or more of the following: stimulation site, stimulation intensity, stimulation pulse waveform, stimulation frequency, and stimulation time.

[0016] Optionally, the frequency calculation method includes: taking any two last points just before the falling edge of the waveform, calculating the difference between the data frame indexes of the two points, and the frequency is 250*

[0017] (Number of falling edges between two points) / I d .

[0018] In some embodiments, the method further comprises determining a predicted outcome of the subject being in a fatigued state, a transitional state, or an awake state.

[0019] In some embodiments, the EEG / EOG signals are preprocessed signals, and the preprocessing method includes: eliminating high-frequency interference, differential amplification, and conversion into digital signals.

[0020] The second aspect of the present application discloses a computer device, which includes: a memory and a processor; the memory is used to store a computer program; and the processor executes the computer program to implement the steps of the above method.

[0021] A third aspect of the present application discloses a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0022] A fourth aspect of the present application discloses a computer program product, including a computer program, which implements the steps of the above method when executed by a processor.

[0023] This application has the following beneficial effects:

[0024] This application innovatively discloses a method for real-time monitoring of fatigue and negative feedback regulation. The method changes the location of data collection, collects EEG and EOG signals based on FP1 and FP2 electrodes, and then separates EEG and EOG signals through a built-in algorithm. Based on EEG signals, EOG signals, and blood oxygen saturation data, real-time monitoring of the fatigue state of the human body is achieved. When the fatigue of the subject reaches the threshold, the electric stimulation module is automatically started to negatively feedback and electrically stimulate the prefrontal cortex FP1, FP2, Fz, F3 brain areas and the auricular vagus nerve to relieve the fatigue of the subject. Thus, a real-time fatigue monitoring and electric stimulation closed-loop system is formed to form a monitoring and stimulation closed-loop system to complete the biological self-feedback regulation function. The stimulation parameters are adjusted according to the state to complete a cycle, and the biofeedback intervention effect is enhanced through adaptive parameter matching. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 is a schematic diagram of a method flow chart provided by the first aspect of an embodiment of the present invention;

[0027] Figure 2 It is a schematic diagram of a system for real-time monitoring and negative feedback control of fatigue provided by an embodiment of the present invention;

[0028] Figure 3 is a schematic diagram of a computer device provided by an embodiment of the present invention;

[0029] Figure 4 is a schematic diagram of the architecture of an exemplary computing device provided by an embodiment of the present invention;

[0030] Figure 5 is a schematic diagram of a storage medium provided by an embodiment of the present invention;

[0031] Figure 6 is a schematic diagram of a prototype provided by an embodiment of the present invention;

[0032] Figure 7 is a schematic diagram of electrode distribution inside a prototype provided by an embodiment of the present invention;

[0033] Figure 8 is a front view of the electrode distribution inside the prototype provided by an embodiment of the present invention;

[0034] Fig. 9 is a functional block diagram of a prototype circuit provided by an embodiment of the present invention;

[0035] Fig.10 is a schematic diagram of device data transmission provided by an embodiment of the present invention;

[0036] Fig.11 is a self-test voltage value waveform diagram of the FP1 channel provided by an embodiment of the present invention;

[0037] Fig.12 It is a schematic diagram of a method for constructing a real-time fatigue state assessment model provided by an embodiment of the present invention.

[0038] In the figure, 1. Head-mounted device; 11. Forehead detection structure; 12. Fixing structure; 13. Blood oxygen and blood temperature collection sensor; 14. Electrical stimulation ear clip; 15. FP1 electrode; 16. F3 electrode; 17. FP1 and F3 protective covers; 18. FP2 electrode; 19. FP2 protective cover; 110. FZ electrode; 111. FZ and blood oxygen protective covers; 112. Protective cover fixing buckle; 2. Main control device. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0040] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0042] Figure 1 The present invention provides a method for real-time fatigue monitoring and negative feedback control. Specifically, the method includes the following steps:

[0043] 101: Obtaining an EEG signal and blood oxygen saturation data of a subject; the EEG signal includes signals at FP1 and FP2 sites; and separating the EEG signal and the EEG signal from the EEG signal.

[0044] In some embodiments, the term "subject" or "test subject" or "test sample" used herein refers to any animal (e.g., mammal), including but not limited to humans, non-human primates, rodents, etc., which will be the recipient of a specific treatment. Generally, the terms "subject" and "patient" are used interchangeably herein when referring to human subjects. Preferably, the subject is a human.

[0045] In some embodiments, the blood oxygen saturation data is collected by a blood oxygen saturation collection sensor.

[0046] In some embodiments, the method for separating the EEG signal and the EOG signal from the EEG signal includes: calculating the long time difference of the EEG signal to measure the amplitude of the EOG fluctuation; obtaining the amplitude envelope of the EEG signal by low-pass filtering, and detecting the starting and ending points of the EOG signal by a double threshold method; decomposing the EEG signal, and introducing the Birgé-Massart strategy to adaptively determine the threshold of the wavelet coefficient; accurately estimating the EOG signal by wavelet reconstruction, and decoupling the estimated EOG signal from the original EEG signal to separate the EEG signal from the EOG signal.

[0047] In some embodiments, the EEG / EOG signals are preprocessed signals, and the preprocessing method includes: eliminating high-frequency interference, differential amplification, and conversion into digital signals.

[0048] 102: Calculating fatigue level based on the electroencephalogram signal, the electrooculogram signal and the blood oxygen saturation data;

[0049] In some embodiments, the method for calculating the fatigue degree includes: inputting the EEG signal, EOG signal and blood oxygen saturation data into a real-time fatigue state assessment model to calculate the fatigue degree; the real-time fatigue state assessment model is constructed based on a few-channel fatigue decoding algorithm of Transformer and attention mechanism; the real-time fatigue state assessment model is defined as an MFFN model.

[0050] Optionally, the method for constructing the real-time fatigue status assessment model includes: obtaining a data set for executing the PVT paradigm of the training set, specifically the AINS-FA2 data set; inputting the data set into the CNN structure to extract local features of the EEG signal, and using the Transformer to extract the long-distance correlation of the EEG signal, while paying attention to the local and global features of the data; introducing a channel attention mechanism and a spatial attention mechanism to measure the importance weights of different channels and different spatial positions of features for the decoding task; adopting a feature fusion strategy to make the model pay more attention to information that is more helpful for the decoding task, and extract feature maps containing multiple layers of information; inputting the feature map into the classifier for continuous optimization to obtain a constructed real-time fatigue status assessment model. The algorithm can achieve a recognition accuracy of 86.7% on data with few channels (containing only FP1 and FP2 electrodes). Specifically, Fig.12 shown.

[0051] Among them, the PVT paradigm is a psychomotor vigilance task. The PVT task is a simple key-pressing task, which has the characteristics of simple operation, no practice effect, and high sensitivity to sleep loss and circadian rhythm changes. During the experiment, a red circular pattern will be randomly displayed on the screen, and the subjects need to quickly press the space bar to respond to the appearance of the pattern. This task is not only highly repetitive, but also requires the subjects to maintain a high degree of mental concentration throughout the experiment. The repetitiveness of this task will cause the subjects' cognitive abilities to gradually decline, thereby inducing fatigue in the subjects. During the experiment, the reaction time of the subjects from each time the screen displays a pattern to the time the subjects press the space bar to respond is recorded. The changes in the subjects' reaction time during the entire experiment can assist in verifying the process of the subjects entering a fatigue state.

[0052] For the entire experimental process, a sliding window is used to calculate the average reaction time within the window. The sliding window is set to cover 10 experiments each time, and the sliding step is 1 experiment, and the average reaction time change graph of the subjects is obtained. As the experiment progresses, the average reaction time of the subjects increases, and the degree of fatigue of the subjects gradually deepens.

[0053] 103: When fatigue reaches a set threshold, transcranial electrical stimulation is performed on the subject by triggering electrodes located in the prefrontal cortex;

[0054] In some embodiments, the prefrontal cortex includes any two or more stimulation sites located at the following locations: FP1, FP2, F3, Fz; the materials used for the FP1, FP2, F3, and Fz electrodes are the same, but their functions are different.

[0055] Optionally, the stimulation sites of the transcranial electrical stimulation also include: the vagus nerve of the left auricle and the right auricle; relieving the subject's fatigue state and restoring cognitive function.

[0056] In some embodiments, the transcranial electrical stimulation scheme includes any one or more of the following: stimulation site, stimulation intensity, stimulation pulse waveform, stimulation frequency, and stimulation time.

[0057] Optionally, the frequency calculation method includes: taking any two last points just before the falling edge of the waveform, calculating the difference between the data frame indexes of the two points, and the frequency is 250*

[0058] (Number of falling edges between two points) / I d .

[0059] In some embodiments, the method further comprises determining a predicted outcome of the subject being in a fatigued state, a transitional state, or an awake state.

[0060] In some embodiments, the set threshold is obtained through training of training set samples, which may be a specific threshold or an interval range. The specific form is not specifically limited in this embodiment.

[0061] In some embodiments, the electrodes and blood oxygen saturation acquisition sensors in the present application are arranged Figure 6 , Figure 7 and Figure 8 In the prototype, the prototype includes a head-mounted device, and the head-mounted device includes a connected forehead detection structure and a fixing structure; the forehead detection structure is provided with electrodes for collecting EEG and electrical stimulation, a blood oxygen saturation collection sensor and an electrical stimulation ear clip, the electrodes and the blood oxygen saturation collection sensor are in close contact with the forehead skin, and the electrical stimulation ear clip is clamped in the vagus nerve area of ​​the left auricle; the electrodes include at least FP1 electrodes and FP2 electrodes for collecting EEG signals of FP1 and FP2 brain regions and negative feedback electrical stimulation of the prefrontal cortex FP1 and FP2 brain regions, and the blood oxygen saturation collection sensor is located between the FP1 electrode and the FP2 electrode.

[0062] In some embodiments, the EEG signal collected by the FP1 electrode and the FP2 electrode includes an EOG signal, and the EOG signal is decoupled from the collected EEG signal using a decoupling algorithm.

[0063] In some embodiments, the electrodes further include at least an FP3 electrical stimulation electrode and an FZ electrical stimulation electrode located in the FP3 and FZ brain regions. In the front view state, the line connecting the center of the FP3 electrical stimulation electrode and the center of the FZ electrical stimulation electrode is parallel to the horizontal line; the line connecting the center of the FP3 electrical stimulation electrode and the center of the FP1 electrical stimulation electrode is perpendicular to the horizontal line, and the FZ electrical stimulation electrode is located at the upper end of the blood oxygen saturation acquisition sensor.

[0064] In some embodiments, the forehead detection structure is also provided with a protective cover for protecting the electrode, and the protective cover is fixed to the forehead detection structure by a fixing buckle; specifically, the protective cover mainly includes FP1 and FP3 protective covers, Fz and blood oxygen protective covers, and FP2 protective cover.

[0065] In some embodiments, the forehead detection structure further includes a reference electrode, which is attached to the skin of the left and right ears through a conductive gel;

[0066] Optionally, the inner surface of the forehead detection structure is consistent with the curvature of the forehead.

[0067] In some embodiments, the fixing structure is configured as an arc-shaped structure that is consistent with the curvature of the back of the head; a protective pad is provided on the inner surface of the fixing structure to ensure the wearing comfort of the subject.

[0068] In some embodiments, the arc-shaped structure is also provided with a knob for adjusting tightness to suit different head circumferences and ensure comfort.

[0069] In some embodiments, the head-mounted device also includes an anti-slip structure, and both ends of the anti-slip structure can rotate relative to the forehead detection structure or the fixed structure; the two ends of the anti-slip structure are symmetrically arranged on both sides of the forehead detection structure or the fixed structure or the connection between the two; the anti-slip structure is arranged in an integrated headband shape or a separate arc shape.

[0070] Optionally, a protective pad is provided on the inner surface of the anti-fall-off structure to improve wearing comfort and prevent falling off.

[0071] In some embodiments, the device also includes a main control device and a host computer, the head-mounted device is connected to the main control device, and the main control device and the host computer are connected to each other.

[0072] In a specific implementation scheme, the main control device in the head-mounted device includes an EEG acquisition module, an electrical stimulation module, and a data conversion module; the host computer is installed with data processing software (EEG signal processing, fatigue decoding algorithm, and electrical stimulation self-feedback), the data processing software receives EEG data and blood oxygen data, and collects, displays, and calculates in real time, and calculates the fatigue level of the subject in real time. When the subject reaches the fatigue threshold, the negative feedback electrical stimulation module applies electrical pulse stimulation to the subject to relieve the subject's fatigue state and restore cognitive function.

[0073] In another specific embodiment, the connection between the head-mounted device and the main control device, and between the main control device and the host computer includes a communication connection, which is a connection method that forms communication between connected devices through model transmission interaction. The communication connection includes wired connection and wireless connection. Wired connection is such as connection through a data cable, and wireless connection is such as connection through Bluetooth and wifi. Electrical connection is one of the wired connections. Electrical connection can be understood as a form of connection between different components in the circuit structure through physical lines such as PCB copper foil or wires that can transmit signals. It is mainly for completing data interaction.

[0074] In one embodiment, the protective pad is a sponge pad or a silicone pad, which is not limited here.

[0075] Figure 3 is a schematic diagram of a computer device provided by an embodiment of the present invention, such as Figure 3 As shown, the device 2000 may include: one or more processors 2010, and one or more memories 2020; wherein the memories store computer-readable codes, and when the computer-readable codes are run by the one or more processors, the method described above may be executed.

[0076] The processor in this embodiment can be an integrated circuit chip with signal processing capabilities. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, operations and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can be an X86 architecture or an ARM architecture.

[0077] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the disclosed embodiments are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0078] For example, the method or device according to the embodiment of the present disclosure may also be implemented by Figure 4 The architecture of the computing device 3000 shown in FIG. Figure 4As shown, the computing device 3000 may include a bus 3010, one or more CPUs 3020, a read-only memory (ROM) 3030, a random access memory (RAM) 3040, a communication port 3050 connected to a network, an input / output component 3060, a hard disk 3070, etc. The storage device in the computing device 3000, such as ROM 3030 or hard disk 3070, may store various data or files used for processing and / or communication of the method provided by the present disclosure and program instructions executed by the CPU. The computing device 3000 may also include a user interface 3080. Of course, Figure 4 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 4 One or more components of a computing device are shown.

[0079] The embodiment of the present invention also provides a computer-readable storage medium, such as Figure 5 As shown, it is a schematic diagram of a storage medium 4000 provided in an embodiment of the present invention, and a computer readable instruction 4010 is stored on the computer storage medium 4020. When the computer readable instruction 4010 is executed by a processor, the method according to the embodiment of the present disclosure described with reference to the above figures can be executed. The computer readable storage medium in the embodiment of the present disclosure can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM) or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory. It should be noted that the memory of the methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0080] The embodiments of the present disclosure further provide a computer program product or system, including a computer program, which implements the steps of the above method when executed by a processor.

[0081] In some embodiments, this embodiment also discloses a system for real-time fatigue monitoring and negative feedback control, such as Figure 2 As shown, the system comprises:

[0082] The data acquisition module 201 is used or configured to acquire the EEG signal and blood oxygen saturation data of the subject; the EEG signal includes the signal of the FP1 and FP2 sites; the EEG signal and the EEG signal are separated from the EEG signal;

[0083] A fatigue degree calculation module 202, used for or configured to calculate fatigue degree based on the EEG signal, the EOG signal and the blood oxygen saturation data;

[0084] The control judgment module 203 is used or configured to perform transcranial electrical stimulation on the subject by triggering electrodes located in the prefrontal cortex when the fatigue level reaches a set threshold; Specific embodiment:

[0086] 1. Principle prototype functional design: Real-time monitoring of brain fatigue and self-feedback cognitive enhancement The principle prototype is an advanced wearable intelligent principle prototype. The prototype can accurately collect, process and display in real time the EEG signals, electrooculogram and blood oxygen saturation and other physiological signals of the subjects' FP1 and FP2 brain areas, and couple these signals through the built-in algorithm to monitor the fatigue state of the human body in real time. When the subject's fatigue reaches the fatigue threshold, the electrical stimulation module is automatically started to negatively feedback and electrically stimulate the prefrontal cortex FP1, FP2, Fz, F3 brain areas and the auricular vagus nerve to relieve the subject's fatigue. Thus, a real-time fatigue monitoring and electrical stimulation closed-loop system is formed to form a monitoring and stimulation closed-loop system to complete the biological self-feedback regulation function. The device optimizes the human-computer interaction experience, promotes the deep integration of man and machine, and is suitable for fatigue training in a variety of scenarios.

[0087] The prototype modes include:

[0088] Device form: intelligent monitoring headband; functions and parameters: 8 light sources and 2 detectors fNIR, left and right EEG; synchronous acquisition of fNIR and EEG data streams; maximum sampling rate of 240Hz; data acquisition mode: Type C or Bluetooth; real-time data acquisition, recording left and right EEG α, β, γ, θ power spectra and changes in brain oxygen content in left and right brain regions.

[0089] Device form: smart tablet or laptop; functions and parameters: data interface (USB, Bluetooth); signal processing, analysis, and recording; brain and mental fatigue state algorithm; automatic generation of stimulation schemes and parameters (location, intensity, frequency, waveform, duration, etc.); automatic feedback mode: after starting the stimulation, real-time assessment of fatigue status, adjustment of stimulation parameters according to the status, completion of a cycle, and enhancement of the biofeedback intervention effect through adaptive parameter matching.

[0090] Device form: headgear / headband stimulation electrode; functions and parameters: selection of brain area stimulation site; stimulation intensity modulation (0-10mA); stimulation pulse waveform selection (square wave, triangle wave); stimulation frequency adjustment and duty cycle (0-50HZ); stimulation time (0-30min).

[0091] 2. Composition of the prototype: The prototype consists of three parts: a head-mounted device, a main control module, and a host computer. The head-mounted device consists of a blood oxygen module, an EEG electrode, and a stimulation electrode; the main control module gathers physiological signals and exchanges data with the host computer through USB. The main control module consists of an EEG acquisition module, an electrical stimulation module, a data conversion module, etc. The host computer is installed with self-developed EEG signal processing, fatigue decoding algorithm, and electrical stimulation self-feedback software; this module is used to collect, display, and calculate the hardware data in real time, and calculate the fatigue level of the subject in real time. When the subject reaches the fatigue threshold, the negative feedback electrical stimulation module applies electrical pulse stimulation to the subject to relieve the subject's fatigue state and restore cognitive function.

[0092] 2.1 Prototype appearance: Figure 6 As shown, this is a schematic diagram of the prototype of the principle of real-time monitoring of brain fatigue and self-feedback cognitive enhancement.

[0093] 2.2 Principle prototype structure diagram:

[0094] 2.2.1 Principle prototype structure overview: Figure 6 shown.

[0095] 2.2.2 Principle prototype head-mounted device: After the principle prototype head-mounted device is designed, the head-mounted part that fits tightly to the forehead is 3D printed using silicone (model 8400, Shore hardness 40 degrees). The use of silicone in this part can ensure that the EEG acquisition electrodes, blood oxygen saturation and transcranial electrical stimulation electrodes are well embedded in the head-mounted device, and can ensure the comfort of the subject's skin. The back of the head-mounted device is designed to be a sponge backing to ensure the comfort of the subject. Figure 7 As shown in the figure, it is the electrode distribution diagram inside the prototype; Figure 8 Shown is a front view of the prototype at the forehead.

[0096] 2.2.3 Main control module of the prototype: After the main control module of the prototype is designed, it is 3D printed using UV resin (ultraviolet photosensitive resin). After polishing, it is painted with milky white matte paint, which not only improves the aesthetics, but also effectively protects the circuit board.

[0097] 3. Introduction of each module of the principle prototype:

[0098] 3.1 Overall process and deployment structure diagram of the principle prototype: Fig. 9 As shown, this is the functional block diagram of the prototype circuit.

[0099] 3.1.1 Host computer: 1) Host computer configuration requirements: CPU: no less than i5 quad-core; memory: no less than 8G; hard disk: no less than 512M; display: no less than 1080P; WIN10 system or higher version; 2) Self-feedback software: used to collect, display, and calculate the hardware data in real time, and output the stimulation parameters in a self-feedback manner according to the calculated fatigue level, adjust the user's fatigue level, and record the collection, calculation and self-feedback data in the process to facilitate researchers to study the self-feedback process.

[0100] 3.1.2 Main control module: 1) EEG module; 1.1) EEG acquisition: The module collects EEG signals by using silver chloride dry electrodes and high-precision digital-to-analog conversion front end. The silver chloride dry electrodes in the head-mounted device are connected to the module to input the EEG signals into the module. The EEG signal first passes through a pre-low-pass RC filter circuit to eliminate high-frequency interference in the input signal. Subsequently, the differential amplifier at the analog-to-digital conversion front end differentially amplifies the EEG signal and converts the analog signal into a digital signal through two 24-bit synchronous digital-to-analog converters. In addition, the module introduces a right leg drive circuit to reverse the common-mode signal of the input signal and then feed it to the human body to eliminate common-mode interference.

[0101] 1.2) Electrooculogram (EOG) acquisition: Since the silver chloride dry electrodes in the head-mounted device are arranged on the forehead, the EOG signals will be coupled into the collected EEG signals. Therefore, the module decouples the EOG signals from the collected EEG signals through the decoupling algorithm in the background program.

[0102] The decoupling algorithm is based on the long-time difference amplitude envelope and wavelet transform. The long-time difference of the EEG signal is calculated to measure the degree of electrooculogram fluctuation. Then the amplitude envelope of the signal is obtained by low-pass filtering, and the start and end points of the electrooculogram are accurately detected by the double threshold method. Subsequently, the EEG signal is decomposed by the sym5 wavelet, and the Birgé-Massart strategy is introduced to adaptively determine the threshold of the wavelet coefficient. The electrooculogram signal is accurately estimated through wavelet reconstruction, and the estimated electrooculogram signal is decoupled from the original EEG signal to achieve the separation of the electrooculogram signal and the EEG signal.

[0103] 2) Electrical stimulation module: The electrical stimulation module consists of a power supply, three independent power supplies, a boost circuit, an electrical stimulation main circuit, a control circuit, and a communication circuit. The power supply and control instructions of this module are provided by the USB interface. According to the instructions (see the electrical stimulation module manual for detailed instruction sets), the three independent power supplies of the output module can be controlled, and three waveforms of DC, square wave, and triangle wave can be output as needed.

[0104] 3) Data conversion module: bridges the communication data between the EEG acquisition module and the electrical stimulation module to a USB interface for data interaction with the host computer.

[0105] 3.1.3 Head-mounted device: The head-mounted device consists of a blood oxygen module, EEG electrodes, electrical stimulation electrodes, a shell, etc.

[0106] 1) Blood oxygen module: The module uses a multi-spectral physiological data measurement sensor to collect blood oxygen saturation. The sensor integrates a multi-wavelength LED light source and a high-sensitivity photodetector. The LED light source emits red light and infrared light of specific wavelengths to penetrate the skin and tissues and irradiate the hemoglobin in the blood. Due to the different absorption rates of oxygenated and deoxygenated hemoglobin to these light waves, the photodetector in the module can detect the difference in light intensity after absorption by the blood. These light intensity signals are converted into electrical signals and processed to calculate the blood oxygen saturation by the ratio method.

[0107] 2) EEG electrodes: The electrodes for collecting EEG signals from the FP1 and FP2 brain regions use silver chloride electrodes (shared with the electrical stimulation electrodes) and electrode wires wrapped with shielded wires to reduce signal loss and improve the accuracy of the system's EEG acquisition.

[0108] 3) Electrical stimulation electrodes: Electrodes for electrical stimulation of FP1, FP2, Fz and F3 brain regions (FP1, FP2 and EEG acquisition share two electrodes) silver chloride electrodes (FP1, FP2 and electrical stimulation share electrodes), and ear clip dry electrodes for vagus nerve stimulation of the auricle;

[0109] 4) Shell: The forehead part uses soft silicone to fix the silver chloride electrode and blood oxygen module, and the back of the head uses an adjustable tightness device to adapt to the use of different people;

[0110] 3.2 Principle prototype usage process: The principle prototype usage process is described as follows:

[0111] Connect the device to the host computer; 1) Connect the device to the computer; 2) Wear the prototype headset and adjust the tightness to make the EEG acquisition and electrical stimulation electrodes in good contact with the corresponding brain area of ​​the subject; 3) Start the software and connect the EEG acquisition and electrical stimulation serial ports; 4) Set the fatigue threshold, electrical stimulation parameters, and create a subject document; 5) Start the EEG acquisition module to display the EEG, electrooculogram and blood oxygen saturation in real time; 6) The built-in fatigue algorithm couples the calculation of physiological indicators and displays the fatigue status in real time; 7) When the fatigue reaches the set threshold, the electrical stimulation is automatically started. After the electrical stimulation ends, the biological signal detection module is automatically turned on to calculate the subject's fatigue based on the real-time data. When the fatigue reaches the threshold, a new round of electrical stimulation is started.

[0112] 3.3 Data transmission communication protocol and data description: 3.3.1 Communication protocol;

[0113] 3.3.1.1 Confirm the port number: The acquisition hardware module (hereinafter referred to as the "device") uses the serial communication protocol. As shown in the figure below, connect the device to the computer via a USB cable, open the "Device Manager", and find the serial port number corresponding to the device in the "Port (COM and LPT)" item (the main control module uses CP2102 as the USB to serial port chip, which will be reflected in the device name). If you are not sure which one it is, it is recommended to unplug the USB cable and see which port disappears, which is the corresponding port.

[0114] 3.3.1.2 Detailed information of serial port configuration is shown in Table 1:

[0115] Table 1

[0116] Serial port properties value Baud rate 961200 Data bits 8 Check digit none Stop bits 1 Hardware flow control none

[0117] 3.3.1.3 The equipment instruction table is shown in Table 2:

[0118] Table 2

[0119] instruction Function EEG_SPO_ON Normal acquisition mode, start acquisition OFF Stop acquisition in any mode self_test_ON Entering the self-test mode, the device sends a self-test signal

[0120] Note: The command does not include line feed or carriage return.

[0121] 3.3.2 Equipment data format:

[0122] 3.3.2.1 Device data transmission test: It is recommended to download the "Serial Port Debugging Assistant" in the Microsoft Store to perform the device data transmission test. Set the relevant parameters according to the red box on the left in the figure below, and then send the "EEG_SPO_ON" command. The data sent by the device can be observed on the right. The red box on the right in the figure is a data frame. Fig.10 shown.

[0123] 3.3.2.2 Data frame format and data conversion:

[0124] The data sent by the device consists of several data frames. A data frame is as follows Fig.10 As shown, it includes 11 bytes, which represent the frame header, function code, data length, data bit 1 to data bit 6, and frame tail in sequence. Each data frame transmits 6 data bits (6 bytes), and the transmission rate is about 250 data frames per second. The specific format is shown in Table 3 (all in hexadecimal):

[0125] Table 3

[0126]

[0127] Function code 01 represents two-channel EEG data. The data length is fixed at 6 bytes, that is, the 6 bytes represented by data bits 1 to 6. Among them, data bits 1 to 3 represent FP1 channel data, and data bits 4 to 6 represent FP2 channel data.

[0128] Taking the data frame in the above table as an example, the decimal values corresponding to data bits 1, 2, and 3 are represented as A1, B1, and C1 respectively. First, calculate:

[0129] T1 = 65536 * A1 + 256 * B1 + C1

[0130] If 0 ≤ A1 ≤ 127, then:

[0131] V1 = 0.0401 * T1

[0132] Conversely, if 127 < A1 ≤ 255, then:

[0133] V1 = 0.0401 * (T1 - 16777215)

[0134] The calculated V1 is the EEG signal voltage value corresponding to the FP1 channel.

[0135] The calculation method of the EEG signal voltage value corresponding to the FP2 channel is the same as that of the FP1 channel.

[0136] Taking the data frame in the above table as an example, A1 = 7, B1 = 213, C1 = 223, 0 ≤ A1 ≤ 127, then V1 = 0.0401 * (65536 * 7 + 256 * 213 + 223) = 20591.4703 (unit: μV).

[0137] Similarly, A2 = 3, B2 = 112, C2 = 81, 0 ≤ A2 ≤ 127, V2 = 0.0401 * (65536 * 3 + 256 * 112 + 81) = 9036.9761 (unit: μV).

[0138] Function code 02 represents blood oxygen data. Take the decimal data corresponding to data bit 6 to obtain the blood oxygen saturation at this time. In Table 4, data bit 6 is 0x62 = 98, so the blood oxygen saturation at this time is 98.

[0139] Table 4

[0140]

[0141] The specific connection status is shown in Table 5: C280 is converted to binary as 1100001010000000;

[0142] Table 5

[0143]

[0144] Note that the data bits corresponding to the above [N, P1, P2] represent the connection status of the reference electrode, FP1 electrode, and FP2 electrode respectively. When a bit = 1, the electrode is in the off state. For example, C280 indicates that the FP1 and FP2 electrodes are off. Note: If no electrode is off, function code 03 will not appear.

[0145] The function code of the self-test signal is also 01. The device will send self-test data only after sending the "self_test_ON" command.

[0146] According to the data conversion method introduced above, collect and convert no less than 2500 data frames, analyze the voltage value data of 2 channels and draw a graph, which should be a square wave with a frequency of 0.9765Hz (±4%), an amplitude of 3360μV (±3%), and a duty cycle of 50% (±3%), such as Fig.11 As shown, this is the self-test voltage value waveform of the FP1 channel. The horizontal axis is the data frame index, and the vertical axis is the voltage value. The data frame index of point A is 111, and the voltage value is 1641.27μV. The data frame index of point B is 751, and the voltage value is -1724.01μV. The same is true for point CE.

[0147] The amplitude calculation formula is (waveform maximum value - waveform minimum value), so the measured amplitude is 1642.55-(-1724.01)=3366.56μV, and the error with the nominal value is (3366.56-3360) / 3360*100%=0.195%, which is much less than ±3%;

[0148] The frequency is calculated by taking any two points just before the falling edge of the waveform and calculating the difference in the data frame index of the two points. d , the frequency is 250*(the number of falling edges between two points) / I d ; For example, points A, C, and E in the figure all meet the requirement of "the last point before the falling edge of the waveform". Take points A and E to calculate the frequency, then I d =2049, the frequency is 250*8 / 2049=0.9761Hz, and the deviation from the nominal value of 0.9766Hz is (0.9766-0.9761) / 0.9766*100%=0.05%, which is much lower than ±4%;

[0149] The calculation method of duty cycle is as follows: take the percentage of the high level segment (BC segment) in the whole cycle (BD segment) within a square wave period (the range of three points of BCD is one square wave period). For example, there are 129 data frames in the BC segment and 256 data frames in the BD segment. Therefore, the duty cycle is 129 / 256*100%=50.39%, and the deviation from the nominal value of 50% is (50.39-50) / 50*100%=0.78%, which is much lower than ±3%.

[0150] So far, it can be proved that the self-test signal of the FP1 channel is correct. The verification method of the FP2 channel is the same.

[0151] 3.3.3 Data background reception and processing: The reception of the above data frames, voltage conversion, and calculation of fatigue values ​​and electrooculogram data are completed by the background program. The program can receive EEG data and complete the corresponding voltage conversion, electrooculogram data extraction, and fatigue value calculation. After completing the above work, all data will be sent to the foreground program. Its functions include: handshake, exit program, and obtain fatigue values ​​and electrooculogram data.

[0152] 3.4 Instructions for use of computer software:

[0153] 3.4.1 Introduction to computer software: The software is divided into two modules: 1. Backend fatigue calculation and processing module Backend_pilao.exe, 2. Foreground data acquisition, display and processing module Selffeedback.exe. The background module is automatically called by the foreground module and does not need to be manually run. After being called, it stays in the background and calculates and processes the EEG data sent by the foreground module, and sends the filtered data and fatigue and eye movement data back to the foreground module. Its configuration file is config.ini file. By adjusting the parameters, the background algorithm can be adjusted to calculate the foreground data more accurately.

[0154] 3.4.2 Software Settings: Click the "User Settings" button on the homepage to get the user management dialog box. After entering the user name and serial number on the right side of the dialog box, click Add to add the user to the user list; click a user in the user list, the user information is displayed in the information on the right, and the user information can be deleted and modified. After the user management is completed, click OK to close the dialog box. The user currently selected in the list is the current test user. The user management part of the program is under Settings. Click the "Settings" button to display a dialog box including the EEG serial port number, the electrical stimulation serial port number, and the EEG display buffer size. The list on the right side of the EEG serial port and the electrical stimulation serial port shows the list of serial ports on the computer. The EEG serial port will contain the Silicon string, and the electrical stimulation serial port will contain the CH340 string. The EEG display buffer size is the maximum buffer of the image during real-time acquisition, that is, how long ago the data can be displayed from the current time.

[0155] 3.4.3 EEG, EOG, and blood oxygen saturation collection: On the front page of the front-end program, click "Data Collection Feedback", where the real-time display of EEG FP1 and FP2, EOG, and blood oxygen data is displayed. Among them, EEG FP1 and FP2 and blood oxygen are directly obtained from the helmet device, while EOG data is obtained from the background calculation program.

[0156] After clicking Start, real-time data begins to display. In the real-time graphs of EEG FP1, FP2, and EOG, click the left mouse button and drag to drag the image. Roll the mouse wheel up and down to zoom the y-axis range; roll left and right to zoom the x-axis, i.e., the time range. If you don't have a mouse that can scroll left and right, you can hold down the Ctrl key on the keyboard and roll the mouse wheel up and down to achieve the same effect as rolling the mouse wheel left and right, i.e., zooming the x-axis. After the x-axis is expanded or dragged, it will stop automatically following the image movement. Double-click to restore the x-axis automatic following function. After clicking the "Stop" button, the acquisition activity stops and the image stops updating.

[0157] 3.4.4 Real-time monitoring of fatigue: In the real-time acquisition interface, there is a real-time display of fatigue, which is obtained by the back-end calculation module. After the front-end collects the real-time EEG FP1 and FP2 data, it is sent to the back-end through the interface of the back-end program. After the back-end calculates, the calculation results are sent back to the front-end, and the calculation results include fatigue data.

[0158] 3.4.5 Self-feedback control: To use the self-feedback function of the software, you first need to define the parameters of electrical stimulation. On the homepage of the program, click the "Electrical Stimulation" button. The dialog box displays the stimulation mode in decibels. Channel 1, Channel 2 and Channel 3, the top current, time, frequency and pulse width are displayed below each channel. After setting the parameters of the three channels of electrical stimulation, click the "Start" button to test the output effect of the electrical stimulation parameters. If the parameters are not appropriate, you can click the "Stop" button to stop the electrical stimulation and reset the parameters of the electrical stimulation. For the electrical stimulation parameters, please refer to the electrical stimulation design technology section. After the settings are completed, click OK to save. Then go to the real-time data acquisition interface to adjust the self-feedback parameters. Set the fatigue threshold. When the fatigue exceeds the given threshold, the program automatically turns on the electrical stimulation and provides self-feedback to the user. After the electrical stimulation is over, the program automatically restarts the acquisition and detection operations.

[0159] 4. The use process of the prototype:

[0160] 4.1 Preparation stage: Since the device has an instantaneous working current of about 600mA, the USB interface of an ordinary computer cannot meet the requirements; please connect the device data cable to the high-current USB interface of the computer. If the computer does not have a high-current interface, please connect the USB data cable to the prepared self-powered USB-HUB (USB-HUB power supply is connected to a 5V1A or above adapter), and then connect the HUB to the computer USB interface.

[0161] 4.2 Prototype wearing: Wipe the forehead and the back of the left and right ears with saline to remove grease; wear the headset to the designated brain area, and stick the reference electrodes behind the left and right ears with conductive gel patches; clamp the two electrical stimulation ear clips at the designated positions of the left auricle.

[0162] 4.3 Data collection: Open the foreground program Selffeedback.exe, set the EEG serial port and electrical stimulation serial port, current user, and electrical stimulation parameters according to the above process, then turn on real-time data collection, set the electrical stimulation feedback parameters, and click to start real-time data collection, background data processing, and fatigue self-feedback automatic processing.

[0163] 5. Data management: When the foreground program is running, the EEG FP1, FP2, blood oxygen data obtained during the runtime, the electrooculogram and fatigue data calculated by the background program, and the self-feedback action data during the program running will be saved. The data is saved in the Data directory created under the directory where the currently running program is located.

[0164] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0165] In general, various example embodiments of the present disclosure may be implemented in hardware or dedicated circuits, software, firmware, logic, or any combination thereof. Certain aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device. When various aspects of the disclosed embodiments are illustrated or described as block diagrams, flow charts, or using some other graphical representation, it will be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general purpose hardware or controllers or other computing devices, or some combination thereof as non-limiting examples.

[0166] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0167] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0168] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0169] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0170] The exemplary embodiments of the present disclosure described in detail above are merely illustrative and not restrictive. It should be understood by those skilled in the art that various modifications and combinations may be made to these embodiments or their features without departing from the principles and spirit of the present disclosure, and such modifications should fall within the scope of the present disclosure.

Claims

1. A method for real-time fatigue monitoring and negative feedback control, characterized in that: The method comprises: 101, obtaining an EEG signal and blood oxygen saturation data of a subject; the EEG signal includes signals at FP1 and FP2 sites; and separating the EEG signal and the EEG signal from the EEG signal. 102, calculating fatigue level based on the electroencephalogram signal, the electrooculogram signal and the blood oxygen saturation data; 103. When fatigue reaches a set threshold, transcranial electrical stimulation is administered to the subject by triggering electrodes located in the prefrontal cortex.

2. The method for real-time fatigue monitoring and negative feedback control according to claim 1, characterized in that: The prefrontal cortex includes any two or more stimulation sites located at: FP1, FP2, F3, Fz; Optionally, the stimulation sites of the transcranial electrical stimulation also include: the vagus nerve of the left auricle and the right auricle; relieving the fatigue state of the subject and restoring cognitive function; Optionally, the blood oxygen saturation data is collected by a blood oxygen saturation collection sensor.

3. The method for real-time fatigue monitoring and negative feedback control according to claim 1, characterized in that: The method for separating an electroencephalogram (EEG) signal and an electrooculogram (EOG) signal from an electroencephalogram (EEG) signal comprises: calculating the long time difference of the EEG signal to measure the electrooculogram (EOG) fluctuation amplitude; obtaining the amplitude envelope of the EEG signal by low-pass filtering, and detecting the starting and ending points of the EOG signal by a double threshold method; decomposing the EEG signal and introducing a Birgé-Massart strategy to adaptively determine the threshold of the wavelet coefficient; accurately estimating the EOG signal by wavelet reconstruction, and decoupling the estimated EOG signal from the original EEG signal to separate the EEG signal from the EOG signal.

4. The method for real-time fatigue monitoring and negative feedback control according to claim 1, characterized in that: The fatigue degree calculation method comprises: inputting the EEG signal, the EOG signal and the blood oxygen saturation data into a real-time fatigue state evaluation model to calculate the fatigue degree; the real-time fatigue state evaluation model is constructed based on a few-channel fatigue decoding algorithm of Transformer and attention mechanism; Optionally, the method for constructing the real-time fatigue status assessment model includes: obtaining a data set for executing the PVT paradigm for the training set; inputting the data set into the CNN structure to extract local features of the EEG signal, and using the Transformer to extract the long-distance correlation of the EEG signal, while paying attention to the local and global features of the data; introducing a channel attention mechanism and a spatial attention mechanism to measure the importance weights of different channels and different spatial positions of features for the decoding task; adopting a feature fusion strategy to extract a feature map containing multiple layers of information; inputting the feature map into the classifier for continuous optimization to obtain a constructed real-time fatigue status assessment model.

5. The method for real-time fatigue monitoring and negative feedback control according to claim 1, characterized in that: The transcranial electrical stimulation scheme includes any one or more of the following: stimulation site, stimulation intensity, stimulation pulse waveform, stimulation frequency, and stimulation time; Optionally, the frequency calculation method includes: taking any two last points just before the falling edge of the waveform, calculating the difference in the data frame indexes of the two points, and the frequency is 250*(the number of falling edges between the two points) / I d .

6. The method for real-time fatigue monitoring and negative feedback control according to claim 1, characterized in that: The method also includes determining the predicted results of the subject being in a fatigued state, a transitional state, or an awake state.

7. The method for real-time fatigue monitoring and negative feedback control according to claim 1, characterized in that: The electroencephalogram and electrooculogram signals are preprocessed signals, and the preprocessing method includes: eliminating high-frequency interference, differential amplification, and converting into digital signals.

8. A computer device, characterized in that: The device comprises: a memory and a processor; the memory is used to store a computer program; the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

Images