A bis index correction method of electroencephalogram signal

Abstract

The embodiment of the application discloses a BIS index correction method of electroencephalogram, comprising: acquiring original electroencephalogram in a preset time length; pre-processing the original electroencephalogram to obtain standard electroencephalogram; extracting features from the standard electroencephalogram to obtain multiple target electroencephalograms; calculating the BIS index, sample entropy and LZ complexity of the kth target electroencephalogram; determining a reference standard value according to the sample entropy and the LZ complexity; correcting the BIS index according to the reference standard value to obtain a corrected BIS index; wherein k is an integer greater than 0, and the process is repeated until the corrected BIS index of each target electroencephalogram is obtained. The above method obtains a reference standard value according to the sample entropy and the LZ complexity, corrects the BIS index according to the reference standard value to obtain a corrected BIS index, can improve the stability of the BIS index, improve the accuracy of the BIS index, and thus obtain accurate anesthesia depth monitoring results.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical treatment, in particular to a BIS index correction method of electroencephalogram. BACKGROUND

[0002] Clinically, bispectral index (BIS) or cerebral state index (CSI) of electroencephalogram is often used to monitor the depth of anesthesia of patients during operation, and the depth of anesthesia will directly affect whether the operation can be successfully completed. When the depth of anesthesia is shallow, the patient's body movement, reflex activity, consciousness and other phenomena will be enhanced, which may cause the patient to have a painful memory and psychological barriers. When the depth of anesthesia is too deep, the patient's recovery time will be prolonged, which may cause damage to the central nervous system. Therefore, it is very important to monitor the depth of anesthesia of patients during operation.

[0003] In the prior art, when the BIS index is used to monitor the depth of anesthesia of patients during operation, the BIS index is mainly obtained according to four sub-parameters of burst suppression ratio (BSR), QUAZI suppression index, beta ratio and synch fast slow (SFS) obtained from electroencephalogram, and the BIS index is obtained by calculating the four sub-parameters. However, the BIS index obtained by this method has certain limitations, that is, the BIS index obtained by this method is not completely suitable for all different anesthetic drugs, and the differences between patients, central nervous system diseases and other factors will interfere with the calculation of the BIS index, so that the obtained BIS index is not accurate, thereby affecting the monitoring result of the depth of anesthesia of patients during operation. SUMMARY

[0004] Therefore, it is necessary to propose a BIS index correction method of electroencephalogram to improve the stability of the BIS index, improve the accuracy of the BIS index, and obtain accurate monitoring result of the depth of anesthesia.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides a BIS index correction method of electroencephalogram, which comprises:

[0006] obtaining original electroencephalogram within a preset time length;

[0007] obtaining standard electroencephalogram by preprocessing the original electroencephalogram;

[0008] extracting features of the standard electroencephalogram to obtain a plurality of target electroencephalograms;

[0009] determine a BIS index, a sample entropy and a LZ complexity of the kth target brain electrical signal;

[0010] determine a reference standard value according to the sample entropy and the LZ complexity;

[0011] correct the BIS index according to the reference standard value to obtain a corrected BIS index;

[0012] wherein k is an integer greater than 0, until a corrected BIS index of each target brain electrical signal is obtained.

[0013] Optionally, the preprocessing of the original brain electrical signal to obtain a standard brain electrical signal comprises:

[0014] performing wavelet decomposition and reconstruction processing on the original brain electrical signal based on wavelet transform to obtain a first brain electrical signal free of electro-oculogram interference;

[0015] performing filtering processing on the first brain electrical signal to obtain a second brain electrical signal free of power frequency interference;

[0016] performing wavelet decomposition and reconstruction processing on the second brain electrical signal based on wavelet transform to obtain a third brain electrical signal free of baseline drift;

[0017] performing segmentation and reconstruction processing on the third brain electrical signal using an adaptive threshold method to obtain a fourth brain electrical signal free of electrocautery interference;

[0018] taking the fourth brain electrical signal as the standard brain electrical signal.

[0019] Optionally, the wavelet decomposition and reconstruction processing on the original brain electrical signal based on wavelet transform to obtain a first brain electrical signal free of electro-oculogram interference comprises:

[0020] performing wavelet transform decomposition on the original brain electrical signal to obtain a first low-frequency approximation signal and a first high-frequency detail signal; performing wavelet transform decomposition on the first low-frequency approximation signal to obtain a second low-frequency approximation signal and a second high-frequency detail signal; and performing wavelet transform decomposition on the M-1th low-frequency approximation signal to obtain an Mth low-frequency approximation signal and an Mth high-frequency detail signal, wherein M is a positive integer greater than 1, and M is equal to a first preset threshold value;

[0021] using the formula determine a first frequency f1(Z1) of the Z1th low-frequency approximation signal, and determine a second frequency f2(Z1) of the Z1th high-frequency detail signal according to the first frequency, wherein Z1 is a positive integer greater than 0, and Z1 is equal to M, f s1 is a sampling frequency;

[0022] If the first frequency meets a first preset frequency range, a corresponding low-frequency approximation signal is removed, if the second frequency meets the first preset frequency range, a corresponding high-frequency detail signal is removed, and the remaining low-frequency approximation signal and high-frequency detail signal are reconstructed to obtain the first electroencephalogram signal;

[0023] The wavelet transform is used to decompose and reconstruct the second electroencephalogram signal to obtain a third electroencephalogram signal with baseline drift removed, including:

[0024] The wavelet transform is used to decompose the second electroencephalogram signal to obtain a first low-frequency approximation signal and a first high-frequency detail signal, the wavelet transform is used to decompose the first low-frequency approximation signal to obtain a second low-frequency approximation signal and a second high-frequency detail signal, and so on, the wavelet transform is used to decompose the (L-1)th low-frequency approximation signal to obtain an Lth low-frequency approximation signal and an Lth high-frequency detail signal, where L is a positive integer greater than 1, and L is equal to a second preset threshold value.

[0025] The formula is used The third frequency f3(Z2) of the Z2th low-frequency approximation signal is determined, and the fourth frequency f4(Z2) of the Z2th high-frequency detail signal is determined according to the first frequency, where Z2 is a positive integer greater than 0, and Z2 is equal to L, f s2 is a sampling frequency;

[0026] If the third frequency meets a second preset frequency range, a corresponding low-frequency approximation signal is removed, if the fourth frequency meets the second preset frequency range, a corresponding high-frequency detail signal is removed, and the remaining low-frequency approximation signal and high-frequency detail signal are reconstructed to obtain the third electroencephalogram signal.

[0027] Optionally, the adaptive threshold method is used to segment and reconstruct the third electroencephalogram signal to obtain a fourth electroencephalogram signal with electrotome interference removed, including:

[0028] The third electroencephalogram signal is segmented according to a preset segmentation length to obtain multiple segments of electroencephalogram signals;

[0029] If the variance of the first segment of electroencephalogram signals is greater than an initial threshold value, the first segment of electroencephalogram signals is removed;

[0030] The formula threshold K = F x σ K-1 is used to determine the threshold value threshold K of the Kth segment of electroencephalogram signals;

[0031] If the variance of the Kth segment of electroencephalogram signals is greater than the threshold value of the Kth segment of electroencephalogram signals, the Kth segment of electroencephalogram signals is removed;

[0032] Among them, σ K-1 is the variance of the K-1th segment of the EEG signal, F is the proportional coefficient, and K is a positive integer greater than 1, until K is equal to the number of EEG signal segments;

[0033] The remaining multiple segments of EEG signals are reconstructed to obtain the fourth EEG signal.

[0034] Optionally, the feature extraction of the standard EEG signal to obtain multiple segments of target EEG signals includes:

[0035] The standard EEG signal is segmented and processed according to a preset segmentation interval and a preset signal overlapping time to obtain multiple segments of target EEG signals.

[0036] Optionally, calculating the sample entropy of the k-th target EEG signal includes:

[0037] Sampling and processing the k-th target EEG signal to obtain multiple EEG data;

[0038] When the multiple EEG data are N EEG data {x(1), x(2), x(3), ..., x(N)}, the N EEG data are reconstructed according to the sequence numbers of the N EEG data to obtain N-m+1 m-dimensional vectors {X m (1),X m (2),X m (3),...,X m (N-m+1)}; where X m (h)={x(h),x(h+1),x(h+2),...,x(h+m-1)}, 1≤h≤N-m+1;

[0039] Using formula D i,j = max[|x(i+v)-x(j+v)|] Determine the i-th m-dimensional vector X m (i) with the j-th m-dimensional vector X m (j) The absolute value D of the maximum difference between the corresponding elements v i,j ; Wherein, v is 1, 2, 3, ..., m-1 in sequence, i≠j;

[0040] Count the number of said absolute values ​​less than or equal to the similarity tolerance r num{D i,j ≤r}; where 1≤r≤Nm;

[0041] Using the formula Determine the i-th probability

[0042] Using the formula Get the average value B of all probabilities m (r);

[0043] The dimension of the change vector is m+1, and the above steps are repeated to obtain the average value B m+1 (r);

[0044] The formula is used to obtain the sample entropy SampEn; wherein, ln is the logarithm with e as the base.

[0045] Optionally, the LZ complexity of the kth target electroencephalogram signal is calculated, comprising:

[0046] The kth target electroencephalogram signal is sampled to obtain a plurality of electroencephalogram data;

[0047] When the plurality of electroencephalogram data is N electroencephalogram data {x(1), x(2), x(3),..., x(N)}, the average value X of the N electroencephalogram data is calculated mean ;

[0048] The formula is used to obtain the N electroencephalogram data {y(1), y(2), y(3),..., y(N)} after coarse-grained processing; wherein, n and t are taken as 1, 2, 3,..., N in turn;

[0049] S is defined as {y(1), y(2), y(3),..., y(m)}, Q is defined as {y(m+1), y(m+2), y(m+3),..., y(m+q)}, SQ is obtained by splicing S and Q, SQ1 is obtained by removing the last data of SQ, SQ1 is {y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q-1)};

[0050] Let G1=y(m+q+1), if at least one data in SQ1 is equal to G1, SQ2 is obtained by removing the last data of SQ1, SQ2 is {y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q-2)};

[0051] Similarly, let G g =y(m+q+g), if at least one data in SQ g is equal to G g , SQ g is obtained by removing the last data of SQ g+1= {y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q-g+1)} until SQ g+1 there is no data equal to G g+1 , and taking the value of g as the complexity c(N);

[0052] using the formula to obtain the LZ complexity; wherein, log is the logarithm with base 10.

[0053] Optionally, the BIS index of the kth target electroencephalogram signal is calculated, comprising:

[0054] According to the kth target electroencephalogram signal, a burst suppression ratio, a QUAZI suppression index, a beta ratio, and a synchronization fast-slow ratio are obtained.

[0055] Using the formula BIS = a x BSR + b x QUAZI + c x betaRatio + d x SFS, the BIS index is obtained.

[0056] Wherein, BIS is the BIS index, BSR is the burst suppression ratio, QUAZI is the QUAZI suppression index, betaRatio is the beta ratio, SFS is the synchronization fast-slow ratio, a, b, c, d are respectively the coefficients of the burst suppression ratio, the QUAZI suppression index, the beta ratio, and the synchronization fast-slow ratio, and a + b + c + d = 1.

[0057] Optionally, the reference standard value is determined according to the sample entropy and the LZ complexity, comprising:

[0058] Using the formula T = SampEn 2 - d x LZC + u, the reference standard value T is determined.

[0059] Wherein, SampEn 2 is the square of the sample entropy, LZC is the LZ complexity, and d, u are both constants.

[0060] Optionally, the BIS index is corrected according to the reference standard value to obtain a corrected BIS index, comprising:

[0061] When the reference standard value is greater than a first threshold value and less than or equal to a second threshold value, using the formula BIS f = 0.9 x BIS + a1 x BSR + a2 x QUAZI + b1 x T, the corrected BIS index is obtained.

[0062] When the reference standard value is greater than the second threshold value and less than or equal to a third threshold value, using the formula BIS f= 0.9 x BIS + a3 x betaRatio + b2 x SampEn, wherein BIS is the BIS index, a1 is a coefficient of burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of beta ratio, betaRatio is the beta ratio, b2 is a coefficient of sample entropy, SampEn is the sample entropy, a4 is a coefficient of synchronization fast slow ratio, SFS is the synchronization fast slow ratio, and the modified BIS index is obtained.

[0063] When the reference standard value is greater than the third threshold value and less than or equal to the fourth threshold value, the modified BIS index is obtained by the formula BIS f = 0.95 x BIS + a4 x SFS, wherein BIS is the BIS index, a1 is a coefficient of burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of beta ratio, betaRatio is the beta ratio, b2 is a coefficient of sample entropy, SampEn is the sample entropy, a4 is a coefficient of synchronization fast slow ratio, SFS is the synchronization fast slow ratio, and the modified BIS index is obtained.

[0064] wherein BIS is the BIS index, a1 is a coefficient of burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of beta ratio, betaRatio is the beta ratio, b2 is a coefficient of sample entropy, SampEn is the sample entropy, a4 is a coefficient of synchronization fast slow ratio, SFS is the synchronization fast slow ratio, and the modified BIS index is obtained. f

[0065] To achieve the above object, the present application provides a computer readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method according to any one of the first aspect.

[0066] To achieve the above object, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method according to any one of the first aspect.

[0067] ​The embodiment of the present invention has the following beneficial effects: obtaining raw EEG signals within a preset time length; preprocessing the raw EEG signals to obtain standard EEG signals; performing feature extraction on the standard EEG signals to obtain multiple target EEG signal segments; calculating the BIS index, sample entropy, and LZ complexity of the kth target EEG signal segment; determining a reference standard value based on the sample entropy and LZ complexity; and correcting the BIS index based on the reference standard value to obtain a corrected BIS index; wherein k is sequentially taken as an integer greater than 0 until a corrected BIS index for each target EEG signal segment is obtained. Since sample entropy and LZ complexity are both parameters used to measure signal complexity, and in EEG signals, these two parameters represent the complexity of the EEG signal, and the more complex the EEG signal, the higher the sample entropy and LZ complexity. Therefore, the above method can improve the stability of the BIS index by obtaining a reference standard value based on the sample entropy and LZ complexity, and correcting the BIS index based on the reference standard value to obtain the corrected BIS index, thereby improving the accuracy of the BIS index and obtaining accurate anesthesia depth monitoring results. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0069] in:

[0070] Figure 1 Schematic diagram of a flow chart of a BIS index correction method for an EEG signal in an embodiment of the present application;

[0071] Figure 2 Schematic diagram of performing 6-layer wavelet decomposition on the original EEG signal in an embodiment of the present application;

[0072] Figure 3 Schematic diagram of performing 6-layer wavelet decomposition on the second EEG signal in an embodiment of the present application;

[0073] Figure 4 Schematic diagram of comparison curves of all BIS indices and all revised BIS indices in the examples of the present application;

[0074] Figure 5 1 is a diagram of the internal structure of a computer device in some embodiments. DETAILED DESCRIPTION

[0075] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work are within the scope of protection of the present application.

[0076] Please refer to Figure 1 , a flowchart of a BIS index correction method for an electroencephalogram signal in an embodiment of the present application, the method comprising:

[0077] Step 110: obtaining an original electroencephalogram signal within a preset time length.

[0078] The preset time length is set by an operator according to actual needs. In some embodiments, the preset time length is generally set to 64s. It can be understood that, since the original electroencephalogram signal is obtained by monitoring a patient in surgery in real time, the original electroencephalogram signal obtained within the preset time length is generally used for processing, so as to facilitate processing of the original electroencephalogram signal and improve the processing efficiency of the original electroencephalogram signal.

[0079] Step 120: obtaining a standard electroencephalogram signal by preprocessing the original electroencephalogram signal.

[0080] The preprocessing includes but is not limited to removing electrooculogram interference, filtering out power frequency interference, removing baseline drift, and electrotome interference processing. It can be understood that the purpose of preprocessing is to remove some non-electroencephalogram signal components in the original electroencephalogram signal, so as to avoid the influence of non-electroencephalogram signals on the monitoring result of the electroencephalogram signal and obtain an inaccurate monitoring result of the electroencephalogram signal.

[0081] Step 130: obtaining a plurality of target electroencephalogram signals by performing feature extraction on the standard electroencephalogram signal.

[0082] It should be noted that the feature extraction on the standard electroencephalogram signal is to further reduce the noise level of the standard electroencephalogram signal, improve the authenticity and resolution of the standard electroencephalogram signal, obtain a plurality of useful target electroencephalogram signals, and facilitate the calculation of the BIS index, sample entropy, and LZ complexity of the plurality of electroencephalogram signals, so as to improve the efficiency of calculating the BIS index, sample entropy, and LZ complexity, thereby improving the efficiency of the monitoring result of the electroencephalogram signal.

[0083] Step 140: calculating the BIS index, sample entropy, and LZ complexity of the kth target electroencephalogram signal.

[0084] In some embodiments, for the calculation methods of the BIS index, the sample entropy, and the LZ complexity, the BIS index is calculated by using the burst suppression ratio, the suppression index, the ratio, and the synchronization fast-slow ratio; the sample entropy is calculated by using a signal processing algorithm; and the LZ complexity is calculated by using mean binary. It can be understood that the calculation principles of the BIS index, the sample entropy, and the LZ complexity are known in the prior art.

[0085] Step 150: determining a reference standard value according to the sample entropy and the LZ complexity.

[0086] It should be noted that the sample entropy and the LZ complexity are both parameters for measuring the complexity of a signal, and in the electroencephalogram, the sample entropy and the LZ complexity represent the complexity of the electroencephalogram. The more complex the electroencephalogram is, the higher the sample entropy and the LZ complexity are. Therefore, a reference standard value can be obtained according to the sample entropy and the LZ complexity, that is, the reference standard value is obtained by adding the sample entropy and the LZ complexity according to a certain functional relationship.

[0087] Step 160: correcting the BIS index according to the reference standard value to obtain a corrected BIS index.

[0088] In some embodiments, the BIS index can be corrected according to a certain functional relationship of the burst suppression ratio, the suppression index, the ratio, the synchronization fast-slow ratio, the sample entropy, and the reference standard value according to the value of the reference standard value, so as to obtain a corrected BIS index.

[0089] Step 170: k is sequentially taken as an integer greater than 0 until a corrected BIS index of each target electroencephalogram is obtained.

[0090] For example, if the target electroencephalogram signal obtained in step 130 is 3 segments, then: when k is 1, the BIS index, sample entropy and LZ complexity of the first segment of the target electroencephalogram signal are calculated, the reference standard value is determined according to the sample entropy and the LZ complexity, the BIS index is corrected according to the reference standard value to obtain the corrected BIS index, that is, the corrected BIS index of the first segment of the target electroencephalogram signal is obtained; when k is 2, the BIS index, sample entropy and LZ complexity of the second segment of the target electroencephalogram signal are calculated, the reference standard value is determined according to the sample entropy and the LZ complexity, the BIS index is corrected according to the reference standard value to obtain the corrected BIS index, that is, the corrected BIS index of the second segment of the target electroencephalogram signal is obtained; when k is 3, the BIS index, sample entropy and LZ complexity of the third segment of the target electroencephalogram signal are calculated, the reference standard value is determined according to the sample entropy and the LZ complexity, the BIS index is corrected according to the reference standard value to obtain the corrected BIS index, that is, the corrected BIS index of the third segment of the target electroencephalogram signal is obtained.

[0091] In the embodiment of the present application, by obtaining a reference standard value according to the sample entropy and the LZ complexity, and correcting the BIS index according to the reference standard value to obtain the corrected BIS index, the stability of the BIS index can be improved, the accuracy of the BIS index is improved, and thus an accurate anesthesia depth monitoring result is obtained.

[0092] In a feasible implementation manner, in step 120, the original electroencephalogram signal is preprocessed to obtain the standard electroencephalogram signal, including: performing wavelet decomposition and reconstruction processing on the original electroencephalogram signal based on wavelet transform to obtain a first electroencephalogram signal in which eye interference is removed; performing filtering processing on the first electroencephalogram signal to obtain a second electroencephalogram signal in which power frequency interference is removed; performing wavelet decomposition and reconstruction processing on the second electroencephalogram signal based on wavelet transform to obtain a third electroencephalogram signal in which baseline drift is removed; performing segmentation and reconstruction processing on the third electroencephalogram signal by using an adaptive threshold method to obtain a fourth electroencephalogram signal in which electrotome interference is removed; and taking the fourth electroencephalogram signal as the standard electroencephalogram signal.

[0093] In some embodiments, in order to remove some non-electroencephalogram signal components in the original electroencephalogram signal to avoid the influence of the non-electroencephalogram signal on the monitoring result of the electroencephalogram signal, first, the original electroencephalogram signal is subjected to wavelet decomposition and reconstruction processing based on wavelet transform to obtain a first electroencephalogram signal free of electrooculogram interference, that is, the original electroencephalogram signal is subjected to wavelet decomposition for a preset number of times using wavelet transform, thereby obtaining a plurality of low-frequency approximation signals and a plurality of high-frequency detail signals, the low-frequency approximation signals and the high-frequency detail signals of the non-electroencephalogram signal components (electrooculogram interference) are removed according to the signal frequencies of the low-frequency approximation signals and the high-frequency detail signals, and the remaining low-frequency approximation signals and high-frequency detail signals are subjected to reconstruction processing to obtain the first electroencephalogram signal free of electrooculogram interference; then, the first electroencephalogram signal is subjected to filter processing to obtain a second electroencephalogram signal free of power frequency interference, that is, the first electroencephalogram signal is subjected to filter processing using a wave trap to filter the non-electroencephalogram signal components (power frequency interference) to obtain the second electroencephalogram signal free of power frequency interference, wherein the wave trap can be a 50Hz wave trap; then, the second electroencephalogram signal is subjected to wavelet decomposition and reconstruction processing based on wavelet transform to obtain a third electroencephalogram signal free of baseline drift, that is, the second electroencephalogram signal is subjected to wavelet decomposition for a preset number of times using wavelet transform, thereby obtaining a plurality of low-frequency approximation signals and a plurality of high-frequency detail signals, the low-frequency approximation signals and the high-frequency detail signals of the non-electroencephalogram signal components (baseline drift) are removed according to the signal frequencies of the low-frequency approximation signals and the high-frequency detail signals, and the remaining low-frequency approximation signals and high-frequency detail signals are subjected to reconstruction processing to obtain the third electroencephalogram signal free of baseline drift; finally, the third electroencephalogram signal is subjected to segmentation and reconstruction processing using an adaptive threshold method to obtain a fourth electroencephalogram signal free of electrotome interference, that is, the third electroencephalogram signal is segmented to obtain a plurality of electroencephalogram signals, the threshold value of the current segment of electroencephalogram signal is determined according to the variance of the previous segment of electroencephalogram signal, and when the variance of the current segment of electroencephalogram signal is greater than the threshold value of the current segment of electroencephalogram signal, the current segment of electroencephalogram signal of the non-electroencephalogram signal components (electrotome interference) is removed, and the electroencephalogram signals of the remaining segments are subjected to reconstruction processing to obtain the fourth electroencephalogram signal free of electrotome interference, and the fourth electroencephalogram signal is the standard electroencephalogram signal, wherein each segment of electroencephalogram signal is 0.5s, the first segment of electroencephalogram signal cannot obtain the threshold value of the first segment of electroencephalogram signal because there is no variance of the previous segment of electroencephalogram signal, therefore, a preset threshold value is used as the threshold value of the first segment of electroencephalogram signal in the form of a preset threshold value.

[0094] In an embodiment of the present application, the original EEG signal is first subjected to wavelet decomposition and reconstruction based on wavelet transform to obtain a first EEG signal with eye-electromagnetic interference removed, and then the first EEG signal is filtered to obtain a second EEG signal with power frequency interference filtered out, and then the second EEG signal is subjected to wavelet decomposition and reconstruction based on wavelet transform to obtain a third EEG signal with baseline drift removed, and finally the third EEG signal is segmented and reconstructed using an adaptive threshold method to obtain a fourth EEG signal with electro-tome interference removed, and the fourth EEG signal is used as the standard EEG signal, that is, by removing some non-EEG signal components in the original EEG signal to avoid the non-EEG signal from affecting the monitoring results of the EEG signal, so as to obtain inaccurate EEG signal monitoring results.

[0095] In a feasible implementation, the wavelet decomposition and reconstruction processing of the original EEG signal based on the wavelet transform in the above embodiment to obtain the first EEG signal with eye electrooculogram interference removed, and the wavelet decomposition and reconstruction processing of the second EEG signal based on the wavelet transform to obtain the third EEG signal with baseline drift removed, include the following:

[0096] First, in the above embodiment, the original EEG signal is subjected to wavelet decomposition and reconstruction based on wavelet transform to obtain the first EEG signal without electrooculogram interference, including: performing wavelet transform decomposition on the original EEG signal to obtain a first low-frequency approximate signal and a first high-frequency detail signal; performing wavelet transform decomposition on the first low-frequency approximate signal to obtain a second low-frequency approximate signal and a second high-frequency detail signal; and so on, performing wavelet transform decomposition on the M-1th low-frequency approximate signal to obtain the Mth low-frequency approximate signal and the Mth high-frequency detail signal; wherein M successively takes a positive integer greater than 1 until M is equal to the first preset threshold; using the formula Determine the first frequency f1(Z1) of the Z1th low-frequency approximation signal, and determine the second frequency f2(Z1) of the Z1th high-frequency detail signal based on the first frequency; wherein Z1 successively takes a positive integer greater than 0 until Z1 is equal to M, f s1 is the sampling frequency; if the first frequency satisfies the first preset frequency range, the corresponding low-frequency approximate signal is removed; if the second frequency satisfies the first preset frequency range, the corresponding high-frequency detail signal is removed, and the remaining low-frequency approximate signal and high-frequency detail signal are reconstructed to obtain the first EEG signal.

[0097] The first preset threshold is a preset number of wavelet decomposition, and it can be understood that the first preset threshold is set by an operator according to actual needs. In some embodiments, the first preset threshold is generally set to 6, that is, the original electroencephalogram signal is decomposed by 6 layers of wavelet. The first preset frequency range is also set by an operator according to actual needs. In some embodiments, the first preset frequency range is generally set to 3 Hz to 16 Hz. It can be understood that the frequency range of the electrooculogram interference is 3 Hz to 16 Hz.

[0098] For example, when the first preset threshold is 6, please refer to Figure 2 The first low-frequency approximate signal and the first high-frequency detail signal are obtained by wavelet transform decomposition of the original electroencephalogram signal. The second low-frequency approximate signal and the second high-frequency detail signal are obtained by wavelet transform decomposition of the first low-frequency approximate signal. When M is 3, the third low-frequency approximate signal and the third high-frequency detail signal are obtained by wavelet transform decomposition of the second (i.e., the 2nd low-frequency approximate signal, and the Arabic numerals are replaced by Chinese numerals below) low-frequency approximate signal. When M is 4, the fourth low-frequency approximate signal and the fourth high-frequency detail signal are obtained by wavelet transform decomposition of the third low-frequency approximate signal. When M is 5, the fifth low-frequency approximate signal and the fifth high-frequency detail signal are obtained by wavelet transform decomposition of the fourth low-frequency approximate signal. When M is 6, the sixth low-frequency approximate signal and the sixth high-frequency detail signal are obtained by wavelet transform decomposition of the fifth low-frequency approximate signal. Then, the first frequency of each low-frequency approximate signal is determined by using the formula The first frequency of each low-frequency approximate signal is determined by using the formula. If the first frequency satisfies the frequency range of 3 Hz to 16 Hz, the corresponding low-frequency approximate signal is removed. If the second frequency satisfies the frequency range of 3 Hz to 16 Hz, the corresponding high-frequency detail signal is removed. The remaining low-frequency approximate signal and high-frequency detail signal are reconstructed to obtain the first electroencephalogram signal.

[0099] Then, the second electroencephalogram signal is decomposed and reconstructed by wavelet transform in the above embodiment to obtain the third electroencephalogram signal with baseline drift removed, including: the first low-frequency approximate signal and the first high-frequency detail signal are obtained by wavelet transform decomposition of the second electroencephalogram signal; the second low-frequency approximate signal and the second high-frequency detail signal are obtained by wavelet transform decomposition of the first low-frequency approximate signal; and so on, the Lth low-frequency approximate signal and the Lth high-frequency detail signal are obtained by wavelet transform decomposition of the (L-1)th low-frequency approximate signal; wherein L is a positive integer greater than 1 in turn until L is equal to the second preset threshold; the first frequency of each low-frequency approximate signal is determined by using the formula Determine the third frequency f3(Z2) of the Z2th low-frequency approximation signal, and determine the fourth frequency f4(Z2) of the Z2th high-frequency detail signal based on the first frequency; wherein Z2 successively takes a positive integer greater than 0 until Z2 is equal to L, f s2 is the sampling frequency; if the third frequency meets the second preset frequency range, the corresponding low-frequency approximate signal is removed; if the fourth frequency meets the second preset frequency range, the corresponding high-frequency detail signal is removed, and the remaining low-frequency approximate signal and high-frequency detail signal are reconstructed to obtain the third EEG signal.

[0100] Among them, the second preset threshold is also the preset number of times of wavelet decomposition. It can be understood that the second preset threshold is set by the operator according to actual needs. In some embodiments, the second preset threshold is generally set to 6, that is, the original EEG signal is subjected to 6-layer wavelet decomposition; the second preset frequency range is also set by the operator according to actual needs. In some embodiments, the second preset frequency range is generally set to 0.1Hz to 0.5Hz. It can be understood that the frequency range of baseline drift is 0.1Hz to 0.5Hz.

[0101] For example, when the second preset threshold is 6, please refer to Figure 3 , is a schematic diagram of a 6-layer wavelet decomposition of a second EEG signal in an embodiment of the present application. First, the second EEG signal is decomposed by wavelet transform to obtain a first low-frequency approximate signal and a first high-frequency detail signal. The first low-frequency approximate signal is decomposed by wavelet transform to obtain a second low-frequency approximate signal and a second high-frequency detail signal. When L is 3, the second (i.e., the second low-frequency approximate signal, Chinese numerals are used instead of Arabic numerals hereinafter) low-frequency approximate signal is decomposed by wavelet transform to obtain a third low-frequency approximate signal and a third high-frequency detail signal. When L is 4, the third low-frequency approximate signal is decomposed by wavelet transform to obtain a fourth low-frequency approximate signal and a fourth high-frequency detail signal. When L is 5, the fourth low-frequency approximate signal is decomposed by wavelet transform to obtain a fifth low-frequency approximate signal and a fifth high-frequency detail signal. When L is 6, the fifth low-frequency approximate signal is decomposed by wavelet transform to obtain a sixth low-frequency approximate signal and a sixth high-frequency detail signal. Then, using the formula Determine the third frequency of each low-frequency approximation signal, and determine the corresponding fourth frequency of each high-frequency detail signal based on the third frequency of each low-frequency approximation signal. If the third frequency satisfies the frequency range of 0.1Hz to 0.5Hz, the corresponding low-frequency approximation signal will be removed. If the fourth frequency satisfies the frequency range of 0.1Hz to 0.5Hz, the corresponding high-frequency detail signal will be removed. The remaining low-frequency approximation signal and high-frequency detail signal are reconstructed to obtain the third EEG signal.

[0102] In the embodiment of the present application, the first electroencephalogram signal removing electrooculogram interference is obtained by wavelet decomposition and reconstruction of the original electroencephalogram signal based on wavelet transform in the above embodiment, and the third electroencephalogram signal removing baseline drift is obtained by wavelet decomposition and reconstruction of the second electroencephalogram signal based on wavelet transform, which is an optimized refinement to facilitate the operator to remove electrooculogram interference and baseline drift in this optimized manner, so as to avoid the influence of electrooculogram interference and baseline drift on the monitoring result of the electroencephalogram signal, and obtain an inaccurate monitoring result of the electroencephalogram signal.

[0103] In a feasible implementation, the fourth electroencephalogram signal removing electrotome interference is obtained by segmenting and reconstructing the third electroencephalogram signal based on the adaptive threshold method in the above embodiment, which includes: segmenting the third electroencephalogram signal according to a preset segment duration to obtain a plurality of electroencephalogram signals; if the variance of the first electroencephalogram signal is greater than the initial threshold, the first electroencephalogram signal is removed; using the formula threshold K = F x σ K-1 to determine the threshold threshold K of the Kth electroencephalogram signal; if the variance of the Kth electroencephalogram signal is greater than the threshold of the Kth electroencephalogram signal, the Kth electroencephalogram signal is removed; wherein σ K-1 is the variance of the (K-1)th electroencephalogram signal, F is a proportional coefficient, and K is a positive integer greater than 1 in sequence until K is equal to the number of electroencephalogram signals; the remaining plurality of electroencephalogram signals are reconstructed to obtain the fourth electroencephalogram signal.

[0104] The preset segment duration is set by the operator according to the actual demand, and in some embodiments, the preset segment duration is generally set to 0.5s; and the initial threshold is determined by the operator according to a large amount of historical data or a large amount of historical tests.

[0105] For example, when the preset segment duration is set to 0.5s, the third electroencephalogram signal is segmented into 3 electroencephalogram signals according to 0.5s, first, the variance of the first electroencephalogram signal is calculated (i.e. a plurality of electroencephalogram data are obtained by sampling the first electroencephalogram signal, and the variance of the plurality of electroencephalogram data is calculated), when the variance of the first electroencephalogram signal is greater than the initial threshold, the first electroencephalogram signal is removed, when the variance of the first electroencephalogram signal is less than or equal to the initial threshold, the first electroencephalogram signal is retained; then, when K is 2, the formula threshold K = F x σ K-1Determine the threshold threshold2 of the second segment of the EEG signal, calculate the variance of the second segment of the EEG signal (that is, sample the second EEG signal to obtain multiple EEG data, and calculate the variance of the multiple EEG data). When the variance of the second segment of the EEG signal is greater than the threshold of the second segment of the EEG signal, the second segment of the EEG signal is removed. When the variance of the second segment of the EEG signal is less than or equal to the threshold of the second segment of the EEG signal, the second segment of the EEG signal is retained. Then, when K is 3, use the formula threshold K =F×σ K-1 Determine the threshold threshold3 of the third segment of the EEG signal, calculate the variance of the third segment of the EEG signal (that is, sample the first EEG signal to obtain multiple EEG data, and calculate the variance of the multiple EEG data). When the variance of the third segment of the EEG signal is greater than the threshold of the third segment of the EEG signal, remove the third segment of the EEG signal. When the variance of the third segment of the EEG signal is less than or equal to the threshold of the third segment of the EEG signal, retain the third segment of the EEG signal. Finally, reconstruct the remaining multiple segments of the EEG signal to obtain a fourth EEG signal.

[0106] In an embodiment of the present application, the fourth EEG signal with the electrosurgical interference removed is obtained by segmenting and reconstructing the third EEG signal using the adaptive threshold method in the above embodiment, and is preferably refined so that the operator can remove the electrosurgical interference in this preferred manner to avoid the influence of the electrosurgical interference on the monitoring results of the EEG signal, thereby obtaining inaccurate EEG signal monitoring results.

[0107] In a feasible implementation, in step 130, feature extraction is performed on the standard EEG signal to obtain multiple segments of target EEG signals, including: cutting the standard EEG signal according to a preset cutting interval and a preset signal overlapping time to obtain multiple segments of target EEG signals.

[0108] Among them, the preset cutting interval and the preset signal overlapping time are set by the operator according to actual needs. In some embodiments, the preset cutting interval is generally set to 1.5s, and the preset signal overlapping time is set to 0.5s.

[0109] In an embodiment of the present application, by cutting and processing the standard EEG signal according to a preset cutting interval and a preset signal overlapping time, multiple segments of target EEG signals are obtained, which can reduce the noise level in the standard EEG signal and improve the authenticity and resolution of the standard EEG signal. By obtaining multiple segments of useful target EEG signals, it is convenient to calculate the BIS index, sample entropy, and LZ complexity of the multiple segments of EEG signals respectively, so as to improve the efficiency of calculating the BIS index, sample entropy, and LZ complexity, thereby improving the efficiency of monitoring the EEG signal results.

[0110] In a feasible implementation, at step 140, the sample entropy of the kth target brain electrical signal is calculated, including: sampling the kth target brain electrical signal to obtain a plurality of brain electrical data; when the plurality of brain electrical data is N brain electrical data {x(1), x(2), x(3),..., x(N)}, the N brain electrical data is reconstructed according to the serial number of the N brain electrical data to obtain N-m+1 m-dimensional vectors {X m (1), X m (2), X m (3),..., X m (N-m+1)}; wherein, X m (h) = {x(h), x(h+1), x(h+2),..., x(h+m-1)}, 1≤h≤N-m+1; the maximum difference absolute value D i,j between the i th m-dimensional vector X m (i) and the j th m-dimensional vector X m (j) is determined by the formula D i,j ; wherein, v is 1, 2, 3,..., m-1 in turn, i≠j; the number num{D i,j ≤r} of absolute values less than or equal to the similarity tolerance r is counted; wherein, 1≤r≤N-m; the i th probability P is determined by the formula The average value B m (r) of all probabilities is obtained by the formula The dimension of the vector is changed to m+1, and the above steps are repeated to obtain the average value B m+1 (r); the sample entropy SampEn is obtained by the formula ; wherein, ln is the logarithm with base e.

[0111] In the embodiments of the present application, the sample entropy of the plurality of brain electrical data obtained by sampling the kth target brain electrical signal is calculated, so as to facilitate the operator to calculate the sample entropy by the preferred method.

[0112] In a feasible implementation, at step 140, the sample entropy of the kth target brain electrical signal is calculated, including: sampling the kth target brain electrical signal to obtain a plurality of brain electrical data; when the plurality of brain electrical data is N brain electrical data {x(1), x(2), x(3),..., x(N)}, the average value X mean of the N brain electrical data is calculated; the formula Coarse-grained N electroencephalogram data to obtain coarse-grained N electroencephalogram data {y(1), y(2), y(3),..., y(N)}; wherein, n, t are taken in turn 1, 2, 3,..., N; define S = {y(1), y(2), y(3),..., y(m)}, Q = {y(m+1), y(m+2), y(m+3),..., y(m+q)}, S and Q are spliced to obtain SQ = {y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q)}, remove the last data of SQ to obtain SQ1 = {y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q-1)}; let G1 = y(m+q+1), if at least one data in SQ1 is equal to G1, remove the last data of SQ1 to obtain SQ2 = {y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q-2)}; in this way, let G g =y(m+q+g), if at least one data in SQ g is equal to G g , remove the last data of SQ g to obtain SQ g+1 ={y(1), y(2), y(3),..., y(m), y(m+1), y(m+2), y(m+3),..., y(m+q-g+1)}, until SQ g+1 has no data equal to G g+1 , take the value of g as the complexity c(N); the LZ complexity is obtained by using the formula ; wherein, log is the logarithm with base 10.

[0113] In the embodiments of the application, the LZ complexity is calculated by using the mean value binary of the kth target electroencephalogram signal after sampling processing of the plurality of electroencephalogram data, so as to facilitate the operator to calculate the LZ complexity by the preferred way.

[0114] In a possible implementation, at step 140, the BIS index of the kth target brain electrical signal is calculated, including: obtaining burst suppression ratio, QUAZI suppression index, beta ratio, and synchronization fast-slow ratio from the kth target brain electrical signal; and calculating the BIS index by using the formula BIS=a*BSR+b*QUAZI+c*betaRatio+d*SFS, where BIS is the BIS index, BSR is the burst suppression ratio, QUAZI is the QUAZI suppression index, betaRatio is the beta ratio, SFS is the synchronization fast-slow ratio, a, b, c, and d are coefficients of the burst suppression ratio, the QUAZI suppression index, the beta ratio, and the synchronization fast-slow ratio respectively, and a+b+c+d=1.

[0115] It should be noted that, in the prior art, there is a way to calculate the burst suppression ratio, the QUAZI suppression index, the beta ratio, and the synchronization fast-slow ratio from the brain electrical signal, and thus, details are not repeated here.

[0116] In the embodiments of the present application, the burst suppression ratio, the QUAZI suppression index, the beta ratio, and the synchronization fast-slow ratio are obtained from the kth target brain electrical signal, and the BIS index is calculated based on the four sub-parameters, so as to facilitate the operator to calculate the BIS index by using the preferred way.

[0117] In a possible implementation, at step 150, the reference standard value is determined based on the sample entropy and the LZ complexity, including: determining the reference standard value T by using the formula T=SampEn 2 -d*LZC+u, where SampEn 2 is the square of the sample entropy, LZC is the LZ complexity, and d and u are constants.

[0118] In some embodiments, d can be 0.373, and u can be 0.126. It should be particularly noted that the operator can also adjust the above constants according to the actual situation.

[0119] In the embodiments of the present application, the reference standard value is determined by using the formula, so that the corrected BIS index can be obtained according to the value of the reference standard value, the stability of the BIS index can be improved, the accuracy of the BIS index is improved, and thus an accurate anesthesia depth monitoring result is obtained.

[0120] In a possible implementation, the BIS index is corrected based on the reference standard value to obtain the corrected BIS index, including: when the reference standard value is greater than a first threshold and less than or equal to a second threshold, the corrected BIS index is obtained by using the formula BIS f= 0.9 x BIS + a1 x BSR + a2 x QUAZI + b1 x T, wherein BIS is the modified BIS index, BIS is the BIS index, a1 is a coefficient of the burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of the QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of the beta ratio, betaRatio is the beta ratio, b2 is a coefficient of the sample entropy, SampEn is the sample entropy, and a4 is a coefficient of the synchronization fast-slow ratio, SFS is the synchronization fast-slow ratio. f = 0.9 x BIS + a1 x BSR + a2 x QUAZI + b1 x T, wherein BIS is the modified BIS index, BIS is the BIS index, a1 is a coefficient of the burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of the QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of the beta ratio, betaRatio is the beta ratio, b2 is a coefficient of the sample entropy, SampEn is the sample entropy, and a4 is a coefficient of the synchronization fast-slow ratio, SFS is the synchronization fast-slow ratio. f = 0.9 x BIS + a1 x BSR + a2 x QUAZI + b1 x T, wherein BIS is the modified BIS index, BIS is the BIS index, a1 is a coefficient of the burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of the QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of the beta ratio, betaRatio is the beta ratio, b2 is a coefficient of the sample entropy, SampEn is the sample entropy, and a4 is a coefficient of the synchronization fast-slow ratio, SFS is the synchronization fast-slow ratio. f = 0.9 x BIS + a1 x BSR + a2 x QUAZI + b1 x T, wherein BIS is the modified BIS index, BIS is the BIS index, a1 is a coefficient of the burst suppression ratio, BSR is the burst suppression ratio, a2 is a coefficient of the QUAZI suppression index, QUAZI is the QUAZI suppression index, b1 is a coefficient of the reference standard value, T is the reference standard value, a3 is a coefficient of the beta ratio, betaRatio is the beta ratio, b2 is a coefficient of the sample entropy, SampEn is the sample entropy, and a4 is a coefficient of the synchronization fast-slow ratio, SFS is the synchronization fast-slow ratio.

[0121] In some embodiments, the first threshold is 0, the second threshold is 4, the third threshold is 8, and the fourth threshold is 15. It can be understood that the calculated reference standard value will not exceed the range of 0 to 15. In addition, a1 can be 0.232, a2 can be 0.169, b1 can be 0.102, a3 can be 1.107, b2 can be 0.899, and a4 can be -3.017. It should be particularly noted that the operator can adjust the above coefficients according to the actual situation.

[0122] In the embodiments of the present application, by comparing the value of the reference standard value with the first threshold, the second threshold, the third threshold and the fourth threshold, a corresponding BIS index correction formula is selected to correct the BIS index, so as to obtain the modified BIS index, which can improve the stability of the BIS index and improve the accuracy of the BIS index, thereby obtaining accurate anesthesia depth monitoring results.

[0123] For example, refer to Figure 4 FIG. 2 is a schematic diagram of a comparison curve of all BIS indexes and all modified BIS indexes according to the embodiments of the present application. It can be seen that after the BIS index is corrected, the data fluctuation is small, and more detailed stage changes can be presented, that is, after the BIS index is corrected, the stepwise changes of the anesthesia depth can be observed more clearly, and the brain state and anesthesia degree of the patient can be more clearly and explicitly shown.

[0124] In some embodiments, a computer readable storage medium storing a computer program is provided. The computer program, when executed by a processor, causes the processor to perform the BIS index correction method of an electroencephalogram signal in any of the above method embodiments.

[0125] In some embodiments, a computer device is provided, which includes a memory and a processor. The memory stores a computer program. The computer program, when executed by the processor, causes the processor to perform the BIS index correction method of an electroencephalogram signal in any of the above method embodiments.

[0126] Figure 5 An internal structure diagram of a computer device in some embodiments is shown. The computer device can be a terminal, a server, or a gateway. As shown in the figure, the computer device includes a processor, a memory, and a network interface connected through a system bus. Figure 5

[0127] The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and can also store a computer program. The computer program, when executed by the processor, can cause the processor to implement each step in the above method embodiments. The internal memory can also store a computer program. The computer program, when executed by the processor, can cause the processor to perform each step in the above method embodiments. Those skilled in the art can understand that the computer program can be stored in the internal memory of the computer device, or stored in the non-volatile storage medium, or stored in the internal memory and the non-volatile storage medium at the same time. Figure 5 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0128] Those of ordinary skill in the art can understand that all or part of the processes in the above embodiments can be completed by a computer program instructing related hardware. The program can be stored in a non-volatile computer readable storage medium. When the program is executed, it can include the processes of the above method embodiments.

[0129] ​Any reference to storage, memory, database or other medium herein can include non-volatile and / or volatile storage. Non-volatile storage can include read-only memory (ROM), programmable ROM (PROM), electronically programmable ROM (EPROM), or electrically erasable ROM (EEPROM). Volatile storage can include random-access memory (RAM). By way of illustration, and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct Rambus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM). To provide for interaction with a user, embodiments can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube), LCD (liquid crystal display), LED (light emitting diode) monitor, or the like, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse, trackball, etc., by which the user can provide input to the computer. Input can also be provided to the computer using other input devices not shown in FIG. 1, e.g., a microphone, joystick, game pad, satellite dish, scanner, or the like. The computer can operate in a networked environment using logical connections to one or more remote computers, such as a remote computer. These logical connections are achieved by physical connections to one or more networks, such as a LAN or WAN. The remote computer can be another computer running a comparable desktop environment, or it can be a server, a database, or it can be an application program running on a mainframe computer or in a cloud computing environment. The logic and useful data are transferred from the remote computer to other computers or servers via data information provided by one or more carriers over one or more networks, along with other data and information requested by a user.

[0130] Any combination of one or more computer techniques described above can be employed. In order to make the description concise, not all combinations of the computer techniques described above are described, however, as long as the combinations of the computer techniques do not contradict each other, they should be considered within the scope of the present disclosure.

[0131] The above-described embodiments are merely illustrative for the present application and are not to be used in a limiting manner. It should be understood by those skilled in the art that various modifications and improvements can be made to the present application without departing from the spirit of the present application, which shall fall within the scope of the present application. Therefore, the scope of the present application should be defined by the appended claims.

Claims

1. A method of BIS index correction of an electroencephalogram signal, characterized by, The method comprises: acquiring original electroencephalogram signals within a preset time length; preprocessing the original electroencephalogram signals to obtain standard electroencephalogram signals; extracting features from the standard electroencephalogram signals to obtain multiple segments of target electroencephalogram signals; Computing the first segment target electroencephalogram signal exponent, sample entropy, complexity; determining a reference standard value for the complexity determining a reference standard value for the complexity According to the reference standard value, the index is corrected to obtain a corrected index ​​ Wherein, Take the integer greater than 0 in turn until the modified target electroencephalogram signal of each segment is obtained Exponent.

2. The method of claim 1, wherein, The preprocessing of the original electroencephalogram signals to obtain standard electroencephalogram signals comprises: wavelet decomposition and reconstruction processing of the original electroencephalogram signals based on wavelet transform to obtain first electroencephalogram signals free of electrooculogram interference; filtering processing of the first electroencephalogram signals to obtain second electroencephalogram signals free of power frequency interference; wavelet decomposition and reconstruction processing of the second electroencephalogram signals based on wavelet transform to obtain third electroencephalogram signals free of baseline drift; segmentation and reconstruction processing of the third electroencephalogram signals using an adaptive threshold method to obtain fourth electroencephalogram signals free of electrocautery interference; The fourth electroencephalogram signals are taken as the standard electroencephalogram signals.

3. The method of claim 2, wherein, The wavelet decomposition and reconstruction processing of the original electroencephalogram signals based on wavelet transform to obtain first electroencephalogram signals free of electrooculogram interference comprises: The original brain electrical signal is decomposed by wavelet transform to obtain a first low-frequency approximation signal and a first high-frequency detail signal; the first low-frequency approximation signal is decomposed by wavelet transform to obtain a second low-frequency approximation signal and a second high-frequency detail signal; and the second low-frequency approximation signal is decomposed by wavelet transform to obtain a third low-frequency approximation signal and a third high-frequency detail signal. The third low-frequency approximation signal is decomposed by wavelet transform to obtain a fourth low-frequency approximation signal and a fourth high-frequency detail signal. The fourth low-frequency approximation signal is decomposed by wavelet transform to obtain a fifth low-frequency approximation signal and a fifth high-frequency detail signal. The fifth low-frequency approximation signal is decomposed by wavelet transform to obtain a sixth low-frequency approximation signal and a sixth high-frequency detail signal. A positive integer greater than 1 is sequentially taken until the positive integer is equal to a first preset threshold value. A positive integer greater than 1 is sequentially taken until the positive integer is equal to a first preset threshold value. Using the formula Determine the The first frequency of the low-frequency approximation signal , and determine the first frequency according to the first frequency Second frequency of high-frequency detail signal ;in, Take positive integers greater than 0 in sequence until equal , is the sampling frequency; If the first frequency meets a first preset frequency range, the corresponding low-frequency approximation signal is removed; if the second frequency meets the first preset frequency range, the corresponding high-frequency detail signal is removed; and the remaining low-frequency approximation signal and high-frequency detail signal are reconstructed to obtain the first electroencephalogram signals. The wavelet decomposition and reconstruction processing of the second electroencephalogram signals based on wavelet transform to obtain third electroencephalogram signals free of baseline drift comprises: The second electroencephalogram signal is decomposed by wavelet transform to obtain a first low-frequency approximation signal and a first high-frequency detail signal; the first low-frequency approximation signal is decomposed by wavelet transform to obtain a second low-frequency approximation signal and a second high-frequency detail signal; and the process is repeated to obtain a third low-frequency approximation signal and a third high-frequency detail signal. The second low-frequency approximation signal is decomposed by wavelet transform to obtain a third low-frequency approximation signal and a third high-frequency detail signal. The third low-frequency approximation signal is decomposed by wavelet transform to obtain a fourth low-frequency approximation signal and a fourth high-frequency detail signal. The fourth low-frequency approximation signal is decomposed by wavelet transform to obtain a fifth low-frequency approximation signal and a fifth high-frequency detail signal. A positive integer greater than 1 is sequentially taken until the positive integer is equal to a second preset threshold value. A positive integer greater than 1 is sequentially taken until the positive integer is equal to a second preset threshold value. using the formula determining a first frequency of the low frequency approximation signal and a second frequency of the high frequency detail signal ; wherein, successively taking positive integers greater than 0 until is equal to , is the sampling frequency; If the third frequency meets a second preset frequency range, the corresponding low-frequency approximation signal is removed; if the fourth frequency meets the second preset frequency range, the corresponding high-frequency detail signal is removed; and the remaining low-frequency approximation signal and high-frequency detail signal are reconstructed to obtain the third electroencephalogram signals.

4. The method according to claim 2 or 3, characterized in that, The segmentation and reconstruction processing of the third electroencephalogram signals using an adaptive threshold method to obtain fourth electroencephalogram signals free of electrocautery interference comprises: The third electroencephalogram signals are segmented according to a preset segmentation time length to obtain multiple segments of electroencephalogram signals; If the variance of the first segment of electroencephalogram signals is greater than an initial threshold value, the first segment of electroencephalogram signals is removed; Using the formula determining the first threshold of the brain electrical signal ; Jordi The variance of the EEG signal in the first The threshold of the EEG signal of the first The EEG signal is removed; wherein, is the number of segments, is the variance of the segment of the electroencephalogram signal, is a proportionality coefficient, are positive integers greater than 1 taken in ascending order until is equal to the number of segments of the electroencephalogram signal; The remaining multiple segments of electroencephalogram signals are reconstructed to obtain the fourth electroencephalogram signals.

5. The method of claim 1, wherein, The feature extraction from the standard electroencephalogram signals to obtain multiple segments of target electroencephalogram signals comprises: The standard electroencephalogram signals are cut according to a preset cutting interval and a preset signal overlap time length to obtain multiple segments of target electroencephalogram signals.

6. The method of claim 1, wherein, Calculate the The sample entropy of the target EEG signal is as follows: The method comprises the following steps: Sampling the target brain electrical signals to obtain a plurality of brain electrical data; When multiple EEG data are EEG data When The sequence number of the EEG data will be The EEG data are reconstructed to obtain indivual dimensional vector ;in, , ; Using the formula Determine the indivual dimensional vector With the indivual dimensional vector Corresponding elements The absolute value of the maximum difference ;in, Take in turn , ; counting the number of absolute values less than or equal to a similar tolerance of the number ; wherein ; Using the formula determining the first probability ; Using the formula The average of all probabilities is obtained ; The dimension of the change vector is , repeat the above steps to get the average ; Using the formula the sample entropy is obtained ; wherein is the logarithm to the base .

7. The method of claim 1, wherein, Computing the first Complexity of the segment target electroencephalogram signal, comprising: Complexity of the segment target electroencephalogram signal, comprising: The method comprises the following steps: Sampling the target brain electrical signals to obtain a plurality of brain electrical data; When multiple EEG data are EEG data When calculating The average value of EEG data ; Utilize formula To The rough granulation of 1 brain electrical data is obtained The rough granulation of 1 brain electrical data ; wherein, , All are taken in turn ; Definition , , will and splice to get , remove the last data of get ; Let , if at least one of the data is equal to , then remove the last data of get ; By analogy, let ,like At least one data is equal to , then remove The last data obtained , until There is no data equal to ,Will The value of ; Using the formula obtained the complexity; wherein, is the logarithm to the base 10.

8. The method of claim 1, wherein, Computing the first segment target electroencephalogram signal index, comprising: According to the first segment target brain electrical signal burst suppression ratio, suppression index, ratio, synchronization fast and slow ratio; Using the formula obtained the exponent wherein is the exponent, is the burst suppression ratio, is the suppression exponent, is the ratio, is the synchronization fast-slow ratio, , , , are coefficients of the burst suppression ratio, the suppression exponent, the ratio, the synchronization fast-slow ratio, respectively, and .

9. The method of claim 1, wherein, The complexity determination reference standard value comprises: The complexity determination reference standard value comprises: using the formula determining the reference standard value ; wherein is the square of the sample entropy, is the complexity, , are constants.

10. The method of claim 1, wherein, The index is corrected according to the reference standard value to obtain a corrected index, including: The index is corrected according to the reference standard value to obtain a corrected index, including:​ When the reference standard value is greater than the first threshold value and less than or equal to the second threshold value, the modified index is obtained using the formula exponent;​ When the reference standard value is greater than the second threshold value and less than or equal to a third threshold value, the modified index is obtained using the formula exponent;​ When the reference standard value is greater than the third threshold value and less than or equal to a fourth threshold value, the modified index is obtained using the formula exponent;​ wherein is the modified exponent, is the exponent, is a coefficient of the burst suppression ratio, is the burst suppression ratio, is a coefficient of the suppression exponent, is the suppression exponent, is a coefficient of the reference standard value, is the reference standard value, is a coefficient of the ratio, is the ratio, is a coefficient of the sample entropy, is the sample entropy, is a coefficient of the synchronization speed ratio, is the synchronization speed ratio.

Images