Electroencephalogram analysis method based on LPP

Abstract

The invention discloses an electroencephalogram analysis method based on LPP. The electroencephalogram analysis method comprises the steps that a resting state picture is displayed to a subject according to preset relaxation time; sequentially displaying a plurality of experiment diagram groups to the subject according to preset experiment time, receiving and recording a negative emotion intensity evaluation value input by the subject after each experiment diagram group is displayed, and displaying a next experiment diagram group after blank screen display is performed according to preset blank screen duration; each experiment picture group comprises a resting state picture and a stimulation picture which are arranged in sequence, and the stimulation picture is a social rejection picture or a physiological pain picture; collecting electroencephalogram signals of the subject at the preset relaxation time and the preset experiment time; according to the electroencephalogram signal, dividing a plurality of sliding time windows within 300 milliseconds after the stimulation picture appears to a preset stimulation time, quantifying the average amplitude of the LPP of each electrode channel of the subject in the sliding time window, and comparing; sampling and comparing the electroencephalogram signals within a preset stimulation time after the stimulation picture appears; and carrying out LPP quantization again.

Description

Technical Field

[0001] The present invention relates to the technical field of electroencephalogram analysis, and in particular to an electroencephalogram analysis method based on LPP. Background Art

[0002] Major Depressive Disorder (MDD) is one of the most common mental illnesses, and its core symptoms are significant and persistent depression and loss of interest. Non-suicidal self-injury (NSSI) refers to the deliberate destruction of one's own body tissue without suicidal intent. Common forms include cutting the skin or wrists, beating and burning oneself, etc. Adolescence is a critical period for the occurrence of NSSI behavior. Recent studies have shown that about 18.6-44.8% of adolescents participate in NSSI behavior every year, making NSSI an important public health issue. NSSI behavior is usually discussed in the context of mental illness disorders, such as major depression and borderline personality disorder. It is worth noting that about 41.6% of adolescents with NSSI are diagnosed with MDD. Compared with adolescent MDD patients without NSSI behavior, adolescent MDD patients with NSSI behavior have significantly impaired cognitive function, mainly manifested as memory loss, inattention, and decreased executive function. At the same time, NSSI is also a risk factor for future suicide attempts and completed suicides. In some cases, NSSI is a stronger predictor of suicide attempts than past self-harm behaviors. Therefore, it is crucial to conduct adequate research on depressed adolescents with self-harm to provide information for treatment and prevention efforts. However, the existing EEG research results on NSSI behaviors in depressed adolescents with self-harm are less accurate. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide an electroencephalogram (EEG) analysis method based on LPP to improve the accuracy of EEG analysis of NSSI behavior.

[0004] In order to solve the above technical problems, the purpose of the present invention is achieved through the following technical solutions: to provide an electroencephalogram (EEG) analysis method based on LPP, comprising the following steps: resting state picture display: displaying resting state pictures to the subject according to the preset relaxation time; experimental picture group display: displaying several experimental picture groups to the subject in sequence according to the preset experimental time, and prompting the subject to input a negative emotion intensity evaluation value after each experimental picture group is displayed, after receiving and recording the negative emotion intensity evaluation value input by the subject or after the preset evaluation time, a black screen is displayed according to the preset black screen duration, and the next experimental picture group is displayed after the black screen is displayed for the preset black screen duration until all experimental picture groups have completed the display of the preset display times; wherein each of the experimental picture groups includes a resting state picture and a stimulation picture arranged in sequence, and the stimulation picture is a social exclusion picture or a physiological pain picture; the preset experimental time is the total time of the display time of all experimental picture groups, all preset evaluation times and all preset black screen durations; EEG signal acquisition : Collect the EEG signals of the subjects during the preset relaxation time and the preset experimental time; Compare the LPP of the stimulation pictures: According to the EEG signals, divide several sliding time windows in the time period from 300 milliseconds after the stimulation pictures appear to the preset stimulation time, quantify the average amplitude of the LPP of each electrode channel of the subjects in the sliding time window, compare the LPP corresponding to the social exclusion pictures and the physiological pain pictures of each subject, and obtain the comparison result of the LPP of the stimulation pictures; Point-by-point sampling and comparison of the amplitude of the EEG signals of the stimulation pictures: Sample the EEG signals according to the amplitude of the EEG signals within the preset stimulation time after the stimulation pictures appear, compare the amplitude of the EEG signals corresponding to the social exclusion pictures and the physiological pain pictures of each subject according to each sampling point, and obtain the point-by-point sampling and comparison result of the amplitude of the EEG signals of the stimulation pictures; Quantify LPP again: Divide the electrode channel clusters and time windows according to the comparison result of the LPP of the stimulation pictures and the point-by-point sampling and comparison result of the amplitude of the EEG signals of the stimulation pictures, and quantify the average amplitude of the LPP on the electrode channel clusters in the time window.

[0005] The beneficial technical effects of the present invention are as follows: the LPP-based EEG analysis method of the present invention displays resting-state pictures to the subjects according to a preset relaxation time; displays several experimental picture groups to the subjects in sequence according to a preset experimental time; collects the EEG signals of the subjects at the preset relaxation time and the preset experimental time; divides several sliding time windows in the time period from 300 milliseconds after the stimulation picture appears to the preset stimulation time according to the EEG signals, quantifies the average amplitude of the LPP of each electrode channel of the subjects within the sliding time window, compares the LPP corresponding to the social exclusion pictures and the physiological pain pictures viewed by each subject, and obtains the stimulation picture LPP comparison result; samples the EEG signals according to the amplitude of the EEG signals within the preset stimulation time after the stimulation picture appears, compares the amplitude of the EEG signals corresponding to the social exclusion pictures and the physiological pain pictures viewed by each subject according to each sampling point, and obtains the stimulation picture EEG signal amplitude point-by-point sampling comparison result; divides the electrode channel clusters and the time window according to the stimulation picture LPP comparison result and the stimulation picture EEG signal amplitude point-by-point sampling comparison result, and quantifies the average amplitude of the LPP on the electrode channel clusters within the time window. Acquiring LPP can more accurately analyze the EEG signals of NSSI behavior, which is beneficial to the study of emotion regulation and emotion disorders. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0007] Figure 1 A schematic diagram of a process flow of an LPP-based EEG analysis method provided in an embodiment of the present invention;

[0008] Figure 2 A schematic diagram of a sub-process of an LPP-based EEG analysis method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0009] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0010] See also Figure 1 As shown, Figure 1 A schematic diagram of a flow chart of an electroencephalogram analysis method based on LPP provided in an embodiment of the present invention, wherein the electroencephalogram analysis method based on LPP comprises the following steps:

[0011] Step S10, display of resting state pictures: display resting state pictures to the subject according to the preset relaxation time; wherein the preset relaxation time may be five minutes, and the resting state pictures refer to pictures that put the subject in a resting state, which refers to a state when the brain does not perform specific cognitive tasks, remains quiet, relaxed, and awake. The resting state picture may be a picture with a solid color background and a cross cursor drawn in the center.

[0012] Step S20, display of experimental picture groups: display several experimental picture groups to the subject in sequence according to the preset experimental time, and prompt the subject to input the negative emotion intensity evaluation value after each experimental picture group is displayed, after receiving and recording the negative emotion intensity evaluation value input by the subject or after the preset evaluation time, display the black screen according to the preset black screen duration, and display the next experimental picture group after the black screen displays the preset black screen duration until all experimental picture groups have completed the display of the preset display times; wherein each experimental picture group includes a resting state picture and a stimulating picture arranged in sequence, and the stimulating picture refers to a picture that stimulates the subject, and the stimulating picture is a social rejection picture or a physiological pain picture. The preset experimental time is the total time of the display time of all experimental picture groups, all preset evaluation times and all preset black screen durations. The preset evaluation time refers to the preset time for waiting for the subject to input the negative emotion intensity evaluation value, and the preset display times can be twice. Of course, in some embodiments, the preset display times can be once, three times or more than three times. The total number of experimental picture groups may be seventy, and the number of experimental picture groups whose stimulus pictures are social rejection pictures and the number of experimental picture groups whose stimulus pictures are physical pain pictures may be thirty-five respectively.

[0013] Step S30, EEG signal acquisition: collecting the EEG signals of the subject at the preset relaxation time and the preset experimental time; wherein, the EEG signal acquisition can be performed simultaneously during the display of the resting-state pictures and the display of the experimental picture group, so as to collect the EEG signals of the subject in real time when the subject views the resting-state pictures at the preset relaxation time and the experimental picture group at the preset experimental time.

[0014] Step S40, comparison of LPP of stimulation pictures: according to the EEG signal, a number of sliding time windows are divided in the time period from 300 milliseconds after the stimulation picture appears to the preset stimulation time, the average amplitude of the LPP of each electrode channel of the subject in the sliding time window is quantified, and the LPP corresponding to the social exclusion pictures and the physiological pain pictures of each subject is compared to obtain the comparison result of the LPP of the stimulation pictures;

[0015] Among them, the preset stimulation time is the time for displaying the stimulation picture, which can be 5 seconds. Then, dividing a number of sliding time windows in the time period from 300 milliseconds after the stimulation picture appears to the preset stimulation time means dividing a number of sliding time windows in the time period from 300 milliseconds to 5 seconds after the stimulation picture appears. Event-related potential (ERP) is a special brain evoked potential and an important indicator of neuroelectrophysiological changes in the brain during emotion regulation. Event-related potential analysis reflects the neuroelectrophysiological changes in the brain during the cognitive process from the scalp EEG through the average superposition method, and reflects the dynamic process of cognition in a time-locked manner. The late positive component (LPP) is a component of ERP, which usually starts about 300 milliseconds after the stimulus is presented and can last throughout the entire process of stimulus presentation. It is a neurophysiological indicator for studying emotional responses and an event-related potential that is sensitive to the regulation of emotional responses. LPP is mainly generated in the central and posterior areas of the scalp. It is a positive component and is usually displayed as an increase in scalp positive potential on the EEG. Compared with traditional ERP components such as N1, EPN or P3, LPP can reflect the brain's sustained and in-depth processing of significant stimuli.

[0016] Step S50, point-by-point sampling and comparison of the EEG signal amplitude of the stimulation picture: sampling the EEG signal according to the amplitude of the EEG signal within the preset stimulation time after the stimulation picture appears, comparing the amplitude of the EEG signal corresponding to each subject watching the social exclusion pictures and the physical pain pictures according to each sampling point, and obtaining the point-by-point sampling and comparison result of the EEG signal amplitude of the stimulation picture; wherein the sampling frequency is 500 Hz.

[0017] Step S60, LPP quantification again: divide the electrode channel clusters and time windows according to the LPP comparison result of the stimulation picture and the point-by-point sampling comparison result of the EEG signal amplitude of the stimulation picture, and quantify the average amplitude of the LPP on the electrode channel cluster within the time window.

[0018] The LPP-based EEG analysis method can obtain LPP to more accurately analyze the EEG signals of NSSI behavior, and is beneficial to the study of emotion regulation and emotion disorders.

[0019] Preferably, the quantification in step S40 and step S60 is statistically analyzed by a two-factor mixed effects analysis of variance method, wherein the subjects may include two groups: a depression-associated self-injury group and a depression-not-self-injury group, the group of subjects is a between-subject factor, the category of stimulus pictures is a within-subject factor, and the gender and age of the subjects are covariates.

[0020] Combination Figure 2 , the step S20 specifically includes:

[0021] Step S21, displaying a resting state picture according to a preset resting state display time; wherein the preset resting state display time may be 2 seconds;

[0022] Step S22, displaying a stimulation picture according to a preset stimulation time;

[0023] Step S23, after the stimulus picture is displayed, the subject is prompted to input a negative emotion intensity evaluation value according to a preset evaluation time; the negative emotion intensity evaluation value ranges from 1 to 9, 1 indicates no negative emotion, 9 indicates strong negative emotion, that is, the higher the negative emotion intensity evaluation value, the stronger the negative emotion. The subject can input the negative emotion intensity evaluation value through the keyboard, or the screen displays different negative emotion intensity evaluation values ​​for the subject to click and select.

[0024] Step S24: after receiving and recording the negative emotion intensity evaluation value input by the subject or after a preset evaluation time has passed, a black screen is displayed according to a preset black screen duration;

[0025] Step S25: after the black screen is displayed for a preset black screen duration, determine whether there is an experimental image group that has not been displayed for a preset number of times;

[0026] Step S26: If there is no experimental image group that has not been displayed for the preset number of times, all experimental image groups are displayed for the preset number of times and the display ends.

[0027] Specifically, after step S25, the method further includes:

[0028] If there are experimental image groups that have not been displayed for the preset number of times, steps S21 to S25 are repeated until all experimental image groups have been displayed for the preset number of times.

[0029] Specifically, before step S10, the method further includes:

[0030] Select social exclusion pictures and physical pain pictures from the database, and store the selected social exclusion pictures and physical pain pictures as stimulus pictures.

[0031] Preferably, the step S30 is specifically as follows:

[0032] An electrode cap with 64 silver or silver chloride channels was used to collect and record the EEG signals of the subjects at the preset relaxation time and the preset experimental time through the BrainAmp amplifier and BrainVisionRecorder software; the sampling frequency was 500 Hz.

[0033] Specifically, after step S30, the method further includes:

[0034] The collected EEG signals are preprocessed offline to obtain preprocessed EEG signals. The original EEG signals collected are preprocessed to improve the signal quality and more accurately reflect the activity and state of the brain. Letswave can be used to preprocess EEG signals offline.

[0035] Preferably, the step of performing offline preprocessing on the collected EEG signal to obtain the preprocessed EEG signal specifically includes:

[0036] The eye movement channel of the acquired EEG signal is deleted, and the bad channel is interpolated to obtain the interpolated EEG signal;

[0037] The interpolated EEG signal is filtered using a bandpass filter and a notch filter; the bandpass frequency of the bandpass filter is 1-45 Hz, and the notch frequency of the notch filter is 50 Hz.

[0038] The interpolated EEG signals before and after filtering are respectively extracted from the time period from 0.5 seconds before the start of the experimental picture group display to 0.5 seconds after the end of the experimental picture group display, as the pre-filtered stimulation EEG signal segments and the post-filtered stimulation EEG signal segments; wherein, the experimental picture group display includes the resting state picture display and the stimulation picture display, then 0.5 seconds before the start of the experimental picture group display is 0.5 seconds before the start of the preset resting state display time, and 0.5 seconds after the end of the experimental picture group display is 0.5 seconds after the end of the stimulation picture display. Extracting the signal segments from 0.5 seconds before the start of the experimental picture group display to 0.5 seconds after the end of the experimental picture group display corresponding to each experimental picture group can ensure that the extracted signal segments are relatively complete in the stimulation process. The interpolated EEG signals before and after filtering refer to the interpolated EEG signals before filtering and the interpolated EEG signals after filtering. The pre-filtered stimulation EEG signal segments are extracted from the pre-filtered interpolated EEG signals, and the post-filtered stimulation EEG signal segments are extracted from the post-filtered interpolated EEG signals.

[0039] An independent component analysis algorithm is used to process the obtained filtered stimulation EEG signal fragments to obtain an independent component analysis matrix of the filtered stimulation EEG signal fragments;

[0040] The obtained independent component analysis matrix of the post-filtering stimulation EEG signal segment is assigned to the pre-filtering stimulation EEG signal segment to obtain a comprehensive stimulation EEG signal segment;

[0041] Visually inspect the integrated stimulation EEG signal segments to remove artifacts caused by eye and muscle activity to obtain artifact-free stimulation EEG signal segments;

[0042] All the anti-pseudo-stimulation EEG signal segments are subjected to bandpass filtering and notch filtering to obtain anti-pseudo-filtered stimulation EEG signal segments; wherein the bandpass frequency of the bandpass filter is 0.01-45 Hz, and the notch frequency of the notch filter is 50 Hz.

[0043] Re-reference the anti-aliased filtered stimulation EEG signal segment according to the average value of the amplitude of the left mastoid electrode and the right mastoid electrode to obtain the anti-aliased filtered reference stimulation EEG signal segment;

[0044] The anti-aliased filtered reference stimulation EEG signal segment corresponding to the 0.5 second before the stimulation picture is displayed is used as the baseline, and the anti-aliased filtered reference stimulation EEG signal segment is baseline corrected to obtain the preprocessed EEG signal.

[0045] Specifically, in step S20, if the categories of the stimulus pictures of the adjacent experimental picture groups are different, then the stimulus pictures of one of the adjacent experimental picture groups are social exclusion pictures, and the stimulus pictures of the other experimental picture group are physiological pain pictures. The adjacent experimental picture groups refer to the experimental picture groups corresponding to the two adjacent presentations of the sequentially presented experimental picture groups.

[0046] In summary, the LPP-based EEG analysis method of the present invention is as follows: showing a resting-state picture to the subject according to a preset relaxation time; showing several experimental picture groups to the subject in sequence according to a preset experimental time; collecting the EEG signals of the subject at the preset relaxation time and the preset experimental time; dividing several sliding time windows in the time period from 300 milliseconds after the stimulation picture appears to the preset stimulation time according to the EEG signals, quantifying the average amplitude of the LPP of each electrode channel of the subject within the sliding time window, comparing the LPP corresponding to the social exclusion pictures and the physiological pain pictures viewed by each subject, and obtaining the stimulation picture LPP comparison result; sampling the EEG signals according to the amplitude of the EEG signals within the preset stimulation time after the stimulation picture appears, comparing the amplitude of the EEG signals corresponding to the social exclusion pictures and the physiological pain pictures viewed by each subject according to each sampling point, and obtaining the stimulation picture EEG signal amplitude point-by-point sampling comparison result; dividing the electrode channel clusters and the time windows according to the stimulation picture LPP comparison result and the stimulation picture EEG signal amplitude point-by-point sampling comparison result, and quantifying the average amplitude of the LPP on the electrode channel clusters within the time window. Acquiring LPP can more accurately analyze the EEG signals of NSSI behavior, which is beneficial to the study of emotion regulation and emotion disorders.

[0047] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An electroencephalogram analysis method based on LPP, characterized in that: The following steps are involved: Resting state picture display: Show resting state pictures to the subjects according to the preset relaxation time; Experimental picture group display: display several experimental picture groups to the subject in sequence according to the preset experimental time, and prompt the subject to input the negative emotion intensity evaluation value after each experimental picture group is displayed, after receiving and recording the negative emotion intensity evaluation value input by the subject or after the preset evaluation time, display a black screen according to the preset black screen duration, and display the next experimental picture group after the black screen is displayed for the preset black screen duration until all experimental picture groups have completed the preset display times; wherein each experimental picture group includes a resting state picture and a stimulating picture arranged in sequence, and the stimulating picture is a social rejection picture or a physiological pain picture; the preset experimental time is the total time of the display time of all experimental picture groups, all preset evaluation times and all preset black screen durations; EEG signal collection: collect EEG signals of subjects at preset relaxation time and preset experimental time; Comparison of LPP of stimulus pictures: According to the EEG signals, several sliding time windows were divided in the time period from 300 milliseconds after the stimulus picture appeared to the preset stimulus time, and the average amplitude of LPP of each electrode channel in the whole brain of the subject in the sliding time window was quantified. The LPP corresponding to the social exclusion pictures and the physiological pain pictures of each subject was compared to obtain the comparison results of LPP of stimulus pictures; Point-by-point sampling comparison of the EEG signal amplitude of the stimulation picture: sampling the EEG signal according to the amplitude of the EEG signal within the preset stimulation time after the stimulation picture appears, comparing the amplitude of the EEG signal corresponding to each subject watching the social exclusion picture and the physiological pain picture according to each sampling point, and obtaining the point-by-point sampling comparison result of the EEG signal amplitude of the stimulation picture; LPP is quantified again: the electrode channel clusters and time windows are divided according to the LPP comparison results of the stimulation image and the point-by-point sampling comparison results of the EEG signal amplitude of the stimulation image, and the average amplitude of the LPP on the electrode channel cluster within the time window is quantified.

2. The LPP-based EEG analysis method according to claim 1, characterized in that: The steps shown in the experimental diagram group specifically include: Display the resting state picture according to the preset resting state display time; Display stimulus pictures according to the preset stimulus time; After the stimulus picture is displayed, the subject is prompted to enter the negative emotion intensity assessment value according to the preset assessment time; After receiving and recording the negative emotion intensity evaluation value input by the subject or after a preset evaluation time has passed, a black screen is displayed according to a preset black screen duration; After the black screen is displayed for a preset black screen duration, it is determined whether there is an experimental image group that has not been displayed for a preset number of times; If there is no experimental image group that has not been displayed for the preset number of times, all experimental image groups will complete the display of the preset number of times and end the display.

3. The LPP-based EEG analysis method according to claim 2, characterized in that: After the step of collecting the EEG signal, the method further includes: The collected electroencephalogram (EEG) signals are preprocessed offline to obtain preprocessed EEG signals.

4. The LPP-based EEG analysis method according to claim 3, characterized in that: The step of performing offline preprocessing on the collected EEG signals to obtain the preprocessed EEG signals comprises: Delete the eye movement channel of the acquired EEG signal and interpolate the bad channel; The interpolated EEG signal is filtered using a bandpass filter and a notch filter; Extract the signal segments in the time period from 0.5 seconds before the start of the display of the experimental image group to 0.5 seconds after the end of the display of the experimental image group corresponding to each experimental image group from the interpolated EEG signals before and after filtering, as the stimulation EEG signal segments before filtering and the stimulation EEG signal segments after filtering; An independent component analysis algorithm is used to process the obtained filtered stimulation EEG signal fragments to obtain an independent component analysis matrix of the filtered stimulation EEG signal fragments; The obtained independent component analysis matrix of the filtered stimulation EEG signal segment is assigned to the pre-filter stimulation EEG signal segment to obtain a comprehensive stimulation EEG signal segment; Visually inspect the integrated stimulation EEG signal segments to remove artifacts caused by eye and muscle activity to obtain artifact-free stimulation EEG signal segments; Performing bandpass filtering and notch filtering on all the artifact-free EEG signal segments to obtain artifact-free filtered EEG signal segments; Re-reference the anti-aliased filtered stimulation EEG signal segment according to the average value of the amplitude of the left mastoid electrode and the right mastoid electrode to obtain the anti-aliased filtered reference stimulation EEG signal segment; The anti-aliased filtered reference stimulation EEG signal segment corresponding to the 0.5 second before the stimulation picture is displayed is used as the baseline, and the anti-aliased filtered reference stimulation EEG signal segment is baseline corrected to obtain the preprocessed EEG signal.

5. The LPP-based EEG analysis method according to claim 4, characterized in that: In the step of filtering the interpolated EEG signal using a bandpass filter and a notch filter, the bandpass frequency range of the bandpass filter is 1-45 Hz, and the notch frequency of the notch filter is 50 Hz.

6. The LPP-based EEG analysis method according to claim 4, characterized in that: In the step of performing bandpass filtering and notch filtering on all the artifact-free stimulation EEG signal segments to obtain artifact-free filtered stimulation EEG signal segments, the bandpass frequency range of the bandpass filtering is 0.01-45 Hz, and the notch frequency of the notch filtering is 50 Hz.

7. The LPP-based EEG analysis method according to claim 1, characterized in that: In the step of displaying the experimental image groups, the categories of stimulus images of adjacently displayed experimental image groups are different.

8. The LPP-based EEG analysis method according to claim 1, characterized in that: The step of displaying the resting state picture also includes: Select social exclusion pictures and physical pain pictures from the database, and store the selected social exclusion pictures and physical pain pictures as stimulus pictures.

Images