CN112515686B - Electroencephalogram data processing method and device and computer readable storage medium - Google Patents
Electroencephalogram data processing method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN112515686B CN112515686B CN202011374355.9A CN202011374355A CN112515686B CN 112515686 B CN112515686 B CN 112515686B CN 202011374355 A CN202011374355 A CN 202011374355A CN 112515686 B CN112515686 B CN 112515686B
- Authority
- CN
- China
- Prior art keywords
- data
- group
- correlation
- standard
- electroencephalogram
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
Abstract
The invention discloses an electroencephalogram data processing method, which comprises the following steps: s1: acquiring electroencephalogram data of the standard group and the characteristic group; s2, processing the electroencephalogram data of the standard group and the electroencephalogram data of the characteristic group to obtain a plurality of first-order to fourth-order intrinsic mode function components, and calculating the intrinsic mode function components to obtain a sample entropy set; calculating the time sequence correlation of the electroencephalogram data of the standard group and the characteristic group to obtain a correlation set; s3: processing the sample entropy set and the correlation set of the standard group and the feature group to obtain sample entropy index data correlation index data of the standard group and the feature group; s4: calculating boundary conditions according to the sample entropy indicating data and the correlation indicating data of the standard group and the feature group; s5: the electroencephalogram data of the group to be classified is processed by the method to obtain sample entropy indicating data and correlation indicating data, and the grouping category of the electroencephalogram data of the group to be classified is judged by combining boundary conditions. The invention also discloses an electroencephalogram data processing device and a computer readable storage medium.
Description
Technical Field
The invention relates to the technical field of electroencephalogram data processing, in particular to an electroencephalogram data processing method and device based on characteristic information and a computer readable storage medium.
Background
In the prior art, the study on brain waves is mainly carried out through electroencephalography (EEG), and the electroencephalography study divides the EEG into the following categories according to frequency and amplitude: beta waves, high frequency (between 14 and 30Hz, sometimes up to 50 Hz), small amplitude (about 5 μ V); d-waves, the most typical brainwave rhythm, are between 8 and 13Hz, with an amplitude slightly greater than the beta-waves; theta wave, slightly lower in frequency than d-wave, typically 4 to 7Hz, and larger in amplitude than d-wave; the amplitude of the delta wave, the slowest brain wave rhythm, is usually below 3.5Hz, and is at its maximum, up to 300 μ V.
The current state of the target is judged according to the proportion of the four waves and the frequency and amplitude of the corresponding waves, but the current electroencephalogram data processing method is not high in accuracy and cannot accurately reflect the current state of the target. Therefore, the method for classifying the electroencephalogram data with high accuracy is a technical problem to be solved urgently at present.
Disclosure of Invention
In order to solve the technical problems that the accuracy of an electroencephalogram data processing method in the prior art is not high, and the current state of a target cannot be accurately reflected, the present disclosure provides an electroencephalogram data processing method, which comprises:
s1: acquiring electroencephalogram data of a plurality of positions of a standard group and electroencephalogram data of a plurality of positions of a characteristic group;
s2: respectively processing the electroencephalogram data of each position point of the standard group and the electroencephalogram data of each position point of the feature group by adopting an empirical mode decomposition method to respectively obtain a plurality of first-order to fourth-order intrinsic mode function components of the standard group and a plurality of first-order to fourth-order intrinsic mode function components of the feature group; calculating a plurality of inherent mode function components from the first order to the fourth order of the standard group to obtain a standard group sample entropy set; calculating a plurality of first-order to fourth-order intrinsic mode function components of the feature group to obtain a feature group sample entropy set, wherein each position point has a corresponding first-order to fourth-order intrinsic mode function component;
respectively calculating the time sequence correlation of the electroencephalogram data of the standard group and the electroencephalogram data of the correlated sites in the electroencephalogram data of the characteristic group to obtain a standard group correlation set and a characteristic group correlation set;
s3: processing the standard group sample entropy set, the feature group sample entropy set, the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain standard group sample entropy indicating data, feature group sample entropy indicating data, standard group correlation indicating data and feature group correlation indicating data;
s4: computing a boundary condition from the criteria set of sample entropy indicator data, the feature set of sample entropy indicator data, the criteria set of correlation indicator data, and the feature set of correlation indicator data;
s5: and processing the electroencephalogram data of the group to be classified by adopting the methods from S1 to S4 to obtain sample entropy indicating data and correlation indicating data of the group to be classified, and judging the grouping category of the electroencephalogram data of the group to be classified by combining the sample entropy indicating data and the correlation indicating data of the group to be classified with the boundary conditions.
According to some embodiments of the present disclosure, in S2, the calculating the natural modal function components of the plurality of first to fourth orders of the standard set to obtain a standard set sample entropy set includes:
calculating intrinsic mode function components of a first order to a fourth order obtained by calculation of electroencephalogram data of each position point in the standard group and the characteristic group respectively to obtain multi-component quantity sample entropies;
calculating the average value of the multiple groups of component sample entropies to obtain the sample entropies of the current position, combining all the sample entropies obtained by calculation in the standard group into a standard group sample entropy set, and combining all the sample entropies obtained by calculation in the feature group into a feature group sample entropy set.
According to some embodiments of the disclosure, the plurality of sites are sites within the brain that are symmetrically distributed two by two.
According to some embodiments of the present disclosure, in S2, data of two symmetrically distributed sites in the standard set is extracted, the data of the two symmetrically distributed sites in the standard set is divided into N groups according to a band, and time-series correlation data and a power ratio of the data of the two sites of each band are calculated, where N is a positive integer greater than or equal to 2;
forming a correlation set of the standard group by natural logarithms of all time sequence correlation data and power ratios obtained by calculation in the standard group;
extracting data of two sites corresponding to the extraction sites in the standard group in the feature group, dividing the data of the two sites in the feature group into four groups according to wave bands, and calculating time sequence correlation data and a power ratio of the data of the two sites in each wave band;
and forming a correlation set of the feature set by using all the time sequence correlation data obtained by calculation in the feature set and the natural logarithm of the power ratio.
According to some embodiments of the present disclosure, in S3, a significance test method is adopted to calculate the standard group sample entropy set and the feature group sample entropy set respectively to obtain a plurality of point significance difference data, and sample entropies corresponding to k minimum values in the plurality of point significance difference data are extracted as sample entropy indication data;
and respectively calculating the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain a plurality of site significance difference data, and extracting time sequence correlation data or natural logarithm of power ratio corresponding to k minimum values in the plurality of site significance difference data as correlation indication data, wherein k is a positive integer less than 50% of the number of the sites.
According to some embodiments of the disclosure, the S4 comprises: and calculating boundary conditions according to the sample entropy indicating data and the correlation indicating data of all the electroencephalogram data in the standard group and the sample entropy indicating data and the correlation indicating data of all the electroencephalogram data in the characteristic group.
According to some embodiments of the disclosure, the S5 is followed by: and adding the classified groups of electroencephalogram data into corresponding classified groups and updating the electroencephalogram data.
According to some embodiments of the disclosure, the number of feature groups is at least two.
The present disclosure also provides an electroencephalogram data processing device, which includes a storage module, a reading module, a calculating module and an output module;
the acquisition module is used for acquiring the electroencephalogram data of a plurality of sites of the standard group and the electroencephalogram data of a plurality of sites of the characteristic group;
the first calculation module is used for respectively processing the electroencephalogram data of each position point of the standard group and the electroencephalogram data of each position point of the feature group by adopting an empirical mode decomposition method to respectively obtain a plurality of first-order to fourth-order intrinsic mode function components of the standard group and a plurality of first-order to fourth-order intrinsic mode function components of the feature group; calculating a plurality of inherent mode function components from the first order to the fourth order of the standard group to obtain a standard group sample entropy set; calculating a plurality of first-order to fourth-order intrinsic mode function components of the feature group to obtain a feature group sample entropy set, wherein each position point has a corresponding first-order to fourth-order intrinsic mode function component; respectively calculating the time sequence correlation of the electroencephalogram data of the standard group and the electroencephalogram data of the correlated sites in the electroencephalogram data of the characteristic group to obtain a standard group correlation set and a characteristic group correlation set;
the processing module is used for processing the standard group sample entropy set, the feature group sample entropy set, the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain standard group sample entropy indicating data, feature group sample entropy indicating data, standard group correlation indicating data and feature group correlation indicating data;
a second computing module to compute a boundary condition based on the criteria set of sample entropy indicator data, the feature set of sample entropy indicator data, the criteria set of correlation indicator data, and the feature set of correlation indicator data;
the judgment module is used for processing the electroencephalogram data of the group to be classified to obtain sample entropy indication data and correlation indication data of the group to be classified, and judging the grouping category of the electroencephalogram data of the group to be classified according to the combination of the sample entropy indication data and the correlation indication data of the group to be classified and the boundary condition.
The present disclosure also provides a computer-readable storage medium having a program stored thereon, which when executed by a processor, implements the above-described electroencephalogram data processing method.
By the technical scheme, the sample entropy information and the correlation information of the standard group and the characteristic group are obtained by adopting an empirical mode decomposition method and a significance test method, electroencephalogram characteristic signals can be extracted more comprehensively, correlation analysis is performed from two angles of time domain and energy, characteristics of electroencephalogram can be reflected better, and classification and extraction precision of electroencephalogram data can be improved.
Drawings
FIG. 1 schematically illustrates a flow chart of a brain electrical data processing method of an embodiment of the present disclosure;
fig. 2 schematically shows a flowchart of an electroencephalogram data processing method according to a specific embodiment of the present disclosure.
Detailed Description
In order that the objects, technical solutions and advantages of the present invention will become more apparent, the present invention will be further described in detail with reference to the accompanying drawings in conjunction with the following specific embodiments.
It is to be understood that such description is merely illustrative and not intended to limit the scope of the present invention. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the invention. It may be evident, however, that one or more embodiments may be practiced without these specific details. Furthermore, in the following description, descriptions of well-known technologies are omitted so as to avoid unnecessarily obscuring the concepts of the present invention.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "comprising" as used herein indicates the presence of the features, steps, operations but does not preclude the presence or addition of one or more other features.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
In order to solve the technical problems that the accuracy of an electroencephalogram data processing method in the prior art is not high, and the current state of a target cannot be accurately reflected, the electroencephalogram data processing method comprises the following steps.
S1: and acquiring the electroencephalogram data of the plurality of sites of the standard group and the electroencephalogram data of the plurality of sites of the characteristic group.
According to some embodiments of the present disclosure, the selection of the source of the feature set brain electrical data is determined by actual explicit features, such as mood, the feature set is selected to be the brain electrical of a person exhibiting excited mood, and the standard set is selected to be the brain electrical of a person with calm mind.
S2: respectively processing the electroencephalogram data of each site of the standard group and the electroencephalogram data of each site of the characteristic group by adopting an empirical mode decomposition method to respectively obtain a plurality of first-order to fourth-order intrinsic mode function components of the standard group and a plurality of first-order to fourth-order intrinsic mode function components of the characteristic group; calculating a plurality of inherent mode function components from the first order to the fourth order of the standard group to obtain a standard group sample entropy set; and calculating a plurality of first-order to fourth-order intrinsic mode function components of the feature group to obtain a feature group sample entropy set, wherein each position point has a corresponding first-order to fourth-order intrinsic mode function component.
And respectively calculating the time sequence correlation of the electroencephalogram data of the standard group and the electroencephalogram data of the associated sites in the electroencephalogram data of the characteristic group to obtain a standard group correlation set and a characteristic group correlation set.
Among them, empirical Mode Decomposition (EMD) is a novel adaptive signal time-frequency processing method, and is particularly suitable for analysis processing of nonlinear non-stationary signals.
According to some embodiments of the present disclosure, associated sites refer to sites that have an association relationship with each other, e.g., symmetrically distributed sites, or sites that have a direct or indirect impact, etc.
S3: and processing the standard group sample entropy set, the feature group sample entropy set, the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain standard group sample entropy indicating data, feature group sample entropy indicating data, standard group correlation indicating data and feature group correlation indicating data.
According to some embodiments of the present disclosure, optionally, a T-test (T-text-test) is employed in the present disclosure.
S4: boundary conditions are calculated from the criteria set sample entropy indicative data, the feature set sample entropy indicative data, the criteria set correlation indicative data, and the feature set correlation indicative data.
S5: and processing the electroencephalogram data of the group to be classified by adopting methods such as S1 to S4 to obtain sample entropy indicating data and correlation indicating data of the group to be classified, and judging the grouping category of the electroencephalogram data of the group to be classified according to the sample entropy indicating data and the correlation indicating data of the group to be classified and the boundary conditions.
According to some embodiments of the disclosure, the site may take multiple sites within the frontal, parietal, temporal, occipital lobe area.
According to some embodiments of the present disclosure, each site may be sampled to acquire multiple sets of electroencephalogram data.
According to some embodiments of the present disclosure, the step S2 further includes, before the step S2, a step of preprocessing the electroencephalogram data of the standard group and the electroencephalogram data of the feature group, wherein the preprocessing includes removing an electrooculogram and a damaged region in the original electroencephalogram signal by using an average value of the voltage as a reference, and filtering by using a filter to finally obtain a result after the electroencephalogram data is processed.
According to some embodiments of the present disclosure, filtering may be performed with a 0.5-50Hz filter.
According to some embodiments of the present disclosure, in step S2, calculating the natural mode function components of the plurality of first to fourth orders of the standard set to obtain the standard set sample entropy set includes the following sub-steps.
According to some embodiments of the present disclosure, all electroencephalogram data of each site in the standard group and the feature group are preprocessed, optimized, and combined into a corresponding electroencephalogram set.
According to some embodiments of the present disclosure, the multi-component sample entropy is calculated by calculating the eigenmode function components of the first order to the fourth order of the electroencephalogram data of each of the sites in the standard group and the feature group, respectively.
According to some embodiments of the present disclosure, the average value of the multi-component sample entropies is calculated to obtain the sample entropy of the current site, all the sample entropies calculated in the standard group are combined into a standard group sample entropy set, and all the sample entropies calculated in the feature group are combined into a feature group sample entropy set.
According to some embodiments of the disclosure, the plurality of sites are sites within the brain that are symmetrically distributed two by two.
According to some embodiments of the present disclosure, in step S2, data of two symmetrically distributed sites in the standard set are extracted, the data of the two symmetrically distributed sites in the standard set are divided into N groups according to the wave bands, and time sequence correlation data and a power ratio of the data of the two sites of each wave band are calculated, where N is a positive integer greater than or equal to 2.
According to some embodiments of the present disclosure, the different bands of the N =4,4 set are the delta band (0.5-3.5 Hz), the theta band (3.5-7.5 Hz), the alpha band (7.5-13.5 Hz), and the beta band (13.5-30 Hz), respectively.
According to some embodiments of the present disclosure, the correlation set of the standard set is formed by natural logarithms of all the time-series correlation data and the power ratio values calculated in the standard set.
And extracting data of two sites corresponding to the extraction sites in the standard group in the characteristic group, dividing the data of the two sites in the characteristic group into four groups according to wave bands, and calculating time sequence correlation data and power ratio of the data of the two sites in each wave band.
According to some embodiments of the present disclosure, the correlation set of the feature set is formed by natural logarithms of all the time-series correlation data and the power ratio values calculated in the feature set.
According to some embodiments of the present disclosure, in step S3, a significance test method is adopted to calculate the standard group sample entropy set and the feature group sample entropy set respectively to obtain a plurality of site significance difference data, and sample entropies corresponding to k minimum values in the plurality of site significance difference data are extracted as sample entropy indication data, where k is a positive integer less than 50% of the number of sites.
And respectively calculating the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain a plurality of site significance difference data, and extracting time sequence correlation data or natural logarithm of power ratio corresponding to k minimum values in the plurality of site significance difference data as correlation indication data.
According to some embodiments of the present disclosure, step S4 includes calculating the boundary condition D according to the sample entropy indicating data and the correlation indicating data of all the electroencephalogram data in the standard group and the sample entropy indicating data and the correlation indicating data of all the electroencephalogram data in the feature group.
According to some embodiments of the present disclosure, step S5 is followed by adding the classified groups of brain electrical data to the corresponding classified groups and updating the brain electrical data.
According to some embodiments of the disclosure, the number of feature groups is at least two.
According to some embodiments of the present disclosure, different features or associated features are selected as a plurality of feature groups according to actual needs.
The utility model also provides an electroencephalogram data processing device, which comprises a storage module, a reading module, a calculating module and an output module.
And the acquisition module is used for acquiring the electroencephalogram data of the plurality of sites of the standard group and the electroencephalogram data of the plurality of sites of the characteristic group.
The first calculation module is used for respectively processing the electroencephalogram data of each site of the standard group and the electroencephalogram data of each site of the characteristic group by adopting an empirical mode decomposition method to respectively obtain a plurality of first-order to fourth-order intrinsic mode function components of the standard group and a plurality of first-order to fourth-order intrinsic mode function components of the characteristic group; calculating a plurality of inherent mode function components from the first order to the fourth order of the standard group to obtain a standard group sample entropy set; calculating a plurality of first-order to fourth-order intrinsic mode function components of the feature group to obtain a feature group sample entropy set, wherein each position point has a corresponding first-order to fourth-order intrinsic mode function component; and respectively calculating the time sequence correlation of the electroencephalogram data of the correlated sites in the electroencephalogram data of the standard group and the electroencephalogram data of the characteristic group to obtain a standard group correlation set and a characteristic group correlation set.
And the processing module is used for processing the standard group sample entropy set, the feature group sample entropy set, the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain standard group sample entropy indicating data, feature group sample entropy indicating data, standard group correlation indicating data and feature group correlation indicating data.
And the second calculation module is used for calculating the boundary condition according to the standard group sample entropy indication data, the characteristic group sample entropy indication data, the standard group correlation indication data and the characteristic group correlation indication data.
The judgment module is used for processing the electroencephalogram data of the group to be classified to obtain sample entropy indicating data and correlation indicating data of the group to be classified, and judging the grouping category of the electroencephalogram data of the group to be classified according to the sample entropy indicating data and the correlation indicating data of the group to be classified and the boundary conditions.
According to some embodiments of the present disclosure, the first calculation module is further configured to calculate the first-order to fourth-order eigenmode function components of the electroencephalogram data of each of the positions in the standard group and the feature group, respectively, to obtain multi-component quantity sample entropies.
Calculating the average value of the multi-component sample entropies to obtain the sample entropies of the current site, combining all the sample entropies obtained by calculation in the standard group into a standard group sample entropy set, and combining all the sample entropies obtained by calculation in the feature group into a feature group sample entropy set.
According to some embodiments of the present disclosure, the plurality of sites acquired by the acquisition module are sites distributed in a pairwise symmetric manner in the brain.
According to some embodiments of the present disclosure, the first calculation module is further configured to extract data of two symmetrically distributed sites in the standard set, divide the data of the two symmetrically distributed sites in the standard set into N groups according to a band, and calculate a time-series correlation data and a power ratio of the data of the two sites of each band, where N is a positive integer greater than or equal to 2.
And forming a correlation set of the standard group by using the natural logarithms of all the time sequence correlation data and the power ratio values obtained by calculation in the standard group.
And extracting data of two sites corresponding to the extraction sites in the standard group in the characteristic group, dividing the data of the two sites in the characteristic group into four groups according to wave bands, and calculating time sequence correlation data and power ratio of the data of the two sites in each wave band.
And forming a correlation set of the feature set by using all the time sequence correlation data obtained by calculation in the feature set and the natural logarithm of the power ratio.
According to some embodiments of the disclosure, the processing module is configured to calculate the standard group sample entropy set and the feature group sample entropy set by adopting a significance test method to obtain a plurality of location significance difference data, and extract sample entropies corresponding to i minimum values in the plurality of location significance difference data as sample entropy indication data, where i is a positive integer less than 50% of the number of locations.
And respectively calculating the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain a plurality of site significance difference data, and extracting time sequence correlation data or natural logarithm of power ratio corresponding to i minimum values in the plurality of site significance difference data as correlation indication data.
According to some embodiments of the present disclosure, the second calculation module is configured to calculate the boundary condition D from the sample entropy and relevance indicator data of all the electroencephalogram data in the standard set and the sample entropy and relevance indicator data of all the electroencephalogram data in the feature set.
According to some embodiments of the disclosure, the determination module is further configured to add the classified groups of electroencephalogram data to corresponding classified groups and update the electroencephalogram data.
The technical solutions of the present disclosure are further described below with reference to some specific examples, and it should be understood that these specific examples are only for better explaining the above technical solutions, and do not limit the protection scope of the present disclosure.
Fig. 2 schematically shows a flowchart of an electroencephalogram data processing method according to a specific embodiment of the present disclosure.
According to some embodiments of the present disclosure, as shown in fig. 2, the whole process is divided into two parts, the electroencephalogram feature classification standard establishment and the electroencephalogram feature classification, which are specifically as follows.
Establishing an electroencephalogram feature classification standard: preprocessing electroencephalogram data (a standard group and a characteristic group), and then respectively acquiring sample entropy indication data and correlation indication data through two processes. Obtaining IMF components based on empirical mode decomposition and empirical mode decomposition, calculating sample entropy values of the IMF components of all the sites, and selecting 3 sample entropies with most significant differences by adopting t test as sample entropy indicators; dividing the electroencephalogram filtering into 4 wave bands, randomly intercepting 3 groups of wave bands, calculating a time domain correlation average value, calculating an average power correlation, and selecting 3 correlations with most obvious differences as correlation indications by adopting a t test. Then obtaining the indication data, calculating the boundary condition by using a support vector machine, and storing the indication data position and the boundary condition.
Classification of electroencephalogram features: after preprocessing the electroencephalogram data to be classified, respectively acquiring sample entropy indicating data and correlation indicating data through two processes. Obtaining IMF components based on empirical mode decomposition and empirical mode decomposition, and calculating sample entropy of IMF components of each site; the electroencephalogram filtering is divided into 4 wave bands, 3 groups of wave bands are randomly intercepted, the average value of the time domain correlation is calculated, and the average power correlation is calculated. Obtaining the indication data, and judging the category of the electroencephalogram data to be classified according to the boundary conditions.
The electroencephalogram feature classification standard is established and used for providing indication data sites and boundary conditions for electroencephalogram feature classification.
According to some embodiments of the present disclosure, two different sets of electroencephalogram data are selected, respectively a standard set and a feature set, each set containing a plurality of electroencephalogram data.
According to some embodiments of the present disclosure, the sampling frequency is 1KHz for each brain electrical data.
According to some embodiments of the present disclosure, the source of the location of the electroencephalographic data includes: 12 sites including frontal lobe, parietal lobe, temporal lobe and occipital lobe FP1, FP2, F3, F4, F7, F8, P3, P4, T5, T6, O1 and O2.
According to some embodiments of the present disclosure, the bilateral mastoid sites (A1 and A2) are used as reference points.
According to some embodiments of the present disclosure, vertical electro-oculography and horizontal electro-oculography are employed for excluding the effect of electro-oculography on electroencephalogram data acquisition.
According to some embodiments of the present disclosure, a set of electroencephalogram data in a standard set or a characteristic set is selected, and the electrooculogram and the damaged region in the original electroencephalogram signal are rejected using the average value of the voltages of the two lateral mastoids (A1 and A2) as a reference.
According to some embodiments of the present disclosure, the processed electroencephalogram data is obtained by filtering with a 0.5-50Hz filter.
According to some embodiments of the present disclosure, the processed set of brain electrical data is combined, specifically, the data of 12 sites are grouped into a set E (1) = { X = FP1 、X FP2 、X F3 、X F4 、X F7 、X F8 、X P3 、X P4 、X T5 、X T6 、X O1 、X O2 }。
And repeating the steps, and sequentially processing the electroencephalogram data of all the standard groups and the electroencephalogram data of all the characteristic groups to obtain a set E = { E (1), E (2), E (3) }. Wherein n is a positive integer greater than 2, and represents that the standard group and the characteristic group comprise n groups of electroencephalogram data.
1. And calculating the sample entropy of the electroencephalogram data and the sample entropy indicating data.
According to some embodiments of the present disclosure, one preprocessed electroencephalogram data E (m) is selected, where m < n, m being a positive integer.
According to some embodiments of the disclosure, electroencephalographic data X of a site FP1 is selected FP1 Wherein 3 data sets of 30000 data are randomly selected and recorded as (X) FP1 1 、X FP1 2 、X FP1 3 )。
According to some embodiments of the disclosure, X is selected FP1 1 The signal is defined as x (t), and the electroencephalogram signal is decomposed based on an empirical mode decomposition method to obtain an IMF (intrinsic mode function) component of 1-4 orders. The specific calculation method is as follows.
(11) Acquiring all local maximum values and local minimum values of a signal x, acquiring upper and lower envelope lines by using an interpolation algorithm, then obtaining a mean value m of the upper and lower envelope lines, and defining h = x-m;
(12) Judging whether h meets the IMF decomposition termination condition, if not, making x (t) = h, and returning to the step (1); if yes, entering the next step;
(13) To make imf 1 H, representing the ith IMF component obtained by decomposition, and let ri (t) = ri-ci;
(14) IMF component IMF 1 Separating from the original signal x to obtain a residual signal s = x-IMF1, repeating the steps (11) - (13) 3 times by using the residual signal s as a new original signal to obtain an IMF component IMF of order 1-4 1 ，imf 2 ，imf 3 ，imf 4 ；
(15) From IMF component IMF 1 Calculating sample entropy Xc1 based on IMF component IMF 2 Calculate sample entropy Xc2, from IMF component IMF 3 Calculating sample entropy Xc3 from IMF components IMF 4 Calculating sample entropy Xc4;
(16) Calculating the average value of Xc1, xc2, xc3 and Xc4, and taking the average value as the sample entropy XS of the position FP1 FP1 。
According to some embodiments of the present disclosure, the sample entropy values of other sites are calculated according to steps (11) to (16), respectively, and the sample entropy values of all the sites in the set of electroencephalogram data are combined into a set ES (m) = { XS FP1 、XS FP2 、XS F3 、XS F4 、XS F7 、XS F8 、XS P3 、XS P4 、XS T5 、XS T6 、XS o1 、XS O2 }。
According to some embodiments of the present disclosure, a total sample entropy set ES = { ES (1), ES (2), … … ES (n) } of electroencephalogram data of all groups in the standard group and the feature group is calculated in sequence according to the above steps.
According to some embodiments of the present disclosure, the significance difference between the standard group and the feature group is determined by using T test (method of significance test), and 12 site P values of FP1, FP2, F3, F4, F7, F8, P3, P4, T5, T6, O1, O2, etc. are obtained, wherein the P value represents the minimum significance level which does not accept the original hypothesis, and can be directly compared with the selected significance level. A smaller P value indicates a more significant difference.
According to some embodiments of the present disclosure, the sample entropy values of the smallest 3 points among the P values of the 12 points are recorded as { XS1, XS2, XS3} and the 3 points are recorded as { S1, S2, S3} as sample entropy indicating data.
2. And calculating the correlation indication data of the electroencephalogram data.
According to some embodiments of the present disclosure, one preprocessed brain electrical data E (m) is selected.
According to some embodiments of the disclosure, a electroencephalographic data set (X) of site FP1 and site FP2 is selected FP1 ，X FP2 ) The data is filtered into 4 groups: delta wave band (0.5-3.5 Hz), theta wave band (3.5-7.5 Hz), alpha wave band (7.5-13.5 Hz) and beta wave band (13.5-30 Hz). Is recorded as (X) FP1δ ，X FP2δ )，(X FP1θ ，X FP2θ )，(X FP1α ，X FP2α )，(X FP1β ，X FP2β )。
According to some embodiments of the present disclosure, (X) is selected FP1δ ，X FP2δ ) In which 3 sets of data with length of 30000 are randomly intercepted and recorded as [ (X) FP1 1 ，X FP2δ 1 )，(X FP1δ 2 ，X FP2δ 2 )，(X FP1δ 3 ，X FP2δ 3 )]。
According to some embodiments of the disclosure, the meter is separately measuredCalculate X FP1δ 1 And X FP2δ 1 ，X FP1δ 2 And X FP2δ 2 ，X FP1δ 3 And X FP2δ 3 The time sequence correlation between the two is calculated and averaged, and is recorded as V FP1-FP2-δ 。
According to some embodiments of the present disclosure, (X) is calculated FP1 1 ，X FP2δ 1 )，(X FP1δ 2 ，X FP2δ 2 )，(X FP1δ 3 ，X FP2δ 3 ) Is recorded as (P) FP1 1 ，P FP2δ 1 )，(P FP1δ 2 ，P FP2δ 2 )，(P FP1δ 3 ，P FP2δ 3 ) Respectively calculating the average power ratio of 3 groups of electroencephalogram signals, taking the natural logarithm, and then carrying out average calculation on 3 values to obtain P FP1-FP2-δ 。
According to some embodiments of the disclosure, the data V of other 3 bands are obtained according to the above steps FP1-FP2-θ ，V FP1-FP2-α ，V FP1-FP2-β ，P FP1-FP2-θ ，P FP1-FP2-α ，P FP1-FP2-β 。
According to some embodiments of the present disclosure, a total of 8 sets of data, recorded as D, were obtained for site FP1 and site FP2 FP1-FP2 ＝{V FP1-FP2-δ ，P FP1-FP2-δ ，V FP1-FP2-θ ，P FP1-FP2-θ ，V FP1-FP2-α ，P FP1-FP2-α ，V FP1-FP2-β ，P FP1-FP2-β }。
According to some embodiments of the present disclosure, 5 groups of electroencephalogram signals D, such as F3 and F4, F7 and F8, P3 and P4, T5 and T6, and O1 and O2, are calculated according to the above steps F3-F4 ，D F7-F8 ，D FP1-FP2 ，D P3-P4 ，D T5-T6 ，D O1 - O2 ，D FP1-FP2 。
According to some embodiments of the present disclosure, the above 6 sets of 48 data are recorded as EF.
According to some embodiments of the present disclosure, according to the above steps, all the data in the standard group and the feature group are sequentially calculated, and the results are combined into a set EF { }.
According to some embodiments of the present disclosure, the significance difference between the standard set and the feature set was determined using t-test, yielding 48P-values.
According to some embodiments of the present disclosure, the 3 pieces of correlation data with the minimum 48P values are recorded as { XF1, XF2, XF3} and the 3 bits are recorded as { F1, F2, F3} as the index data.
According to some embodiments of the present disclosure, the indication data { XS } for all of the electroencephalographic data in the criteria set and the feature set 1 、XS 2 、XS 3 、XF 1 、XF 2 、XF 3 Dividing the obtained product into 2 groups, calculating boundary conditions by adopting a least square method or a support vector machine method or other algorithms, and recording the boundary conditions as boundary conditions D for storage and standby. At the same time, the locus information { S corresponding to the indication data 1 、S 2 、S 3 }，{F 1 、F 2 、F 3 Store for standby.
3. And (5) processing the electroencephalogram data to be processed.
According to some embodiments of the disclosure, the electroencephalogram data to be processed is preprocessed, and the corresponding sample entropy indicating data and the corresponding correlation indicating data are calculated by adopting the methods in the first step and the second step, so that a sample entropy set ES (m) = { XS is obtained FP1 、XS FP2 、XS F3 、XS F4 、XS F7 、XS F8 、XS P3 、XS P4 、XS T5 、XS T6 、XS O1 、XS O2 }, extracting { S 1 ，S 2 ，S 3 Sample entropies corresponding to 3 bit points, recorded as { XS } 1 、XS 2 、XS 3 And correlation data EF, extracting { F 1 、F 2 、F 3 }, correlation data for 3 sites, recorded as { XF 1 、XF 2 、XF 3 }。
According to some embodiments of the disclosure, indication data { XS (x type dimension) of electroencephalogram data to be processed is obtained through calculation 1 、XS 2 、XS 3 、XF 1 、XF 2 、XF 3 }. And analyzing whether the electroencephalogram data to be processed belongs to the standard group or the characteristic group according to the boundary condition D, and outputting a result.
The present disclosure also provides a computer-readable storage medium having a program stored thereon, which when executed by a processor implements the above-mentioned electroencephalogram data processing method.
Through the technical scheme, the sample entropy information and the correlation information of the standard group and the characteristic group are obtained by adopting an empirical mode decomposition method and a significance test method, the time domain characteristic, the energy characteristic and the nonlinear characteristic of the electroencephalogram are combined for analysis, the electroencephalogram characteristic signal can be extracted more comprehensively, the signal is decomposed into the sum of a plurality of inherent mode functions in the empirical mode decomposition mode, the local characteristic of the data is highlighted by each decomposed IMF component, the characteristic information of the original data can be mastered more accurately and effectively by analyzing the IMF component, the correlation analysis is carried out from two angles of time domain and energy, the characteristics of the electroencephalogram can be reflected better, and the classification and extraction accuracy of the electroencephalogram data can be improved.
So far, the embodiments of the present disclosure have been described in detail with reference to the accompanying drawings. It is to be noted that, in the attached drawings or in the description, the implementation modes not shown or described are all the modes known by the ordinary skilled person in the field of technology, and are not described in detail. In addition, the above definitions of the components are not limited to the specific structures, shapes or manners mentioned in the embodiments, and those skilled in the art may easily modify or replace them.
It should also be noted that, unless otherwise indicated, the numerical parameters set forth in this specification and attached claims are approximations that can vary depending upon the desired properties sought to be obtained by the present disclosure. In particular, all numbers expressing dimensions, range conditions, and so forth, used in the specification and claims are to be understood as being modified in all instances by the term "about". Generally, the expression is meant to encompass variations of ± 10% in some embodiments, 5% in some embodiments, 1% in some embodiments, 0.5% in some embodiments by the specified amount.
It will be appreciated by a person skilled in the art that various combinations and/or combinations of features described in the various embodiments and/or in the claims of the invention are possible, even if such combinations or combinations are not explicitly described in the invention. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present invention may be made without departing from the spirit or teaching of the invention. All such combinations and/or associations are within the scope of the present invention.
The above-mentioned embodiments, objects, technical solutions and advantages of the present invention are further described in detail, it should be understood that the above-mentioned embodiments are only examples of the present invention, and should not be construed as limiting the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. An electroencephalogram data processing method, characterized by comprising:
s1: acquiring electroencephalogram data of a plurality of positions of a standard group and electroencephalogram data of a plurality of positions of a characteristic group;
s2: respectively processing the electroencephalogram data of each position point of the standard group and the electroencephalogram data of each position point of the feature group by adopting an empirical mode decomposition method to respectively obtain a plurality of first-order to fourth-order intrinsic mode function components of the standard group and a plurality of first-order to fourth-order intrinsic mode function components of the feature group; calculating a plurality of inherent modal function components from the first order to the fourth order of the standard group to obtain a standard group sample entropy set; calculating a plurality of first-order to fourth-order intrinsic mode function components of the feature group to obtain a feature group sample entropy set, wherein each position point has a corresponding first-order to fourth-order intrinsic mode function component;
respectively calculating the time sequence correlation of the electroencephalogram data of the standard group and the electroencephalogram data of the correlated sites in the electroencephalogram data of the characteristic group to obtain a standard group correlation set and a characteristic group correlation set, and the method comprises the following steps:
extracting data of two symmetrically distributed sites in the standard group, dividing the data of the two symmetrically distributed sites in the standard group into N groups according to wave bands, and calculating time sequence correlation data and a power ratio of the data of the two sites of each wave band, wherein N is a positive integer greater than or equal to 2;
forming a correlation set of the standard group by natural logarithms of all time sequence correlation data and power ratios obtained by calculation in the standard group;
extracting data of two sites corresponding to the extraction sites in the standard group in the feature group, dividing the data of the two sites in the feature group into four groups according to wave bands, and calculating time sequence correlation data and a power ratio of the data of the two sites in each wave band;
forming a correlation set of the feature group by using all the time sequence correlation data obtained by calculation in the feature group and natural logarithms of the power ratios;
s3: processing the standard group sample entropy set, the feature group sample entropy set, the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain standard group sample entropy indication data, feature group sample entropy indication data, standard group correlation indication data and feature group correlation indication data;
s4: computing a boundary condition from the criteria set of sample entropy indicator data, the feature set of sample entropy indicator data, the criteria set of correlation indicator data, and the feature set of correlation indicator data;
s5: and processing the electroencephalogram data of the group to be classified by adopting the methods from S1 to S4 to obtain sample entropy indication data and correlation indication data of the group to be classified, and judging the grouping category of the electroencephalogram data of the group to be classified according to the combination of the sample entropy indication data and the correlation indication data of the group to be classified and the boundary condition.
2. The method of processing electroencephalogram data according to claim 1, wherein in S2, the calculating the natural mode function components of a plurality of first to fourth orders of the standard set to obtain a standard set of sample entropies includes:
respectively calculating the EEG data of each position point in the standard group and the characteristic group to obtain the first-order to fourth-order intrinsic mode function components, and calculating the intrinsic mode function components to obtain multi-component sample entropies;
calculating the average value of the multiple groups of component sample entropies to obtain the sample entropies of the current position, combining all the sample entropies obtained by calculation in the standard group into a standard group sample entropy set, and combining all the sample entropies obtained by calculation in the feature group into a feature group sample entropy set.
3. The electroencephalogram data processing method according to claim 2, wherein the plurality of sites are sites distributed in a pairwise symmetry manner in the brain.
4. The electroencephalogram data processing method according to claim 3, wherein, in said S3,
calculating the standard group sample entropy set and the feature group sample entropy set respectively by adopting a significance test method to obtain a plurality of site significance difference data, and extracting sample entropies corresponding to k minimum values in the plurality of site significance difference data as sample entropy indication data;
and respectively calculating the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain a plurality of site significance difference data, and extracting time sequence correlation data or natural logarithm of power ratio corresponding to k minimum values in the plurality of site significance difference data as correlation indication data, wherein k is a positive integer less than 50% of the number of the sites.
5. The electroencephalogram data processing method according to claim 4, wherein the S4 includes:
and calculating boundary conditions according to the sample entropy indicating data and the correlation indicating data of all the electroencephalogram data in the standard group and the sample entropy indicating data and the correlation indicating data of all the electroencephalogram data in the characteristic group.
6. The electroencephalogram data processing method according to claim 5, further comprising, after the S5: and adding the classified groups of electroencephalogram data into corresponding classified groups and updating the electroencephalogram data.
7. The brain electrical data processing method according to any one of claims 1 to 6, wherein the number of the feature groups is at least two.
8. The electroencephalogram data processing device is characterized by comprising a storage module, a reading module, a calculating module and an output module;
the acquisition module is used for acquiring the electroencephalogram data of a plurality of sites of the standard group and the electroencephalogram data of a plurality of sites of the characteristic group;
the first calculation module is used for respectively processing the electroencephalogram data of each position point of the standard group and the electroencephalogram data of each position point of the feature group by adopting an empirical mode decomposition method to respectively obtain a plurality of first-order to fourth-order intrinsic mode function components of the standard group and a plurality of first-order to fourth-order intrinsic mode function components of the feature group; calculating a plurality of inherent mode function components from the first order to the fourth order of the standard group to obtain a standard group sample entropy set; calculating a plurality of first-order to fourth-order intrinsic mode function components of the feature group to obtain a feature group sample entropy set, wherein each position point has a corresponding first-order to fourth-order intrinsic mode function component; respectively calculating the time sequence correlation of the electroencephalogram data of the standard group and the electroencephalogram data of the correlated sites in the electroencephalogram data of the characteristic group to obtain a standard group correlation set and a characteristic group correlation set, wherein:
extracting data of two symmetrically distributed sites in the standard group, dividing the data of the two symmetrically distributed sites in the standard group into N groups according to wave bands, and calculating time sequence correlation data and a power ratio of the data of the two sites of each wave band, wherein N is a positive integer greater than or equal to 2;
forming a correlation set of the standard group by natural logarithms of all time sequence correlation data and power ratios obtained by calculation in the standard group;
extracting data of two sites corresponding to the extraction sites in the standard group in the feature group, dividing the data of the two sites in the feature group into four groups according to wave bands, and calculating time sequence correlation data and a power ratio of the data of the two sites in each wave band;
forming a correlation set of the feature group by using all the time sequence correlation data obtained by calculation in the feature group and the natural logarithm of the power ratio;
the processing module is used for processing the standard group sample entropy set, the feature group sample entropy set, the standard group correlation set and the feature group correlation set by adopting a significance test method to obtain standard group sample entropy indicating data, feature group sample entropy indicating data, standard group correlation indicating data and feature group correlation indicating data;
a second computing module to compute a boundary condition based on the criteria set of sample entropy indicator data, the feature set of sample entropy indicator data, the criteria set of correlation indicator data, and the feature set of correlation indicator data;
the judgment module is used for processing the electroencephalogram data of the group to be classified to obtain sample entropy indicating data and correlation indicating data of the group to be classified, and judging the grouping category of the electroencephalogram data of the group to be classified according to the combination of the sample entropy indicating data and the correlation indicating data of the group to be classified and the boundary conditions.
9. A computer-readable storage medium, having a program stored thereon, which, when being executed by a processor, carries out the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011374355.9A CN112515686B (en) | 2020-11-30 | 2020-11-30 | Electroencephalogram data processing method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011374355.9A CN112515686B (en) | 2020-11-30 | 2020-11-30 | Electroencephalogram data processing method and device and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112515686A CN112515686A (en) | 2021-03-19 |
| CN112515686B true CN112515686B (en) | 2022-12-30 |
Family
ID=74995153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011374355.9A Active CN112515686B (en) | 2020-11-30 | 2020-11-30 | Electroencephalogram data processing method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112515686B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007044104A (en) * | 2005-08-08 | 2007-02-22 | Shimadzu Corp | Apparatus and method for signal analysis |
| WO2014191805A1 (en) * | 2013-05-27 | 2014-12-04 | Tata Consultancy Services Limited | Devices and methods for determination of cognitive load |
| CN106803095A (en) * | 2016-12-22 | 2017-06-06 | 辽宁师范大学 | Based on the brain electricity emotion identification method that assemblage characteristic is extracted |
| JP2017209267A (en) * | 2016-05-25 | 2017-11-30 | 国立大学法人鳥取大学 | Apparatus, program, and method for diagnosis support of convulsion invagination type acute encephalopathy |
| CN110367980A (en) * | 2019-07-10 | 2019-10-25 | 南京邮电大学 | EEG signals Emotion identification method based on polynary empirical mode decomposition |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009139119A1 (en) * | 2008-05-15 | 2009-11-19 | パナソニック株式会社 | Device, method, and program which regulate method for identifying electroencephalogram signal |
| US20130060125A1 (en) * | 2010-04-16 | 2013-03-07 | Applied Brain And Vision Sciences Inc. | Encephalography method and apparatus incorporating independent component analysis and a spectral shaping filter |
| TWI489298B (en) * | 2014-01-06 | 2015-06-21 | 國立中央大學 | A multi-scales intrinsic entropies analysis method |
-
2020
- 2020-11-30 CN CN202011374355.9A patent/CN112515686B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007044104A (en) * | 2005-08-08 | 2007-02-22 | Shimadzu Corp | Apparatus and method for signal analysis |
| WO2014191805A1 (en) * | 2013-05-27 | 2014-12-04 | Tata Consultancy Services Limited | Devices and methods for determination of cognitive load |
| JP2017209267A (en) * | 2016-05-25 | 2017-11-30 | 国立大学法人鳥取大学 | Apparatus, program, and method for diagnosis support of convulsion invagination type acute encephalopathy |
| CN106803095A (en) * | 2016-12-22 | 2017-06-06 | 辽宁师范大学 | Based on the brain electricity emotion identification method that assemblage characteristic is extracted |
| CN110367980A (en) * | 2019-07-10 | 2019-10-25 | 南京邮电大学 | EEG signals Emotion identification method based on polynary empirical mode decomposition |
Non-Patent Citations (4)
| Title |
|---|
| Analysis of EEG via Multivariate Empirical Mode Decomposition for Depth of Anesthesia Based on Sample Entropy;Qin Wei ET AL;《Entropy》;20130830;第15卷(第9期);第3458-3470页 * |
| EEG在交通安全领域中的应用研究综述;谭墍元 等;《自动化博览》;20211215;第28-33页 * |
| 信息刺激条件下驾驶人警觉性改善规律;王磊等;《中国公路学报》;20180415(第04期);第28-33页 * |
| 抑郁症静息脑电的小波包节点功率谱熵分析;盖淑萍 等;《传感器与微系统》;20170303;第6-9页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112515686A (en) | 2021-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Singh et al. | Classification of focal and nonfocal EEG signals using features derived from Fourier-based rhythms | |
| Murugappan et al. | Human emotion recognition through short time Electroencephalogram (EEG) signals using Fast Fourier Transform (FFT) | |
| Kumari et al. | Seizure detection in EEG using time frequency analysis and SVM | |
| CN106108894A (en) | A kind of emotion electroencephalogramrecognition recognition method improving Emotion identification model time robustness | |
| Mert | ECG feature extraction based on the bandwidth properties of variational mode decomposition | |
| Klug et al. | Zapline‐plus: A Zapline extension for automatic and adaptive removal of frequency‐specific noise artifacts in M/EEG | |
| Boashash et al. | A review of time–frequency matched filter design with application to seizure detection in multichannel newborn EEG | |
| CN105559779A (en) | Method for carrying out cognitive evaluation through electroencephalo-graph frequency spectrum | |
| CN102824172A (en) | EEG (electroencephalogram) feature extraction method | |
| Al-Salman et al. | Detection of EEG K-complexes using fractal dimension of time frequency images technique coupled with undirected graph features | |
| Al-Salman et al. | An efficient approach for EEG sleep spindles detection based on fractal dimension coupled with time frequency image | |
| Dash et al. | Hidden Markov model based epileptic seizure detection using tunable Q wavelet transform | |
| Zhang et al. | Automatic detection of interictal epileptiform discharges based on time-series sequence merging method | |
| Vempati et al. | A systematic review on automated human emotion recognition using electroencephalogram signals and artificial intelligence | |
| Mei et al. | Identifying sleep spindles with multichannel EEG and classification optimization | |
| Fattah et al. | An approach for classifying alcoholic and non-alcoholic persons based on time domain features extracted from EEG signals | |
| Zhao et al. | Improving generalization based on l1-norm regularization for EEG-based motor imagery classification | |
| Anandaraj et al. | Tree based Ensemble for Enhanced Prediction (TEEP) of epileptic seizures | |
| CN114129163A (en) | Electroencephalogram signal-based emotion analysis method and system for multi-view deep learning | |
| Yuan et al. | Automatic seizure prediction based on modified Stockwell transform and tensor decomposition | |
| CN113558644A (en) | Emotion classification method, medium and equipment for 3D matrix and multidimensional convolution network | |
| CN112515686B (en) | Electroencephalogram data processing method and device and computer readable storage medium | |
| Campanharo et al. | Automated EEG signals analysis using quantile graphs | |
| Torabi et al. | Semantic category-based classification using nonlinear features and wavelet coefficients of brain signals | |
| Shahnaz et al. | Seizure detection exploiting EMD-wavelet analysis of EEG signals |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |