Quality determination method, device, terminal and medium based on electroencephalogram signal

Abstract

The embodiment of the present invention provides a quality determination method, device, terminal and medium based on EEG signals, which belongs to the field of EEG signal technology. The method includes: collecting an initial EEG signal, and preprocessing the initial EEG signal to obtain a target EEG signal; performing a full rank detection on the target EEG signal to obtain a first abnormal detection result corresponding to the target EEG signal; performing a discrete degree detection on the target EEG signal to obtain a second abnormal detection result corresponding to the target EEG signal; performing a power supply interference detection on the target EEG signal to obtain a third abnormal detection result corresponding to the target EEG signal; performing a fluctuation detection on the target EEG signal to obtain a fourth abnormal detection result corresponding to the target EEG signal; and fusing the first abnormal detection result, the second abnormal detection result, the third abnormal detection result and the fourth abnormal detection result to determine the quality detection result corresponding to the initial EEG signal.

Description

Technical Field

[0001] The present invention relates to the field of electroencephalogram (EEG) signal technology, and in particular to a method, device, terminal and medium for determining quality based on EEG signals. Background Art

[0002] When evaluating the quality of EEG signals, the signals collected by wet electrodes are usually used as the gold standard, but there are many problems in the practical application of wet electrodes. First, the preparation process of wet electrodes is cumbersome and time-consuming; second, skin treatment may bring additional infection risks; third, conductive gel will contaminate the hair, making wet electrodes difficult to accept in daily life; even if the skin is not treated, conductive gel may cause skin discomfort or allergic reactions; as the use time increases, the conductive gel will dehydrate and harden, resulting in an increase in the contact impedance between the scalp and the electrode, limiting the application of wet electrodes in long-term EEG signal recording; in multi-electrode high-density EEG signal recording, the conductive gel may also accidentally diffuse or leak to form a bridge, affecting the signal quality. Therefore, the quality assessment of EEG signals in the prior art mainly relies on the operator's experience, lacks standardization and consistency, and the waveform change trend is difficult to accurately quantify, which in turn brings uncertainty to the subsequent EEG signal analysis and processing. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide a method, device, terminal and medium for determining quality based on EEG signals, aiming to solve the problem that the quality assessment of EEG signals in related technologies mainly relies on the experience of the operator, lacks standardization and consistency, and the waveform change trend is difficult to accurately quantify, which brings uncertainty to the subsequent EEG signal analysis and processing.

[0004] In a first aspect, an embodiment of the present invention provides a quality determination method based on an electroencephalogram signal, comprising:

[0005] Collecting initial EEG signals, and preprocessing the initial EEG signals to obtain target EEG signals;

[0006] Performing full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal;

[0007] Performing a discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal;

[0008] Performing power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal;

[0009] Performing fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal;

[0010] The first abnormality detection result, the second abnormality detection result, the third abnormality detection result and the fourth abnormality detection result are integrated to determine a quality detection result corresponding to the initial EEG signal.

[0011] In a second aspect, an embodiment of the present invention provides a quality determination device based on an electroencephalogram signal, comprising:

[0012] A data processing module is used to collect initial EEG signals and pre-process the initial EEG signals to obtain target EEG signals;

[0013] A first detection module, configured to perform a full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal;

[0014] A second detection module is used to perform a discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal;

[0015] A third detection module, used for performing power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal;

[0016] A fourth detection module, configured to perform fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal;

[0017] A quality determination module is used to fuse the first abnormality detection result, the second abnormality detection result, the third abnormality detection result and the fourth abnormality detection result to determine the quality detection result corresponding to the initial EEG signal.

[0018] In the third aspect, an embodiment of the present invention further provides a terminal device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of any one of the methods for determining quality based on EEG signals provided in the specification of the present invention are implemented.

[0019] In a fourth aspect, an embodiment of the present invention further provides a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any one of the methods for determining quality based on EEG signals provided in the specification of the present invention.

[0020] The embodiment of the present invention provides a quality determination method, device, terminal and medium based on EEG signals, the method comprising: collecting an initial EEG signal, and preprocessing the initial EEG signal to obtain a target EEG signal; performing full rank detection on the target EEG signal to obtain a first abnormal detection result corresponding to the target EEG signal; performing discrete degree detection on the target EEG signal to obtain a second abnormal detection result corresponding to the target EEG signal; performing power supply interference detection on the target EEG signal to obtain a third abnormal detection result corresponding to the target EEG signal; performing fluctuation detection on the target EEG signal to obtain a fourth abnormal detection result corresponding to the target EEG signal; fusing the first abnormal detection result, the second abnormal detection result, the third abnormal detection result and the fourth abnormal detection result to determine the quality detection result corresponding to the initial EEG signal. The method can comprehensively evaluate the quality of EEG signals through analysis of multiple dimensions such as full rank detection, discrete degree detection, power supply interference detection and fluctuation detection. Each detection step analyzes the signal from different angles (such as signal integrity, stability, interference, etc.), thereby providing a more comprehensive quality assessment. Therefore, by fusing multiple anomaly detection results, anomalies in EEG signals can be identified more accurately, avoiding possible misjudgments or omissions of a single detection method. Each detection step of this method provides independent verification of signal quality, ensuring that anomalies can be effectively identified under different conditions (such as signal noise, equipment interference, etc.). This multiple verification mechanism enhances the robustness of signal processing, so that EEG signals collected even in complex environments can be reliably evaluated for quality. By fusing multiple anomaly detection results, a comprehensive quality detection result can be generated, which provides an intelligent decision-making basis for subsequent signal processing. It also solves the problem that the quality assessment of EEG signals in related technologies mainly relies on the experience of the operator, lacks standardization and consistency, and the trend of waveform changes is difficult to accurately quantify, which in turn brings uncertainty to the subsequent EEG signal analysis and processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 A flow chart of a method for determining quality based on an electroencephalogram signal provided by an embodiment of the present invention;

[0023] Figure 2 A schematic diagram of the module structure of a device for determining quality based on EEG signals provided by an embodiment of the present invention;

[0024] Figure 3A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0027] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0028] The embodiment of the present invention provides a method, device, terminal and medium for determining quality based on electroencephalogram signals. The method for determining quality based on electroencephalogram signals can be applied to a terminal device, which can be an electronic device such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant and a wearable device. The terminal device can be a server or a server cluster.

[0029] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0030] Please refer to Figure 1 , Figure 1 A flowchart of a method for determining quality based on an electroencephalogram signal is provided in an embodiment of the present invention.

[0031] like Figure 1 As shown, the quality determination method based on EEG signals includes steps S101 to S106.

[0032] Step S101: collecting initial EEG signals, and preprocessing the initial EEG signals to obtain target EEG signals.

[0033] Exemplarily, an electroencephalogram (EEG) acquisition device is used to acquire the initial EEG signal corresponding to the target user, and then a filter is used to remove unnecessary frequency components in the initial EEG signal, and baseline drift in the signal is removed by baseline correction or a high-pass filter to ensure the stability of the initial EEG signal, or independent component analysis (ICA) or other methods are used to detect and remove eye movement artifacts, etc., so as to obtain the target EEG signal corresponding to the pre-processed initial EEG signal.

[0034] In some embodiments, the preprocessing of the initial EEG signal to obtain the target EEG signal includes: performing bandpass filtering on the initial EEG signal to obtain a filtered EEG signal; and performing notch filtering on the filtered EEG signal to obtain the target EEG signal.

[0035] Exemplarily, a bandpass filter is selected. Common bandpass filters include Butterworth filter, Chebyshev filter, etc. Butterworth filter has a flat passband and a gradual cutoff characteristic, which is suitable for most EEG signal processing. Determine the cutoff frequency of the filter. For EEG signals, a passband range of 1-50Hz is usually selected to retain the main EEG wave bands such as alpha waves (8-12Hz) and beta waves (13-30Hz), while removing high-frequency noise and extremely low-frequency drift, and then applying the bandpass filter to the initial EEG signal to remove unnecessary frequency components to obtain a filtered EEG signal.

[0036] Exemplarily, a notch filter, i.e., a band-stop filter, is determined. Common notch filters include infinite impulse response (IIR) notch filters and finite impulse response (FIR) notch filters, and then the center frequency and bandwidth of the notch filter are determined. Common center frequencies are 50 Hz (applicable to 50 Hz power supply interference) or 60 Hz (applicable to 60 Hz power supply interference), and the bandwidth is usually a few hertz (such as 2-3 Hz). Then, the notch filter is used to perform notch processing on the filtered EEG signal to remove the power supply interference of a specific frequency to obtain the target EEG signal.

[0037] Specifically, through the above steps, the initial EEG signal can be effectively bandpass filtered and notched to remove high-frequency noise, very low-frequency drift and power supply interference, and obtain high-quality target EEG signals. These processing steps can significantly improve the quality of EEG signals and provide a reliable data basis for subsequent analysis and application.

[0038] Step S102: Perform full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal.

[0039] Exemplarily, the target EEG signal data is constructed as a matrix, in which each row represents the signal of an electrode and each column represents a time point. Assume that the size of the matrix is ​​n*m, where n is the number of electrodes and m is the number of time points, and then use the method of linear algebra to calculate the rank of the matrix. The rank of a matrix represents the number of its linearly independent rows or columns. Compare the calculated rank with the minimum of n and m. If the rank is equal to the minimum of n and m, the matrix is ​​full rank, and there is no obvious redundant information or linear correlation. If the rank is less than the minimum of n and m, the matrix is ​​not full rank, and there may be redundant information or linear correlation. Thus, if the matrix is ​​full rank, it is marked as normal, that is, the first abnormal detection result is "normal". If the matrix is ​​not full rank, it is marked as abnormal, that is, the first abnormal detection result is "abnormal". Full rank detection can help identify redundant information or linear correlation in the signal matrix, thereby providing an important basis for the quality assessment of EEG signals.

[0040] In some embodiments, the full rank detection of the target EEG signal to obtain the first abnormality detection result corresponding to the target EEG signal includes: selecting a baseline time period, and determining a target baseline signal corresponding to the baseline time period; removing the target baseline signal from the target EEG signal to obtain a baseline-removed signal corresponding to the target EEG signal; establishing a corresponding target signal matrix based on the baseline-removed signal, and calculating a covariance matrix corresponding to the baseline-removed signal based on the target signal matrix; performing matrix rank solution on the covariance matrix to obtain a target rank corresponding to the target EEG signal; and determining the first abnormality detection result corresponding to the target EEG signal based on the target rank and the target number of channels corresponding to the target EEG signal.

[0041] Exemplarily, for a multi-channel target EEG signal, if the rank of the target EEG signal is lower than the number of channels, it indicates that there are linearly correlated redundant signals, which may be caused by poor electrode contact or equipment failure. Then, by calculating the target baseline signal corresponding to the baseline-removed target EEG signal, the rank of the covariance matrix corresponding to the target baseline signal is calculated to determine the linear independence of the target EEG signal. If the rank is less than the number of channels, it indicates that there is a problematic channel, and the corresponding first abnormal detection result is determined to be unqualified. This analysis method ensures the independence of the signal, improves the validity of the signal and the credibility of the analysis results.

[0042] Exemplarily, a time period is selected from the target EEG signal as a baseline. The baseline is usually a signal in a task-free or resting state, which is used to remove the baseline drift in the signal. The baseline time period is determined, such as the first few seconds before the start of the target EEG signal, and then the EEG signal corresponding to the baseline time period is used as the target baseline signal.

[0043] Exemplarily, the target baseline signal is subtracted from the target EEG signal to obtain a baseline-removed signal, which is then integrated to obtain a target signal matrix, wherein each row in the target signal matrix represents the baseline-removed signal of an electrode, and each column represents a time point.

[0044] Exemplarily, the covariance matrix is ​​calculated based on the target signal matrix. The covariance matrix is ​​used to describe the correlation between different electrode signals, so that the target rank of the covariance matrix is ​​calculated using linear algebra methods. The rank of a matrix represents the number of its linearly independent rows or columns.

[0045] For example, among the many algorithms based on matrix decomposition, covariance matrix decomposition is indispensable, which can be basically divided into eigenvalue decomposition of conventional matrices and singular value decomposition of singular matrices. Taking task-related component analysis (TRCA) as an example, TRCA is an effective method for extracting visual evoked potential features. For example, according to the following formula:

[0046]

[0047] in, Represents the projection matrix of the multi-channel target EEG signal, is the sum of the covariances between the multi-trial calibration data for each target, is the sum of the covariances of the multi-trial calibration data for all targets. The optimization goal is to find a projection matrix , so that each target can maximize the task-related components of the goal. It maximizes the repeatability of the task components between trials by maximizing the covariance between trials. The above formula is common in linear discriminant analysis and can be solved by the generalized Rayleigh quotient method. In order to obtain a unique solution and accelerate the algorithm solution process, constraints are usually added to transform the unconstrained problem into a constrained problem, and the Lagrangian function is constructed to solve it. Here is the constraint ,but It is the solution that maximizes the sum of trial covariances. For dry electrode signals, when inverting the covariance matrix of all stimulation categories, if the electrode falls off or the wire is disconnected, the matrix will be non-full rank, which will cause inversion errors.

[0048] In addition, according to the assumption of EEG signal additive noise, the observed EEG signal By source signal Signal with noise composition, . Generally, noise is random, and signal There should be no correlation, and there is also a low correlation between trials. Therefore, when TRCA maximizes the repeatability of task-related components between trials, it is also equivalent to maximizing the removal of noise components in the observed signal. Among the current BCI decoding algorithms, in addition to simple time domain filtering, power supply noise removal and blinking operations, the feature extraction part is also a noise removal operation.

[0049] Exemplarily, the calculated target rank is compared with the target number of channels corresponding to the target EEG signal. If the target rank is equal to the target number of channels, the matrix is ​​full rank, and there is no obvious redundant information or linear correlation. If the target rank is less than the target number of channels, the matrix is ​​not full rank, and there may be redundant information or linear correlation. Thus, if the matrix is ​​full rank, it is marked as normal, that is, the first anomaly detection result is "normal". If the matrix is ​​not full rank, it is marked as abnormal, that is, the first anomaly detection result is "abnormal".

[0050] Step S103: Perform a discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal.

[0051] Exemplarily, a dispersion index is determined, and common indexes include standard deviation, variance, and interquartile range. Standard deviation and variance can measure the volatility of the signal, and interquartile range can measure the distribution range of the signal. Thus, a dispersion index is calculated for the target EEG signal, and a threshold is set based on historical data or experience, and then the data corresponding to the calculated dispersion index is compared with the threshold. If the dispersion index of all electrodes in the target EEG signal does not exceed the threshold, it is marked as normal, and the second abnormal detection result is "normal". If there is an electrode in the target EEG signal whose dispersion index exceeds the threshold, it is marked as abnormal, and the second abnormal detection result is "abnormal".

[0052] In some embodiments, the discrete degree detection of the target EEG signal to obtain the second abnormal detection result corresponding to the target EEG signal includes: performing difference calculation between adjacent sampling points of the target EEG signal to obtain a target difference; determining a target gradient corresponding to the target EEG signal according to the target difference; obtaining a maximum gradient corresponding to the target EEG signal from the target gradient; determining a first discrete value corresponding to the target EEG signal according to the maximum gradient and a preset gradient; performing root mean square detection on the target EEG signal to obtain a second discrete value corresponding to the target EEG signal; and determining the second abnormal detection result corresponding to the target EEG signal according to the first discrete value and the second discrete value.

[0053] Exemplarily, a difference calculation is performed on each adjacent sampling point of the target EEG signal to obtain a target difference between two adjacent sampling points, and then a target gradient is obtained by dividing the target difference by the sampling time difference between the adjacent sampling points. The target gradient reflects the rate of change of the target EEG signal on the time axis.

[0054] Exemplarily, the maximum value of the target gradient is solved to obtain the maximum gradient, and a preset gradient is set according to experimental requirements or experience. Then, the maximum gradient is compared with the preset gradient to obtain a first discrete value. When the maximum gradient is greater than the preset gradient, the first discrete value is equal to the data corresponding to the maximum gradient divided by the preset gradient; when the maximum gradient is less than or equal to the preset gradient, the first discrete value is equal to 1.

[0055] Exemplarily, the root mean square value of the target EEG signal is calculated, where the root mean square is an indicator for measuring signal volatility, so that the root mean square value of the target EEG signal is determined as the second discrete value.

[0056] For example, if the first discrete value is greater than the first threshold or the second discrete value is greater than the second threshold, the target EEG signal is considered to be abnormal, and the second abnormality detection result is "abnormal". If the first discrete value is less than or equal to the first threshold and the second discrete value is less than or equal to the second threshold, the target EEG signal is considered to be normal, and the second abnormality detection result is "normal".

[0057] Step S104: Perform power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal.

[0058] Exemplarily, a target EEG signal that has been preprocessed is loaded, a Fourier transform is performed on the target EEG signal, the target EEG signal is converted from the time domain to the frequency domain, and then the spectrum after the Fourier transform is analyzed to identify the main frequency components in the target EEG signal.

[0059] Exemplarily, power supply interference usually manifests as noise of a specific frequency, and the common power supply frequency is 50 Hz or 60 Hz. The frequency components corresponding to the power supply frequency (50 Hz or 60 Hz) and its harmonics (such as 100 Hz, 150 Hz, etc.) are searched in the main frequency components, and the amplitudes of these frequency components are calculated, so as to set a threshold for power supply interference based on historical data or experience, and then compare the amplitude of the power supply frequency component with the set threshold to determine whether there is power supply interference. If the amplitude corresponding to the power supply frequency component is greater than the threshold, it is considered that there is power supply interference in the target EEG signal, and the third abnormal detection result is "abnormal". If the amplitude corresponding to the power supply frequency component is less than or equal to the threshold, it is considered that the target EEG signal is normal, and the third abnormal detection result is "normal".

[0060] In some embodiments, the power supply interference detection on the target EEG signal to obtain the third abnormal detection result corresponding to the target EEG signal includes: determining a first frequency corresponding to the power supply interference, and performing spectral analysis on the target EEG signal to obtain target spectrum information corresponding to the target EEG signal; obtaining a first power spectrum density corresponding to the first frequency from the target spectrum information, and determining a first energy corresponding to the first frequency based on the first power spectrum density; determining a second frequency corresponding to the alpha wave, and obtaining a second power spectrum density corresponding to the second frequency from the target spectrum information; determining a second energy corresponding to the second frequency based on the second power spectrum density; and determining the third abnormal detection result corresponding to the target EEG signal based on the first energy and the second energy.

[0061] Exemplarily, power supply interference usually manifests itself as a fixed frequency component, and a common power supply frequency is 50 Hz or 60 Hz. The first frequency corresponding to the power supply interference is then determined, and the target EEG signal is subjected to a Fourier transform, converting the time domain signal into a frequency domain signal to obtain target spectrum information. The first power spectrum density at the first frequency corresponding to the power supply interference is then extracted from the target spectrum information, and the first energy corresponding to the first frequency is calculated based on the first power spectrum density.

[0062] Exemplarily, alpha waves are a characteristic frequency range in brain activity, usually between 8-13 Hz. Therefore, according to the characteristics of the target EEG signal, the center frequency of the alpha wave, such as 10 Hz, is selected as the second frequency. The second power spectrum density at the second frequency is extracted from the target spectrum information, and then the second energy corresponding to the second frequency is calculated according to the second power spectrum density.

[0063] Exemplarily, a target energy ratio between the first energy and the second energy is obtained, and a corresponding preset energy ratio is set when the wearer is not worn or is not worn correctly, so as to calculate the difference between the target energy ratio and the preset energy ratio. If the difference between the target energy ratio and the preset energy ratio is large, the third abnormality detection result is determined to be normal; if the difference between the target energy ratio and the preset energy ratio is small, the third abnormality detection result is determined to be abnormal.

[0064] Step S105: Perform fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal.

[0065] Exemplarily, a target EEG signal that has been preprocessed is loaded. The rate of change of the target EEG signal in the time domain is calculated by difference or gradient, and a statistical threshold for fluctuation detection is set based on historical data or experience. The fluctuation amplitude of the differential sequence is compared with the set threshold to determine whether there is an abnormal fluctuation. If the absolute value of the differential sequence exceeds the threshold, the number of time points exceeding the threshold is counted, and the proportion of the number of time points exceeding the threshold to the total number of time points is calculated, so as to determine whether there is an abnormal fluctuation in the target EEG signal based on the fluctuation ratio. If the fluctuation ratio exceeds the preset ratio threshold, it is considered that the target EEG signal has abnormal fluctuations, and the fourth abnormal detection result is "abnormal". If the fluctuation ratio does not exceed the ratio threshold, the target EEG signal is considered normal, and the fourth abnormal detection result is "normal".

[0066] In some embodiments, the fluctuation detection of the target EEG signal to obtain the fourth abnormal detection result corresponding to the target EEG signal includes: determining the target signal mean corresponding to the target EEG signal, and calculating the absolute deviation between the target EEG signal and the target signal mean; determining the average amplitude difference corresponding to the target EEG signal according to the absolute deviation; calculating the fourth-order central moment corresponding to the target EEG signal and the target variance corresponding to the target EEG signal; determining the kurtosis corresponding to the target EEG signal according to the fourth-order central moment and the target variance; calculating the third-order central moment corresponding to the target EEG signal and the target standard deviation corresponding to the target EEG signal; determining the skewness corresponding to the target EEG signal according to the third-order central moment and the target standard deviation; performing fluctuation detection on the target EEG signal according to the average amplitude difference, the kurtosis, and the skewness to obtain the fourth abnormal detection result.

[0067] Exemplarily, the mean of the target EEG signal is calculated to obtain the target signal mean, and then the absolute deviation between the data at each time point in the target EEG signal and the target signal mean is calculated. The mean of the absolute deviation sequence is then calculated as the average amplitude difference of the target EEG signal.

[0068] Illustratively, the fourth-order central moment is a statistic used to measure the peak degree of the signal distribution. The target variance is a statistic that measures the degree of signal fluctuation. Kurtosis is a statistic that describes the peak degree of the signal distribution. The third-order central moment is a statistic that measures the asymmetry of the signal distribution. The standard deviation is the square root of the variance and is used to measure the degree of signal fluctuation. Skewness is a statistic that describes the degree of asymmetry of the signal distribution.

[0069] Exemplarily, the fourth-order central moment corresponding to the target EEG signal and the target variance corresponding to the target EEG signal are calculated; the kurtosis corresponding to the target EEG signal is determined based on the fourth-order central moment and the target variance; the third-order central moment corresponding to the target EEG signal and the target standard deviation corresponding to the target EEG signal are calculated; and the skewness corresponding to the target EEG signal is determined based on the third-order central moment and the target standard deviation.

[0070] Exemplarily, the target EEG signal is judged whether it has abnormal fluctuations in combination with the average amplitude difference, kurtosis and skewness. If the average amplitude difference is significantly higher than the normal range (for example, exceeding a certain multiple of the target mean), it is considered that the target EEG signal has abnormal fluctuations. If the kurtosis is significantly higher than the normal range (for example, exceeding twice the typical kurtosis), the target EEG signal distribution peak degree is considered abnormal. If the skewness deviates significantly from zero (for example, exceeds a certain threshold), the target EEG signal distribution is considered to be asymmetric. Therefore, if at least one of the average amplitude difference, kurtosis and skewness is significantly abnormal, it is considered that the target EEG signal has abnormal fluctuations, and the fourth abnormal detection result is "abnormal". Otherwise, the target EEG signal is considered normal, and the fourth abnormal detection result is "normal".

[0071] Step S106: The first abnormality detection result, the second abnormality detection result, the third abnormality detection result and the fourth abnormality detection result are integrated to determine a quality detection result corresponding to the initial EEG signal.

[0072] Exemplarily, the number of "abnormalities" in each test result is counted. If the number of "abnormalities" exceeds a preset threshold (for example, greater than or equal to two), the quality test result is "abnormal", otherwise the quality test result is "normal".

[0073] In some embodiments, the fusion of the first abnormal detection result, the second abnormal detection result, the third abnormal detection result, and the fourth abnormal detection result to determine the quality detection result corresponding to the initial EEG signal includes: determining a preset rule; fusing the first abnormal detection result, the second abnormal detection result, the third abnormal detection result, and the fourth abnormal detection result according to the preset rule to obtain a target fusion result; making a quality judgment on the initial EEG signal according to the target fusion result to obtain the quality detection result corresponding to the initial EEG signal.

[0074] Exemplarily, the preset rules are determined according to a specific combination of logical conditions, for example: if the first detection result is "abnormal" and the second detection result is "abnormal", then the fusion result is "abnormal", or, if any two detection results are "abnormal", then the fusion result is "abnormal".

[0075] Exemplarily, the first anomaly detection result, the second anomaly detection result, the third anomaly detection result, and the fourth anomaly detection result are fused according to a preset rule to obtain multiple target fusion results, thereby counting the number of "anomalies" in the multiple target fusion results. If the number of "anomalies" exceeds a preset threshold (for example, greater than or equal to two), the quality detection result is "abnormal"; otherwise, the quality detection result is "normal".

[0076] The embodiment of the present invention provides a quality determination method, device, terminal and medium based on EEG signals, the method comprising: collecting an initial EEG signal, and preprocessing the initial EEG signal to obtain a target EEG signal; performing full rank detection on the target EEG signal to obtain a first abnormal detection result corresponding to the target EEG signal; performing discrete degree detection on the target EEG signal to obtain a second abnormal detection result corresponding to the target EEG signal; performing power supply interference detection on the target EEG signal to obtain a third abnormal detection result corresponding to the target EEG signal; performing fluctuation detection on the target EEG signal to obtain a fourth abnormal detection result corresponding to the target EEG signal; fusing the first abnormal detection result, the second abnormal detection result, the third abnormal detection result and the fourth abnormal detection result to determine the quality detection result corresponding to the initial EEG signal. The method can comprehensively evaluate the quality of EEG signals through analysis of multiple dimensions such as full rank detection, discrete degree detection, power supply interference detection and fluctuation detection. Each detection step analyzes the signal from different angles (such as signal integrity, stability, interference, etc.), thereby providing a more comprehensive quality assessment. Therefore, by fusing multiple anomaly detection results, anomalies in EEG signals can be identified more accurately, avoiding possible misjudgments or omissions of a single detection method. Each detection step of this method provides independent verification of signal quality, ensuring that anomalies can be effectively identified under different conditions (such as signal noise, equipment interference, etc.). This multiple verification mechanism enhances the robustness of signal processing, so that EEG signals collected even in complex environments can be reliably evaluated for quality. By fusing multiple anomaly detection results, a comprehensive quality detection result can be generated, which provides an intelligent decision-making basis for subsequent signal processing. It also solves the problem that the quality assessment of EEG signals in related technologies mainly relies on the experience of the operator, lacks standardization and consistency, and the trend of waveform changes is difficult to accurately quantify, which in turn brings uncertainty to the subsequent EEG signal analysis and processing.

[0077] See also Figure 2 , Figure 2A quality determination device 200 based on EEG signals is provided in an embodiment of the present application. The quality determination device 200 based on EEG signals includes a data processing module 201, a first detection module 202, a second detection module 203, a third detection module 204, a fourth detection module 205, and a quality determination module 206, wherein the data processing module 201 is used to collect initial EEG signals and pre-process the initial EEG signals to obtain target EEG signals; the first detection module 202 is used to perform full rank detection on the target EEG signals to obtain a first abnormality detection result corresponding to the target EEG signals; the second detection module 203 is used to Perform a discrete degree detection on the target EEG signal to obtain a second abnormal detection result corresponding to the target EEG signal; a third detection module 204 is used to perform power supply interference detection on the target EEG signal to obtain a third abnormal detection result corresponding to the target EEG signal; a fourth detection module 205 is used to perform fluctuation detection on the target EEG signal to obtain a fourth abnormal detection result corresponding to the target EEG signal; a quality determination module 206 is used to fuse the first abnormal detection result, the second abnormal detection result, the third abnormal detection result and the fourth abnormal detection result to determine the quality detection result corresponding to the initial EEG signal.

[0078] In some implementations, the data processing module 201 performs, during the process of preprocessing the initial EEG signal to obtain the target EEG signal:

[0079] Performing bandpass filtering on the initial EEG signal to obtain a filtered EEG signal;

[0080] The target EEG signal is obtained by performing notch processing on the filtered EEG signal.

[0081] In some implementations, the first detection module 202, during the process of performing full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal, performs:

[0082] Selecting a baseline time period, and determining a target baseline signal corresponding to the baseline time period;

[0083] After removing the target baseline signal from the target EEG signal, obtain a baseline-removed signal corresponding to the target EEG signal;

[0084] Establishing a corresponding target signal matrix according to the baseline-removed signal, and calculating a covariance matrix corresponding to the baseline-removed signal according to the target signal matrix;

[0085] Performing matrix rank solution on the covariance matrix to obtain a target rank corresponding to the target EEG signal;

[0086] The first abnormality detection result corresponding to the target EEG signal is determined according to the target rank and the target number of channels corresponding to the target EEG signal.

[0087] In some implementations, the second detection module 203 performs, during the process of performing discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal:

[0088] Calculating the difference between adjacent sampling points of the target EEG signal to obtain a target difference;

[0089] Determining a target gradient corresponding to the target EEG signal according to the target difference;

[0090] Obtaining the maximum gradient corresponding to the target EEG signal from the target gradient;

[0091] Determining a first discrete value corresponding to the target EEG signal according to the maximum gradient and the preset gradient;

[0092] Performing a root mean square detection on the target EEG signal to obtain a second discrete value corresponding to the target EEG signal;

[0093] The second abnormality detection result corresponding to the target EEG signal is determined according to the first discrete value and the second discrete value.

[0094] In some implementations, the third detection module 204 performs, during the process of performing power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal:

[0095] Determine a first frequency corresponding to the power supply interference, and perform spectrum analysis on the target EEG signal to obtain target spectrum information corresponding to the target EEG signal;

[0096] Obtaining a first power spectrum density corresponding to the first frequency from the target spectrum information, and determining a first energy corresponding to the first frequency according to the first power spectrum density;

[0097] Determine a second frequency corresponding to the alpha wave, and obtain a second power spectrum density corresponding to the second frequency from the target spectrum information;

[0098] determining a second energy corresponding to the second frequency according to the second power spectral density;

[0099] The third abnormality detection result corresponding to the target EEG signal is determined according to the first energy and the second energy.

[0100] In some implementations, the fourth detection module 205, during the process of performing fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal, performs:

[0101] Determining a target signal mean value corresponding to the target EEG signal, and calculating an absolute deviation between the target EEG signal and the target signal mean value;

[0102] Determine the average amplitude difference corresponding to the target EEG signal according to the absolute deviation;

[0103] Calculating the fourth-order central moment corresponding to the target EEG signal and the target variance corresponding to the target EEG signal;

[0104] Determine the kurtosis corresponding to the target EEG signal according to the fourth-order central moment and the target variance;

[0105] Calculate the third-order central moment corresponding to the target EEG signal and the target standard deviation corresponding to the target EEG signal;

[0106] Determining the skewness corresponding to the target EEG signal according to the third-order central moment and the target standard deviation;

[0107] The fourth abnormality detection result is obtained by performing fluctuation detection on the target EEG signal according to the average amplitude difference, the kurtosis and the skewness.

[0108] In some implementations, the quality determination module 206, in the process of fusing the first abnormality detection result, the second abnormality detection result, the third abnormality detection result, and the fourth abnormality detection result to determine the quality detection result corresponding to the initial EEG signal, performs:

[0109] Determine the preset rules;

[0110] According to the preset rule, the first anomaly detection result, the second anomaly detection result, the third anomaly detection result and the fourth anomaly detection result are fused to obtain a target fusion result;

[0111] The quality of the initial EEG signal is judged according to the target fusion result to obtain the quality detection result corresponding to the initial EEG signal.

[0112] In some implementations, the device 200 for determining quality based on an electroencephalogram signal may be applied to a terminal device.

[0113] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the device 200 based on EEG signals for quality determination described above can refer to the corresponding process in the aforementioned embodiment of the method for quality determination based on EEG signals, and will not be repeated here.

[0114] See also Figure 3 , Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0115] like Figure 3 As shown, the terminal device 300 includes a processor 301 and a memory 302 , and the processor 301 and the memory 302 are connected via a bus 303 , such as an I2C (Inter-integrated Circuit) bus.

[0116] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal device. The processor 301 can be a central processing unit (CPU), and the processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0117] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0118] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present invention, and does not constitute a limitation on the terminal device to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0119] The processor is used to run a computer program stored in the memory, and implement any one of the methods for determining quality based on EEG signals provided by the embodiments of the present invention when executing the computer program.

[0120] In one embodiment, the processor is used to run a computer program stored in the memory, and implements the following steps when executing the computer program:

[0121] Collecting initial EEG signals, and preprocessing the initial EEG signals to obtain target EEG signals;

[0122] Performing full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal;

[0123] Performing a discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal;

[0124] Performing power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal;

[0125] Performing fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal;

[0126] The first abnormality detection result, the second abnormality detection result, the third abnormality detection result and the fourth abnormality detection result are integrated to determine a quality detection result corresponding to the initial EEG signal.

[0127] In some implementations, the processor 301, during the process of preprocessing the initial EEG signal to obtain the target EEG signal, executes:

[0128] Performing bandpass filtering on the initial EEG signal to obtain a filtered EEG signal;

[0129] The target EEG signal is obtained by performing notch processing on the filtered EEG signal.

[0130] In some implementations, the processor 301, during the process of performing full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal, executes:

[0131] Selecting a baseline time period, and determining a target baseline signal corresponding to the baseline time period;

[0132] After removing the target baseline signal from the target EEG signal, obtain a baseline-removed signal corresponding to the target EEG signal;

[0133] Establishing a corresponding target signal matrix according to the baseline-removed signal, and calculating a covariance matrix corresponding to the baseline-removed signal according to the target signal matrix;

[0134] Performing matrix rank solution on the covariance matrix to obtain a target rank corresponding to the target EEG signal;

[0135] The first abnormality detection result corresponding to the target EEG signal is determined according to the target rank and the target number of channels corresponding to the target EEG signal.

[0136] In some implementations, the processor 301, during the process of performing discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal, executes:

[0137] Calculating the difference between adjacent sampling points of the target EEG signal to obtain a target difference;

[0138] Determining a target gradient corresponding to the target EEG signal according to the target difference;

[0139] Obtaining the maximum gradient corresponding to the target EEG signal from the target gradient;

[0140] Determining a first discrete value corresponding to the target EEG signal according to the maximum gradient and the preset gradient;

[0141] Performing a root mean square detection on the target EEG signal to obtain a second discrete value corresponding to the target EEG signal;

[0142] The second abnormality detection result corresponding to the target EEG signal is determined according to the first discrete value and the second discrete value.

[0143] In some implementations, the processor 301, during the process of performing power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal, executes:

[0144] Determine a first frequency corresponding to the power supply interference, and perform spectrum analysis on the target EEG signal to obtain target spectrum information corresponding to the target EEG signal;

[0145] Obtaining a first power spectrum density corresponding to the first frequency from the target spectrum information, and determining a first energy corresponding to the first frequency according to the first power spectrum density;

[0146] Determine a second frequency corresponding to the alpha wave, and obtain a second power spectrum density corresponding to the second frequency from the target spectrum information;

[0147] determining a second energy corresponding to the second frequency according to the second power spectral density;

[0148] The third abnormality detection result corresponding to the target EEG signal is determined according to the first energy and the second energy.

[0149] In some implementations, the processor 301, during the process of performing fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal, executes:

[0150] Determining a target signal mean value corresponding to the target EEG signal, and calculating an absolute deviation between the target EEG signal and the target signal mean value;

[0151] Determine the average amplitude difference corresponding to the target EEG signal according to the absolute deviation;

[0152] Calculating the fourth-order central moment corresponding to the target EEG signal and the target variance corresponding to the target EEG signal;

[0153] Determine the kurtosis corresponding to the target EEG signal according to the fourth-order central moment and the target variance;

[0154] Calculate the third-order central moment corresponding to the target EEG signal and the target standard deviation corresponding to the target EEG signal;

[0155] Determining the skewness corresponding to the target EEG signal according to the third-order central moment and the target standard deviation;

[0156] The fourth abnormality detection result is obtained by performing fluctuation detection on the target EEG signal according to the average amplitude difference, the kurtosis and the skewness.

[0157] In some implementations, the processor 301, in the process of fusing the first abnormality detection result, the second abnormality detection result, the third abnormality detection result, and the fourth abnormality detection result to determine the quality detection result corresponding to the initial EEG signal, performs:

[0158] Determine the preset rules;

[0159] According to the preset rule, the first anomaly detection result, the second anomaly detection result, the third anomaly detection result and the fourth anomaly detection result are fused to obtain a target fusion result;

[0160] The quality of the initial EEG signal is judged according to the target fusion result to obtain the quality detection result corresponding to the initial EEG signal.

[0161] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the terminal device described above can refer to the corresponding process in the aforementioned embodiment of the quality determination method based on EEG signals, and will not be repeated here.

[0162] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any one of the methods for determining quality based on EEG signals provided in the description of the embodiment of the present invention.

[0163] The storage medium may be an internal storage unit of the terminal device described in the foregoing embodiment, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc., equipped on the terminal device.

[0164] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0165] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0166] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A quality determination method based on electroencephalogram signals, characterized in that: The quality determination method based on EEG signal comprises: Collecting initial EEG signals, and preprocessing the initial EEG signals to obtain target EEG signals; Performing full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal includes: Selecting a baseline time period, and determining a target baseline signal corresponding to the baseline time period; After removing the target baseline signal from the target EEG signal, obtain a baseline-removed signal corresponding to the target EEG signal; Establishing a corresponding target signal matrix according to the baseline-removed signal, and calculating a covariance matrix corresponding to the baseline-removed signal according to the target signal matrix; Performing matrix rank solution on the covariance matrix to obtain a target rank corresponding to the target EEG signal; Determine the first abnormality detection result corresponding to the target EEG signal according to the target rank and the target number of channels corresponding to the target EEG signal; Performing a discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal includes: Calculating the difference between adjacent sampling points of the target EEG signal to obtain a target difference; Determining a target gradient corresponding to the target EEG signal according to the target difference; Obtaining the maximum gradient corresponding to the target EEG signal from the target gradient; Determining a first discrete value corresponding to the target EEG signal according to the maximum gradient and the preset gradient; Performing a root mean square test on the target EEG signal to obtain a second discrete value corresponding to the target EEG signal; Determine the second abnormality detection result corresponding to the target EEG signal according to the first discrete value and the second discrete value; Performing power supply interference detection on the target EEG signal to obtain a third abnormality detection result corresponding to the target EEG signal includes: Determine a first frequency corresponding to the power supply interference, and perform spectrum analysis on the target EEG signal to obtain target spectrum information corresponding to the target EEG signal; Obtaining a first power spectrum density corresponding to the first frequency from the target spectrum information, and determining a first energy corresponding to the first frequency according to the first power spectrum density; Determine a second frequency corresponding to the alpha wave, and obtain a second power spectrum density corresponding to the second frequency from the target spectrum information; determining a second energy corresponding to the second frequency according to the second power spectral density; Determine the third abnormality detection result corresponding to the target EEG signal according to the first energy and the second energy; Performing fluctuation detection on the target EEG signal to obtain a fourth abnormality detection result corresponding to the target EEG signal includes: Determining a target signal mean value corresponding to the target EEG signal, and calculating an absolute deviation between the target EEG signal and the target signal mean value; Determine the average amplitude difference corresponding to the target EEG signal according to the absolute deviation; Calculating the fourth-order central moment corresponding to the target EEG signal and the target variance corresponding to the target EEG signal; Determine the kurtosis corresponding to the target EEG signal according to the fourth-order central moment and the target variance; Calculate the third-order central moment corresponding to the target EEG signal and the target standard deviation corresponding to the target EEG signal; Determining the skewness corresponding to the target EEG signal according to the third-order central moment and the target standard deviation; Performing fluctuation detection on the target EEG signal according to the average amplitude difference, the kurtosis and the skewness to obtain the fourth abnormality detection result; The first abnormality detection result, the second abnormality detection result, the third abnormality detection result and the fourth abnormality detection result are integrated to determine a quality detection result corresponding to the initial EEG signal.

2. The method for determining quality based on EEG signals according to claim 1, characterized in that: The preprocessing of the initial EEG signal to obtain a target EEG signal includes: Performing bandpass filtering on the initial EEG signal to obtain a filtered EEG signal; The target EEG signal is obtained by performing notch processing on the filtered EEG signal.

3. The method for determining quality based on EEG signals according to claim 1, characterized in that: The fusing the first abnormality detection result, the second abnormality detection result, the third abnormality detection result, and the fourth abnormality detection result to determine the quality detection result corresponding to the initial EEG signal includes: Determine the preset rules; According to the preset rule, the first anomaly detection result, the second anomaly detection result, the third anomaly detection result and the fourth anomaly detection result are fused to obtain a target fusion result; The quality of the initial EEG signal is judged according to the target fusion result to obtain the quality detection result corresponding to the initial EEG signal.

4. A quality determination device based on electroencephalogram signals, characterized in that: include: A data processing module is used to collect initial EEG signals and pre-process the initial EEG signals to obtain target EEG signals; The first detection module is used to perform full rank detection on the target EEG signal to obtain a first abnormality detection result corresponding to the target EEG signal, including: selecting a baseline time period, and determining a target baseline signal corresponding to the baseline time period; removing the target baseline signal from the target EEG signal to obtain a baseline-removed signal corresponding to the target EEG signal; establishing a corresponding target signal matrix according to the baseline-removed signal, and calculating a covariance matrix corresponding to the baseline-removed signal according to the target signal matrix; performing matrix rank solution on the covariance matrix to obtain a target rank corresponding to the target EEG signal; and determining the first abnormality detection result corresponding to the target EEG signal according to the target rank and the target number of channels corresponding to the target EEG signal; The second detection module is used to perform discrete degree detection on the target EEG signal to obtain a second abnormality detection result corresponding to the target EEG signal, including: performing difference calculation between adjacent sampling points of the target EEG signal to obtain a target difference; determining a target gradient corresponding to the target EEG signal according to the target difference; obtaining a maximum gradient corresponding to the target EEG signal from the target gradient; determining a first discrete value corresponding to the target EEG signal according to the maximum gradient and a preset gradient; performing root mean square detection on the target EEG signal to obtain a second discrete value corresponding to the target EEG signal; and determining the second abnormality detection result corresponding to the target EEG signal according to the first discrete value and the second discrete value. A third detection module is used to perform power supply interference detection on the target EEG signal to obtain a third abnormal detection result corresponding to the target EEG signal, including: determining a first frequency corresponding to the power supply interference, and performing spectrum analysis on the target EEG signal to obtain target spectrum information corresponding to the target EEG signal; obtaining a first power spectrum density corresponding to the first frequency from the target spectrum information, and determining a first energy corresponding to the first frequency according to the first power spectrum density; determining a second frequency corresponding to an alpha wave, and obtaining a second power spectrum density corresponding to the second frequency from the target spectrum information; determining a second energy corresponding to the second frequency according to the second power spectrum density; and determining the third abnormal detection result corresponding to the target EEG signal according to the first energy and the second energy; The fourth detection module is used to perform fluctuation detection on the target EEG signal to obtain a fourth abnormal detection result corresponding to the target EEG signal, including: determining the target signal mean corresponding to the target EEG signal, and calculating the absolute deviation between the target EEG signal and the target signal mean; determining the average amplitude difference corresponding to the target EEG signal according to the absolute deviation; calculating the fourth-order central moment corresponding to the target EEG signal and the target variance corresponding to the target EEG signal; determining the kurtosis corresponding to the target EEG signal according to the fourth-order central moment and the target variance; calculating the third-order central moment corresponding to the target EEG signal and the target standard deviation corresponding to the target EEG signal; determining the skewness corresponding to the target EEG signal according to the third-order central moment and the target standard deviation; performing fluctuation detection on the target EEG signal according to the average amplitude difference, the kurtosis and the skewness to obtain the fourth abnormal detection result; A quality determination module is used to fuse the first abnormality detection result, the second abnormality detection result, the third abnormality detection result and the fourth abnormality detection result to determine the quality detection result corresponding to the initial EEG signal.

5. A terminal device, characterized in that: The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the quality determination method based on EEG signals according to any one of claims 1 to 3 when executing the computer program.

6. A computer storage medium, characterized in that: The computer storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the quality determination method based on EEG signals according to any one of claims 1 to 3.

Images