Electroencephalogram signal data processing method and device

Abstract

The application discloses an electroencephalogram signal data processing method and device. The method comprises the following steps: acquiring multi-channel physiological signal observation data in a preset time window, performing independent component analysis to obtain a corresponding confusion matrix and multi-dimensional independent signals; dividing the preset time window into multiple sub-time windows, restoring the physiological signal of each channel in each sub-time window according to the confusion matrix and the independent signal of each dimension, determining a global artifact signal or a local artifact signal according to the amplitude of the physiological signal; and / or determining an electrooculogram artifact signal according to the target row elements corresponding to the independent signal of each dimension in the confusion matrix; and / or determining an electrocardiogram artifact signal according to a preset electrocardiogram channel signal or a Q wave recognition algorithm; removing the artifact signal in the independent signal, and determining a target electroencephalogram signal according to the confusion matrix and the independent signal after removing the artifact. The application solves the technical problem that it is difficult to accurately remove the artifact in the electroencephalogram signal in the related art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to an electroencephalogram signal data processing method and device. BACKGROUND

[0002] Electroencephalogram (EEG) is a non-invasive brain activity observation technology, which has the advantages of little harm to patients and simple operation, and is widely used in the diagnosis of nervous system diseases, fatigue and depression detection, and brain function area research. However, the amplitude of the brain electrical signal generated by the brain activity is weak, usually in the order of hundreds of microvolts, and the collected signal is easily disturbed by electrocardiogram, electrooculogram, electromyogram and other physiological signals, and the sliding and poor contact of the collection device itself, showing the characteristics of difficult-to-distinguish artifacts (also known as artifacts, artifacts, etc.). High-precision artifact removal is crucial for the research and clinical interpretation of electroencephalogram signals.

[0003] At present, the mainstream electroencephalogram artifact removal scheme includes BSS (Blind Source Separation, blind source separation) algorithm and deep learning-based noise reduction method. Among them, the traditional BSS algorithm mostly relies on ICA (Independent Component Analysis, independent component analysis) technology, and uses artificial or semi-artificial way to select artifact components; most automatic BSS algorithms are only for eye movement signals, and algorithms are designed to identify high-amplitude baseline drift; the deep learning-based noise reduction method mainly faces two problems of no guarantee of data authenticity and difficulty in constructing training set data, which may cause data distortion, energy drift and other problems after artifact removal. At the same time, when reconstructing the EEG signal by inverse ICA, the traditional method usually sets the identified artifact components to 0 as a whole, but this may interfere and suppress normal electroencephalogram components, affecting the reliability of the final processing result.

[0004] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0005] The embodiments of the present application provide an electroencephalogram signal data processing method and device, to at least solve the technical problem that it is difficult to accurately remove artifacts in electroencephalogram signals in the related art.

[0006] According to one aspect of the embodiments of this application, a method for processing electroencephalogram (EEG) signal data is provided, comprising: acquiring multi-channel physiological signal observation data within a preset time window; performing independent component analysis on the physiological signal observation data to obtain a confusion matrix and multi-dimensional independent signals corresponding to the physiological signal observation data; determining artifact signals from the multi-dimensional independent signals, comprising: dividing the preset time window into multiple sub-time windows; reconstructing the physiological signal of each channel within each sub-time window based on the confusion matrix and the independent signals of each dimension; determining whether the independent signals of each dimension are global artifact signals based on the amplitude of each physiological signal and a preset global amplitude threshold; or determining local artifact signals in the independent signals of each dimension based on the amplitude of each physiological signal and a preset local amplitude threshold; and / or determining whether the independent signals of each dimension are electrooculogram (EOG) artifact signals based on the target row element corresponding to the independent signals of each dimension in the confusion matrix; and / or determining ECG artifact signals in the independent signals of each dimension based on a preset ECG channel signal or a Q-wave recognition algorithm; removing artifact signals from the multi-dimensional independent signals; and determining a target EEG signal based on the confusion matrix and the independent signals after removing artifact signals.

[0007] Optionally, multi-channel physiological signal observation data within a preset time window are acquired, and independent component analysis is performed on the physiological signal observation data to obtain the confusion matrix and multi-dimensional independent signals corresponding to the physiological signal observation data. This includes: acquiring multi-channel physiological signal observation data x(t) within a preset time window, denoted as: x(t) =

[0008] [x1(t),…,x p (t),x p+1 (t),…,x p+q (t)] T ,t=1,…,T,where x i (t) represents the physiological signal observation data of the i-th dimension at time t, i = 1, ..., p + q, p represents the number of dimensions of EEG signals in the physiological signal observation data, q represents the number of dimensions of non-EEG signals in the physiological signal observation data, and T represents the length of the preset time window, which is greater than the preset duration; the signal data relationship is determined as: x(t) = As(t), where,

[0009] s(t)=[s1(t),…,s p (t),s p+1 (t),…,s p+q (t)] T Let represent the multidimensional independent signal to be calculated, and A represent the confusion matrix to be calculated. The confusion matrix and the multidimensional independent signal are solved using the Fast Independent Component Analysis (Fast ICA) method.

[0010] Optionally, determining whether each dimension independent signal is a global artifact signal according to the amplitude of each physiological signal and a preset global amplitude threshold comprises: for each dimension independent signal, determining a first maximum amplitude in the amplitudes of the physiological signals of each channel in each sub-time window according to the independent signal; and if the maximum amplitude is greater than the preset global amplitude threshold, determining that the independent signal is a global artifact signal.

[0011] Optionally, determining a local artifact signal in each dimension independent signal according to the amplitude of each physiological signal and a preset local amplitude threshold comprises: determining a second maximum amplitude of the physiological signal observation data in each channel and each sub-time window; for each sub-time window, determining a target quantile of the second maximum amplitudes of the multiple channels corresponding to the sub-time window, and taking the maximum value of the target quantile and a preset local amplitude threshold as a dynamic threshold corresponding to the sub-time window; for each dimension independent signal, determining a third maximum amplitude in the amplitudes of the physiological signals of each channel in each sub-time window according to the independent signal; and for each sub-time window, if the third maximum amplitude is greater than the dynamic threshold corresponding to the sub-time window, determining that a local signal of the independent signal in the sub-time window is a local artifact signal.

[0012] Optionally, determining whether each dimension independent signal is an electro-oculogram artifact signal according to a target row element corresponding to each dimension independent signal in the confusion matrix comprises: for each dimension independent signal, determining a column vector corresponding to the independent signal in the confusion matrix, determining a first sum value of a first preset number of front target row elements in the column vector, determining a second sum value of all row elements in the column vector, and determining a ratio of the first sum value to the second sum value; sorting the ratios corresponding to the independent signals of each dimension, and determining that the independent signals corresponding to a second preset number of ratios at the front of the sorting are electro-oculogram artifact signals.

[0013] Optionally, the ECG artifact signal in each dimension independent signal is determined according to a preset ECG channel signal or a Q-wave recognition algorithm, including: for each dimension independent signal, determining a correlation coefficient between the independent signal and the preset ECG channel signal, and if the correlation coefficient is greater than a preset correlation coefficient threshold, determining that the independent signal is an ECG artifact signal, wherein the type of the correlation coefficient includes at least one of the following: a Pearson correlation coefficient, a nonlinear correlation coefficient; or for each dimension independent signal, determining a Q-wave occurrence time in the independent signal by using the Q-wave recognition algorithm; for each two adjacent Q-wave occurrence times, determining a sampling number of the physiological signal observation data in each sub-time window and a time difference between the adjacent Q-wave occurrence times, and determining a ratio of the sampling number to the time difference as a set of real-time heart rate data corresponding to a previous Q-wave occurrence time among the adjacent Q-wave occurrence times; for each sub-time window, determining a ratio of a standard deviation to a mean value of a plurality of sets of real-time heart rate data corresponding to a plurality of Q-wave occurrence times in the sub-time window as a coefficient of variation corresponding to the sub-time window, and if the coefficient of variation is less than a preset coefficient of variation threshold, determining that a local signal of the independent signal in the sub-time window is an ECG artifact signal.

[0014] Optionally, the artifact signal in the multi-dimensional independent signal is removed, including: for each dimension independent signal, if the independent signal is determined to be a global artifact signal or an electrooculogram artifact signal or an ECG artifact signal, assigning the independent signal as 0; if the local signal in at least one sub-time window of the independent signal is determined to be a local artifact signal or an ECG artifact signal, performing smoothing processing on the local signal in the independent signal by using a preset window function, wherein the type of the window function includes at least one of the following: a rectangular window function, a trapezoidal window function, a sine window function, a maximum value normalized kernel function.

[0015] According to another aspect of the embodiments of the present application, a brain electrical signal data processing apparatus is also provided, comprising: an acquisition module configured to acquire multi-channel physiological signal observation data in a preset time window, perform independent component analysis on the physiological signal observation data, and obtain a confusion matrix corresponding to the physiological signal observation data and multi-dimensional independent signals; a first determination module configured to determine artifact signals from the multi-dimensional independent signals, including: dividing the preset time window into a plurality of sub-time windows, restoring physiological signals of each channel in each sub-time window according to the confusion matrix and the independent signals of each dimension, determining whether the independent signals of each dimension are global artifact signals according to amplitudes of each physiological signal and a preset global amplitude threshold, or determining local artifact signals in the independent signals of each dimension according to amplitudes of each physiological signal and a preset local amplitude threshold; and / or determining whether the independent signals of each dimension are electro-oculogram artifact signals according to target row elements corresponding to the independent signals of each dimension in the confusion matrix; and / or determining electrocardiogram artifact signals in the independent signals of each dimension according to a preset electrocardiogram channel signal or a Q-wave recognition algorithm; and a second determination module configured to remove the artifact signals from the multi-dimensional independent signals, and determine target brain electrical signals according to the confusion matrix and the independent signals after removing the artifact signals.

[0016] According to another aspect of the embodiments of the present application, a computer program product is also provided, comprising: a computer program, wherein the computer program is executed by a processor to implement the brain electrical signal data processing method described above.

[0017] According to another aspect of the embodiments of the present application, an electronic device is also provided, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the brain electrical signal data processing method described above by the computer program.

[0018] In the embodiments of the present application, ICA is solved by using large window width data, and artifacts are removed locally by using small window width data, wherein the large window width ensures the accuracy of ICA solving, and can depict global noise, and the small window width ensures the accuracy of artifact removal, and can avoid excessive influence on the brain electrical signal; further, considering the time-space characteristics and frequency domain information of the physiological signal observation data, the present application classifies and removes electrocardiogram signals, electro-oculogram signals and other artifact components in the physiological signal observation data based on the signal characteristics of the independent signals themselves and the brain electrical signals reconstructed by inverse ICA transformation of the independent signals, effectively solving the technical problem that artifacts in the brain electrical signal are difficult to accurately remove in the related art. BRIEF DESCRIPTION OF DRAWINGS

[0019] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0020] Figure 1 is a flow diagram of an optional electroencephalogram signal data processing method according to an embodiment of the application;

[0021] Figure 2 is an electroencephalogram signal diagram after removing global artifact signals according to an embodiment of the application;

[0022] Figure 3 is an electroencephalogram signal diagram after removing local artifact signals according to an embodiment of the application;

[0023] Figure 4 is a structural diagram of an optional electroencephalogram signal data processing device according to an embodiment of the application;

[0024] Figure 5 is a structural diagram of an optional electronic device according to an embodiment of the application. DETAILED DESCRIPTION

[0025] In order to make the personnel in the art better understand the scheme of the application, the technical scheme in the embodiments of the application will be clearly and completely described below in conjunction with the drawings in the embodiments of the application. Obviously, the described embodiments are only a part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those of ordinary skill in the art without creative labor should be within the scope of protection of the application.

[0026] It should be noted that the terms "first", "second", and the like in the specification and claims of the application and the drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0027] Embodiment 1

[0028] According to the embodiment of the present application, a method for processing electroencephalogram signal data is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0029] Figure 1 is a flowchart of a method for processing electroencephalogram signal data according to an embodiment of the present application, as shown in Figure 1 , the method comprises the following steps:

[0030] Step S102, acquiring multi-channel physiological signal observation data in a preset time window, performing independent component analysis on the physiological signal observation data to obtain a confusion matrix corresponding to the physiological signal observation data and multi-dimensional independent signals;

[0031] Step S104, determining artifact signals from the multi-dimensional independent signals, comprising: dividing the preset time window into a plurality of sub-time windows, restoring physiological signals of each channel in each sub-time window according to the confusion matrix and the independent signals of each dimension, determining whether the independent signals of each dimension are global artifact signals according to the amplitudes of each physiological signal and a preset global amplitude threshold, or determining local artifact signals in the independent signals of each dimension according to the amplitudes of each physiological signal and a preset local amplitude threshold; and / or, determining whether the independent signals of each dimension are electrooculogram artifact signals according to target row elements corresponding to the independent signals of each dimension in the confusion matrix; and / or, determining electrocardiogram artifact signals in the independent signals of each dimension according to a preset electrocardiogram channel signal or a Q wave recognition algorithm;

[0032] Step S106, removing the artifact signals in the multi-dimensional independent signals, and determining target electroencephalogram signals according to the confusion matrix and the independent signals after removing the artifact signals.

[0033] The steps of the method for processing electroencephalogram signal data will be described in detail below in combination with a specific implementation process.

[0034] Optionally, first, multi-channel physiological signal observation data x(t) in a preset time window is acquired, denoted as:

[0035] x(t) = [x1(t), …, x p (t), x p+1 (t), …, x p+q (t)] T , t = 1, …, T

[0036] In the formula, x i(t) represents the physiological signal observation data of the i-th dimension at the t-th time point, i = 1, …, p + q, p represents the dimension number of the electroencephalogram signal in the physiological signal observation data, q represents the dimension number of the non-electroencephalogram signal in the physiological signal observation data, q > 0, corresponding to the possible electro-oculogram, electrocardiogram, electromyogram and the like, T represents the length of a preset time window, T is greater than a preset time length, which is usually large, and can be adjusted according to actual conditions, which is not limited here.

[0037] When the ICA analysis is performed, the signal data relationship can be determined as:

[0038] x(t) = As(t)

[0039] In the formula, s(t) = [s1(t), …, sq(t)] represents the physiological signal observation data of the i-th dimension at the t-th time point, i = 1, …, p + q, p represents the dimension number of the electroencephalogram signal in the physiological signal observation data, q represents the dimension number of the non-electroencephalogram signal in the physiological signal observation data, q > 0, corresponding to the possible electro-oculogram, electrocardiogram, electromyogram and the like, T represents the length of a preset time window, T is greater than a preset time length, which is usually large, and can be adjusted according to actual conditions, which is not limited here. p (t), s p+1 (t), …, s p+q (t)] T , represents the independent signal to be calculated in multiple dimensions, which is spatially independent but allows autocorrelation in the time dimension, and A represents the confusion matrix to be calculated. The present application follows the general assumption of ICA, i.e. no model error, the independent signal dimension is equal to the observation data dimension.

[0040] Then, the confusion matrix and the independent signal in multiple dimensions can be solved by using the Fast ICA method.

[0041] Without loss of generality, the physiological signal observation data x(t) can be transformed into spatially independent data z(t) as follows:

[0042]

[0043] In the formula, , represents the sample covariance matrix of x(t), at this time, the covariance matrix of z(t) is I p+q , so the space is not related, and because s(t) is spatially independent, it can be known that H is an orthogonal matrix. If s(t) contains at most one normal component and the standard orthogonal of H is limited, the ICA analysis problem can be solved by maximizing the negative entropy, maximizing the non-normal degree represented by the fourth moment, etc. An optional solving process is as follows:

[0044] First, let s(t) = Wz(t), in which W = H -1 , is the inverse confusion matrix.

[0045] The probability density function of s(t) is determined as f(s(t)), and the differential entropy of s(t) is defined as:

[0046]

[0047] Further, the negative entropy of s(t) is determined based on differential entropy as follows:

[0048]

[0049] where s gauss (t) is a random variable with the same mean and covariance matrix as s(t). Since the negative entropy is linear transformation invariant, for any non-zero matrix A, there exists J(x(t)) = J(As(t)) = J(s(t)).

[0050] The correlation between random variables can be represented by cross entropy, and for independent random variables, the cross entropy can be represented as:

[0051]

[0052] According to the linear transformation invariance of the negative entropy, maximizing the variable independence is equivalent to minimizing the cross entropy, which is equivalent to maximizing J(s i (t)) for each dimension separately. To achieve this, the following approximation can be used:

[0053] J(s i (t)) ≈ c(E(G(s i (t)) / )-E(G(v))

[0054] where c is a preset constant, E(u) represents the expectation of variable u, G(u) = logcoshu, and v is a random variable of a standard normal distribution. At this time, the ICA analysis problem is converted into an optimization problem of the following objective function:

[0055]

[0056] where w i i represents the i-th row vector of the inverse mixing matrix W, and δ jk is the Kronecker delta function, δ jk = 1 only when j = k, otherwise δ jk = 0.

[0057] Alternatively, the Newton method can be used to iteratively solve the above objective function to obtain the target inverse mixing matrix Specifically, in each iteration process, w i is updated by the following objective formula:

[0058]

[0059] where w represents the updated w iTo avoid solving p+q converge to the same point, the updated each is processed to make it orthogonal to each other by using Gram-Schmidt orthogonalization method.

[0060] After the completion of this iteration update, if and w i difference is less than a predetermined threshold, or the number of iterations has reached a predetermined number, stop iteration and determine the target inverse confusion matrix according to the current Otherwise, continue to the next iteration, wherein the w i to be updated in the next iteration is the

[0061] Finally, the multi-dimensional independent signal is determined as: The confusion matrix is determined as:

[0062] According to the above estimation process, the sample covariance matrix and the expected formula in Newton method require a higher sample size for accurate estimation, that is, the accuracy of ICA estimation is affected by the length of the estimated signal. When the signal is stationary, increasing the signal length can improve the accuracy of the estimated and In practical applications, for physiological signal observation data collected at 500Hz, physiological signal observation data with a large window width of a preset time window B=20 minutes can be selected for ICA analysis to obtain high-precision ICA estimation parameters and

[0063] In the analysis of artifact signals, the embodiments of the present application comprehensively consider the spatiotemporal characteristics and frequency domain information of different physiological signals, and based on the signal characteristics of the independent signal itself and the electroencephalogram reconstructed by inverse ICA transformation of the independent signal, classify and remove various artifact components in the physiological signal observation data.

[0064] Optionally, in the analysis of global artifact signals, the following method can be used: for each dimension of the independent signal, determine the first maximum amplitude of the amplitude of the physiological signal in each sub-time window according to the independent signal restoration of each channel; if the maximum amplitude is greater than a predetermined global amplitude threshold, the independent signal is determined as a global artifact signal.

[0065] Since the global artifact usually comes from poor electrode contact, wire friction, sweat of the subject or large motion interference, etc., it has the characteristics of low frequency, high amplitude and dense occurrence in a period of time with the EOG artifact, so it should be globally removed by using its spatial and temporal characteristics. For example, the estimated The vector of the i-th independent signal along time is defined as Since has been normalized, its amplitude can be identified in the multi-channel physiological signal recovered after its inverse ICA transformation.

[0066] Specifically, determine is the i-th independent signal restored multi-channel physiological signal matrix, where is the i-th column vector of the matrix The total observation time, i.e. the preset time window B, is divided into small windows b, where the k-th observation period, i.e. the sub-time window, is Π k =,(k-1)b+1,kb-, at this time, can be defined The interval amplitude of different channels and different observation periods is:

[0067]

[0068] In the formula, represents the maximum amplitude of the i-th independent signal restored multi-channel physiological signal in the j-th electrode channel and the k-th sub-time window, represents the maximum amplitude of the i-th independent signal restored multi-channel physiological signal in the k-th sub-time window, represents the maximum amplitude of the i-th independent signal restored multi-channel physiological signal in the entire preset time window.

[0069] In identifying the global artifact signal, a preset global amplitude threshold τ1 can be determined, and the independent signal of is the global noise component, which can be assigned a value of 0 for denoising in the future. As an example, τ1 = 2000 μV can be taken, which has achieved good results in practice, as shown in Figure 2 The gray solid line represents the original signal, and the black solid line represents the signal after removing the artifact, and it can be seen that the high amplitude oscillation is eliminated after processing, while the original physiological signal is disturbed very little.

[0070] Optionally, when analyzing electrooculography (EOG) artifact signals, the following method can be used: For each independent signal in each dimension, determine the column vector corresponding to the independent signal in the confusion matrix, and determine the first sum of the first preset number of target row elements in the column vector, determine the second sum of all row elements in the column vector, and determine the ratio of the first sum to the second sum; sort the ratios corresponding to the independent signals in each dimension, and determine the independent signal corresponding to the second preset number of ratios that are at the top of the sort as the EOG artifact signal.

[0071] The identification of electrooculography (EOG) artifacts also utilizes their spatial characteristics. Since the influence of EOG signals is most pronounced in the frontal pole region and gradually weakens posteriorly, it is possible to utilize... Extracting prior knowledge about the spatial distribution of coefficients. For example, assuming the electrode channel corresponding to the front pole region is... The first row of the vector Second row The corresponding channels for electrodes Fp1 and Fp2 in the commonly used international 10-20 system can then be calculated.

[0072]

[0073] In the formula, α FP,i express The frontal pole coefficient, as a measure of the energy proportion in the EEG signal channel, can be measured by selecting α. FP,i The two largest independent signals are used as electrooculography artifact signals, and subsequent signals can be assigned a value of 0 for noise reduction. If the physiological signal observation data comes from high-density electrodes, other electrode channels corresponding to the frontal pole region can be selected accordingly, which will not be elaborated here.

[0074] Alternatively, analysis of ECG artifact signals can be performed in the following two ways.

[0075] In the first method, for each dimension of independent signal, the correlation coefficient between the independent signal and the preset ECG channel signal is determined. If the correlation coefficient is greater than the preset correlation coefficient threshold, the independent signal is determined to be an ECG artifact signal. The type of correlation coefficient includes at least one of the following: Pearson correlation coefficient, nonlinear correlation coefficient.

[0076] Specifically, if the ECG channels in the data are known, the identification of ECG signal artifacts can be achieved by calculating the global correlation coefficient. For example, the i-th independent signal can be calculated. With ECG channel [x j (1),…,x j (T)] TThe Pearson correlation coefficient or nonlinear correlation coefficient between the signals is used to select independent signals with a correlation coefficient greater than a preset correlation coefficient threshold τ2 as ECG artifact signals, which can then be assigned an overall value of 0 for noise reduction. τ2 can be set empirically; for example, a setting of 0.7 has shown good results in practice.

[0077] In the second approach, for each independent signal in each dimension, the Q-wave recognition algorithm is used to determine the Q-wave occurrence time in the independent signal. For every two adjacent Q-wave occurrence times, the number of samples of physiological signal observation data within each sub-time window and the time difference between adjacent Q-wave occurrence times are determined. The ratio of the number of samples to the time difference is determined as a set of real-time heart rate data corresponding to the previous Q-wave occurrence time among adjacent Q-wave occurrence times. For each sub-time window, the ratio of the standard deviation to the mean of multiple sets of real-time heart rate data corresponding to multiple Q-wave occurrence times within the sub-time window is determined as the coefficient of variation corresponding to the sub-time window. If the coefficient of variation is less than a preset coefficient of variation threshold, the local signal of the independent signal within the sub-time window is determined to be an ECG artifact signal.

[0078] Specifically, if the ECG channel is unknown, the periodic rhythm characteristics of the ECG can be used to select the corresponding independent signal. For example, the classic Pan-Tompkins algorithm for identifying Q waves in ECG can be considered; for the i-th independent signal... The Pan-Tompkins algorithm can be used to extract the time corresponding to the Q wave in the signal, denoted as *t. i1 ,…,t iK For any two adjacent moments, the real-time heart rate within each sub-time window can be calculated based on their time difference:

[0079]

[0080] In the formula, f s denoted by b, represents the sampling rate per second of the physiological signal observation data, and b is the length of the sub-time window with a small window width.

[0081] Based on real-time heart rate data*f i1 ,…,f i(K-1) The + operator can calculate the coefficient of variation (COP) within each sub-time window, which is the ratio of the standard deviation to the mean. It can select independent signals from sub-time windows with COPs less than a preset COP threshold as ECG artifact signals and perform noise reduction on them. The preset COP threshold can be set empirically, such as 0.2.

[0082] Optionally, in analyzing the local artifact signal, the following can be performed: determining a second maximum amplitude of the physiological signal observation data in each channel and each sub-time window; for each sub-time window, determining a target quantile of the second maximum amplitudes of the multiple channels corresponding to the sub-time window, and taking the maximum value of the target quantile and a preset local amplitude threshold as a dynamic threshold corresponding to the sub-time window; for each independent signal of each dimension, determining a third maximum amplitude of the amplitudes of the physiological signal in each sub-time window of each channel restored according to the independent signal, and for each sub-time window, if the third maximum amplitude is greater than the dynamic threshold corresponding to the sub-time window, determining that the local signal of the independent signal in the sub-time window is a local artifact signal.

[0083] Since the above artifact selection methods are all for specific electro-oculogram, electrocardiogram or large-scale baseline drift cases, more delicate local artifact removal algorithms are needed for more common local artifacts or electromyogram artifacts. The embodiment of the present application considers the variability of electroencephalogram signals, and using a simple fixed threshold method may affect intense brain activity, such as erroneously removing electroencephalogram signals during epileptic seizures. Therefore, a dynamic threshold method is proposed, which uses local features of artifacts to identify them.

[0084] For example, the maximum amplitude of the ith independent signal in the kth sub-time window can be considered and a dynamic threshold is used to select the local artifact period. The maximum amplitude of the jth channel in the kth sub-time window of the original physiological signal observation sequence can be defined as is the original physiological signal observation sequence, and the maximum amplitude of the jth channel in the kth sub-time window is calculated similarly:

[0085]

[0086] For each sub-time window, the maximum amplitude of the jth channel in the kth sub-time window can be considered in the upper 90% quantile of multiple channels

[0087] Setting a dynamic threshold in the time domain According to the dynamic threshold, the independent signal is segmented and detected, which can retain the electroencephalogram signal in intense brain activity and avoid the impact of direct overall removal on active period electroencephalogram. Among them, τ3 is a preset local amplitude threshold, which can be set according to experience, such as setting 800 μV, which has shown good results in practice.

[0088] As an optional implementation, when removing artifact signals from multi-dimensional independent signals, the following method can be used: For each dimension of independent signal, if the independent signal is determined to be a global artifact signal, an electrooculogram (EOG) artifact signal, or an electrocardiogram (ECG) artifact signal, the independent signal is assigned a value of 0; if the local signal corresponding to at least one sub-time window in the independent signal is determined to be a local artifact signal or an ECG artifact signal, the local signal in the independent signal is smoothed using a preset window function, wherein the type of window function includes at least one of the following: a rectangular window function, a trapezoidal window function, a sine window function, and a kernel function with maximum value normalization.

[0089] Understandably, global artifact signals can be directly denoised by setting the entire signal to 0. However, for local artifact signals, it's necessary to consider how to maintain the continuity of the restored signal after artifact removal. To achieve this, a window function in the time domain can be used to suppress the identified local artifact signals and smooth the attenuation intensity in the time dimension.

[0090] For example, if the i-th independent signal is identified as an artifact signal in the k-th sub-time window, a window function can be used to smoothly attenuate the selected independent signal. The specific formula is as follows:

[0091]

[0092] Some commonly used window functions are shown in Table 1. Various kernel functions that normalize the maximum value can also be selected as smoothing windows to process independent signals. Figure 3 This is a schematic diagram of an optional EEG signal after removing local artifacts.

[0093] Table 1

[0094]

[0095] In this embodiment, a large window width is used to optimize the ICA solution, while a small window width is used to locally remove artifacts. The large window width ensures the accuracy of the ICA solution and can characterize global noise, while the small window width ensures the accuracy of artifact removal and avoids excessive influence on the EEG signal. Furthermore, considering the spatiotemporal characteristics and frequency domain information of physiological signal observation data, this application classifies, identifies, and removes artifact components such as ECG and EEG signals in physiological signal observation data based on the independent signal itself and the signal characteristics of the EEG signal reconstructed using the inverse ICA transformation of the independent signal. This effectively solves the technical problem of accurately removing artifacts from EEG signals in related technologies.

[0096] Example 2

[0097] According to an embodiment of this application, an electroencephalogram (EEG) signal data processing device for implementing the EEG signal data processing method in Embodiment 1 is also provided, such as... Figure 4 As shown, the EEG signal data processing device includes at least: an acquisition module 41, a first determination module 42, and a second determination module 43, wherein:

[0098] The acquisition module 41 is used to acquire multi-channel physiological signal observation data within a preset time window, perform independent component analysis on the physiological signal observation data, and obtain the confusion matrix and multi-dimensional independent signals corresponding to the physiological signal observation data.

[0099] The first determining module 42 is used to determine artifact signals from multi-dimensional independent signals, including: dividing a preset time window into multiple sub-time windows; reconstructing the physiological signal of each channel within each sub-time window based on the confusion matrix and the independent signal of each dimension; determining whether the independent signal of each dimension is a global artifact signal based on the amplitude of each physiological signal and a preset global amplitude threshold; or determining the local artifact signal in the independent signal of each dimension based on the amplitude of each physiological signal and a preset local amplitude threshold; and / or determining whether the independent signal of each dimension is an electrooculogram artifact signal based on the target row element corresponding to the independent signal of each dimension in the confusion matrix; and / or determining the electrocardiogram artifact signal in the independent signal of each dimension based on a preset electrocardiogram channel signal or a Q-wave recognition algorithm.

[0100] The second determining module 43 is used to remove artifact signals from the multi-dimensional independent signals and determine the target EEG signal based on the confusion matrix and the independent signals after removing artifact signals.

[0101] The following section describes the functions of each module of the EEG signal data processing device in conjunction with the specific implementation process.

[0102] Optionally, the acquisition module can acquire multi-channel physiological signal observation data x(t) within a preset time window, denoted as: x(t)=[x1(t),…,x p (t),x p+1 (t),…,x p+q (t)] T ,t=1,…,T,where x i (t) represents the physiological signal observation data of the i-th dimension at time t, i = 1, ..., p + q, p represents the number of dimensions of EEG signals in the physiological signal observation data, q represents the number of dimensions of non-EEG signals in the physiological signal observation data, and T represents the length of the preset time window, which is greater than the preset duration; the signal data relationship is determined as: x(t) = As(t), where s(t) = [s1(t), ..., s p (t),s p+1 (t),…,s p+q(t)] T , represents the multi-dimension independent signals to be calculated, A represents the confusion matrix to be calculated; the confusion matrix and the multi-dimension independent signals are solved by using a Fast ICA method.

[0103] Optionally, the first determining module can analyze the global artifact signals in the following manner: for each dimension of the independent signals, a first maximum amplitude of the amplitudes of the physiological signals of each channel in each sub-time window restored according to the independent signals is determined; if the maximum amplitude is greater than a preset global amplitude threshold, the independent signal is determined as a global artifact signal.

[0104] Optionally, the first determining module can analyze the local artifact signals in the following manner: a second maximum amplitude of the physiological signal observation data in each channel and each sub-time window is determined; for each sub-time window, a target quantile of the second maximum amplitudes of the multiple channels corresponding to the sub-time window is determined, and the maximum value of the target quantile and a preset local amplitude threshold is taken as a dynamic threshold corresponding to the sub-time window; for each dimension of the independent signals, a third maximum amplitude of the amplitudes of the physiological signals of each channel in each sub-time window restored according to the independent signals is determined, and for each sub-time window, if the third maximum amplitude is greater than the dynamic threshold corresponding to the sub-time window, the local signal of the independent signal in the sub-time window is determined as a local artifact signal.

[0105] Optionally, the first determining module can analyze the electro-oculogram artifact signals in the following manner: for each dimension of the independent signals, a column vector corresponding to the independent signal in the confusion matrix is determined, a first sum value of a first preset number of target row elements in the column vector is determined, a second sum value of all row elements in the column vector is determined, and a ratio of the first sum value to the second sum value is determined; the ratios corresponding to the independent signals of each dimension are sorted, and the independent signals corresponding to the second preset number of ratios in the front of the sorting are determined as the electro-oculogram artifact signals.

[0106] Optionally, the first determining module can analyze the electrocardio artifact signal in the following manner: for each dimension of the independent signal, determining a correlation coefficient between the independent signal and a preset electrocardio channel signal, and if the correlation coefficient is greater than a preset correlation coefficient threshold, determining that the independent signal is an electrocardio artifact signal, wherein the type of the correlation coefficient includes at least one of the following: a Pearson correlation coefficient, a nonlinear correlation coefficient; or for each dimension of the independent signal, determining a Q-wave occurrence time in the independent signal using a Q-wave identification algorithm; for each two adjacent Q-wave occurrence times, determining a sampling number of the physiological signal observation data in each sub-time window and a time difference between the adjacent Q-wave occurrence times, and determining a ratio of the sampling number to the time difference as a set of real-time heart rate data corresponding to a previous Q-wave occurrence time among the adjacent Q-wave occurrence times; for each sub-time window, determining a ratio of a standard deviation to a mean of a plurality of sets of real-time heart rate data corresponding to a plurality of Q-wave occurrence times in the sub-time window as a coefficient of variation corresponding to the sub-time window, and if the coefficient of variation is less than a preset coefficient of variation threshold, determining that a local signal of the independent signal in the sub-time window is an electrocardio artifact signal.

[0107] Optionally, the second determining module can remove the artifact signal in the multi-dimension independent signal in the following manner: for each dimension of the independent signal, if the independent signal is determined to be a global artifact signal or an electrooculogram artifact signal or an electrocardio artifact signal, assigning the independent signal a value of 0; if a local signal corresponding to at least one sub-time window in the independent signal is determined to be a local artifact signal or an electrocardio artifact signal, performing smoothing processing on the local signal in the independent signal using a preset window function, wherein the type of the window function includes at least one of the following: a rectangular window function, a trapezoidal window function, a sinusoidal window function, a maximum value normalized kernel function.

[0108] It should be noted that each module in the electroencephalogram signal data processing apparatus in the embodiments of the present application corresponds to each implementation step of the electroencephalogram signal data processing method in Embodiment 1. Since Embodiment 1 has been described in detail, the details not embodied in the present embodiment can be referred to Embodiment 1, and will not be described in more detail here.

[0109] Embodiment 3

[0110] According to the embodiments of the present application, a computer program product is also provided, which includes a computer program, wherein the computer program is executed by a processor to implement the electroencephalogram signal data processing method in Embodiment 1.

[0111] According to the embodiments of the present application, a non-volatile storage medium is also provided, which includes a stored computer program, wherein a device in which the non-volatile storage medium is located executes the electroencephalogram signal data processing method in Embodiment 1 by running the computer program.

[0112] According to an embodiment of the present application, a processor for running a computer program is further provided, wherein the computer program performs the brain electrical signal data processing method in embodiment 1 when running.

[0113] According to an embodiment of the present application, an electronic device is further provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to perform the brain electrical signal data processing method in embodiment 1 through the computer program.

[0114] Specifically, the computer program performs the following steps when running: obtaining multi-channel physiological signal observation data in a preset time window, performing independent component analysis on the physiological signal observation data to obtain a confusion matrix corresponding to the physiological signal observation data and multi-dimensional independent signals; determining artifact signals from the multi-dimensional independent signals, including: dividing the preset time window into a plurality of sub-time windows, restoring physiological signals of each channel in each sub-time window according to the confusion matrix and the independent signal of each dimension, determining whether the independent signal of each dimension is a global artifact signal according to the amplitude of each physiological signal and a preset global amplitude threshold, or determining a local artifact signal in the independent signal of each dimension according to the amplitude of each physiological signal and a preset local amplitude threshold; and / or determining whether the independent signal of each dimension is an electrooculogram artifact signal according to a target row element corresponding to the independent signal of each dimension in the confusion matrix; and / or determining an electrocardiogram artifact signal in the independent signal of each dimension according to a preset electrocardiogram channel signal or Q-wave recognition algorithm; removing the artifact signals in the multi-dimensional independent signals, and determining a target brain electrical signal according to the confusion matrix and the independent signals after removing the artifact signals.

[0115] As an optional implementation, the electronic device can exist in the form of a mobile terminal, a computer terminal or a similar computing device. Figure 5 A hardware structure block diagram of an electronic device for implementing a brain electrical signal data processing method is shown. As shown in the figure, Figure 5 The electronic device 50 can include one or more processors 502 (the processor 502 can include but is not limited to a microprocessor MCU or a programmable logic device FPGA processing device), a memory 504 for storing data, and a transmission device 506 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that, Figure 5 The structure shown in the figure is only schematic, and does not limit the structure of the above-mentioned electronic device. For example, the electronic device 50 can also include moreFigure 5 The more or fewer components shown, or having the same Figure 5 The different configurations shown.

[0116] It should be noted that the aforementioned one or more processors 502 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element of the electronic device 50. As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0117] The memory 504 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the EEG signal data processing method in this embodiment. The processor 502 executes various functional applications and data processing by running the software programs and modules stored in the memory 504, thereby implementing the above-mentioned application vulnerability detection method. The memory 504 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 504 may further include memory remotely located relative to the processor 502, and these remote memories can be connected to the electronic device 50 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0118] The transmission device 506 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the electronic device 50. In one example, the transmission device 506 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 506 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0119] The display may be, for example, a touchscreen liquid crystal display (LCD), which allows a user to interact with the user interface of the electronic device 50.

[0120] The sequence numbers of the above embodiments are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0121] In the above-described embodiments of the present application, the description of each embodiment focuses on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0122] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other manners. Among them, the above-described device embodiments are only schematic, for example, the division of units can be a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, and can be electrical or other forms.

[0123] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed to multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0124] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0125] If the integrated unit is realized in the form of software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part of the prior art that makes a contribution or the whole or part of the technical solutions can be embodied in the form of software product, which is stored in a storage medium and includes a plurality of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the embodiments of the present application. The foregoing storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and various program codes that can be stored in the medium.

[0126] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be regarded as the protection scope of the present application.

Claims

1. A method for processing electroencephalogram (EEG) signal data, characterized in that, include: Acquire multi-channel physiological signal observation data within a preset time window, perform independent component analysis on the physiological signal observation data, and obtain the confusion matrix and multi-dimensional independent signals corresponding to the physiological signal observation data; Determining artifact signals from the multi-dimensional independent signals includes: dividing the preset time window into multiple sub-time windows; reconstructing the physiological signal of each channel within each sub-time window based on the confusion matrix and the independent signals of each dimension; determining whether the independent signal of each dimension is a global artifact signal based on the amplitude of each physiological signal and a preset global amplitude threshold, or determining local artifact signals in the independent signals of each dimension based on the amplitude of each physiological signal and a preset local amplitude threshold; and determining whether the independent signal of each dimension is a global artifact signal based on the target row element corresponding to the independent signal of each dimension in the confusion matrix. Whether it is an electrooculogram (EOG) artifact signal, wherein, for an independent signal in each dimension, the column vector corresponding to the independent signal in the confusion matrix is ​​determined, and the first sum of the elements of the target row in the first preset number of rows in the column vector is determined, the second sum of all row elements in the column vector is determined, and the ratio of the first sum to the second sum is determined; the ratios corresponding to the independent signals in each dimension are sorted, and the independent signals corresponding to the second preset number of ratios in the sorting are determined to be EOG artifact signals; the ECG artifact signals in the independent signals of each dimension are determined according to a preset ECG channel signal or Q-wave recognition algorithm; Artifacts are removed from the multidimensional independent signals, and the target EEG signal is determined based on the confusion matrix and the independent signals after artifact removal.

2. The method according to claim 1, characterized in that, Acquire multi-channel physiological signal observation data within a preset time window, perform independent component analysis on the physiological signal observation data to obtain the confusion matrix and multi-dimensional independent signals corresponding to the physiological signal observation data, including: Acquire multi-channel physiological signal observation data within a preset time window. , denoted as: ; In the formula, Indicates the first Each dimension Physiological signal observation data at time, , This indicates the number of dimensions of the electroencephalogram (EEG) signals in the physiological signal observation data. This indicates the number of dimensions of non-EEG signals in the physiological signal observation data. This indicates the length of the preset time window, where T is greater than the preset duration; The signal data relationship is determined as follows: ; In the formula, , represents the multi-dimensional independent signal to be calculated, and A represents the confusion matrix to be calculated; The confusion matrix and multidimensional independent signals are solved using the FastICA method.

3. The method according to claim 1, characterized in that, Determining whether an independent signal in each dimension is a global artifact signal based on the amplitude of each physiological signal and a preset global amplitude threshold includes: For each independent signal, determine the first maximum amplitude of the physiological signal amplitude in each channel within each sub-time window, based on the independent signal reconstruction. If the maximum amplitude is greater than the preset global amplitude threshold, the independent signal is determined to be a global artifact signal.

4. The method according to claim 1, characterized in that, Determining local artifact signals in the independent signals of each dimension based on the amplitude of each physiological signal and a preset local amplitude threshold includes: Determine the second maximum amplitude of the physiological signal observation data within each channel and each sub-time window; For each sub-time window, the target quantile of the second maximum amplitude of the multiple channels corresponding to the sub-time window is determined, and the maximum value of the target quantile and the preset local amplitude threshold is used as the dynamic threshold corresponding to the sub-time window. For each independent signal, the third maximum amplitude corresponding to each sub-time window is determined among the amplitudes of the physiological signals of each channel reconstructed based on the independent signal in each sub-time window. For each sub-time window, if the third maximum amplitude is greater than the dynamic threshold corresponding to the sub-time window, the local signal of the independent signal in the sub-time window is determined to be a local artifact signal.

5. The method according to claim 1, characterized in that, Based on a preset ECG channel signal or Q-wave recognition algorithm, determine the ECG artifact signals in the independent signal of each dimension, including: For each independent signal, a correlation coefficient is determined between the independent signal and the preset ECG channel signal. If the correlation coefficient is greater than a preset correlation coefficient threshold, the independent signal is determined to be an ECG artifact signal. The type of the correlation coefficient includes at least one of the following: Pearson correlation coefficient, nonlinear correlation coefficient; or... For each independent signal, the Q-wave occurrence time in the independent signal is determined using a Q-wave recognition algorithm. For every two adjacent Q-wave occurrence times, the number of samples of the physiological signal observation data within each sub-time window and the time difference between the adjacent Q-wave occurrence times are determined. The ratio of the number of samples to the time difference is determined as a set of real-time heart rate data corresponding to the previous Q-wave occurrence time among the adjacent Q-wave occurrence times. For each sub-time window, the ratio of the standard deviation to the mean of multiple sets of real-time heart rate data corresponding to multiple Q-wave occurrence times within the sub-time window is determined as the coefficient of variation corresponding to the sub-time window. If the coefficient of variation is less than a preset coefficient of variation threshold, the local signal of the independent signal within the sub-time window is determined to be an ECG artifact signal.

6. The method according to claim 1, characterized in that, Removing artifact signals from the multi-dimensional independent signals includes: For each independent signal, if the independent signal is determined to be a global artifact signal, an electrooculogram artifact signal, or an electrocardiogram artifact signal, the independent signal is assigned a value of 0. If the local signal corresponding to at least one sub-time window in the independent signal is determined to be a local artifact signal or an ECG artifact signal, the local signal in the independent signal is smoothed using a preset window function, wherein the type of the window function includes at least one of the following: rectangular window function, trapezoidal window function, sine window function, and kernel function with maximum value normalization.

7. A brainwave signal data processing device, characterized in that, include: The acquisition module is used to acquire multi-channel physiological signal observation data within a preset time window, perform independent component analysis on the physiological signal observation data, and obtain the confusion matrix and multi-dimensional independent signals corresponding to the physiological signal observation data. The first determining module is used to determine artifact signals from the multi-dimensional independent signals, including: dividing the preset time window into multiple sub-time windows; reconstructing the physiological signal of each channel within each sub-time window based on the confusion matrix and the independent signals of each dimension; determining whether the independent signal of each dimension is a global artifact signal based on the amplitude of each physiological signal and a preset global amplitude threshold, or determining local artifact signals in the independent signals of each dimension based on the amplitude of each physiological signal and a preset local amplitude threshold; and determining the target row element corresponding to the independent signal of each dimension in the confusion matrix. Whether an independent signal is an electrooculogram (EOG) artifact signal is determined as follows: For each independent signal in a dimension, the column vector corresponding to the independent signal in the confusion matrix is ​​determined, and the first sum of the elements of the target row in the first preset number of rows in the column vector is determined. The second sum of all row elements in the column vector is determined, and the ratio of the first sum to the second sum is determined. The ratios corresponding to the independent signals in each dimension are sorted, and the independent signals corresponding to the second preset number of ratios in the sorting are determined to be EOG artifact signals. The ECG artifact signals in each independent signal are determined according to a preset ECG channel signal or Q-wave recognition algorithm. The second determining module is used to remove artifact signals from the multi-dimensional independent signals and determine the target EEG signal based on the confusion matrix and the independent signals after removing artifact signals.

8. A computer program product, characterized in that, include: A computer program, wherein when executed by a processor, the computer program implements the electroencephalogram (EEG) signal data processing method according to any one of claims 1 to 6.

9. An electronic device, characterized in that, include: A memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the electroencephalogram (EEG) signal data processing method according to any one of claims 1 to 6 via the computer program.

Images