Clustering integration-based electroencephalogram identification analysis method under complex visual stimulation

Abstract

A clustering integration-based electroencephalogram identification analysis method under complex visual stimulation comprises the following steps of designing an experimental normal form of electroencephalogram signal acquisition under a complex visual stimulation condition, selecting N subjects, and respectively operating according to the experimental normal form, synchronously collecting and recording multi-channel electroencephalogram signal data of N subjects under complex visual stimulation in the experimental normal form; starting an experiment, preprocessing electroencephalogram signals of the N subjects, obtaining preprocessed electroencephalogram signal data corresponding to the N subjects, and forming a rapid sequence to present an electroencephalogram signal data set; carrying out clustering analysis on the electroencephalogram signal data of the N subjects by utilizing a spectral clustering algorithm to obtain N sub-clusters; fusing the N tested sub-clusters by using a spectral clustering integrated optimization strategy based on hypergraph division; and testing and evaluating the fusion model to obtain a performance attribute value of the fusion model. According to the method, decision-making layer integration of a plurality of tested brain visual cognition image classification models is realized, so that a cross-individual electroencephalogram signal analysis method under a visual stimulation target with relatively high generalization, accuracy and robustness is constructed.

Description

Technical Field

[0001] The invention relates to an electroencephalogram recognition and analysis method under complex visual stimulation based on clustering integration, and belongs to the technical field of intelligent information processing. Background Art

[0002] The human visual system has a significantly better recognition effect than the current best machine vision system under statistical learning theory. In particular, when processing visual information in some harsh environments, traditional methods encounter great difficulties. The human visual system has a strong anti-interference ability and can complete visual recognition tasks in various complex environments. Therefore, how to study target recognition and computer vision algorithms from the perspective of visual cognition has become a very important and challenging task. From the perspective of neuroelectrophysiology, the evolution mechanism of EEG signals under complex visual stimulation is studied, and the knowledge in the field of cutting-edge artificial intelligence is used to decode the visual cognition process of the brain. This can provide new ideas for understanding the mechanism of visual cognition and provide a theoretical basis and solution for intelligent image processing technology for complex target recognition.

[0003] Considering that the intelligent image processing model based on the visual cognitive process of a single subject can effectively approximate and learn the visual cognitive skills of a single subject, it is inevitable that there are problems such as overfitting and poor generalization to a certain extent. How to comprehensively utilize the visual cognitive advantages of different subjects to construct an effective, reliable and highly generalized image processing and classification model is a very important research content. Based on the ensemble clustering theory, the present invention develops and designs an effective ensemble clustering algorithm, and uses multiple different subjects trained synchronously to realize the decision-making layer fusion of the EEG signal recognition model under visual stimulation of multiple subjects, and constructs a cross-individual EEG signal recognition and analysis method under complex visual stimulation with high generalization, accuracy and robustness. Summary of the invention

[0004] The technical problem solved by the present invention is that different subjects have different visual cognition differences for complex visual stimulation images, so the representation ability and generalization of the EEG signal decoding model based on the personalized visual stimulation of a single subject is weak.

[0005] The present invention provides an electroencephalogram analysis method under complex visual stimulation based on cluster integration, comprising the following steps:

[0006] S1. Design an experimental paradigm for EEG signal acquisition under complex visual stimulation conditions, select N subjects to perform the operation according to the experimental paradigm, and synchronously acquire and record the multi-channel EEG signal data of N subjects under complex visual stimulation in the experimental paradigm;

[0007] S2. Start the experiment, pre-process the EEG signals of N subjects, obtain the pre-processed EEG signal data corresponding to the N subjects, and form a rapid sequence presentation EEG signal data set;

[0008] S3. Use the spectral clustering algorithm to perform cluster analysis on the EEG signal data of N subjects to obtain N sub-clusterers;

[0009] S4. Use the spectral clustering ensemble optimization strategy based on hypergraph partitioning to fuse the N tested sub-clusterers;

[0010] S5. Test and evaluate the fusion model to obtain the performance attribute value of the fusion model.

[0011] In the rapid serial visual presentation experiment, the subject's task is to pay attention to the target picture and press a button as soon as possible after it appears. By analyzing the subject's EEG response, it is determined whether the subject has noticed the target picture, thereby achieving information transmission. The experimental setting parameters include stimulation type, stimulation duration, stimulation interval, target stimulation, target stimulation frequency, and response method.

[0012] In step S1, the rapid serial visual presentation experimental paradigm and parameters with a presentation frequency of 12.5 Hz are as follows: the visual stimulus image is selected according to the task requirements; the stimulus duration is set to 80 ms; the stimulus interval is set to 0; the target stimulus frequency is set to 5%; and the response mode is set to the subject counting the number of targets.

[0013] In this experiment, the subjects wore an electrode cap and sat 1 meter away from a 1920×1080 resolution monitor. A fixed cross mark was added in the middle of the visual stimulus image as a fixation point. The subjects pressed the space bar on the keyboard to start viewing the image sequence. Each experimental stage had 10 sequences, each sequence contained 200 pictures, and was repeated 5 rounds; the interval between two sequences was controlled by the subjects, and pressing the space bar started the presentation of the next sequence of images.

[0014] The pre-processing includes resetting the reference electrode and filtering.

[0015] The reference electrodes were reset with the left mastoid channel M1 and the right mastoid channel M2 as references; and the band-pass FIR digital filters with cut-off frequencies of 0.3 Hz and 65 Hz were used for filtering.

[0016] In step S3, an attribute weighted spectral clustering algorithm based on knowledge entropy is adopted. The algorithm uses the related concept of knowledge entropy in rough set to evaluate the importance of each EEG signal, and then assigns it as a weight to the corresponding feature, and then applies the spectral clustering method to cluster the data points.

[0017] The specific steps of the algorithm are as follows:

[0018] S31. Construct a similarity matrix: construct a similarity matrix based on the data, where each element in the matrix represents the similarity between two data points. Commonly used similarity metrics include Gaussian kernel function and k nearest neighbors, where k is the number of clusters;

[0019] S32. Construct a graph: construct a graph based on the similarity matrix, where each data point represents a node and the weight of each edge represents the similarity between two nodes;

[0020] S33. Calculate the Laplacian matrix: Calculate the Laplacian matrix of the graph. The Laplacian matrix can be used to represent the structural information of the graph;

[0021] S34. Calculate eigenvalues ​​and eigenvectors: Calculate the eigenvalues ​​and eigenvectors of the Laplace matrix, and select the eigenvectors corresponding to the first k smallest eigenvalues;

[0022] S35. Clustering: Use K-Means algorithm to cluster feature vectors.

[0023] The fusion in step S4 includes the following steps:

[0024] S41. Use the spectral clustering algorithm to train multiple sub-clusterers, and store the prediction results of the multiple sub-clusterers in the ensemble list;

[0025] S42. Transpose the ensemble list so that each row represents the prediction result of a sample in all sub-clusterers;

[0026] S43. Create a hypergraph object and add the prediction result of each row as an edge to the hypergraph; in this hypergraph, each node represents a sample, each edge represents the prediction result of a sub-clusterer, and the nodes connected by the edge indicate that these nodes are classified into the same cluster in the sub-clusterer;

[0027] S44. Use the hypergraph partitioning algorithm to determine the final cluster label for each sample.

[0028] In the evaluation, the silhouette coefficient, Dunn index, and CH index are introduced; if the true clustering label of the data is already known, one or more of the adjusted Rand index, adjusted mutual information, homogeneity, completeness, and V-measure score are introduced in the review.

[0029] Compared with the prior art, the present invention has the following beneficial effects: based on the ensemble clustering theory, the present invention develops and designs an effective ensemble clustering algorithm and decision fusion strategy, and utilizes a plurality of different subjects who are trained synchronously to realize the decision-layer integration of the visual cognitive image classification models of the brains of a plurality of subjects, thereby constructing a cross-individual EEG signal analysis method under visual stimulation targets with high generalization, accuracy and robustness. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a basic framework schematic diagram of the present invention;

[0031] Figure 2 Schematic diagram of the experimental paradigm for EEG signal acquisition based on RSVP

[0032] Figure 3 EEG signal acquisition flow chart

[0033] Figure 4 Waveform comparison before and after channel filtering

[0034] Figure 5 Time domain waveform and frequency domain power spectrum of the original acquisition signal FP1 channel

[0035] Figure 6 Time domain waveform and frequency domain power spectrum of FP1 channel after processing

[0036] Figure 7 Flowchart of EEG signal preprocessing under complex visual stimulation DETAILED DESCRIPTION

[0037] like Figure 1 As shown, the present invention provides an EEG analysis method under complex visual stimulation based on cluster integration, comprising the following steps:

[0038] S1. Design an experimental paradigm for EEG signal acquisition under complex visual stimulation conditions, select N subjects to perform operations according to the experimental paradigm, and synchronously acquire and record the multi-channel EEG signal data of the N subjects under complex visual stimulation in the experimental paradigm.

[0039] The present invention is aimed at the demand for EEG signal recognition and decoding under complex visual stimulation. By analyzing the theoretical techniques and methods of designing a neural signal acquisition experimental paradigm under typical rapid serial visual presentation (RSVP) visual stimulation, the present invention takes target recognition in complex scenes as the experimental task, designs a rapid serial visual presentation EEG signal acquisition experimental paradigm under this task, and completes the EEG signal acquisition of N subjects.

[0040] (1) Experimental paradigm design of EEG signal acquisition based on RSVP

[0041] Rapid serial visual presentation is a commonly used experimental paradigm to measure attention and cognitive processing ability. In this experimental paradigm, the subjects need to watch a series of rapidly presented visual stimulus images in a short period of time and look for the target stimulus among them. In the RSVP experiment, the subject's task is to pay attention to the target image and press the button as soon as possible after it appears. By analyzing the subject's EEG response, it can be determined whether the subject has noticed the target image, thereby achieving information transmission.

[0042] In order to study the visual cognition of the visual system under the rapid presentation of visual stimuli, the present invention designs a rapid serial visual presentation experimental paradigm in which each image is presented for a certain period of time, such as an image presentation time of 80sm, that is, a presentation frequency of 12.5Hz. Visual stimulus images use complex target images to study the perception ability of the human visual system to related visual stimulus images. The advantages of the RSVP experimental paradigm are simplicity, efficiency, and ease of operation, and it is suitable for a variety of application scenarios, such as image retrieval, safety testing, cognitive assessment, etc. When designing the RSVP experimental paradigm, it is usually necessary to consider factors such as stimulus type, stimulus duration, stimulus interval, target stimulus, target stimulus frequency, and reaction mode, which are specifically described as follows:

[0043] Stimulus type: The stimulus type should be selected according to the research purpose and hypothesis. For example, if you want to study language processing, you can choose words or sentences as stimuli; if you want to study visual processing, you can choose pictures or graphics as stimuli. The stimulus type will also affect the characteristics of the EEG signal. For example, language stimulation will cause components such as N400 and P600, while visual stimulation will cause components such as P1, N1 and P3.

[0044] Stimulus duration: Stimulus duration refers to the time each stimulus is presented on the screen, which determines the depth and difficulty of the subject's processing of the stimulus. Generally speaking, the shorter the stimulus duration, the more difficult the processing, the higher the attention, and the stronger the EEG signal. However, if the stimulus duration is too short, it may cause the subject to be unable to recognize the stimulus or produce a visual aftereffect, affecting the experimental effect. Therefore, the stimulus duration should be appropriately adjusted according to the stimulus type and task requirements, usually between 100-500 milliseconds.

[0045] Interstimulus interval: The interstimulus interval refers to the time interval between two consecutive stimuli, which affects the subject's memory and comparison of the previous and subsequent stimuli. Generally speaking, the shorter the interstimulus interval, the more difficult the memory and comparison, the higher the attention, and the stronger the EEG signal. However, if the interstimulus interval is too short, it may cause the subject to be unable to distinguish the previous and subsequent stimuli or produce visual fusion, affecting the experimental effect. Therefore, the interstimulus interval should also be appropriately adjusted according to the stimulus type and task requirements, usually between 0-200 milliseconds.

[0046] Target stimulus: Target stimulus refers to the stimulus that the subject needs to pay special attention to or respond to in RSVP. It can be used to measure the subject's attention, reaction speed, accuracy and other indicators. Target stimulus can be set with different characteristics according to different tasks, such as color, shape, size, position, semantics, etc. Target stimulus will also cause specific EEG components, such as P3, N2pc and LPC.

[0047] Target stimulation frequency: The target stimulation frequency refers to the probability of the target stimulation appearing in RSVP, which affects the subject's expectations and degree of surprise. Generally speaking, the lower the target stimulation frequency, the lower the subject's expectations, the higher the surprise, and the stronger the EEG signal. However, if the target stimulation frequency is too low, it may cause the subject to lose interest or attention, affecting the experimental effect. Therefore, the target stimulation frequency should be selected according to the research purpose and hypothesis, usually between 5-40%.

[0048] Response mode: Response mode refers to the way the subject responds to the target stimulus. It can be used to measure indicators such as the subject's response speed and accuracy. Response modes can take many forms, such as key presses, eye movements, and spoken language. Response modes will also affect the characteristics of EEG signals. For example, key presses will cause motor evoked potentials (MEPs), eye movement reactions will cause electrooculograms (EOGs), and spoken language reactions will cause electromyograms (EMGs).

[0049] Aiming at the problem of rapid visual stimulation image target recognition to be solved by the invention, an RSVP experimental paradigm and key parameters with a presentation frequency of 12.5Hz are designed. The visual stimulation images are selected according to the task requirements; the stimulation duration is set to 80ms; the stimulation interval is set to 0 according to the conventional RSVP paradigm, that is, there is no interval between the two visual stimulation images, and the next visual stimulation image appears immediately after the presentation of one image; the target stimulation frequency is set to 5%, that is, there are only 5 target images in every 100 visual stimulation images, and the others are non-target images; the response mode is set to the subject counting the number of targets to avoid interference from motion signals generated by key presses.

[0050] In this experiment, participants wore electrode caps and sat 1 meter away from a 1920*1080 resolution monitor. In order to reduce the influence of the subjects' eye movements during the experiment, a fixed cross mark was added in the middle of the visual stimulus image as the fixation point. The subjects pressed the space bar on the keyboard to start viewing the image sequence. Each experimental stage included 10 sequences, each sequence contained 200 pictures, and was repeated 5 times. The interval between two sequences was controlled by the subject, and pressing the space bar started the next sequence image presentation. The experimental process is as follows Figure 2 shown.

[0051] (2) EEG signal acquisition under visual stimulation

[0052] The main characteristics of EEG signals are: ① The signal amplitude is weak and needs to be amplified by about tens of thousands of times; ② The signal frequency range is low (within 100 Hz, and extremely low frequency drift also needs to be filtered out), and high-frequency noise needs to be filtered out; ③ The signal source impedance is high (there is a resistance of several thousand ohms between the scalp and the skull), and the preamplifier circuit input impedance needs to be very high, generally greater than 10 megohms; ④ The power supply frequency interference is common-mode interference, and the amplifier common-mode suppression needs to be relatively high; ⑤ The signal-to-noise ratio is low and the randomness is strong.

[0053] The experiment uses an EEG acquisition device specially used for measuring and recording EEG to collect EEG signals. The working process is as follows: the electrodes placed on the scalp can detect weak EEG signals, usually 5-100μV, which are coupled to the differential amplifier through the electrode leads for amplification by about 10000-100000 times, and then converted into digital signals through the data acquisition card in the computer host, and finally the signal data is recorded through the recording system on the matching PC.

[0054] The EEG acquisition equipment of the present invention adopts the 64-lead device of Australia's NeuroScan SynAmps, which is specially designed for the acquisition of weak electrophysiological signals and shielded from the outside world. It uses standardized and reasonable operating methods to collect EEG signals in a laboratory environment that is specially designed for the acquisition of weak electrophysiological signals. The following are the following: ① To ensure good conductivity between the scalp and the sensor, the subject should first wash his hair before acquisition; ② The EEG acquisition equipment is connected and debugged, and the amplifier, AD converter, data transmission optical fiber and data recording notebook are correctly connected so that the EEG signals can be effectively collected; ③ Assist the subject to wear the EEG cap to ensure that the electrode position matches the corresponding area of ​​the scalp; ④ Explain the experimental process to the subject to confirm that the subject has a full understanding of the experimental process and operation; ⑤ Start the experiment. During the signal recording process, the experimenter always pays attention to the subject's reaction, prevents and promptly handles and resolves emergencies such as equipment failure and subject discomfort during the experiment, and supervises the smooth progress of the entire experiment; ⑥ After the experiment, organize the equipment and keep good experimental records. The signal acquisition flow chart is as follows Figure 3 shown.

[0055] S2. Start the experiment, pre-process the EEG signals of N subjects, obtain the pre-processed EEG signal data corresponding to the N subjects, and form a rapid serial presentation EEG signal data set.

[0056] The present invention designs an EEG signal preprocessing method to obtain high-quality EEG signal data from a sufficient number of subjects, form a rapid sequence presentation EEG signal data set, and provide a key data basis for subsequent EEG signal processing under complex visual stimulation. The EEG signal preprocessing method proposed by the present invention mainly includes two parts: resetting the reference electrode and filtering.

[0057] (1) Reset the reference electrode

[0058] The electrophysiological amplifier used for EEG acquisition is a differential amplifier, which generally uses a unipolar lead. Each position electrode is connected to one input terminal of the amplifier, and the other input terminal is connected to a common reference electrode, so the potential of each electrode should be the result of subtracting its absolute potential from the absolute potential of the same reference electrode. The ideal reference electrode point should be a constant potential, but because the human body is a volume conductor, the bioelectricity of each part is changing at any time, so theoretically an ideal reference electrode does not exist. The present invention uses the left and right mastoid channels (M1 and M2 channels) as a reference to reset the reference electrode.

[0059] (2) Filtering

[0060] There are two main reasons for filtering the collected raw EEG data: ① According to the Nyquist theorem, if the collected raw signal contains a signal with a frequency higher than 2 times the sampling frequency, frequency aliasing will occur, so a lower frequency needs to be selected as the cutoff frequency. In general, the sampling frequency is at least 3 times the cutoff frequency in practical applications; ② Some noise contained in EEG has a large frequency difference from the useful signal. In cognitive neuroscience experiments, the frequency range of most experimentally relevant useful components in EEG signals is between 0.01Hz and 60Hz, so a low-pass filter with a cutoff frequency of around 60Hz can be set to filter out noise information and retain effective components. The filter should be used reasonably, otherwise it will cause significant signal distortion in severe cases. In addition to filtering out high-frequency signals, filters are also needed in many experiments to attenuate extremely low-frequency signals to remove slow voltage changes caused by non-neural activities during the recording process. For example, sweating, electrode resistance drift, etc. will cause slow and continuous changes in the EEG baseline voltage, resulting in slow and long-term voltage drift, which will cause large distortion of the waveform. A high-pass filter with a cutoff frequency of about 0.1Hz can be used to filter out those extremely low frequencies.

[0061] The present invention uses a bandpass FIR digital filter with cutoff frequencies of 0.3 Hz and 65 Hz to filter out low-frequency and high-frequency noise. The waveforms of the EEG data collected by a subject before and after channel filtering are shown in the figure below. Figure 4 As shown, by comparison, the high-frequency noise components are obviously filtered out.

[0062] The original EEG signal FP1 channel time domain waveform and frequency domain power spectrum are as follows Figure 5 As shown in the figure, it can be seen that the original signal collected contains complex noise interference, and the signal amplitude is large, with an order of magnitude of 10 4 Microvolt, wide frequency band, occupying [0, 100Hz], effective weak electrophysiological signals are submerged.

[0063] The time domain waveform and frequency domain power spectrum of the corresponding channel signal after filtering are shown in Figure 6It can be seen that the amplitude of the extracted weak electrophysiological signal is on the order of tens of microvolts, and the frequency band occupies [0, 20Hz], which is consistent with the characteristics of electrophysiological signals and verifies the effectiveness of the weak signal processing method.

[0064] S3. Use the spectral clustering algorithm to perform cluster analysis on the EEG signal data of N subjects to obtain N sub-clusterers.

[0065] According to the characteristics of key EEG information under different visual stimuli, the present invention proposes an unsupervised EEG information decoding and target classification method based on high-dimensional spectral clustering, and explores the potential characteristics of EEG signals under different visual stimuli, in order to accurately identify EEG signals under different stimuli. At present, graph theory has been proved to be a very useful tool for solving important combinatorial problems in the fields of geometry, number theory, operations research and optimization. In recent years, with the popularization of computers and the increase in the scale of integrated circuits, the research on hypergraph theory and its application has received more and more attention. Spectral clustering is a method of clustering using the principle of graph segmentation. It converts the data clustering problem into a graph partitioning problem for solution, which is particularly suitable for the case where the data set is non-convex. Each point in the data set is used as a vertex of the graph, and the similarity value between any two points is used as the weight of the edge connecting the two vertices, so that an undirected weighted graph is constructed. According to the objective function, the graph can be divided into several unconnected subgraphs, the connection weight within the subgraph is maximized, and the connection weight between each subgraph is minimized. The point set contained in the subgraph is the cluster generated after clustering. Spectral clustering theory provides a new approach to solving clustering problems, and it can effectively handle many practical problems faced in unsupervised learning.

[0066] In view of the fact that the traditional spectral clustering method performs well on some low-dimensional and small-scale data sets, it is easily disturbed by noise and irrelevant attributes in the application scenarios of high-dimensional data. In order to make full use of the information contained in each attribute and strengthen the role of important attributes, the present invention designs an attribute weighted spectral clustering algorithm based on knowledge entropy. The algorithm uses the relevant concept of knowledge entropy in the rough set to evaluate the importance of each EEG feature, and then assigns it as a weight to the corresponding feature, and then applies the spectral clustering method to cluster the data points. The algorithm can better handle high-dimensional data and has strong robustness and generalization ability. The basic idea of ​​the spectral clustering algorithm is to regard the data points as nodes in the graph, and then establish edges between the nodes according to the similarity between the data points; then divide the data into different clusters by cutting the graph. The specific steps of the algorithm are as follows:

[0067] (1) Constructing a similarity matrix: A similarity matrix is ​​constructed based on the data, where each element in the matrix represents the similarity between two data points. Commonly used similarity metrics include Gaussian kernel function and k-nearest neighbor.

[0068] (2) Graph construction: A graph is constructed based on the similarity matrix, where each data point represents a node and the weight of each edge represents the similarity between two nodes.

[0069] (3) Calculate the Laplacian matrix: Calculate the Laplacian matrix of the graph. The Laplacian matrix can be used to represent the structural information of the graph.

[0070] (4) Calculate eigenvalues ​​and eigenvectors: Calculate the eigenvalues ​​and eigenvectors of the Laplacian matrix, and select the eigenvectors corresponding to the first k smallest eigenvalues.

[0071] (5) Clustering: Use K-Means or other clustering algorithms to cluster feature vectors.

[0072] S4. Use the spectral clustering ensemble optimization strategy based on hypergraph partitioning to fuse the sub-clusterers of N subjects.

[0073] The present invention designs a spectral clustering ensemble optimization method based on hypergraph partitioning. Generally, the edge of a graph has only two vertices, while a hyperedge of a hypergraph can have any number of vertices. Cluster members can be represented by a hypergraph: a hyperedge represents a cluster, and the vertices of the hyperedge represent data points belonging to the cluster. Clustering integration is converted into a minimum cut problem of a hypergraph, and clustering integration is performed using a clustering algorithm based on graph theory. Hypergraph partitioning is a method of dividing a hypergraph into multiple parts, which can be used to determine the final clustering label of each sample in a clustering ensemble. First, multiple sub-clusterers are trained using a spectral clustering algorithm, and their prediction results are stored in an ensemble list. Then, the ensemble list is transposed so that each row represents the prediction result of a sample in all sub-clusterers. Next, a hypergraph object is created, and the prediction result of each row is added to the hypergraph as an edge. In this hypergraph, each node represents a sample, and each edge represents the prediction result of a sub-clusterer. Nodes connected by edges indicate that these nodes are classified into the same cluster in the sub-clusterer. Therefore, this hypergraph captures the relationship between the predictions of all sub-clusterers. Finally, the hypergraph partitioning algorithm is used to determine the final cluster label for each sample. The hypergraph partitioning algorithm attempts to partition the hypergraph into multiple parts so that the sum of the edge weights within each part is maximized, while the sum of the edge weights between different parts is minimized. In this case, each part represents a final cluster. Further, based on this construction, Figure 1 The highly generalized image analysis model framework based on clustering ensemble is shown.

[0074] S5. Test the fusion model to obtain performance attribute values ​​of the fusion model.

[0075] In order to evaluate the effectiveness of the proposed method, the present invention analyzes common evaluation indicators of clustering and performs quantitative evaluation on the proposed method. Clustering, as an unsupervised learning algorithm, has no real value to verify the results. This makes it complicated to evaluate the quality of clustering. However, commonly used indicators such as silhouette coefficient, Dunn index, Calinski-Harabasz index, etc. can be used to evaluate the performance of clustering models. Different indicators are defined as follows:

[0076] Silhouette Coefficient: The silhouette coefficient measures the cohesion within a cluster and the separation between clusters. Its value is between -1 and 1, and the larger the value, the better the clustering quality. The silhouette coefficient is a commonly used clustering evaluation indicator.

[0077] Dunn's Index: Dunn's Index measures the compactness within a cluster and the separation between clusters. The larger its value, the better the clustering quality.

[0078] Calinski-Harabasz Index: Calinski-Harabasz Index is also called the variance ratio criterion. It measures the compactness within a cluster and the separation between clusters. The larger its value, the better the clustering quality.

[0079] In addition, if the real clustering labels of the data are already known, then the effect of the clustering algorithm can also be evaluated using indicators such as Adjusted Rand Index (ARD), Adjusted Mutual Information (AMI), Homogeneity (Homo), Completeness (Comp) and V-measure score (V_m). They can all be used to measure the consistency between the clustering results and the real labels. The larger the value, the better the clustering effect. The evaluation of the clustering effect of EEG signals of different subjects is shown in the following table:

[0080] Table 1 Clustering and integration results of EEG signals of different subjects

[0081]

[0082]

[0083] Parts not described in detail in the present invention belong to common knowledge of those skilled in the art.

Claims

1. A method for analyzing EEG under complex visual stimulation based on clustering integration, characterized in that: The following steps are involved: S1. Design an experimental paradigm for EEG signal acquisition under complex visual stimulation conditions, select N subjects to perform the operation according to the experimental paradigm, and synchronously acquire and record the multi-channel EEG signal data of N subjects under complex visual stimulation in the experimental paradigm; S2. Start the experiment, pre-process the EEG signals of N subjects, obtain the pre-processed EEG signal data corresponding to the N subjects, and form a rapid sequence presentation EEG signal data set; S3. Perform cluster analysis on the EEG signal data of N subjects using a spectral clustering algorithm to obtain N sub-clusterers; S4. Use the spectral clustering ensemble optimization strategy based on hypergraph partitioning to fuse the N tested sub-clusterers; S5. Test and evaluate the fusion model to obtain the performance attribute value of the fusion model.

2. The method according to claim 1, characterized in that In the rapid serial visual presentation experiment, the subject's task is to pay attention to the target picture and press a button as soon as possible after it appears. By analyzing the subject's EEG response, it is determined whether the subject has noticed the target picture, thereby achieving information transmission. The experimental setting parameters include stimulation type, stimulation duration, stimulation interval, target stimulation, target stimulation frequency, and response method.

3. The method according to claim 2, characterized in that In step S1, the rapid serial visual presentation experimental paradigm and parameters with a presentation frequency of 12.5 Hz are as follows: the visual stimulus image is selected according to the task requirements; the stimulus duration is set to 80 ms; the stimulus interval is set to 0; the target stimulus frequency is set to 5%; and the response mode is set to the subject counting the number of targets.

4. The method according to claim 3, characterized in that In this experiment, the subjects wore an electrode cap and sat 1 meter away from a 1920×1080 resolution monitor. A fixed cross mark was added in the middle of the visual stimulus image as a fixation point. The subjects pressed the space bar on the keyboard to start viewing the image sequence. Each experimental stage had 10 sequences, each sequence contained 200 pictures, and was repeated 5 rounds; the interval between two sequences was controlled by the subjects, and pressing the space bar started the presentation of the next sequence of images.

5. The method according to claim 4, characterized in that The pre-processing includes resetting the reference electrode and filtering.

6. The method according to claim 5, characterized in that The reference electrodes were reset with the left mastoid channel M1 and the right mastoid channel M2 as references; and the band-pass FIR digital filters with cut-off frequencies of 0.3 Hz and 65 Hz were used for filtering.

7. The method according to claim 6, characterized in that In step S3, an attribute weighted spectral clustering algorithm based on knowledge entropy is adopted. The algorithm uses the related concept of knowledge entropy in rough set to evaluate the importance of each EEG signal, and then assigns it as a weight to the corresponding feature, and then applies the spectral clustering method to cluster the data points.

8. The method according to claim 7, characterized in that The specific steps of the algorithm are as follows: S31. Construct a similarity matrix: construct a similarity matrix based on the data, where each element in the matrix represents the similarity between two data points. Commonly used similarity metrics include Gaussian kernel function and k nearest neighbors, where k is the number of clusters; S32. Construct a graph: construct a graph based on the similarity matrix, where each data point represents a node and the weight of each edge represents the similarity between two nodes; S33. Calculate the Laplacian matrix: Calculate the Laplacian matrix of the graph. The Laplacian matrix can be used to represent the structural information of the graph; S34. Calculate eigenvalues ​​and eigenvectors: Calculate the eigenvalues ​​and eigenvectors of the Laplace matrix, and select the eigenvectors corresponding to the first k smallest eigenvalues; S35. Clustering: Use K-Means algorithm to cluster feature vectors.

9. The method according to claim 8, characterized in that The fusion in step S4 includes the following steps: S41. Use the spectral clustering algorithm to train multiple sub-clusterers, and store the prediction results of the multiple sub-clusterers in the ensemble list; S42. Transpose the ensemble list so that each row represents the prediction result of a sample in all sub-clusterers; S43. Create a hypergraph object and add the prediction result of each row as an edge to the hypergraph; in this hypergraph, each node represents a sample, each edge represents the prediction result of a sub-clusterer, and the nodes connected by the edge indicate that these nodes are classified into the same cluster in the sub-clusterer; S44. Use the hypergraph partitioning algorithm to determine the final cluster label for each sample.

10. The method according to claim 9, characterized in that In the evaluation, the silhouette coefficient, Dunn index, and CH index are introduced; if the true clustering label of the data is already known, one or more of the adjusted Rand index, adjusted mutual information, homogeneity, completeness, and V-measure score are introduced in the review.

Images