Electroencephalogram analysis method for evaluating auditory rhythm stimulation effect

Abstract

The invention discloses an electroencephalogram analysis method for evaluating an auditory rhythm stimulation effect, which comprises the following steps: step 1, applying auditory rhythm stimulation to a subject, and collecting resting state electroencephalogram signals of the subject after different stimulation; 2, electrode positioning, reference conversion, filtering and denoising are conducted on the resting-state electroencephalogram signals in sequence, and preprocessed resting-state electroencephalogram signals are obtained; 3, analyzing the preprocessed resting-state electroencephalogram signals to obtain micro-state characteristic parameters; 4, on the basis of the micro-state characteristic parameters stimulated by different auditory takt, the correlation between the behavior performance data under the stimulation of different auditory takt and the micro-state characteristic parameters is obtained through calculation; and then behavior performance changes corresponding to the micro-state characteristic parameter changes are analyzed according to the Pearson's correlation coefficient, and the auditory rhythm stimulation effect is evaluated. According to the method, the micro-state analysis feature result of the resting-state electroencephalogram signal and the behavior performance result related to the attention function are combined, operation is easy and convenient, and non-invasiveness is high.

Description

Technical Field

[0001] This invention belongs to the field of neuromodulation technology, specifically an electroencephalogram (EEG) analysis method for evaluating the effects of auditory beat stimulation. Background Technology

[0002] In recent years, electroencephalography (EEG), as a widely available, low-cost, non-invasive neuroimaging technique with high temporal resolution, has been extensively applied in the study of neural mechanisms related to cognitive function. It can effectively characterize the brain's neural oscillatory activity and its synchronicity. Attention is a crucial component of the cognitive system, playing a key role in information selection and processing, learning, and the performance of daily living tasks. Existing research indicates that impairment of attention function can lead to various cognitive problems and even develop into various disorders, such as sleep disorders, attention deficit hyperactivity disorder (ADHD), etc. Given the widespread and severe nature of attention impairment, there is an urgent need to find effective methods to improve it.

[0003] Auditory beat stimulation, as a non-invasive neuromodulation method, has the advantages of being simple, portable, highly manipulable, and inexpensive. It restores or enhances specific brain functions by regulating the activity of neurons and their networks, thereby influencing brain oscillations. Research suggests that auditory beat stimulation can induce brainwave frequencies to follow the stimulation frequency through pulsed sound stimulation, thus affecting and regulating neural activity and cognitive functions of the brain. Current research methods on the neurotic mechanisms modulated by auditory beat stimulation mainly include power spectrum analysis (Fujikawa I, Fujikawa J, Fujiwara M, Takagi Y, Morigaki R. Effect of inaudible 40 Hz binaural beats on attention. Exp BrainRes. 2025 May 29;243(7):158.), subjective scale assessment of behavioral performance and brain functional connectivity (Dos Anjos T, Di Rienzo F, Benoit CE, et al. Brain wave modulation and EEG power changes during auditory beat stimulation. Neuroscience. 2024; 554: 156-166.), etc., but the research results are highly controversial and do not fully utilize the nonlinear characteristics and spatial information contained in EEG signals. Isochronous pitch and binaural frequency difference are currently widely used methods for auditory beat stimulation modulation, showing certain application potential in areas such as attention enhancement, working memory function regulation, sleep quality improvement, and emotion regulation, and are gradually being applied to research and practice in mental health intervention and cognitive training.

[0004] However, existing auditory beat stimulation modulation techniques have some limitations: on the one hand, auditory beat stimulation modulation lacks systematic scientific validation; on the other hand, different individuals respond differently to auditory beat stimulation, making it difficult to subjectively perceive a significant improvement in cognitive function brought about by the stimulation. This limits its application in the field of cognitive function regulation and improvement. Therefore, there is an urgent need for an analysis method based on electroencephalogram (EEG) signals to systematically compare and identify the specific effects of different auditory beat stimuli on attentional function, thereby providing technical support for non-invasive cognitive modulation and individualized intervention. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an electroencephalogram (EEG) analysis method for evaluating the effects of auditory beat stimulation.

[0006] The technical solution of the present invention to solve the aforementioned technical problem is to provide an electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation, characterized in that the method includes the following steps: Step 1: Apply auditory beat stimulation to the subject and collect the subject's resting-state EEG signals after different stimuli; Step 2: Perform electrode localization, reference conversion, filtering, and noise reduction on the resting-state EEG signal acquired in Step 1 to obtain the preprocessed resting-state EEG signal. Step 3: Analyze the preprocessed resting-state EEG signal obtained in Step 2 to obtain microstate characteristic parameters; Step 4: Based on the microstate characteristic parameters after different auditory beat stimuli, use the Pearson correlation analysis method to calculate the correlation between behavioral performance data and microstate characteristic parameters under different auditory beat stimuli; then analyze the behavioral performance changes corresponding to changes in microstate characteristic parameters according to the Pearson correlation coefficient to evaluate the effect of auditory beat stimulation.

[0007] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The microstate-based EEG analysis method proposed in this invention can capture the instantaneous stable pattern of EEG spatial distribution at high temporal resolution, reflecting the global dynamic changes of the brain. Traditional feature extraction methods only analyze and focus on local frequency energy, which is difficult to comprehensively characterize brain activity. Microstate analysis reflects the instantaneous activation and state switching patterns of brain networks, and can directly reveal the dynamic information of brain state transitions. Through indicators such as the average duration of microstates, time coverage ratio, occurrence frequency and transition probability, the regulatory effect of auditory beat stimulation is revealed from an objective perspective, which helps to evaluate the effectiveness of auditory beat stimulation.

[0008] (2) This invention is easy to operate and non-invasive. It combines the micro-state analysis results of resting-state EEG signals with the results of attention-related behavioral performance, making full use of the rich physiological information contained in resting-state EEG signals. It can be applied to attention intervention, cognitive training, auxiliary diagnosis of attention deficit and hyperactivity disorder, and has broad scientific research and clinical application prospects.

[0009] (3) In step 1, by combining attention task research on the effect of auditory beat stimulation on brain function, the effect of auditory beat stimulation on attention function can be evaluated more objectively and systematically from the two levels of neural activity and cognitive performance; combining behavioral performance data with EEG signals for analysis improves the scientificity and reliability of the results of this invention and provides objective and clear scientific evidence; the clear experimental scheme makes this invention have better conditions for promotion and replication.

[0010] (4) In step 2, the lower electrode impedance avoids signal distortion caused by electrode impedance mismatch, ensuring the accuracy and reliability of EEG data; the filtering process provides a high-quality signal basis for subsequent data analysis; by combining independent component analysis and artifact subspace reconstruction technology, the effective information in the EEG signal can be preserved to the maximum extent, ensuring the accuracy and credibility of EEG data in the next step of analysis.

[0011] (5) In step 3, the four microstate parameters mentioned and used in this invention can comprehensively evaluate the stability, frequency, and inter-state transition relationships of each microstate. The time coverage ratio and average duration help to understand the persistence and stability of different microstates, while the frequency of occurrence reveals the universality of the neural activity pattern corresponding to a certain microstate in the brain. The transition probability can reveal the dynamic transition relationship between microstates, provide the state transition pattern of the brain in different cognitive processes, and improve the understanding of brain neural activity. Through precise microstate division and feature parameter extraction, this invention can efficiently monitor and analyze brain activity in different cognitive states.

[0012] (6) In step 4, the Pearson correlation coefficient can intuitively and accurately reflect the direction and strength of the linear relationship between the performance results of the attention network test and the microstate parameters, which helps to clarify the regulatory role of different microstate characteristics on the performance of the attention network. By establishing the correlation between microstate parameters and behavioral performance data, it can help reveal the influence of different auditory beat stimuli on attention level, and also provide a scientific basis for the assessment of attention level based on EEG characteristics. Attached Figure Description

[0013] Figure 1 This is an overall flowchart of the present invention; Figure 2This is a graph showing the difference in average reaction time of subjects performing attention network tests after applying different auditory beat stimuli in Embodiment 1 of the present invention; in the graph, * indicates p<0.05, ** indicates p<0.01, and *** indicates p<0.001; Figure 3 This is a graph showing the difference in accuracy of subjects performing attention network tests after applying different auditory beat stimuli in Embodiment 1 of the present invention. Figure 4 This is a graph showing the differences in attentional network effects in subjects after applying different auditory beat stimuli in Example 1 of the present invention. Figure 5 This is a microstate clustering result diagram of the resting-state EEG signals of the subjects after applying different auditory beat stimulation in Embodiment 1 of the present invention; Figure 6 This is a graph showing the difference in the microstate time coverage ratio after applying different auditory beat stimuli in Embodiment 1 of the present invention; Figure 7 This is a graph showing the difference in the average duration of microstates after applying different auditory beat stimuli in Embodiment 1 of the present invention; Figure 8 This is a graph showing the difference in the frequency of microstate occurrence after applying different auditory beat stimuli in Embodiment 1 of the present invention; Figure 9 This is a graph showing the difference in microstate transition probabilities after applying different auditory beat stimuli in Embodiment 1 of the present invention; Figure 10 The correlation between the transition probability from microstate B to microstate C and the directional network effect after applying gamma-band binaural difference frequency stimulation in Embodiment 1 of the present invention. Detailed Implementation

[0014] Specific embodiments of the present invention are given below. These specific embodiments are only used to further illustrate the present invention in detail and do not limit the scope of protection of the present invention.

[0015] This invention provides an electroencephalogram (EEG) analysis method (hereinafter referred to as the method) for evaluating the effect of auditory beat stimulation, characterized in that the method includes the following steps: Step 1: Apply auditory beat stimulation to the subject and collect the subject's resting-state EEG signals after different stimuli; Preferably, the specific steps of step 1 are as follows: S11. Randomly apply auditory beat stimuli to the subjects, while the subjects perform an attention network test; Preferably, in step S11, the auditory beat stimulation includes pink noise stimulation, gamma band binaural difference frequency stimulation, and gamma band isochronous tone stimulation.

[0016] Preferably, in step S11, the gamma band binaural difference frequency stimulation is a binaural difference frequency audio produced with a carrier frequency of 300Hz and a difference frequency of 40Hz. Specifically, the left ear audio is 300Hz and the right ear audio is 340Hz. The gamma-band isochronous tone stimulus is a 300Hz pure tone played once every 0.025s with a duty cycle of 50% to produce an audio frequency of 40Hz.

[0017] Preferably, in step S11, the attention network test specifically involves the following steps: At the start of each attention network test, the fixation point is displayed in the center of the screen at a variable time interval of 400-1600ms; subsequently, the cue stimulus is presented for 200ms, followed by the fixation point for 400ms, and then the target is presented until the subject responds; if the subject does not respond, the target is presented for a maximum of 1700ms; then the fixation point is presented again in the center of the screen, and the next test begins; each test lasts 4-5 seconds, depending on the subject's reaction time; there are 50 practice tests before the formal test begins, and the subject needs to perform 220 tests for each sound stimulus for a total of 15-16 minutes, with a total test duration of 1-1.5 hours.

[0018] S12. After each auditory beat stimulation, the subject closes their eyes and obtains resting EEG data for 3-5 minutes. Finally, the resting EEG signal after auditory beat stimulation and the behavioral performance data during the stimulation process are obtained.

[0019] Preferably, in step S12, the behavioral performance data includes: the accuracy rate of the attention network test, the average reaction time, and the effects of the alertness network, the orientation network, and the executive control network.

[0020] Preferably, in step S12, the calculation methods for the effects of the three attention sub-networks are as follows: The vigilance network effect = unawakened alertness (RT) - bilaterally cued alertness (RT): Compared to the two, the vigilance effect is increased over time, resulting in the vigilance network effect. Oriented network effect = Central cueing RT - Spatial cueing RT: Compared to the former, the latter provides more effective prediction of the target stimulus, thus yielding the oriented network effect; Executive control network effect = Inconsistent stimulus RT - Consistent stimulus RT: Compared with consistent stimuli, inconsistent target stimuli require additional conflict monitoring and conflict resolution, which can yield executive control network effect.

[0021] Step 2: Perform electrode localization, reference conversion, filtering, and noise reduction on the resting-state EEG signal acquired in Step 1 to obtain the preprocessed resting-state EEG signal. Preferably, step 2 includes the following steps: S21. Perform electrode localization on the resting-state EEG signals acquired in step 1; Preferably, in step S21, the experiment uses a 64-lead electrode cap that strictly conforms to the international standard lead 10-20 system; the electrode cap is equipped with Ag / AgCl electrodes, and all electrode impedances are <5kΩ during resting-state EEG signal recording.

[0022] S22. Perform a reference conversion on the resting-state EEG signals collected by all electrodes in step S21 to obtain the reference-converted resting-state EEG signals. Preferably, in step S22, the reference transformation is as shown in equation (1): (1) In equation (1), V represents the potential of the i-th electrode after reference switching; i V represents the potential of the i-th electrode; N is the total number of electrodes; j This represents the potential of the j-th electrode.

[0023] Preferably, in step S22, the reference conversion uses a whole-brain average reference method, which helps the EEG to more accurately reflect spatial characteristics during microstate analysis, reduces dependence on reference electrodes, and can more realistically reflect brain neural activity patterns.

[0024] S23. Since the induced gamma band activity in the electroencephalogram data recorded by the scalp usually manifests as bursts of high-frequency (30~80Hz) oscillatory activity, the resting-state EEG signal obtained in step S22 is filtered to improve signal quality and remove power frequency interference and baseline drift, thus obtaining the filtered resting-state EEG signal. Preferably, in step S23, the filtering is a 0.5Hz low-pass filter, a 50Hz dip filter, and an 80Hz high-pass filter.

[0025] S24. Based on the filtered resting-state EEG signal obtained in step S23, the artifact components in the EEG signal are removed using independent component analysis and artifact subspace reconstruction techniques. Then, the remaining artifacts are manually checked to obtain the preprocessed resting-state EEG signal.

[0026] Preferably, in step S24, the components removed are horizontal and vertical eye movements, bad electrodes, and muscle movement artifacts.

[0027] Preferably, in step S24, the independent component analysis aims to decompose the raw data into a set of mutually independent components, each component corresponding to a source signal in the data; The artifact subspace reconstruction technique identifies artifact components in the data by analyzing the feature vectors and feature values ​​of the data, and then removes them by setting them to zero.

[0028] Step 3: Analyze the preprocessed resting-state EEG signal obtained in Step 2 to obtain microstate characteristic parameters; Preferably, step 3 includes the following steps: S31. Calculate the global field power GFP of the preprocessed resting-state EEG signal obtained in step 2, and then calculate the topographic map based on the global field power GFP. The topographic map can enable people to clearly assess the electric field characteristics of their brain without depending on the electrode location or the selection of the reference electrode. Preferably, in step S31, the global field power GFP can quantify the intensity of the scalp potential, that is, the standard deviation of the potential of all electrode channels at each sampling point, which can reflect the instantaneous electric field intensity of the brain, as shown in equation (2): (2) In equation (2), c represents the total number of electrodes, u i This represents the voltage value of the i-th electrode. This represents the average voltage value of the electrodes.

[0029] S32. Use the k-means algorithm to perform micro-state clustering on the topographic map obtained in step S31 to obtain micro-state clustering templates with different numbers of clusters; then, determine the optimal number of micro-state clusters by the number of clusters corresponding to the maximum value of the global explained variance between the topographic map and each number of micro-state clusters, and the micro-state clustering template corresponding to the optimal number of micro-state clusters is the optimal micro-state clustering template; then, import the optimal micro-state clustering template into the resting-state EEG signal and perform backtracking fitting to obtain the resting-state EEG signal corresponding to each micro-state category; Preferably, in step S32, the global explained variance GEV is calculated as the standard deviation of all electrodes of the resting-state EEG signal of the nth sample, and its calculation formula is shown in equation (3): (3) In equation (3), to calculate the global explained variance of a given prototype, it is necessary to sum the global explained variance of each of its members; x n L represents the topographic map of the nth sample. n Represents a template topographic map, GFP n This represents the global field power of the nth sample.

[0030] S33. Based on the resting-state EEG signals corresponding to different microstate categories obtained in step S32, calculate the microstate feature parameters of the resting-state EEG signals under different microstate categories.

[0031] Preferably, in step S33, the micro-state characteristic parameters include time coverage ratio, average duration, occurrence frequency, and transition probability.

[0032] Time coverage ratio: This represents the proportion of time each microstate category lasts during the entire experiment. This metric can provide information on the persistence and stability of microstates throughout the experiment, which helps to understand the behavior patterns and trends of microstates. Average duration: This refers to the average duration of a given microstate, usually measured in milliseconds. The length of the duration reflects the stability and persistence of the microstate; a longer duration microstate may correspond to a more stable pattern of neural activity.

[0033] Frequency of occurrence: This reflects the average number of times a microstate dominates per second. Microstates that occur more frequently may correspond to neural activity patterns that occur frequently in the brain, while microstates that occur less frequently may correspond to rarer or more specific neural activity patterns. Transition probability: The probability of transitioning from one microstate to another; transition probability can reveal the interrelationship between microstates and reflect the brain's switching patterns between different neural activity states.

[0034] Step 4: Based on the microstate characteristic parameters after different auditory beat stimuli, use the Pearson correlation analysis method to calculate the correlation between behavioral performance data and microstate characteristic parameters under different auditory beat stimuli; then analyze the behavioral performance changes corresponding to changes in microstate characteristic parameters according to the Pearson correlation coefficient to evaluate the effect of auditory beat stimulation.

[0035] Example 1: In step S11, the gamma band binaural difference frequency stimulation is a binaural difference frequency audio produced with a carrier frequency of 300Hz and a difference frequency of 40Hz. The left ear audio is 300Hz and the right ear audio is 340Hz. The gamma-band isochronous tone stimulus is a 300Hz pure tone played once every 0.025s with a duty cycle of 50% to produce an audio frequency of 40Hz.

[0036] In step S12, the subjects pay attention to the behavioral results of the network test, such as... Figures 2-4 As shown, one-way ANOVA was used to statistically analyze the accuracy, average reaction time, and network effect values ​​of each attentional subnetwork under three types of sound stimuli: pink noise, binaural difference frequency in the gamma band, and isochronous pitch in the gamma band. When the data did not conform to a normal distribution, the Friedman test was used for analysis.

[0037] Statistical analysis showed no statistically significant difference in the accuracy of ANT (p=0.105). The main effect of the total reaction time of ANT was significant (F(2,78)=8.632, p<0.001, η). 2=0.181). The total reaction time under the IT condition was significantly lower than that under the PN condition (p=0.001). One-way repeated measures ANOVA was used to analyze the effects of different auditory stimuli on the effects of each subnetwork of the attention network. In the vigilance network, the main effect of stimulus group was significant (F(2,78)=15.876, p<0.001, η). 2 =0.289). The alarm network effect under IT stimulation was significantly higher than that under PN stimulation (p<0.001) and BB stimulation (p=0.006), and the alarm network effect under BB stimulation was also significantly higher than that under PN stimulation (p=0.012). In the orientation network, the main effect of stimulus group was significant (F(2,78)=5.032, p=0.009, η). 2 =0.114). The orientation network effect under IT stimulation was significantly higher than that under PN stimulation (p=0.026). In the executive control network, the main effect of stimulation group was significant (F(2,78)=17.824, p<0.001, η). 2 =0.314). The executive control network effect under IT stimulation was significantly lower than that under PN stimulation (p<0.001) and BB stimulation (p=0.027), and the alertness network effect under BB stimulation was also significantly lower than that under PN stimulation (p=0.002). Statistical analysis results showed that both binaural difference frequency and isochronous sound stimulation in the gamma band could improve the attention network efficiency of the subjects, but isochronous sound stimulation had a better improvement effect.

[0038] In step S32, the number of optimal microstate clustering templates is 4. Therefore, the microstate topographic map categories are divided and named as microstate A, microstate B, microstate C and microstate D.

[0039] In step S33, the microstate topography and changes in various microstate parameters after different auditory beat stimuli were statistically analyzed. The obtained microstate topography is shown below. Figure 5 As shown, the global interpretation variance of all topographic maps is between 66% and 88%, indicating high reliability.

[0040] The analysis results of the time coverage ratio of different categories of microstates are as follows: Figure 6As shown, the temporal coverage ratio of microstate B differed significantly after pink noise, gamma-band binaural difference frequency, and gamma-band isochronous tone stimulation (F(2,78)=14.600, p<0.001). The temporal coverage ratio of microstate B after gamma-band isochronous tone stimulation was significantly lower than that after pink noise stimulation. The temporal coverage ratio of microstate D also differed significantly after pink noise, gamma-band binaural difference frequency, and gamma-band isochronous tone stimulation (F(2,78)=11.450, p=0.003). The temporal coverage ratio of microstate D after gamma-band isochronous tone stimulation was significantly higher than that after pink noise (p=0.004) and gamma-band binaural difference frequency stimulation (p=0.042).

[0041] The analysis results of the average duration of different categories of microstates are as follows: Figure 7 As shown, the mean duration of microstate B differed significantly among the three sound stimuli (F(2,78)=5.056, p=0.009, η). 2 =0.115). The mean duration of isochronous tone stimulation in the gamma band was significantly shorter than that after pink noise stimulation (p=0.018).

[0042] The results of the analysis of the occurrence frequency of different categories of microstates are as follows: Figure 8 As shown, the frequency of microstate A differed significantly after pink noise stimulation, gamma-band binaural difference stimulation, and gamma-band isochronous tone stimulation (p=0.013). After gamma-band binaural difference stimulation, the frequency of microstate A was significantly higher than after pink noise stimulation (p=0.002). The frequency of microstate B also differed significantly under different stimuli (F(2,78)=11.450, p=0.003). Furthermore, the frequency of microstate B after pink noise stimulation was significantly higher than after gamma-band binaural difference stimulation (p=0.042) and gamma-band isochronous tone stimulation (p=0.004). Additionally, the frequency of microstate D also differed significantly among the three stimuli (F(2,78)=23.750, p<0.001). The frequency of microstate D after gamma band isochronous tone stimulation was significantly higher than that after pink noise stimulation (p<0.001) and gamma band binaural difference frequency stimulation (p<0.001).

[0043] The results of the analysis of the transition probabilities between different categories of microstates are as follows: Figure 9 As shown, TP A→B The transition probability from microstate B to microstate A showed significant differences after pink noise, binaural difference frequency in the gamma band, and isochronous tone stimulation in the gamma band (F(2,78)=7.800, p=0.020). The TP after isochronous tone stimulation in the gamma band was significantly higher. A→B Significantly higher than after pink noise stimulation (p=0.022). TP B→CSignificant differences were also observed after stimulation (F(2,78)=10.850, p=0.004), with TP after gamma-band isochronous pitch stimulation. B→C The TPB was significantly increased compared to stimulation with pink noise (p=0.011) and gamma-band binaural difference frequency (p=0.016). TPB under different stimuli →D There is a significant difference (F(2,78)=7.543, p=0.001, η). 2 =0.162), TP after isochronous pitch stimulation in the gamma band B→D Significantly higher than after binaural difference-frequency stimulation in the gamma band (p<0.001). Meanwhile, TP... C→B Significant differences were also observed after tonal stimulation at pink noise, gamma band binaural difference frequency, and gamma band frequencies (F(2,78)=18.231, p<0.001, η). 2 =0.319), significantly higher after pink noise stimulation than after gamma band isochronous pitch (p<0.001) and gamma band binaural difference frequency stimulation (p<0.001). TP D→B Similarly, significant differences were observed after tonal stimulation at pink noise, gamma band binaural difference frequency, and gamma band (F(2,78)=6.192, p=0.003, η). 2 =0.137), TP after isochronous pitch stimulation in the gamma band D→B The difference was significantly higher than that after binaural differential stimulation in the gamma band (p=0.001).

[0044] Microstate analysis can reflect the dynamic changes in the brain under different states and reveal the underlying cognitive processes. Our results show that, compared with pink noise stimulation, the frequency of microstate B significantly decreased after gamma-band binaural difference-frequency stimulation; after gamma-band isochronous pitch stimulation, the temporal coverage, average duration, and frequency of microstate B all significantly decreased, while the temporal coverage and frequency of microstate D significantly increased. Correlation analysis revealed that after gamma-band binaural difference-frequency stimulation, TP... B→C It shows a significant positive correlation with the orientation network effect. The functional significance and interaction of microstate categories are not yet clear, but many studies have shown that microstate A is closely related to the auditory network; microstate B is related to the visual network; microstate C corresponds to the salience network; and microstate D is related to the attention network.

[0045] Microstate B is associated with the visual resting-state network. Studies have found that characteristics of microstate B are negatively correlated with vigilance levels, and the average duration of microstate B increases during rest. Furthermore, a decreased frequency of microstate B may indicate reduced brain processing of irrelevant visual distractions, and enhanced attentional stability and selective attention. Microstate D is closely related to attentional function, playing a crucial role in maintaining and regulating attention, cognitive flexibility, and vigilance levels. Recent research suggests that microstate D originates in the frontal and parietal lobes of the brain and is believed to be related to the dorsal attentional network; higher microstate D eigenvalues ​​are associated with better regulation of attentional function and faster reaction times. TP B→C These findings can reflect changes in the efficiency of distinguishing significant stimuli after visual perception processing, which may be related to changes in attentional resource regulation patterns. These analyses of microstate characteristics indicate that binaural difference frequency and isochronous tone stimulation in the gamma band can improve performance on attentional tasks, helping to understand the neural regulation patterns of individuals performing attentional tasks and providing a new perspective for studying the neural mechanisms by which auditory beat stimulation regulates attentional function.

[0046] In step 4, the correlation analysis results between the subjects' attention network test performance and microstate parameters are as follows: Figure 10 As shown, TP after binaural difference-frequency stimulation in the gamma band B→C It showed a significant positive correlation with the directional network effect (r=0.328, p=0.039).

[0047] Any aspects not covered in this invention are applicable to existing technologies.

Claims

1. An electroencephalogram (EEG) analysis method for evaluating the effects of auditory beat stimulation, characterized in that, The method includes the following steps: Step 1: Apply auditory beat stimulation to the subject and collect the subject's resting-state EEG signals after different stimuli; Step 2: Perform electrode localization, reference conversion, filtering, and noise reduction on the resting-state EEG signal acquired in Step 1 to obtain the preprocessed resting-state EEG signal. Step 3: Analyze the preprocessed resting-state EEG signal obtained in Step 2 to obtain microstate characteristic parameters; Step 4: Based on the microstate characteristic parameters after different auditory beat stimuli, use the Pearson correlation analysis method to calculate the correlation between behavioral performance data and microstate characteristic parameters under different auditory beat stimuli; then analyze the behavioral performance changes corresponding to changes in microstate characteristic parameters according to the Pearson correlation coefficient to evaluate the effect of auditory beat stimulation.

2. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 1, characterized in that, The specific steps for step 1 are as follows: S11. Randomly apply auditory beat stimuli to the subjects, while the subjects perform an attention network test; S12. After each auditory beat stimulation, the subject closes their eyes and obtains resting EEG data for 3-5 minutes. Finally, the resting EEG signal after auditory beat stimulation and the behavioral performance data during the stimulation process are obtained.

3. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 2, characterized in that, In step S11, the auditory beat stimulation includes pink noise stimulation, gamma band binaural difference frequency stimulation, and gamma band isochronous tone stimulation. In step S11, the attention network test is specifically as follows: at the beginning of each attention network test, the fixation point is displayed in the center of the screen at a variable time interval of 400 to 1600 ms; then, the cue stimulus is presented for 200 ms, then the fixation point is presented for 400 ms, and then the target is presented until the subject responds. If the subject does not respond, the target is presented for a maximum of 1700ms; then the fixation point is presented again in the center of the screen, and the next trial begins; each trial lasts 4-5 seconds, depending on the subject's reaction time; there are 50 practice trials before the formal experiment begins, and the subject needs to perform 220 trials for each sound stimulus for a total of 15-16 minutes, with a total experiment duration of 1-1.5 hours. In step S12, the behavioral performance data includes: the accuracy of the attention network test, the average reaction time, and the effects of the vigilance network, the orientation network, and the executive control network.

4. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 3, characterized in that, In step S11, the gamma band binaural difference frequency stimulation is a binaural difference frequency audio produced with a carrier frequency of 300Hz and a difference frequency of 40Hz. The left ear audio is 300Hz and the right ear audio is 340Hz. The gamma-band isochronous tone stimulus is a 300Hz pure tone played once every 0.025s with a duty cycle of 50% to produce an audio frequency of 40Hz.

5. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 1, characterized in that, Step 2 includes the following steps: S21. Perform electrode localization on the resting-state EEG signals acquired in step 1; S22. Perform a reference conversion on the resting-state EEG signals collected by all electrodes in step S21 to obtain the reference-converted resting-state EEG signals. S23. Filter the reference-converted resting-state EEG signal obtained in step S22 to improve signal quality and remove power frequency interference and baseline drift, and obtain the filtered resting-state EEG signal. S24. Based on the filtered resting-state EEG signal obtained in step S23, the artifact components in the EEG signal are removed using independent component analysis and artifact subspace reconstruction techniques. Then, the remaining artifacts are manually checked to obtain the preprocessed resting-state EEG signal.

6. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 5, characterized in that, In step S21, the experiment used a 64-lead electrode cap conforming to the international standard lead 10-20 system; the electrode cap was equipped with Ag / AgCl electrodes, and all electrode impedances were <5kΩ during resting-state EEG signal recording. In step S22, the reference transformation is as shown in equation (1): （1） In equation (1), V represents the potential of the i-th electrode after reference switching; i V represents the potential of the i-th electrode; N is the total number of electrodes; j This represents the potential of the j-th electrode; In step S22, the reference transformation uses the whole-brain average reference method; In step S23, the filtering is a 0.5Hz low-pass filter, a 50Hz dip filter, and an 80Hz high-pass filter. In step S24, the components removed are horizontal and vertical eye movements, bad electrodes, and muscle movement artifacts.

7. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 1, characterized in that, Step 3 includes the following steps: S31. Calculate the global field power GFP of the preprocessed resting-state EEG signal obtained in step 2, and then calculate the topographic map based on the global field power GFP. S32. Use the k-means algorithm to perform micro-state clustering on the topographic map obtained in step S31 to obtain micro-state clustering templates with different numbers of clusters; then, determine the optimal number of micro-state clusters by the number of clusters corresponding to the maximum value of the global explained variance between the topographic map and each number of micro-state clusters, and the micro-state clustering template corresponding to the optimal number of micro-state clusters is the optimal micro-state clustering template; then, import the optimal micro-state clustering template into the resting-state EEG signal and perform backtracking fitting to obtain the resting-state EEG signal corresponding to each micro-state category; S33. Based on the resting-state EEG signals corresponding to different microstate categories obtained in step S32, calculate the microstate feature parameters of the resting-state EEG signals under different microstate categories.

8. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 7, characterized in that, In step S31, the global field power GFP is as shown in equation (2): （2） In equation (2), c represents the total number of electrodes, u i This represents the voltage value of the i-th electrode. This represents the average voltage value of the electrodes.

9. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 7, characterized in that, In step S32, the global explained variance GEV is shown in equation (3): （3） In equation (3), to calculate the global explained variance of a given prototype, it is necessary to sum the global explained variance of each of its members; x n L represents the topographic map of the nth sample. n Represents a template topographic map, GFP n This represents the global field power of the nth sample.

10. The electroencephalogram (EEG) analysis method for evaluating the effect of auditory beat stimulation according to claim 7, characterized in that, In step S33, the micro-state characteristic parameters include time coverage ratio, average duration, occurrence frequency, and transition probability.

Images