Method and apparatus for processing electroencephalogram signals and electroencephalogram electrode cap

Abstract

The application discloses a kind of electroencephalogram processing method, device and brain electrode cap, it is related to brain-computer interface technical field, main purpose is to solve the problem of existing electroencephalogram processing efficiency is poor, inaccurate.Method includes: when the pressure value between electrode piece and detection site is determined by force feedback sensor matches preset pressure threshold value, the electroencephalogram of detection site is collected by electrode piece, force feedback sensor is used to drive motor in detection body and pneumatic push rod adjusts the contact angle and contact intensity between electrode piece and detection site when pressure value does not match preset pressure threshold value;Residual brain electrical component of electroencephalogram is extracted based on deep residual shrinkage network model, and the main component of electroencephalogram is reconstructed with feature information;The multi-channel electroencephalogram of reconstructed electroencephalogram is uniformly scaled and decomposed, and sub-band features are obtained, and view feature processing is carried out based on sub-band features, and a plurality of view feature sets are obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of brain-computer interface, and particularly relates to a brain electrical signal processing method and device. BACKGROUND

[0002] The brain-computer interface is a signal exchange mode for brain nerve activity to communicate with the outside world. The user's intention of the brain-computer interface is recognized and judged by analyzing brain signals and recognizing and analyzing the brain signals, so as to realize the interaction with the external environment. Among them, due to the difficulty of brain electrical signal reception, there are two existing recording methods for brain electrical signals, namely implanted electrodes and scalp electrodes. Since the scalp electrode detection has non-invasiveness and simple detection method, the scalp electrode detection is widely used.

[0003] At present, when the existing brain electrical signal device using the scalp electrode collects the brain electrical signal, the electrode cap is worn on the patient's head, the electrode sheet is used to contact the patient's head to collect the brain electrical signal, and the brain electrical signal is analyzed and processed based on the brain electrical signal, for example, frequency spectrum analysis. However, since the shapes of different patient's heads are quite different, the electrode sheet on the electrode cap will not be tightly attached to the patient's head, which will cause a large difference between the collected brain electrical signal and the real brain reflection information. At the same time, the frequency spectrum analysis of the brain electrical signal alone cannot accurately obtain the brain activity information, and therefore, a brain electrical signal processing method is needed to solve the above problems. SUMMARY

[0004] Therefore, the present application provides a brain electrical signal processing method and device and a brain electrode cap, which mainly aims to solve the problems of poor processing efficiency and inaccuracy of the existing brain electrical signal processing.

[0005] According to one aspect of the present application, a brain electrical signal processing method is provided, comprising:

[0006] When the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches the preset pressure threshold value, the brain electrical signal of the detection part is collected through the electrode sheet. The pressure value is collected by the force feedback sensor in the detection body. The force feedback sensor is also used to drive the motor and the pneumatic push rod in the detection body to adjust the contact angle and the contact strength between the electrode sheet and the detection part when the collected pressure value does not match the preset pressure threshold value, until the pressure value matches the preset pressure threshold value.

[0007] The deep residual shrinkage network model based on the completed model training is used to extract the feature of the residual brain electrical component of the brain electrical signal, and the feature information after the feature extraction is reconstructed with the main component of the brain electrical signal to obtain the reconstructed brain electrical signal.

[0008] Uniform scale decomposition is performed on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and view feature processing is performed based on the decomposition result to obtain a multi-view feature set.

[0009] Furthermore, before extracting features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model that has completed model training, the method further includes:

[0010] Performing singular spectrum analysis on the EEG signal to obtain multiple signal subcomponents;

[0011] A main component is determined from the multiple signal subcomponents, and the main component is deleted from the EEG signal to obtain a residual EEG component.

[0012] Further, determining the main component from the multiple signal sub-components includes:

[0013] extracting relevant sub-components from the plurality of signal sub-components based on a preset autocorrelation coefficient threshold, and reconstructing the relevant sub-components to obtain the main component;

[0014] The signal subcomponents include frequency domain subcomponents, time domain subcomponents, spatial subcomponents, physiological and artifact shadow components, equipment artifact signal subcomponents, and nonlinear subcomponents.

[0015] Furthermore, the feature information after feature extraction is reconstructed with the main components of the EEG signal to obtain the reconstructed EEG signal, which includes:

[0016] Obtaining reconstruction weights, wherein the reconstruction weights include subject weights and feature weights;

[0017] Performing weighted sum calculation based on the reconstruction weight, the main component, and the feature information to obtain a reconstructed EEG signal;

[0018] The subject weight is determined based on the signal variance of the subject component and the signal variance of the feature information, and the sum of the feature weight and the subject weight is 1.

[0019] Furthermore, the main components include frequency domain main components, time domain main components, and spatial main components, and the weighted sum calculation based on the reconstruction weights, the main components, and the feature information to obtain the reconstructed EEG signal includes:

[0020] Determining a first main weight and a first characteristic weight of the frequency domain main component, and performing weighting and calculation based on the first main weight, the first characteristic weight, the main component, and the characteristic information to obtain a first reconstructed EEG signal, wherein the first characteristic weight is a frequency domain dynamic weight function;

[0021] Determining a second subject weight and a second feature weight of the time-domain subject component, and performing weighted sum calculation based on the second subject weight, the second feature weight, the subject component, and the feature information to obtain a second reconstructed EEG signal, where the second subject weight is 1 and the second feature weight is a time-varying gain function;

[0022] Determine a third subject weight and a third feature weight of the spatial subject component, and perform weighting and calculation based on the third subject weight, the third feature weight, the subject component, and the feature information to obtain a third reconstructed EEG signal.

[0023] Furthermore, before extracting features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model that has completed model training, the method further includes:

[0024] Constructing a deep residual contraction network, wherein the input of the deep residual contraction network is a multi-scale input, and the deep residual contraction network model is embedded with a filtering unit, an attention mechanism unit, and a recurrent unit;

[0025] The deep residual shrinkage network model is trained based on the residual EEG feature training samples to obtain a deep residual shrinkage network model, wherein the dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise.

[0026] Furthermore, before performing uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal, the method further includes:

[0027] Determine the mutual information between the reconstructed EEG signal and the standard EEG signal, and screen the multi-channel EEG signal of the reconstructed EEG signal based on the mutual information, wherein the mutual information is used to characterize the degree of correlation between the reconstructed EEG signal and the standard EEG signal.

[0028] Furthermore, after screening the multi-channel EEG signal of the reconstructed EEG signal based on the mutual information, the method further includes:

[0029] The multi-channel EEG signal is filtered based on a spatial filter to perform uniform scale decomposition based on the filtered multi-channel EEG signal, wherein a weight matrix in the spatial filter is determined based on a deep learning network and a common spatial pattern.

[0030] Furthermore, performing uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result includes:

[0031] performing modal decomposition on the multi-channel electroencephalogram signal to obtain a modal decomposition result, and screening a sub-band set from the modal decomposition result;

[0032] calculating energy of each sub-band in the sub-band set, and determining an optimal sub-band based on a dynamic correlation weight and the sub-band energy;

[0033] performing feature extraction on the optimal sub-band based on a multi-layer convolution model that has completed model training to obtain a decomposition result of sub-band features.

[0034] Further, the view feature processing based on the decomposition result to obtain a multi-view feature set comprises:

[0035] performing feature classification on the decomposition result according to feature types, and generating a view set of the classified decomposition result according to time, frequency and time domain, wherein the feature types include linear types, nonlinear types and energy types.

[0036] According to another aspect of the present application, a processing device for electroencephalogram signals is provided, comprising:

[0037] The acquisition module is configured to acquire the electroencephalogram signals of the detection part through the electrode sheet when the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches the preset pressure threshold value, wherein the pressure value is collected by the force feedback sensor in the detection body, and the force feedback sensor is further configured to drive the motor and the pneumatic push rod in the detection body to adjust the contact angle and the contact strength between the electrode sheet and the detection part when the collected pressure value does not match the preset pressure threshold value, until the pressure value matches the preset pressure threshold value.

[0038] The extraction module is configured to perform feature extraction on the residual electroencephalogram components of the electroencephalogram signals based on a deep residual shrinkage network model that has completed model training, and reconstruct the feature information after feature extraction and the main components of the electroencephalogram signals to obtain the reconstructed electroencephalogram signals.

[0039] The processing module is configured to perform uniform scale decomposition on the multi-channel electroencephalogram signals of the reconstructed electroencephalogram signals to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set.

[0040] Further, the device further comprises:

[0041] The determination module is configured to perform singular spectrum analysis on the electroencephalogram signals to obtain a plurality of signal sub-components, determine a main component from the plurality of signal sub-components, and delete the main component from the electroencephalogram signals to obtain residual electroencephalogram components.

[0042] Further, the determining module is specifically configured to extract a relevant subcomponent from the plurality of signal subcomponents based on a preset autocorrelation coefficient threshold, and reconstruct the relevant subcomponent to obtain the main component; wherein the signal subcomponent includes a frequency domain subcomponent, a time domain subcomponent, a spatial subcomponent, a physiological and artifact subcomponent, a device artifact signal subcomponent, and a nonlinear subcomponent.

[0043] Further, the extracting module is specifically configured to obtain a reconstruction weight value, the reconstruction weight value including a main weight value and a feature weight value; and perform weight sum calculation based on the reconstruction weight value, the main component, and the feature information to obtain a reconstructed electroencephalogram signal; wherein the main weight value is determined based on signal variance of the main component and signal variance of the feature information, and the sum of the main weight value and the feature weight value is 1.

[0044] Further, the main component includes a frequency domain main component, a time domain main component, and a spatial main component, and the extracting module is specifically configured to determine a first main weight value and a first feature weight value of the frequency domain main component, and perform weight sum calculation based on the first main weight value, the first feature weight value, the main component, and the feature information to obtain a first reconstructed electroencephalogram signal, the first feature weight value being a frequency domain dynamic weight function; determine a second main weight value and a second feature weight value of the time domain main component, and perform weight sum calculation based on the second main weight value, the second feature weight value, the main component, and the feature information to obtain a second reconstructed electroencephalogram signal, the second main weight value being 1 and the second feature weight value being a time-varying gain function; and determine a third main weight value and a third feature weight value of the spatial main component, and perform weight sum calculation based on the third main weight value, the third feature weight value, the main component, and the feature information to obtain a third reconstructed electroencephalogram signal.

[0045] Further, the apparatus further comprises:

[0046] The training module is configured to construct a deep residual shrinkage network, an input of the deep residual shrinkage network being a multi-scale input, and the deep residual shrinkage network model embedding a filter unit, an attention mechanism unit, and a cycle unit; and perform model training on the deep residual shrinkage network model based on a residual electroencephalogram feature training sample to obtain a deep residual shrinkage network model, wherein a dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise.

[0047] Further, the determining module is further configured to determine mutual information between the reconstructed electroencephalogram signal and a standard electroencephalogram signal, and filter a multi-channel electroencephalogram signal of the reconstructed electroencephalogram signal based on the mutual information, the mutual information being used to represent a degree of correlation between the reconstructed electroencephalogram signal and the standard electroencephalogram signal.

[0048] Furthermore, the processing module is also used to filter the multi-channel EEG signal based on a spatial filter to perform uniform scale decomposition based on the filtered multi-channel EEG signal, and the weight matrix in the spatial filter is determined based on a deep learning network and a common spatial pattern.

[0049] Furthermore, the processing module is specifically used to perform modal decomposition on the multi-channel EEG signal to obtain a modal decomposition result, and filter out a subband set from the modal decomposition result; calculate the energy of each subband in the subband set, and determine the optimal subband based on the dynamic correlation weight and the subband energy; perform feature extraction on the optimal subband based on the multi-layer convolution model that has completed model training to obtain a decomposition result of the subband feature.

[0050] Furthermore, the processing module is specifically used to perform feature classification on the decomposition results according to feature types, and generate a view set of the classified decomposition results according to time, frequency, and time domain, and the feature types include linear type, nonlinear type, and energy type.

[0051] According to another aspect of the present invention, there is provided a brain electrode cap, comprising: a cap body, a detection body, a signal processor,

[0052] The cap body is provided with a plurality of cavities, the end of the detection body is placed in the cavity, the detection body includes a force feedback sensor, a motor and a pneumatic push rod, and the hemispherical front end of the detection body is provided with an electrode sheet on the outside.

[0053] The force feedback sensor exchanges data with the motor and the pneumatic push rod to collect the pressure value between the electrode sheet and the detection part, and sends a control instruction to the motor and the pneumatic push rod when the pressure value does not match the preset pressure threshold;

[0054] The motor and the pneumatic push rod are used to adjust the contact angle and contact strength between the electrode sheet and the detection part based on the control instruction;

[0055] The signal processor is used to collect the EEG signal of the detection part through the electrode sheet when the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches a preset pressure threshold; perform feature extraction on the residual EEG component of the EEG signal based on the deep residual shrinkage network model that has completed model training, and reconstruct the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal; perform uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set.

[0056] Further, the force feedback sensor is arranged inside the hemispherical front end of the detection body, and the detection surface of the force feedback sensor is connected with the electrode sheet, a contraction groove is arranged on the inner wall of the side adjacent to the detection body of the cap body, a connecting assembly is arranged on the side adjacent to the contraction groove of the detection body, one side of the connecting assembly extends into the contraction groove and slides in the contraction groove, and a compression spring is fixedly connected between the side of the connecting assembly away from the detection body and the inner wall of the contraction groove.

[0057] Further, a driving shaft is fixedly connected to the output shaft of the motor, the opposite end of the driving shaft rotates and penetrates into the contraction groove, a sleeve rod is slidably arranged on the outer surface of the driving shaft, two limiting sliding grooves are arranged on the inner wall of the sleeve rod, and a limiting sliding block is fixedly connected to the outer surface of the driving shaft, and the opposite sides of the limiting sliding block extend into the limiting sliding grooves and slide in the limiting sliding grooves.

[0058] Further, a control cavity is arranged in the cap body, and the pneumatic push rod and the motor are fixedly installed in the control cavity, and one end of the telescopic rod of the pneumatic push rod slidably penetrates into the contraction groove and is in contact with the connecting shell.

[0059] Further, a connecting column is fixedly connected to the side of the connecting shell facing the detection body, an active groove is arranged on the surface of the side of the detection body connected to the cap body, an active ball is arranged in the active groove, the active ball rotates in the active groove, one side of the active ball extends out of the active groove and is fixedly connected with the connecting column, a plurality of control rods arranged in a circumferential array are arranged in the connecting shell, one end of the control rod slidably penetrates out of the connecting shell, a contact ball is fixedly connected to the end of the control rod located outside the connecting shell, the contact ball is in contact with the surface of the detection body, an extrusion body is fixedly connected to the end of the control rod located in the connecting shell, a supporting spring is fixedly connected between the extrusion body and the inner wall of the connecting shell, and the supporting spring is sleeved on the control rod.

[0060] Further, a rotating plate is arranged in the connecting shell, the rotating plate rotates in the connecting shell, a sleeve rod is fixedly connected to the surface of the side of the rotating plate away from the control rod, and the other end of the sleeve rod rotates and penetrates out of the outer surface of the connecting shell.

[0061] Further, a pushing block is fixedly connected to the surface of the side of the rotating plate facing the extrusion body, and the extrusion body is in contact with and slides on the inclined surface of the pushing block.

[0062] According to another aspect of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, wherein the executable instruction enables a processor to execute an operation corresponding to the above-mentioned method for processing EEG signals.

[0063] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0064] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute an operation corresponding to the above-mentioned method for processing EEG signals.

[0065] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0066] The present invention provides a method and device for processing brain electrical signals. Compared with the prior art, the embodiment of the present invention collects brain electrical signals of the detection part through the electrode sheet when the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches a preset pressure threshold. The pressure value is collected by the force feedback sensor in the detection body. The force feedback sensor is also used to drive the motor and the pneumatic push rod in the detection body to adjust the contact angle and contact strength between the electrode sheet and the detection part when the collected pressure value does not match the preset pressure threshold until the pressure value matches the preset pressure threshold; based on the deep residual shrinkage network model for which model training has been completed, feature extraction is performed on the residual brain electrical components of the brain electrical signal. The feature information after feature extraction is reconstructed with the main components of the EEG signal to obtain a reconstructed EEG signal; the multi-channel EEG signal of the reconstructed EEG signal is uniformly decomposed to obtain a decomposition result, and view feature processing is performed based on the decomposition result to obtain a multi-view feature set, thereby achieving the purpose of accurately collecting EEG signals while adjusting the pressure between the electrode and the head in real time, ensuring that the collected EEG signals can better reflect brain activities. At the same time, by extracting features from the collected EEG signals, reconstructing them, and then performing multi-scale decomposition on the reconstructed EEG signals, the analysis requirements of brain activity characteristics in different dimensions are met, and the processing accuracy of EEG signals is greatly improved to achieve accurate identification of brain activities.

[0067] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0069] Figure 1 A flowchart of a method for processing an EEG signal provided by an embodiment of the present invention is shown;

[0070] Figure 2 A block diagram of the composition of a brain electrode cap provided by an embodiment of the present invention is shown;

[0071] Figure 3 A schematic diagram of a cap structure provided by an embodiment of the present invention is shown;

[0072] Figure 4 A schematic diagram of the structure of a detection body provided by an embodiment of the present invention is shown;

[0073] Figure 5 A schematic diagram of the structure of a rotating plate provided by an embodiment of the present invention is shown;

[0074] Figure 6 The following is a block diagram showing the composition of an electroencephalogram signal processing device provided by an embodiment of the present invention;

[0075] Figure 7 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown;

[0076] Among them, 1-electrode cap; 2-detection body; 3-electrode sheet; 4-force feedback sensor; 5-contraction groove; 6-connecting shell; 7-compression spring; 8-control chamber; 9-pneumatic push rod; 10-motor; 11-connecting column; 12-movable groove; 13-movable ball; 14-control rod; 15-contact ball; 16-extrusion body; 17-support spring; 18-drive shaft; 19-rotating plate; 20-sleeve rod; 21-limiting slide groove; 22-limiting slider; 23-pushing block. DETAILED DESCRIPTION

[0077] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0078] The embodiment of the present invention provides a method for processing an electroencephalogram signal. Figure 1 As shown, the method includes:

[0079] 101. When the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches a preset pressure threshold, the electroencephalogram signal of the detection part is collected through the electrode sheet.

[0080] In an embodiment of the present invention, a user whose EEG signal is to be collected wears an electroencephalogram (EBG) cap to collect the signal. The EBG cap includes a cap body, a detection body, and a signal processor that performs steps 101-103. A force feedback sensor collects the pressure between the electrode sheet and the detection site (e.g., a certain collection point on the head) and compares it with a preset pressure threshold. To ensure that the EEG signal clearly and accurately reflects brain activity, the preset pressure threshold can be configured based on different collection requirements or human characteristics. The preset pressure threshold can be a numerical range to prevent the collected EEG signal from being used as the basis for EEG signal processing when the pressure value is equal to or greater than the preset pressure threshold (excessive pressure) or less than the preset pressure threshold (excessive pressure). Furthermore, the current execution end of the signal processor collects the EEG signal via the electrode sheet when the force feedback sensor determines that the pressure value matches the preset pressure threshold. In addition, the force feedback sensor is also used to drive the motor and pneumatic push rod in the detection body to adjust the contact angle and contact strength between the electrode sheet and the detection part when the collected pressure value does not match the preset pressure threshold, until the pressure value matches the preset pressure threshold, and the motor and pneumatic push rod are no longer driven to adjust the contact angle and contact strength.

[0081] It should be noted that the cap body of the brain electrode cap is provided with multiple cavities, and the end of the detection body is placed in the cavity. The detection body includes a force feedback sensor, a motor and a pneumatic push rod. The hemispherical front end of the detection body is provided with an electrode sheet on the outside. The force feedback sensor interacts with the motor and the pneumatic push rod for data collection, and is used to collect the pressure value between the electrode sheet and the detection part, and sends a control instruction to the motor and the pneumatic push rod when the pressure value does not match the preset pressure threshold. The motor and the pneumatic push rod are used to adjust the contact angle and contact strength between the electrode sheet and the detection part based on the control instruction until the collected pressure value matches the preset pressure threshold.

[0082] 102. Based on the deep residual shrinkage network model for which model training has been completed, feature extraction is performed on the residual EEG component of the EEG signal, and the feature information after feature extraction is reconstructed with the main component of the EEG signal to obtain a reconstructed EEG signal.

[0083] In an embodiment of the present invention, in order to extract part of the EEG signal that is useful for signal analysis, the current execution end first determines the residual EEG component of the EEG signal. At this time, the residual EEG component is the remaining component after the main component is subtracted from the collected EEG signal. Then, the deep residual shrinkage network model is used to extract features of the residual EEG component to obtain feature information. The feature information at this time is used to characterize the high-frequency or abnormal signal features with signal analysis significance in the residual EEG signal, including but not limited to abnormal waveform features, artifact features, EEG high-frequency features, and detail texture features. The embodiment of the present invention does not make specific limitations. Then, this feature information is reconstructed with the main component of the EEG signal to obtain the reconstructed EEG signal as the object for actual EEG signal analysis. Among them, the main component is the content of the EEG signal related to the main brain activity to reflect the core signal of the brain's functional activity. At this time, the main component includes the frequency domain main component, the time domain main component, and the spatial main component. The embodiment of the present invention does not make specific limitations.

[0084] 103. Perform uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set.

[0085] In an embodiment of the present invention, in order to effectively analyze EEG signals and thus display the functional activities in the user's brain, the current execution end first performs multi-channel EEG signal extraction on the reconstructed EEG signals, and then performs uniform scale decomposition on the multi-channel EEG signals to obtain sub-band features. Among them, uniform scale decomposition is an adaptive, multi-scale, and intelligent decomposition method for multi-channel EEG signals, which is suitable for high-complexity signal processing methods. At this time, the characteristic changes of the EEG signals in different sub-bands or components after uniform scale decomposition can reflect the microscopic characteristics and potential significance of the brain activity of the signal, mainly including frequency characteristics, energy distribution, time domain fluctuations, nonlinear dynamics, and spatial distribution. The embodiment of the present invention does not make specific limitations. In addition, in order to describe the characteristics of the signal in multiple dimensions, the current execution end performs view feature processing based on the decomposition results to obtain a multi-view feature set for comprehensively reflecting the details and regularities in the EEG signals. Among them, view feature processing includes view generation of features corresponding to linear type, nonlinear type, and energy type, thereby displaying multi-dimensional brain features.

[0086] In another embodiment of the present invention, for further definition and explanation, before the step of extracting features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model that has completed model training, the method further includes:

[0087] Performing singular spectrum analysis on the EEG signal to obtain multiple signal subcomponents;

[0088] A main component is determined from the multiple signal subcomponents, and the main component is deleted from the EEG signal to obtain a residual EEG component.

[0089] To improve the accuracy of EEG signals in reflecting brain activity, the current execution end first processes the collected EEG signals using singular spectrum analysis before feature extraction to obtain multiple signal subcomponents. These signal subcomponents include frequency domain subcomponents, time domain subcomponents, spatial subcomponents, physiological and artifact components, device artifact signal subcomponents, and nonlinear subcomponents. Specifically, singular spectrum analysis constructs a trajectory matrix based on the time series of the EEG signals, and then decomposes and reconstructs the trajectory matrix to obtain frequency domain subcomponents, time domain subcomponents, spatial subcomponents, physiological and artifact components, device artifact signal subcomponents, and nonlinear subcomponents. Among them, the frequency domain subcomponents include: Delta waves (0.5-4 Hz) related to deep sleep and brain damage, Theta waves (4-8 Hz) related to light sleep, relaxation, and attention fluctuations, Alpha waves (8-13 Hz) reflecting the relaxed state during wakefulness, Beta waves (13-30 Hz) related to thinking activities, alertness, and anxiety, and Gamma waves (>30 Hz) related to activities such as higher cognitive functions and memory processing. The time domain subcomponents include: burst signals related to epilepsy or abnormal brain activity (such as spikes or transient waveforms), background noise of non-EEG activity signals (low-amplitude baseline fluctuations). The spatial distribution subcomponents include independent features such as visual cortex, motor cortex, or frontal lobe activity. The physiological and artifact shadow components include physiological signal artifacts such as electrocardiogram, electromyography, or eye movement signals, such as device artifact signals introduced by poor device contact, power supply interference, etc. The nonlinear subcomponents include, for example, chaotic EEG activity (such as epileptic discharges) or sudden strong non-periodic signals, which are not specifically limited in this embodiment of the present invention. Furthermore, a main component that mainly reflects brain activity is determined from the multiple signal sub-components, and this main component is deleted from the EEG signal to obtain a residual EEG component.

[0090] In another embodiment of the present invention, for further definition and illustration, the step of determining a main component from the plurality of signal subcomponents includes:

[0091] Correlated subcomponents are extracted from the multiple signal subcomponents based on a preset autocorrelation coefficient threshold, and the correlated subcomponents are reconstructed to obtain the main component.

[0092] To improve the effectiveness of EEG signal decomposition and thereby achieve the accuracy of EEG principal component reconstruction, thereby improving the precision of brain activity analysis, the current execution end, when determining the principal component, specifically, first extracts relevant sub-components from multiple signal sub-components based on a pre-configured autocorrelation coefficient threshold. The autocorrelation coefficient is calculated by calculating the correlation coefficient between two adjacent time series values, that is, the autocorrelation value between each two signal sub-components is calculated and compared with the preset autocorrelation coefficient threshold. Signal sub-components with autocorrelation coefficients greater than the preset autocorrelation coefficient threshold are used as relevant value components for reconstruction. Since the signal sub-components include frequency domain sub-components, time domain sub-components, spatial sub-components, physiological and artifact components, device artifact signal sub-components, and nonlinear sub-components, when reconstructing after extracting the relevant sub-components, the reconstruction process of the frequency domain sub-components, time domain sub-components, spatial sub-components, and nonlinear sub-components includes decomposition and reconstruction, and the reconstruction process of the physiological and artifact components includes artifact removal and reconstruction.

[0093] In a specific implementation scenario, the reconstruction process of the frequency domain sub-components is as follows:

[0094] 1. Decomposition process: EEG signal is , expressed as The EEG signal x collected at time can be filtered by the frequency domain filter right Decompose the signal value components into multiple frequency bands: ,in, and are Fourier transform and inverse Fourier transform, respectively. is a frequency selective filter defined as: , each frequency value component Corresponding to Delta, Theta, Alpha, Beta or Gamma waves, are the lowest frequency and the highest frequency respectively, i=1,2...M, M is a positive integer.

[0095] 2. Reconstruction process: Reconstruct the complete frequency domain signal by weighted summation of sub-components in different frequency bands:

[0096] ;in, is the weight of the frequency domain value component, which is dynamically adjusted according to specific tasks (such as sleep detection or cognitive status assessment). is the number of frequency bands.

[0097] In a specific implementation scenario, the reconstruction process of the time domain sub-component is as follows:

[0098] 1. Decomposition process: EEG signals contain burst signals and background noise , extract burst signals through sparse decomposition model: ; Burst signal extraction based on sparse representation optimization: ;in, is a sparse regularization term, which indicates the sparsity of the burst signal. is the reconstruction error constraint term.

[0099] 2. Reconstruction process: The burst signal and background noise are combined in a weighted manner during reconstruction: , weight and Dynamically adjust according to noise removal requirements.

[0100] In a specific implementation scenario, the reconstruction process of the spatial sub-components is as follows:

[0101] 1. Decomposition process: multi-channel EEG signals (Number of time points T, number of channels N) can be decomposed into subcomponents related to different brain regions by spatial filtering: ,in, is the spatial filter matrix, which is obtained by the common spatial pattern (CSP) algorithm. The optimization goal is to maximize the difference between classes: , and are the inter-class and intra-class covariance matrices of EEG signals, respectively.

[0102] 2. Reconstruction process: through the inverse filter matrix Reconstruct the spatial classification back to the original signal: .

[0103] In a specific implementation scenario, the reconstruction process of the nonlinear sub-component is as follows:

[0104] 1. Decomposition process: Use empirical mode decomposition (EMD) to decompose the signal into intrinsic mode fractions (IMFs): ,in, is the classification of each mode, is the residual term, where is the selected modal classification set, k=1, 2...K.

[0105] 2. Reconstruction process: Select the main modal classification for reconstruction: .

[0106] In a specific implementation scenario, the reconstruction process of physiological and artifact shadow components is as follows:

[0107] 1. Decomposition process: Separate the artifact signal A and the main component B of the EEG signal through independent component analysis (ICA), expressed as ,in, is the mixing matrix.

[0108] 2. Reconstruction process: Reconstruct by selecting the component of interest B and performing inverse matrix operation:

[0109] .

[0110] In summary, combining the above reconstruction models of frequency domain, time domain, spatial distribution and nonlinear components, the weighted reconstruction of the signal is achieved through weighting: , where each item can be dynamically weighted and adjusted to meet specific task requirements, and ,the artifact removal operation runs through the entire reconstruction process to ensure ,the signal purity.

[0111] It should be noted that in a specific implementation scenario, the main components obtained by reconstructing the relevant sub-components are the frequency domain main components, the time domain main components, and the spatial main components. At this time, the frequency domain main components are the main components in the frequency domain sense, which are the main frequency components related to the normal EEG waveform characteristics, reflecting the basic activity state and main physiological functions of the brain, such as: Delta waves (0.5-4 Hz) representing low-frequency states such as deep sleep and brain recovery, Theta waves (4-8 Hz) representing attention fluctuations and mild relaxation states, Alpha waves (8-13 Hz) representing relaxation waves in the awake rest state, Beta waves (13-30 Hz) representing alertness, anxiety, and concentrated thinking, and Gamma waves (>30 Hz) representing high-level cognitive and memory processing. Among these frequency components, the part related to the target task is preferably the main component. For example, if the goal is to analyze the attention level, Beta waves and Alpha waves may be the main components, which is not specifically limited in the embodiment of the present invention. The temporal main component is the main component in the time domain, which manifests as a smooth, periodic, or regular waveform in the time domain. These signal components are related to specific brain activities (such as motor planning, visual response) or event-related potentials (ERPs), and are not specifically limited in this embodiment of the present invention. The spatial main component is the main component in the sense of spatial distribution, which can be understood as spatial features in the signal related to the activity of certain specific brain regions. For example, channel signals related to the motor cortex or visual cortex separated by spatial filtering (such as cospatial pattern, CSP) or independent component analysis (ICA) are not specifically limited in this embodiment of the present invention.

[0112] In another embodiment of the present invention, for further definition and explanation, before the step of extracting features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model that has completed model training, the method further includes:

[0113] Construct a deep residual shrinkage network;

[0114] The deep residual shrinkage network model is trained based on the residual EEG feature training samples to obtain a deep residual shrinkage network model.

[0115] In order to improve the accurate extraction of useful features in EEG signals, the current execution end pre-constructs a deep residual network, and uses the residual EEG feature training samples to train the deep residual shrinkage network model to obtain a deep residual shrinkage network model. Among them, the input of the deep residual shrinkage network is a multi-scale input. The deep residual shrinkage network model is embedded with a filter unit, an attention mechanism unit, and a recurrent unit. The dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise. Specifically, as a multi-scale input, time series features of different scales are simultaneously processed through multi-resolution convolution kernels (such as 3×3, 5×5, 7×7, etc.) to obtain a multi-scale residual signal, which is expressed as , where M represents the number of scales, is the convolution kernel size of different scales, Represents feature splicing. For the filtering unit, that is, a wavelet transform or filtering module is embedded in the network, the signal is decomposed into different frequency bands, processed separately and then merged, which can be expressed as , is the number of frequency bands, is the weight of each frequency band. In addition, in order to solve the non-stationary and high noise characteristics of EEG signals, a dynamic shrinkage mechanism is adopted to adjust the local noise standard deviation. The adaptive shrinkage mechanism is expressed as , is the network output, is the network output after contraction, the threshold Determined by the statistical characteristics of the EEG signal, it can be automatically adjusted according to the local variance or energy of the signal. ,in, is the scale factor, is the standard deviation of local noise. For the attention mechanism unit, in order to enhance the ability to extract signal detail features, the current execution end embeds the attention mechanism in the deep residual contraction network to weight key features: , are query, key, and value matrices respectively, is a dynamic threshold. At this point, the attention mechanism can locate key features in the EEG signal, such as burst discharges and abnormal waveforms. Furthermore, to leverage the temporal correlation of EEG signals for feature extraction, the current implementation embeds recurrent units in a deep residual contraction network (such as an LSTM or Transformer) to model the dependencies of the time series, expressed as: , is the hidden state at the current moment, is the input signal, is the input layer weight, is the recurrent unit weight, and b is the bias term.

[0116] It should be noted that in order to carry out effective learning and extract features from the residual EEG components, the residual EEG feature training samples store high-frequency or abnormal features that need to be learned, including but not limited to abnormal waveforms (such as spike waves in epileptic patients, K complex waves and sleep spindles in the sleep stage), artifact features (such as high-frequency artifact features of electromyographic signals, periodic fluctuations caused by ECG signal interference), EEG high-frequency features (such as Gamma waves related to advanced cognition and memory functions), non-linear features (such as chaotic characteristics of signals, non-periodic burst activities), detailed textures, etc. corresponding signal samples, which are not specifically limited in the embodiments of the present invention.

[0117] In another embodiment of the present invention, for further definition and explanation, the step of reconstructing the feature information after feature extraction with the main components of the EEG signal to obtain the reconstructed EEG signal includes:

[0118] Get reconstruction weights;

[0119] A weighted sum calculation is performed based on the reconstruction weight, the main component, and the feature information to obtain a reconstructed EEG signal.

[0120] In order to reconstruct the EEG signal that can represent brain activity, taking into account the characteristics of time domain, frequency domain, spatial distribution, etc., to ensure that the reconstructed signal retains the main components and integrates the detailed features, while avoiding the backflow of artifacts or noise, the current execution end first obtains the reconstruction weight. At this time, the reconstruction weight includes the main weight and the feature weight, so that the reconstruction weight is weighted and calculated based on the main components and feature information to obtain the reconstructed EEG signal. Specifically, For computer signals, is the main component obtained by the main extraction module (such as singular spectrum analysis); For the feature information extracted by the deep residual network, the reconstruction formula is: ,in, is the subject weight, is the feature weight, which is used to balance the contribution of the main component and the feature information in the final signal. At this time, the main weight is determined based on the signal variance of the main component and the signal variance of the feature information. The sum of the feature weight and the main weight is 1.

[0121] In another embodiment of the present invention, for further definition and explanation, the step of performing weighted sum calculation based on the reconstruction weight, the main component, and the feature information to obtain the reconstructed EEG signal includes:

[0122] Determining a first main weight and a first characteristic weight of the frequency domain main component, and performing weighting and calculation based on the first main weight, the first characteristic weight, the main component, and the characteristic information to obtain a first reconstructed EEG signal, wherein the first characteristic weight is a frequency domain dynamic weight function;

[0123] Determining a second subject weight and a second feature weight of the time-domain subject component, and performing weighted sum calculation based on the second subject weight, the second feature weight, the subject component, and the feature information to obtain a second reconstructed EEG signal, where the second subject weight is 1 and the second feature weight is a time-varying gain function;

[0124] Determine a third subject weight and a third feature weight of the spatial subject component, and perform weighting and calculation based on the third subject weight, the third feature weight, the subject component, and the feature information to obtain a third reconstructed EEG signal.

[0125] In order to improve the diversity and effectiveness of reconstructed EEG signals and meet the reconstruction of different subject components, the current execution end can adjust the subject weights and feature weights based on task requirements, such as denoising priority (for EEG signals with low signal-to-noise ratio, setting , which relies more on the main components) and detail enhancement (for scenarios where abnormal or weak signals need to be identified, such as epileptic seizure detection, setting , strengthen feature details). Specifically, the weight adjustment method is ; ,in, (·) represents the signal variance.

[0126] In a specific implementation scenario, for the reconstruction of the main components in the frequency domain, the main components and feature information are decomposed into the frequency domain and then combined according to the frequency bands: , and It is the representation of the main components and characteristic information in the frequency domain. and is the frequency domain weight function, which can be adjusted dynamically: , , which can highlight the main components in the low-frequency part and enhance the detailed features in the high-frequency part.

[0127] In a specific implementation scenario, for the reconstruction of the main components in the time domain, the points are added in the time domain. , as the second feature weight is a time-varying gain function used to dynamically adjust the contribution of detail features. The second subject weight is 1. For example, when the intensity of feature information is high (such as spike waves or abnormal signals), the gain Increase, amplify the details, and in the normal band, Reduce and highlight the main components.

[0128] In a specific implementation scenario, for the reconstruction of the spatial main component, if the signal is a multi-channel EEG signal , the combination of main components and feature information needs to be processed separately in the channel dimension and time dimension: , the multi-channel signals can be further decomposed by independent component analysis (ICA), and the combined signals can be mapped to the spatial distribution related to the brain regions.

[0129] It should be noted that in the embodiment of the present invention, in order to achieve flexible adjustment of weight values ​​according to different task requirements, the current execution end can introduce a dynamic weighting strategy to adjust the weights in real time according to the local characteristics of the signal (such as signal-to-noise ratio, spike intensity, etc.): , , (·) represents the signal-to-noise ratio, and the dynamic weight can more flexibly adapt to the local characteristics of the signal. At the same time, the attention mechanism can be introduced in the combination process to automatically locate the part that contributes more to the overall signal, such as , where the attention weight is determined by the temporal and spatial characteristics of the signal, ensuring that the key features are strengthened. Ultimately, the quality of the reconstructed signal can be gradually improved through multiple iterative combinations and optimizations: 1. The first-level signal obtained by the initial combination 2. Further optimization using deep learning models: , It is a deep learning algorithm and is not specifically limited in the embodiment of the present invention.

[0130] In another embodiment of the present invention, for further definition and explanation, before performing uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal, the method further includes:

[0131] Mutual information between the reconstructed EEG signal and a standard EEG signal is determined, and a multi-channel EEG signal of the reconstructed EEG signal is screened based on the mutual information.

[0132] In order to accurately know the correlation or degree of association between the EEG signal and the standard signal, the mutual information between the real-time EEG signal and the standard signal is used as a screening indicator to more comprehensively capture the importance of the channel. The current execution end first obtains the mutual information. At this time, the mutual information is used to characterize the degree of association between the reconstructed EEG signal and the standard EEG signal, and is defined as: , where X and Y represent EEG signals and standard signals respectively. is a joint probability distribution, which indicates the probability of two variables occurring at the same time. and is the edge probability distribution. At this time, the mutual information measures the amount of information shared between X and Y, and high mutual information indicates strong dependence and high correlation between the two, and low mutual information indicates that the two are almost independent.

[0133] It should be noted that when screening the electroencephalogram channel based on mutual information, the mutual information between the collected or reconstructed electroencephalogram signal and the standard electroencephalogram signal can be calculated to screen the channel most related to the target task (such as cognitive state, motor intention, etc.): , wherein is the mutual information threshold, which can be adjusted according to the specific application scenario, and N is the total number of channels. At this time, the channels that meet the condition are retained, and the remaining channels are removed. At the same time, since the electroencephalogram signal often contains artifacts such as electromyogram, electrocardiogram, and environmental noise, by calculating the mutual information, the channels unrelated or with low dependence to the standard signal can be effectively identified and removed. For example, the channel with low mutual information reflects the artifact or invalid data, and the channel with high mutual information is the target electroencephalogram signal.

[0134] In another embodiment of the present application, in order to further limit and illustrate, after the step of screening the multi-channel electroencephalogram signal of the reconstructed electroencephalogram signal based on the mutual information, the method further comprises:

[0135] filtering the multi-channel electroencephalogram signal based on a spatial filter to perform uniform scale decomposition based on the filtered multi-channel electroencephalogram signal.

[0136] In order to improve the processing accuracy of the signal, the current execution end filters the multi-channel electroencephalogram signal to enhance the target signal and suppress noise and artifacts in the spatial domain, and optimize the feature extraction and classification ability of the multi-channel electroencephalogram signal. Specifically, the multi-channel electroencephalogram signal can be filtered by a spatial filter. At this time, the weight matrix in the spatial filter is determined based on a deep learning network and a common spatial pattern. Specifically, first, determine the weight matrix W of the spatial filter, and determine the training target as maximizing the discrimination degree of task-related information and noise, and then perform model training of the deep learning network. The training steps include: 1. Calculate the covariance matrix , of each task category: ; 2. Jointly optimize the weight matrix W: ​At this time, the common spatial pattern (CSP) used is to extract linear spatial features that distinguish different tasks, aiming to maximize the covariance difference between task categories. Meanwhile, when using a deep learning network, the weights of the filter are automatically learned using a deep neural network to capture more complex nonlinear spatial features. The specific steps include: the input of the deep neural network is the multi-channel electroencephalogram X, the hidden layer includes a convolutional layer for capturing local features between channels and an attention mechanism for dynamic weight allocation, and the output layer is the optimized spatial weight .

[0137] In specific implementation scenarios, the filtering object and target are not the same for different principal component signals. For example, the filtering object of the multi-channel electroencephalogram is the multi-channel signal filtered by mutual information, and the filtering target is to enhance the contribution of the target channel signal to suppress the influence of the task-irrelevant or low-correlation channel signal. The filtering object of the brain region-specific signal is the signal from a specific brain region (such as the occipital channel signal), and the filtering target is to extract the spatial features of the target brain region to suppress the cross interference with other brain region signals. The filtering object of the time-spatial joint signal is the signal formed after spatial filtering, which can be further input to the time series analysis module, and the filtering target is to optimize the spatial features and subsequent time domain analysis (such as event-related potential extraction) to provide high-quality signals.

[0138] In another embodiment of the present application, to further limit and illustrate, the step of uniformly scaling the multi-channel electroencephalogram after reconstruction to obtain a decomposition result including:

[0139] The multi-channel electroencephalogram is modally decomposed to obtain a modal decomposition result, and a sub-band set is selected from the modal decomposition result;

[0140] The energy of each sub-band in the sub-band set is calculated, and the optimal sub-band is determined based on the dynamic correlation weight and the sub-band energy;

[0141] The optimal sub-band is feature-extracted based on the multi-layer convolutional model trained to obtain a decomposition result of the sub-band features.

[0142] To improve the comprehensive capture of time domain, frequency domain and other features in the electroencephalogram, a modal decomposition method is used to process the multi-channel electroencephalogram to select a sub-band set from the modal decomposition result. When the multi-channel electroencephalogram is modally decomposed, specifically, the preprocessed multi-channel electroencephalogram Wavelet decomposition and empirical mode decomposition are performed simultaneously to form a multi-modal decomposition result. In wavelet decomposition, a compactly supported wavelet sensitive to electroencephalogram signals (such as Daubechies wavelet) is determined, and the signal is decomposed into J layers to obtain sub-bands as detail components and approximate components : , At this time, the frequency band range corresponding to each layer component is: ,in, is the sampling frequency. At the same time, the EMD first adaptively decomposes the EEG signal into a set of intrinsic mode components (IMFs) and residuals: , at this time, the frequency range estimation estimates the instantaneous frequency of each IMF through Hilbert transform After decomposition, the subbands of wavelet decomposition are Decomposition with the IMF Merge to form a collection of modal decomposition results: . Then, the subband set is selected from the modal decomposition results. Specifically, a dynamic optimization mechanism is introduced. First, the energy of each subband in the subband set is calculated, and the optimal subband is determined based on the dynamic correlation weight and the subband energy. Specifically, for each subband Calculation Energy , high-energy sub-bands usually contain important features of the target signal. At the same time, the task relevance weight is introduced, and the relevance of each sub-band to the task is calculated in combination with the task goal (such as motion intention decoding, epilepsy feature detection, etc.), which is used as the dynamic relevance weight , Task is the task relevance weight, which is used to calculate the task relevance of each subband and help select the optimal subband. At this time, the contribution of different subbands can be weighted according to specific tasks (such as motion intention decoding or epilepsy detection). The dynamic optimization mechanism based on energy and task relevance weights enhances system adaptability and ensures the efficiency and accuracy of subband selection. When selecting the optimal subband, N optimal subbands are dynamically selected based on real-time task requirements and computing resource constraints: , Top is the optimal sub-band set selected by the dynamic optimization mechanism. After the optimal sub-band is obtained, in order to improve the decomposition accuracy, realize real-time reconstruction and noise suppression, the optimal sub-band is extracted based on the multi-layer convolution model that has completed the model training, and the decomposition result of the sub-band feature is obtained. At this time, the optimal sub-band is Input into the deep neural network DNN, at this time, the hidden layer of the deep neural network DNN is used to capture the local spatiotemporal features between sub-bands to output the reconstructed optimized EEG signal as the decomposition result of the sub-band features, that is, .

[0143] In another embodiment of the present invention, for further definition and explanation, the step of performing view feature processing based on the decomposition result to obtain a multi-view feature set includes:

[0144] The decomposition result is classified according to feature types, and a view set of the classified decomposition result is generated according to time, frequency and time domain.

[0145] In order to show different properties of different electroencephalogram signals, and to describe the characteristics of the electroencephalogram signals, so as to comprehensively reflect the details and rules of the brain activities in the electroencephalogram signals, when the current execution end processes the views of the decomposition result to obtain a multi-view feature set, specifically, the decomposition result is classified according to feature types. The feature types include linear types, nonlinear types and energy types. The linear types refer to classification of linear features, that is, statistical indicators are extracted based on a linear model of a signal, to reflect properties such as mean, variance and covariance of the signal, including time-domain linear features (such as mean, variance, autocorrelation function and root mean square value), frequency-domain linear features (such as power spectral density), covariance matrix and the like, which are not limited in the embodiments of the present application. The nonlinear types refer to classification of nonlinear features, that is, based on complexity, chaos and dynamic change mode of the signal, which is suitable for processing non-stationarity and nonlinear characteristics of the electroencephalogram signals, including entropy-related features (such as sample entropy and approximate entropy), fractal dimension, chaos features and the like, which are not limited in the embodiments of the present application. The energy types refer to classification of energy features reflecting energy distribution in different time periods or frequency ranges of the signal, which is used to represent signal intensity and frequency spectrum characteristics, including total energy, sub-band energy, instantaneous energy, energy distribution diagram and the like, which are not limited in the embodiments of the present application. Further, a view set of the classified decomposition result is generated according to time, frequency and time domain. At this time, the view set comprehensively describes image content of the electroencephalogram signals through time, frequency and spatial dimensions, and the characteristics of the electroencephalogram signals are embodied as a set of multi-modal view features. The content of the view features can include time views, frequency views and spatial views. The time views are time-domain features (such as mean, variance and amplitude) extracted to reflect global trend and time series characteristics of the signal; the frequency views are frequency-domain features (such as power spectral density and frequency band energy) extracted to describe distribution characteristics of the signal at different frequencies; the spatial views are spatial features (such as covariance matrix and common spatial pattern feature) of the multi-channel signal extracted to reflect functional distribution of different brain regions; the nonlinear views are chaos and complexity features (such as sample entropy and fractal dimension) extracted to describe non-stationary dynamics of the signal; and the joint views are unified feature expressions formed by integrating multiple views through feature fusion techniques (such as deep learning and dimension reduction methods), which are not limited in the embodiments of the present application.

[0146] An embodiment of the present invention provides a method for processing EEG signals, which achieves the purpose of accurately collecting EEG signals while adjusting the pressure between the electrode and the head in real time, ensuring that the collected EEG signals can better reflect brain activity. At the same time, by extracting features from the collected EEG signals, reconstructing them, and then performing multi-scale decomposition on the reconstructed EEG signals, the analysis requirements of brain activity characteristics in different dimensions are met, which greatly improves the processing accuracy of EEG signals and realizes accurate identification of brain activities.

[0147] Furthermore, as a response to the above Figure 1 To implement the method shown in FIG, an embodiment of the present invention provides a brain electrode cap, such as Figure 2 As shown, it includes: a cap body 1, a detection body 2, a signal processor 24,

[0148] The cap body 1 is provided with a plurality of cavities, the end of the detection body 2 is placed in the cavity, the detection body 2 includes a force feedback sensor 4, a motor 10 and a pneumatic push rod 9, and the hemispherical front end of the detection body 2 is provided with an electrode sheet 3.

[0149] The force feedback sensor 4 exchanges data with the motor 10 and the pneumatic push rod 9 to collect the pressure value between the electrode sheet 3 and the detection part, and sends a control instruction to the motor 10 and the pneumatic push rod 9 when the pressure value does not match the preset pressure threshold;

[0150] The motor 10 and the pneumatic push rod 9 are used to adjust the contact angle and contact strength between the electrode sheet 3 and the detection part based on the control instruction;

[0151] The signal processor 24 is used to collect the EEG signal of the detection part through the electrode piece 3 when the force feedback sensor 4 determines that the pressure value between the electrode piece 3 and the detection part matches the preset pressure threshold; perform feature extraction on the residual EEG component of the EEG signal based on the deep residual shrinkage network model that has completed model training, and reconstruct the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal; perform uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set.

[0152] In the embodiment of the present invention, Figure 3As shown in the schematic diagram of the cap body 1 structure, the pneumatic push rod 9 primarily functions to adjust the contact strength between the EEG electrode 3 and the head at the detection site. Dynamic adjustment of the contact pressure is achieved through linear motion, ensuring that the pressure value detected by the force feedback sensor 4 remains within the normal range, improving signal acquisition quality and patient comfort. When the force feedback sensor 4 detects abnormal pressure (excessive or insufficient), the feedback trigger module controls the pneumatic push rod 9 to extend and retract. This extension and retraction of the pneumatic push rod 9 involves both extension and retraction. Extension increases the pressure between the electrode 3 and the head at the detection site, ensuring closer contact, while retraction reduces the pressure between the electrode 3 and the head at the detection site, preventing discomfort caused by excessive pressure. Furthermore, when the force feedback sensor 4 detects that the pressure has returned to the normal range, the pneumatic push rod 9 ceases operation. Furthermore, the pneumatic push rod 9 operates as a linear push-pull motion, adjusting the contact pressure. This does not involve angle adjustment or the driving of rotating components. The motor 10 adjusts the contact angle between the electrode sheet 3 and the head through rotational motion, thereby accurately collecting EEG signals while adjusting the pressure between the electrode sheet 3 and the head in real time, ensuring that the collected EEG signals can better reflect brain activity. In addition, the signal processor 24 can be a terminal processor configured on the outside of the cap body 1 of the EEG cap, or it can be a remote processor configured on a remote workstation, which is not specifically limited in the embodiments of the present invention.

[0153] Furthermore, in order to enable the detection body 2 to effectively move up and down when embedded in the cap body 1, so as to ensure that the detection body 2 can effectively adjust the angle and force inside the cap body 1, the force feedback sensor 4 is placed inside the hemispherical front end of the detection body 2, and the detection surface of the force feedback sensor 4 is connected to the electrode sheet 3. A contraction groove 5 is provided on the inner wall of the cap body 1 on the side adjacent to the detection body 2, and a connecting component is provided on the side of the detection body 2 adjacent to the contraction groove 5. One side of the connecting component extends into the contraction groove 5 and slides in the contraction groove 5. A compression spring 7 is fixedly connected between the side of the connecting component facing away from the detection body 2 and the inner wall of the contraction groove 5.

[0154] Furthermore, if Figure 4The detection body 2 shown, in order to ensure that the driving shaft 18 can effectively drive the motor 10 to adjust the contact angle between the electrode piece 3 and the detection part, so as to coordinate the pressure and angle between the electrode piece 3 and the detection part, ensure the effectiveness of the adjustment process, and ensure the safety of the detection part, the output shaft of the motor 10 is fixedly connected with the driving shaft 18, and the opposite end of the driving shaft 18 rotates and passes through the shrinkage groove 5, and the sleeve rod 20 is slidably mounted on the outer surface of the driving shaft 18. Two limiting grooves 21 are provided on the inner wall of the sleeve rod 20, and the outer surface of the driving shaft 18 is fixedly connected with a limiting slider 22, and the opposite side of the limiting slider 22 extends into the limiting groove 21 and slides in the limiting groove 21.

[0155] Furthermore, in order to enable the pneumatic push rod 9 and the motor 10 to effectively adjust the contact angle and contact pressure in the cap body 1, a control cavity 8 is opened in the cap body 1, and the pneumatic push rod 9 and the motor 10 are fixedly installed in the control cavity 8. One end of the telescopic rod of the pneumatic push rod 9 slides through the contraction groove 5 and contacts the connecting shell 6.

[0156] Furthermore, in order to ensure that the detection body 2 embedded in the cap body 1 is stable and can be flexibly adjusted when adjusting the contact angle and contact pressure between the electrode sheet 3 and the detection part, a connecting column 11 is fixedly connected to the side of the connecting shell 6 facing the detection body 2, and a movable groove 12 is provided on the surface of the side of the detection body 2 connected to the cap body 1, and a movable ball 13 is provided in the movable groove 12. The movable ball 13 rotates in the movable groove 12, and one side of the movable ball 13 extends out of the movable groove 12 and is fixedly connected to the connecting column 11. A plurality of control rods 14 arranged in a circular array are provided in the connecting shell 6, and one end of the control rod 14 slides through the connecting shell 6. The end of the control rod 14 located outside the connecting shell 6 is fixedly connected to a contact ball 15, and the contact ball 15 contacts the surface of the detection body 2, and the end of the control rod 14 located inside the connecting shell 6 is fixedly connected to an extrusion body 16, and a support spring 17 is fixedly connected between the extrusion body 16 and the inner wall of the connecting shell 6, and the support spring 17 is sleeved on the control rod 14.

[0157] Furthermore, in order to ensure that the connecting shell 6 can be effectively supported during the process of adjusting the angle and pressure, a rotating plate 19 is provided in the connecting shell 6, and the rotating plate 19 rotates in the connecting shell 6. A sleeve rod 20 is fixedly connected to the side surface of the rotating plate 19 facing away from the control rod 14, and the other end of the sleeve rod 20 rotates to pass through the outer surface of the connecting shell 6.

[0158] Furthermore, if Figure 5As shown, in order to increase the rotational stability of the rotating panel, a push block 23 is fixedly connected to the surface of the rotating plate 19 facing the extrusion body 16 , and the extrusion body 16 contacts and slides with the inclined surface of the push block 23 .

[0159] With respect to the above embodiment, the telescopic rod of the pneumatic push rod 9 pushes the connecting shell 6, and one side of the connecting shell 6 extends into the contraction groove 5. The telescopic movement of the pneumatic push rod 9 causes the connecting shell 6 to be displaced in the contraction groove 5. This displacement changes the contact pressure between the detection body 2 and the patient's head. At the same time, the output shaft of the motor 10 is fixedly connected to the drive shaft 18, and the drive shaft 18 drives the rotating plate 19 to operate, that is, the drive shaft 18 is an extension component of the output shaft of the motor 10, which is directly driven by the motor 10 to realize rotation transmission. At the same time, when the drive shaft 18 rotates, it drives the rotating plate 19 to rotate. The rotating plate 19 further controls the angle adjustment of the detection body 2, that is, the rotating plate 19 squeezes the extrusion body 16 through the push block 23, acting on the control rod 14. The movement of the control rod 14 causes the detection body 2 to tilt, thereby adjusting the contact angle of the electrode sheet 3.

[0160] It should be noted that the pneumatic push rod 9 is linearly movable, used to adjust the contact pressure between the electrode sheet 3 and the head, acting on the connecting housing 6. The drive shaft 18, on the other hand, is rotationally movable, used to adjust the contact angle between the electrode sheet 3 and the head, acting on the rotating plate 19 and the detection body 2. In this embodiment of the present invention, these two forces are coordinated via the force feedback sensor 4 to ensure that both the contact angle and pressure are optimal.

[0161] An embodiment of the present invention provides an electroencephalogram (EEG) cap that can accurately collect EEG signals while adjusting the pressure between the electrode and the head in real time, thereby ensuring that the collected EEG signals can better reflect brain activity.

[0162] Furthermore, as a response to the above Figure 1 The embodiment of the present invention provides a device for processing EEG signals, such as Figure 6 As shown, the device includes:

[0163] an acquisition module 31 configured to acquire an electroencephalogram signal at a detection site via the electrode sheet when a force feedback sensor determines that a pressure value between the electrode sheet and the detection site matches a preset pressure threshold, wherein the pressure value is acquired by the force feedback sensor in the detection body; and further configured to drive a motor and a pneumatic push rod in the detection body to adjust a contact angle and contact strength between the electrode sheet and the detection site until the pressure value matches the preset pressure threshold when the acquired pressure value does not match the preset pressure threshold;

[0164] An extraction module 32 is configured to extract features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model for which model training has been completed, and reconstruct the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal;

[0165] The processing module 33 is configured to perform uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set.

[0166] Furthermore, the device further comprises:

[0167] The determination module is used to perform singular spectrum analysis on the EEG signal to obtain multiple signal sub-components; determine the main component from the multiple signal sub-components, and delete the main component from the EEG signal to obtain a residual EEG component.

[0168] Furthermore, the determination module is specifically used to extract correlated sub-components from the multiple signal sub-components based on a preset autocorrelation coefficient threshold, and reconstruct the correlated sub-components to obtain the main components; wherein the signal sub-components include frequency domain sub-components, time domain sub-components, spatial sub-components, physiological and artifact components, equipment artifact signal sub-components, and nonlinear sub-components.

[0169] Furthermore, the device further comprises:

[0170] A training module is used to construct a deep residual shrinkage network, wherein the input of the deep residual shrinkage network is a multi-scale input, and the deep residual shrinkage network model is embedded with a filtering unit, an attention mechanism unit, and a recurrent unit; the deep residual shrinkage network model is trained based on residual EEG feature training samples to obtain a deep residual shrinkage network model, wherein the dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise.

[0171] Furthermore, the extraction module is specifically used to obtain reconstruction weights, which include subject weights and feature weights; weights and calculations are performed based on the reconstruction weights, the subject components, and the feature information to obtain reconstructed EEG signals; wherein the subject weights are determined based on the signal variance of the subject components and the signal variance of the feature information, and the sum of the feature weights and the subject weights is 1.

[0172] Furthermore, the main components include frequency domain main components, time domain main components, and spatial main components. The extraction module is specifically used to determine the first main weight and the first characteristic weight of the frequency domain main component, and perform weight and calculation based on the first main weight, the first characteristic weight, the main component, and the characteristic information to obtain a first reconstructed EEG signal, where the first characteristic weight is a frequency domain dynamic weight function; determine the second main weight and the second characteristic weight of the time domain main component, and perform weight and calculation based on the second main weight, the second characteristic weight, the main component, and the characteristic information to obtain a second reconstructed EEG signal, where the second main weight is 1, and the second characteristic weight is a time-varying gain function; determine the third main weight and the third characteristic weight of the spatial main component, and perform weight and calculation based on the third main weight, the third characteristic weight, the main component, and the characteristic information to obtain a third reconstructed EEG signal.

[0173] Furthermore, the determination module is also used to determine the mutual information between the reconstructed EEG signal and the standard EEG signal, and to screen the multi-channel EEG signal of the reconstructed EEG signal based on the mutual information, and the mutual information is used to characterize the degree of correlation between the reconstructed EEG signal and the standard EEG signal.

[0174] Furthermore, the processing module is also used to filter the multi-channel EEG signal based on a spatial filter to perform uniform scale decomposition based on the filtered multi-channel EEG signal, and the weight matrix in the spatial filter is determined based on a deep learning network and a common spatial pattern.

[0175] Furthermore, the processing module is specifically used to perform modal decomposition on the multi-channel EEG signal to obtain a modal decomposition result, and filter out a subband set from the modal decomposition result; calculate the energy of each subband in the subband set, and determine the optimal subband based on the dynamic correlation weight and the subband energy; perform feature extraction on the optimal subband based on the multi-layer convolution model that has completed model training to obtain a decomposition result of the subband feature.

[0176] Furthermore, the processing module is specifically used to perform feature classification on the decomposition results according to feature types, and generate a view set of the classified decomposition results according to time, frequency, and time domain, and the feature types include linear type, nonlinear type, and energy type.

[0177] An embodiment of the present invention provides a device for processing EEG signals, which achieves the purpose of accurately collecting EEG signals while adjusting the pressure between the electrode and the head in real time, ensuring that the collected EEG signals can better reflect brain activities. At the same time, by extracting features from the collected EEG signals, reconstructing them, and then performing multi-scale decomposition on the reconstructed EEG signals, the analysis requirements of brain activity characteristics in different dimensions are met, greatly improving the processing accuracy of EEG signals, and realizing accurate identification of brain activities.

[0178] According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the method for processing EEG signals in any of the above method embodiments.

[0179] Figure 7 A schematic structural diagram of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0180] like Figure 7 As shown, the terminal may include: a processor (processor) 402 , a communications interface (Communications Interface) 404 , a memory (memory) 406 , and a communication bus 408 .

[0181] The processor 402 , the communication interface 404 , and the memory 406 communicate with each other via a communication bus 408 .

[0182] The communication interface 404 is used for network communication with other devices such as a client or other servers.

[0183] The processor 402 is configured to execute the program 410 , and specifically to execute the relevant steps in the above-mentioned method embodiment for EEG signals.

[0184] Specifically, the program 410 may include program codes, which include computer operation instructions.

[0185] Processor 402 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs, or processors of different types, such as one or more CPUs and one or more ASICs.

[0186] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0187] The program 410 may be specifically configured to cause the processor 402 to perform the following operations:

[0188] When the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches a preset pressure threshold, the electrode sheet is used to collect an electroencephalogram signal of the detection part, and the pressure value is collected by the force feedback sensor in the detection body. The force feedback sensor is further used to drive the motor and the pneumatic push rod in the detection body to adjust the contact angle and contact strength between the electrode sheet and the detection part until the pressure value matches the preset pressure threshold when the collected pressure value does not match the preset pressure threshold;

[0189] Extracting features of the residual EEG components of the EEG signal based on a deep residual shrinkage network model for which model training has been completed, and reconstructing the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal;

[0190] Uniform scale decomposition is performed on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and view feature processing is performed based on the decomposition result to obtain a multi-view feature set.

[0191] Obviously, those skilled in the art will appreciate that the various modules or steps of the present invention described above can be implemented using a general-purpose computing device, centralized on a single computing device, or distributed across a network of multiple computing devices. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. In some cases, the steps shown or described can be performed in a different order than that shown, or can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0192] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for processing an electroencephalogram signal, characterized in that: include: When the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches a preset pressure threshold, the electroencephalogram signal of the detection part is collected through the electrode sheet. The force feedback sensor is further configured to drive a motor and a pneumatic push rod in the detection body to adjust the contact angle and contact strength between the electrode sheet and the detection part when the collected pressure value does not match the preset pressure threshold, until the pressure value matches the preset pressure threshold; Extracting features of the residual EEG components of the EEG signal based on a deep residual shrinkage network model for which model training has been completed, and reconstructing the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal; performing uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and performing view feature processing based on the decomposition result to obtain a multi-view feature set; Before extracting features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model that has completed model training, the method further includes: Constructing a deep residual contraction network, wherein the input of the deep residual contraction network is a multi-scale input, and the deep residual contraction network model is embedded with a filtering unit, an attention mechanism unit, and a recurrent unit; The deep residual shrinkage network model is trained based on the residual EEG feature training samples to obtain a deep residual shrinkage network model, wherein the dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise.

2. The method according to claim 1, characterized in that Before extracting features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model that has completed model training, the method further includes: Performing singular spectrum analysis on the EEG signal to obtain multiple signal subcomponents; A main component is determined from the multiple signal subcomponents, and the main component is deleted from the EEG signal to obtain a residual EEG component.

3. The method according to claim 2, characterized in that Determining the main component from the plurality of signal subcomponents comprises: extracting relevant sub-components from the plurality of signal sub-components based on a preset autocorrelation coefficient threshold, and reconstructing the relevant sub-components to obtain the main component; The signal subcomponents include frequency domain subcomponents, time domain subcomponents, spatial subcomponents, physiological and artifact shadow components, equipment artifact signal subcomponents, and nonlinear subcomponents.

4. The method according to claim 1, wherein The step of reconstructing the feature information after feature extraction and the main components of the EEG signal to obtain the reconstructed EEG signal includes: Obtaining reconstruction weights, wherein the reconstruction weights include subject weights and feature weights; Performing weighted sum calculation based on the reconstruction weight, the main component, and the feature information to obtain a reconstructed EEG signal; The subject weight is determined based on the signal variance of the subject component and the signal variance of the feature information, and the sum of the feature weight and the subject weight is 1.

5. The method according to claim 4, characterized in that The main components include frequency domain main components, time domain main components, and space main components. The weighted sum calculation based on the reconstruction weights, the main components, and the feature information to obtain the reconstructed EEG signal includes: Determining a first main weight and a first characteristic weight of the frequency domain main component, and performing weighting and calculation based on the first main weight, the first characteristic weight, the main component, and the characteristic information to obtain a first reconstructed EEG signal, wherein the first characteristic weight is a frequency domain dynamic weight function; Determining a second subject weight and a second feature weight of the time-domain subject component, and performing weighted sum calculation based on the second subject weight, the second feature weight, the subject component, and the feature information to obtain a second reconstructed EEG signal, where the second subject weight is 1 and the second feature weight is a time-varying gain function; Determine a third subject weight and a third feature weight of the spatial subject component, and perform weighting and calculation based on the third subject weight, the third feature weight, the subject component, and the feature information to obtain a third reconstructed EEG signal.

6. The method according to claim 1, characterized in that Before uniformly scaling the multi-channel EEG signals of the reconstructed EEG signals, the method further includes: Determine the mutual information between the reconstructed EEG signal and the standard EEG signal, and screen the multi-channel EEG signal of the reconstructed EEG signal based on the mutual information, wherein the mutual information is used to characterize the degree of correlation between the reconstructed EEG signal and the standard EEG signal.

7. The method according to claim 6, characterized in that After screening the multi-channel EEG signal of the reconstructed EEG signal based on the mutual information, the method further includes: The multi-channel EEG signal is filtered based on a spatial filter to perform uniform scale decomposition based on the filtered multi-channel EEG signal, wherein a weight matrix in the spatial filter is determined based on a deep learning network and a common spatial pattern.

8. The method according to claim 6, characterized in that The performing uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result includes: Performing modal decomposition on the multi-channel EEG signal to obtain a modal decomposition result, and screening a subband set from the modal decomposition result; Calculating the energy of each subband in the subband set, and determining an optimal subband based on a dynamic correlation weight and the subband energy; The optimal sub-band is subjected to feature extraction based on the multi-layer convolutional model that has completed model training, and a decomposition result of the sub-band feature is obtained.

9. The method according to any one of claims 1 to 8, characterized in that The performing view feature processing based on the decomposition result to obtain a multi-view feature set includes: The decomposition results are classified according to feature types, and a view set of the classified decomposition results is generated according to time, frequency, and time domain. The feature types include linear type, nonlinear type, and energy type.

10. A device for processing electroencephalogram signals, characterized in that: The method according to claim 1 is performed, wherein the apparatus comprises: an acquisition module, configured to acquire an electroencephalogram signal of the detection part through the electrode sheet when the force feedback sensor determines that the pressure value between the electrode sheet and the detection part matches a preset pressure threshold, the pressure value being acquired by the force feedback sensor in the detection body, and the force feedback sensor being further configured to drive a motor and a pneumatic push rod in the detection body to adjust a contact angle and a contact strength between the electrode sheet and the detection part until the pressure value matches the preset pressure threshold when the acquired pressure value does not match the preset pressure threshold; An extraction module is used to extract features of the residual EEG components of the EEG signal based on the deep residual shrinkage network model for which model training has been completed, and reconstruct the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal; a processing module, configured to perform uniform scale decomposition on the multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set; The device further comprises: A training module is used to construct a deep residual shrinkage network, wherein the input of the deep residual shrinkage network is a multi-scale input, and the deep residual shrinkage network model is embedded with a filtering unit, an attention mechanism unit, and a recurrent unit; the deep residual shrinkage network model is trained based on residual EEG feature training samples to obtain a deep residual shrinkage network model, wherein the dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise.

11. A brain electrode cap, characterized in that: include: Cap body, detection body, signal processor, The cap body is provided with a plurality of cavities, the end of the detection body is placed in the cavity, the detection body includes a force feedback sensor, a motor and a pneumatic push rod, and the hemispherical front end of the detection body is provided with an electrode sheet on the outside. The force feedback sensor exchanges data with the motor and the pneumatic push rod to collect the pressure value between the electrode sheet and the detection part, and sends a control instruction to the motor and the pneumatic push rod when the pressure value does not match the preset pressure threshold; The motor and the pneumatic push rod are used to adjust the contact angle and contact strength between the electrode sheet and the detection part based on the control instruction; The signal processor is configured to collect an EEG signal from the detection site through the electrode sheet when the force feedback sensor determines that the pressure value between the electrode sheet and the detection site matches a preset pressure threshold; perform feature extraction on the residual EEG component of the EEG signal based on a deep residual shrinkage network model for which model training has been completed, and reconstruct the feature information after feature extraction with the main components of the EEG signal to obtain a reconstructed EEG signal; perform uniform scale decomposition on a multi-channel EEG signal of the reconstructed EEG signal to obtain a decomposition result, and perform view feature processing based on the decomposition result to obtain a multi-view feature set; The signal processor is also used to construct a deep residual shrinkage network, the input of the deep residual shrinkage network is a multi-scale input, and the deep residual shrinkage network model is embedded with a filtering unit, an attention mechanism unit, and a recurrent unit; the deep residual shrinkage network model is trained based on the residual EEG feature training sample to obtain a deep residual shrinkage network model, wherein the dynamic threshold in the deep residual shrinkage network model is used to eliminate signal artifacts and noise.

12. The electrode cap according to claim 11, characterized in that: The force feedback sensor is placed inside the hemispherical front end of the detection body, and the detection surface of the force feedback sensor is connected to the electrode sheet. A contraction groove is provided on the inner wall of the cap body adjacent to the detection body. A connecting component is provided on the side of the detection body adjacent to the contraction groove. One side of the connecting component extends into the contraction groove and slides in the contraction groove. A compression spring is fixedly connected between the side of the connecting component facing away from the detection body and the inner wall of the contraction groove.

13. The electrode cap according to claim 12, characterized in that: A driving shaft is fixedly connected to the output shaft of the motor, and the opposite end of the driving shaft rotates and passes through the contraction groove. The sleeve rod is slidably sleeved on the outer surface of the driving shaft, and two limiting grooves are provided on the inner wall of the sleeve rod. A limiting slider is fixedly connected to the outer surface of the driving shaft, and the opposite side of the limiting slider extends into the limiting groove and slides in the limiting groove.

14. The electrode cap according to claim 12, wherein: A control cavity is defined in the cap body, and the pneumatic push rod and the motor are both fixedly mounted in the control cavity. One end of the telescopic rod of the pneumatic push rod slides through the contraction groove and contacts the connecting shell.

15. The electrode cap according to claim 14, characterized in that: A connecting column is fixedly connected to the side of the connecting shell facing the detection body, and a movable groove is opened on the surface of the side of the detection body connected to the cap body, and a movable ball is provided in the movable groove. The movable ball rotates in the movable groove, and one side of the movable ball extends out of the movable groove and is fixedly connected to the connecting column. A number of control rods arranged in a circular array are provided in the connecting shell, and one end of the control rod slides through the connecting shell, and the end of the control rod located outside the connecting shell is fixedly connected to a contact ball, and the contact ball contacts the surface of the detection body, and the end of the control rod located inside the connecting shell is fixedly connected to an extrusion body, and a support spring is fixedly connected between the extrusion body and the inner wall of the connecting shell, and the support spring is sleeved on the control rod.

16. The electrode cap according to claim 15, characterized in that: A rotating plate is provided in the connecting shell and rotates in the connecting shell. A sleeve rod is fixedly connected to a surface of the rotating plate facing away from the control rod, and the other end of the sleeve rod rotates and passes through the outer surface of the connecting shell.

17. The electrode cap according to claim 16, characterized in that: A pushing block is fixedly connected to a surface of the rotating plate facing the extrusion body, and the extrusion body contacts and slides on the inclined surface of the pushing block.

18. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to claim 1 are implemented.

19. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to claim 1.

Images