Electroencephalogram lead selection method and device, computer device and storage medium

Abstract

The present application relates to the field of aerospace technology, and discloses an electroencephalogram lead selection method and device, computer equipment and a storage medium. The present application obtains ear pulse information set and first electroencephalogram signal set of a subject, filters first sub-ear pulse signals in the first ear pulse signal, and filters a relevant electroencephalogram signal subset from the electroencephalogram signals of the first electroencephalogram signal subset according to the time period information of the first sub-ear pulse signals. According to the pre-acquired sampling frequency of the first electroencephalogram signal subset, each electroencephalogram signal in the relevant electroencephalogram signal subset is divided into multiple sub-electroencephalogram signal sets, each sub-electroencephalogram signal set corresponds to one lead, each sub-electroencephalogram signal in the first sub-electroencephalogram signal set is analyzed, the autocorrelation information of the lead corresponding to the first sub-electroencephalogram signal set under the corresponding load is obtained, the correlation information between the load and the lead is determined according to the autocorrelation information of the lead of the subject under the corresponding load and the corresponding load, and the lead is determined according to the correlation information and the preset condition.

Description

Technical Field

[0001] This invention relates to the field of EEG lead selection technology, specifically to EEG lead selection methods, devices, computer equipment, and storage media. Background Technology

[0002] In the aerospace field, it is necessary to analyze the state of subjects under +Gz load during flight and study changes in physiological signals, including electroencephalograms (EEGs). However, during the dynamic acquisition of EEG signals under +Gz load in a centrifuge, the EEG signals are susceptible to various interferences, and the installation and fixation of electrodes and lead wires, as well as the installation and fixation of the measuring instrument and its components, require special methods. Therefore, it is necessary to identify the leads that are highly correlated with changes in G-load and contribute significantly to EEG changes under centrifugation, i.e., those that are highly representative, and then sort and simplify the leads based on these leads. Summary of the Invention

[0003] In view of this, the present invention provides a method, apparatus, computer device, and storage medium for selecting electroencephalogram (EEG) leads to solve the above problems.

[0004] In a first aspect, the present invention provides a method for selecting electroencephalogram (EEG) leads, the method comprising:

[0005] Acquire at least one set of audible pulse information and a first set of EEG signals from a subject during training under different loads; the first set of EEG signals includes multiple subsets of EEG signals, each subset of EEG signals in the first set of EEG signals is acquired from multiple leads corresponding to the subject, and the number of EEG signal subsets is the same as the number of leads.

[0006] Among the first ear pulse signals, the first sub-ear pulse signal with the largest rate of change is selected. The first ear pulse signal is any one of the ear pulse information sets.

[0007] Based on the time period information corresponding to the first sub-auricular pulse signal, relevant EEG signal subsets within the corresponding time period information are selected from the EEG signals corresponding to the first EEG signal subset. The first EEG signal subset and the first auricular pulse signal are from the same subject.

[0008] Based on the sampling frequency of the first subset of pre-acquired EEG signals, each EEG signal in the relevant EEG signal subset is divided into multiple sub-EEG signal sets, and each sub-EEG signal set corresponds to a lead;

[0009] Analyze each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the lead corresponding to the first sub-EEG signal set under the corresponding load. The first sub-EEG signal set is any one of multiple sub-EEG signal sets.

[0010] Based on the autocorrelation information of the leads corresponding to all test subjects under the corresponding load, and the load corresponding to the autocorrelation information, determine the correlation information between the load and the lead;

[0011] Based on the associated information and preset conditions, determine the leads that meet the criteria.

[0012] Beneficial effects include obtaining at least one set of ear pulse information and a first set of EEG signals from a subject during different load training processes; selecting the first sub-ear pulse signal with the highest rate of change from the first ear pulse signal; selecting a subset of relevant EEG signals within the corresponding time period from the EEG signals corresponding to the first subset of EEG signals based on the time period information corresponding to the first subset of EEG signals; accurately identifying a portion of EEG signals with strong correlation based on the ear pulse signals, and further, based on the sampling frequency of the pre-acquired first subset of EEG signals... Each EEG signal in the relevant EEG signal subset is divided into multiple sub-EEG signal sets, and each sub-EEG signal set corresponds to a lead. This allows analysis of each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the lead corresponding to the first sub-EEG signal set under the corresponding load. Subsequently, based on the autocorrelation information of the leads of all subjects under the corresponding load and the load corresponding to the autocorrelation information, the correlation information between the load and the lead is determined. Based on the correlation information and preset conditions, the leads that meet the conditions are determined, thereby identifying the leads with the highest correlation to the EEG signal, reducing the number of leads and simplifying the lead sorting.

[0013] In one optional implementation, acquiring a first set of electroencephalogram (EEG) signals specifically includes:

[0014] Acquire the second set of EEG signals from the subject;

[0015] The second set of EEG signals was subjected to interference processing to obtain the third set of EEG signals.

[0016] The third set of EEG signals was amplified to obtain the fourth set of EEG signals;

[0017] The fourth set of EEG signals was denoised and filtered to obtain the first set of EEG signals.

[0018] Beneficial effects include improved accuracy of EEG signals and enhanced accuracy of subsequent lead correlation analysis.

[0019] In one optional implementation, the correlation information between load and lead is determined based on the autocorrelation information of the leads corresponding to all test subjects under the corresponding load, and the load corresponding to the autocorrelation information. Specifically, this includes:

[0020] Based on the autocorrelation information and the corresponding load, determine the correlation coefficient between the load and the lead;

[0021] Based on the correlation coefficient of each lead corresponding to different loads, the correlation information between the load and the lead is determined.

[0022] The beneficial effect is to more accurately determine the correlation between leads and loads.

[0023] In one optional implementation, leads that meet the conditions are determined based on association information and preset conditions, specifically including:

[0024] The leads are sorted according to the correlation information to obtain the lead sequence;

[0025] Based on the ranking of leads in the lead sequence, weight information is assigned to the leads in the lead sequence;

[0026] Based on weight information and preset conditions, select leads that meet the criteria.

[0027] Beneficial effects include assigning weight information to leads in the lead sequence, which in turn assigns leads with stronger relevant information, thereby improving the accuracy of lead-related information. Furthermore, based on the weight information and preset conditions, it is possible to accurately identify the leads that ultimately meet the preset conditions.

[0028] In an optional implementation, the method further includes: assigning weight information to the leads in the lead sequence according to the ranking of the leads in the lead sequence, specifically including:

[0029] Based on the ranking of the leads in the lead sequence, initial weight information is assigned to the leads in the lead sequence;

[0030] The final weight information of the leads is determined based on the initial weight information of all subjects' leads.

[0031] Beneficial effects: Based on the ranking in the lead sequence, initial weight information is assigned, which reduces the impact of randomness in EEG signals. Furthermore, based on the initial weight information of the leads corresponding to all subjects, the final weight information of the leads is determined, which reduces the influence of unstable factors of individual subjects and improves the accuracy.

[0032] In one optional implementation, the leads are sorted according to the absolute value of the association information to obtain a lead sequence.

[0033] Beneficial effects, eliminating the instability and specificity of lead changes.

[0034] Secondly, the present invention provides an electroencephalogram (EEG) lead selection device, the device comprising:

[0035] The information acquisition module is used to acquire at least one subject's ear pulse information set and first EEG signal set during different load training processes; the first EEG signal set includes multiple EEG signal subsets, and each EEG signal subset in the first EEG signal set is acquired by multiple leads corresponding to the subject, and the number of EEG signal subsets is the same as the number of leads.

[0036] The first filtering sub-information module is used to filter the first sub-ear pulse signal with the largest change rate in the first ear pulse signal from the first ear pulse signal. The first ear pulse signal is any one in the ear pulse information set.

[0037] The second screening sub-information module is used to filter the relevant EEG signal subset within the corresponding time period information from the EEG signals corresponding to the first EEG signal subset based on the time period information corresponding to the first sub-ear pulse signal. The first EEG signal subset and the first ear pulse signal are from the same subject.

[0038] The sub-signal segmentation module is used to divide each EEG signal in the relevant EEG signal subset into multiple sub-EEG signal sets according to the sampling frequency of the pre-acquired first EEG signal subset. Each sub-EEG signal set corresponds to a lead.

[0039] The correlation analysis module is used to analyze each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the lead corresponding to the first sub-EEG signal set under the corresponding load. The first sub-EEG signal set is any one of multiple sub-EEG signal sets.

[0040] The relationship determination module is used to determine the relationship information between load and lead based on the autocorrelation information of all subjects' leads under corresponding loads and the loads corresponding to the autocorrelation information.

[0041] The lead selection module is used to determine the leads that meet the criteria based on the associated information and preset conditions.

[0042] In one optional implementation, the information acquisition module specifically includes:

[0043] The signal acquisition unit is used to acquire the subject's second set of electroencephalogram (EEG) signals.

[0044] An interference processing unit is used to process the second set of EEG signals to obtain a third set of EEG signals.

[0045] The amplification processing unit is used to amplify the third set of EEG signals to obtain the fourth set of EEG signals;

[0046] The filtering unit is used to perform noise reduction filtering on the fourth EEG signal set to obtain the first EEG signal set.

[0047] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the EEG lead selection method of the first aspect or any corresponding embodiment described above.

[0048] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the EEG lead selection method of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0049] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0050] Figure 1 This is a flowchart of the EEG lead selection method according to an embodiment of the present invention;

[0051] Figure 2 This is a schematic diagram of the EEG lead selection method applicable to embodiments of the present invention;

[0052] Figure 3 This is a flowchart of the EEG lead selection method according to an embodiment of the present invention;

[0053] Figure 4 This is a schematic diagram of an EEG lead selection method according to an embodiment of the present invention;

[0054] Figure 5 This is a structural block diagram of an EEG lead selection device according to an embodiment of the present invention;

[0055] Figure 6 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] According to an embodiment of the present invention, an embodiment of an EEG lead selection method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0058] This embodiment provides a method for selecting EEG leads. Figure 1 This is a flowchart of an EEG lead selection method according to an embodiment of the present invention. The EEG lead selection method in this embodiment of the present invention is applied in the fields of aerospace medicine and biomedicine, specifically in the process of analyzing the EEG signals of subjects in a manned centrifuge.

[0059] In the aerospace field, to gain a more comprehensive and in-depth understanding of the state of subjects under +Gz (continuous positive acceleration) during flight, an increasing number of researchers are conducting experiments using manned centrifuges. They are studying changes in physiological signals, including electroencephalograms (EEGs), under +Gz conditions. Since EEGs directly reflect the electrophysiological activity of the cerebral cortex, researchers are attempting to identify characteristic changes in EEGs through appropriate variations, and then use these characteristics to provide early warnings for G-LOC (gravity-induced loss of consciousness) and in-flight incapacitation. Currently, the commonly used electrode numbers for EEG collection include 16, 32, and 64 leads. EEG measurements are mostly performed under static conditions. To gain a more comprehensive understanding of information from various brain regions, the number of electrodes is trending upwards, potentially expanding to 128, 256, or even 512 leads.

[0060] However, under dynamic conditions, especially under centrifugal and high-frequency (GZ) conditions, EEG signals are susceptible to various interferences. The installation and fixation of electrodes and lead wires, as well as the installation and fixation of the measuring instrument and its components, require specialized methods. The workload for electrode installation and fixation is relatively large, and the installation process is complex and cumbersome. Furthermore, not all lead data is necessary; some leads may carry useless or redundant information. There are also certain correlations and mutual information between leads. Therefore, selecting an appropriate number of leads can reduce the amount of EEG data processing and improve efficiency.

[0061] Therefore, it is necessary to identify the leads that are highly correlated with changes in G load and contribute significantly to EEG changes under centrifugation, i.e., those that are more representative. Based on this, the leads should be sorted and simplified. The simplification techniques will have significant application value in improving EEG measurement technology and signal analysis in this special environment under +Gz load.

[0062] Correlation analysis of electroencephalogram (EEG) data is a crucial part of EEG data analysis and forms the basis for other analyses. Calculating the correlation coefficient is one method of correlation analysis. Correlation coefficient algorithms are now widely used in energy, fault detection, and medical signal research. Currently, lead selection methods are generally based on physiological principles or experimental findings. Physiological principles include the fact that specific motor imagery can induce EEG changes in corresponding brain regions; for example, imagining movement of one upper limb activates the limb function area of ​​the contralateral sensory cortex, causing EEG changes. Researchers select leads based on these principles. Experimental findings involve repeated experiments to identify leads that contribute significantly to the extraction of specific EEG signal features.

[0063] Based on the above problems, the EEG lead selection method of the present invention, such as Figure 1 As shown, it includes the following steps:

[0064] Step S101: Obtain at least one set of ear pulse information and a first set of electroencephalogram (EEG) signals from a subject during different load training processes.

[0065] The first EEG signal set includes multiple EEG signal subsets. Each EEG signal subset in the first EEG signal set is acquired from multiple leads corresponding to the subject. The number of EEG signal subsets is the same as the number of leads.

[0066] For example, ear pulse information sets and first EEG signal sets are collected from multiple subjects under different loads. For instance, ear pulse and EEG signals are collected from 5 subjects under different loads. Each subject has 16 leads, and one EEG signal is collected from each lead. Each subject will have 16 sets of EEG signals and 1 set of ear pulse signals collected under the corresponding load. When the corresponding different loads include five sets of loads (using five sets as an example, but in reality, the number of loads for each subject may be different, determined according to the subject's G-load tolerance), a total of 16×5×5 sets of EEG signals and 1×5×5 sets of ear pulse signals will be collected. That is, after completing one training session for each subject, multiple sets of EEG signals (number of training sessions, each including 16 EEG signals) and multiple ear pulse signals will be collected.

[0067] To more accurately analyze ear pulse and electroencephalogram (EEG) signals, electrocardiogram (ECG) signals and corresponding facial expressions of the subjects can also be collected during training. The subjects' endurance can be comprehensively assessed by combining ECG signals and facial expressions. For example, endurance can be judged based on the subjects' visual perception of the ambient and central lights in the cockpit, such as reports of being well, ambient lights dimming, or ambient lights disappearing. This assessment can be combined with the subjects' facial expressions, such as normal expression without significant changes, a blank expression, or a state of near-fainting or already fainting, to make a comprehensive judgment on endurance. This allows for subsequent adjustments to the training load based on actual conditions.

[0068] In one specific embodiment, when collecting various data from subjects during different load training processes, the main equipment for the corresponding experimental operation mode is a novel triaxial high-performance manned centrifuge. The centrifuge's main arm is 8m long and has 3-axis acceleration. The specific settings for the acceleration curve of each run are as follows: when the centrifuge is stationary, the acceleration felt by the subject is 1G.

[0069] When the centrifuge is started, it initially accelerates at a rate of 1 G / s to reach the baseline for several seconds. Then, it accelerates at a rate of 3 G / s to reach the maximum G value set for each run, maintaining this rate for 10-15 seconds. Finally, it accelerates at a rate of 3 G / s to return to the initial state, and the centrifuge stops. Directly exposing subjects to high G values ​​is dangerous; therefore, the maximum G value for each run starts at 2.5 G and increases in increments of 0.5 G until the subject reaches their endurance endpoint or exhibits signs of needing to stop the centrifuge.

[0070] During data collection from subjects, the ear pulse signal first enters the physiological signal recording system in the centrifuge cabin, then is transmitted via a slip ring to the recording computer at the centrifuge medical station. The slip ring signal transmission requires voltage / current conversion and current / voltage conversion. Current loop transmission is used primarily because voltage transmission lines are more susceptible to interference, and the distributed resistance of the transmission lines can cause voltage drops; high resistance significantly affects transmission accuracy. Furthermore, providing the instrument's operating voltage on-site is also a challenge. Current loop transmission minimizes errors, resulting in higher, more accurate, and faster control, and enabling automatic control. Electroencephalogram (EEG) signals are collected and recorded using a portable EEG device, which is installed and fixed inside the centrifuge cabin.

[0071] When collecting EEG signals, an EEG instrument is used for acquisition and recording. The initially acquired EEG signals are subject to interference from other noises, and their accuracy may not meet the corresponding requirements when directly used for correlation studies. Therefore, in a preferred embodiment, a first EEG signal set is obtained, specifically including:

[0072] The second set of EEG signals from the subject was acquired; the second set of EEG signals was subjected to interference processing to obtain the third set of EEG signals; the third set of EEG signals was amplified to obtain the fourth set of EEG signals; the fourth set of EEG signals was subjected to noise reduction filtering to obtain the first set of EEG signals.

[0073] For example, the EEG signal first enters the in-cabin interference box for preliminary interference processing, and then enters the in-cabin amplification and recording box, where it is recorded in a specific format onto an SD card for subsequent processing. The recorded EEG data is first subjected to waveform playback, followed by data conversion, denoising, and filtering preprocessing.

[0074] Because EEG signals are very weak and dynamic EEG signals contain various artifacts, it is necessary to select appropriate methods to preprocess them to remove the artifacts and improve the performance and effectiveness of data feature extraction.

[0075] Wavelet filtering, a type of time-frequency filtering, achieves optimal time and frequency resolution for different parts of the signal, thus decomposing and reconstructing the signal at various spatial levels. It is well-suited for analyzing the transient and time-varying characteristics of non-stationary signals. Specifically, it offers high frequency resolution and low time resolution in the low-frequency range, while providing high time resolution and low frequency resolution in the high-frequency range.

[0076] The emerging wavelet packet method offers more refined decomposition than wavelet decomposition, allowing for simultaneous decomposition in both low and high frequencies. This means it can divide the frequency band into multiple levels, further decomposing the high-frequency components that were not subdivided in multi-resolution analysis. Furthermore, it can adaptively select appropriate frequency bands based on the characteristics of the analyzed signal, matching them to the signal's spectrum and thus improving time-frequency resolution.

[0077] The wavelet packet method can be used to denoise the EEG signal. The EEG artifacts in this invention mainly include electromyography (EMG), electrooculography (EOG), body movement, rotation, baseline drift, and power supply.

[0078] When subjects experience +Gz exposure inside the cabin, they inevitably become tense and make certain resistance movements, thus being significantly affected by electromyography (EMG), primarily in the 35.8-51Hz band. EOG is difficult to remove, as it may be mixed with multiple EEG frequency bands, easily causing loss of useful information during the removal process. Therefore, based on experimental experience, signals in the 0.5-1Hz band can be primarily removed. The AC power frequency is concentrated around 50Hz. Electrode fixation and baseline drift easily generate low-frequency slow waves below 0.8Hz and 0.2Hz. The interference generated by centrifuge rotation is strong and cannot be ignored. Based on the maximum achievable G value, centrifuge arm radius, and the conversion formula for centripetal acceleration, it can be calculated that the interference of centrifuge rotation on the signal is mainly below 0.5Hz. In addition, power supply, magnetic fields, and body movements also have varying degrees of influence. Considering all these factors, the lower limit of the filter is set at approximately 1Hz, and the upper limit at approximately 35Hz. Based on the sampling frequency and the characteristics of dynamic EEG signals, the Daubechies5 wavelet is selected to perform a 6-level decomposition on the original signal. Its minimum resolution can be estimated by the following formula, where f s The sampling frequency.

[0079]

[0080] Based on the filtered results, the wavelet packet method can effectively remove EMG, power frequency, and other high-frequency interference from EEG signals under +Gz influence. It also has a significant effect on slow wave interference in the low-frequency band, such as baseline drift and electrode interference, and the signal-to-noise ratio is greatly improved after denoising.

[0081] Step S102: In the first ear pulse signal, select the first sub-ear pulse signal with the largest rate of change of the first ear pulse signal.

[0082] The first ear pulse signal is any one of the ear pulse information sets.

[0083] Step S103: Based on the time period information corresponding to the first sub-ear pulse signal, filter the relevant EEG signal subset within the corresponding time period information from the EEG signals corresponding to the first EEG signal subset.

[0084] The first subset of EEG signals and the first auricular pulse signal were from the same subject.

[0085] For example, during the loading period, in order to select data with more obvious changes in EEG signals for correlation analysis, data with obvious changes in EEG signals are more conducive to analyzing the correlation of EEG signals.

[0086] In this embodiment of the invention, the G-load application time is 10 seconds. The ear pulse signals within 10 seconds are replayed and observed. Because the ear pulse and head-level blood pressure are consistent, the ear pulse is often used as an objective indicator for judging the +Gz endurance endpoint. A decrease in ear pulse amplitude indicates a decrease in head-level blood pressure; when head-level blood pressure drops to 0, the ear pulse flattens completely. Typically, ear pulse flattening for 1-2 seconds is considered a warning sign of blackouts or loss of consciousness. Therefore, a 1-second time period within the 10 seconds where the ear pulse signal shows a significant decrease is identified, and the corresponding EEG signal within this 1-second time period is selected for subsequent analysis.

[0087] Step S104: Based on the sampling frequency of the pre-acquired first EEG signal subset, each EEG signal in the relevant EEG signal subset is divided into multiple sub-EEG signal sets, and each sub-EEG signal set corresponds to a lead.

[0088] For example, after determining the first subset of EEG signals, each EEG signal in the 16 related EEG signal subsets included in the first subset of EEG signals is divided into multiple sub-EEG signals. The division criteria can be based on the sampling frequency of the EEG signals, as detailed below.

[0089] Step S105: Analyze each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the leads corresponding to the first sub-EEG signal set under the corresponding load.

[0090] The first sub-set of EEG signals is any one of the multiple sub-sets of EEG signals.

[0091] For example, autocorrelation analysis is performed on selected EEG signals within 1 second. The basic principle is based on the Speed ​​Borrell Transform (FFT). Autocorrelation can be explained by the following mathematical formula. Autocorrelation refers to the dependence between the instantaneous value of a signal at one moment and its instantaneous value at another moment; it is a time-domain description of a random signal. It represents the calculated value obtained by comparing the data sequence X of a certain electrode with its own data after time lags of k Δt and τ. This can be expressed by the formula:

[0092]

[0093] In this embodiment, the sampling rate is 128Hz, and correlation analysis is performed with 128 time delays, each with a delay of 0.0078, for a total time lag of 1 second. This means that 1 second of data is collected during the load application, and the filtered data is divided into multiple continuous segments, i.e., multiple blocks, each lasting 1 second. Therefore, each block of the acquired EEG data has 128 sampling points, and each block is recorded for 1 second. Autocorrelation analysis was performed on data from 27 participants across 5 subjects, with each of the 16 leads and each electrode within one block. For each subject's 16 leads, EEG autocorrelation analysis was performed under different G loads, yielding a series of autocorrelation values ​​C for each subject's 16 leads corresponding to different load values ​​G.

[0094] Step S106: Based on the autocorrelation information of the leads corresponding to all test subjects under the corresponding load, and the load corresponding to the autocorrelation information, determine the correlation information between the load and the lead.

[0095] For example, the correlation analysis between load and EEG autocorrelation is represented by the correlation coefficient r between them. The correlation coefficient is a very important statistical indicator of the degree of correlation between two phenomena, and is usually represented by r.

[0096] The Pearson correlation coefficient used (inputting G-load and EEG autocorrelation) is based on the product-moment method, or product-moment method, developed by British statistician Pearson. This method reflects the degree of correlation between two variables by multiplying the deviations of their respective means. The absolute value |r| reflects the strength of the correlation; the closer to 1, the stronger the correlation, and the closer to 0, the weaker the correlation. r > 0 indicates that the two correlated phenomena change in the same direction; if r < 0, it indicates a negative correlation, meaning the two correlated phenomena change in opposite directions. r = 0 indicates no correlation. Calculation formula:

[0097]

[0098] Where n represents the number of sequences and sums.

[0099] In a preferred embodiment, step 106 specifically includes:

[0100] Based on the autocorrelation information and the corresponding load, determine the correlation coefficient between the load and the lead;

[0101] Based on the correlation coefficient of each lead corresponding to different loads, the correlation information between the load and the lead is determined.

[0102] For example, based on the autocorrelation analysis results of each EEG lead, different loads and their corresponding autocorrelation values ​​can be obtained. Different G loads and their corresponding autocorrelation values ​​are treated as two variables. Correlation coefficients between different loads and their corresponding autocorrelation values ​​are calculated for each of the 16 EEG leads. Data is selected from 1 second (one block) of data where the corresponding ear pulse shows significant changes during the load application period. To observe overall trend changes and reduce the randomness of EEG changes, the autocorrelation value is taken as the mean of one block. Based on different loads G and their corresponding autocorrelation values ​​C, the correlation coefficient r between different G loads and their corresponding EEG autocorrelation values ​​C is calculated for each subject's 16 leads. A statistical table of the correlation coefficients between G loads and EEG autocorrelation is created. Based on the table, a topographic map distribution of the correlation coefficients between G loads and EEG autocorrelation is then generated, more clearly showing the magnitude of the correlation between G loads and EEG autocorrelation changes in different parts of the brain, and observing the pattern of changes in the 16 EEG leads with load changes. Correlation topographic maps are generated for each of the five subjects. The correlation coefficient topographic map distribution for one subject is shown below. Figure 2 As shown. Based on the distribution of correlation coefficients on the topographic map, the leads are then sorted and weighted.

[0103] Step S107: Based on the associated information and preset conditions, determine the leads that meet the conditions.

[0104] For example, the preset conditions could be to select the leads with the highest correlation information from the correlation information as the final leads, and then sort them according to the positional relationship between the corresponding leads.

[0105] In a preferred embodiment, such as Figure 3 As shown, step 107 specifically includes:

[0106] Step 1071: Sort the leads according to the correlation information to obtain the lead sequence;

[0107] Step 1072: Assign weight information to the leads in the lead sequence according to their ranking.

[0108] For example, after obtaining the corresponding correlation topography map for each subject in the above embodiments, the 16 leads are ranked according to the absolute value of the correlation coefficient. The ranking can be based on the absolute value of the correlation coefficient because, after obtaining the correlation coefficient, it is found that the values ​​are both positive and negative, indicating that under the influence of +Gz, EEG changes and load changes show inconsistency. Since most leads are negatively correlated with load changes, some leads are positively correlated. The correlation coefficient between the same lead and load shows roughly consistent positive and negative values ​​for different subjects. Therefore, the ranking focuses on the closeness of the correlation between the lead and load changes, i.e., the sensitivity of the lead to load changes, and the magnitude of the correlation coefficient relative to load changes, rather than focusing more on the positive or negative value of the correlation coefficient. This eliminates the instability and specificity of lead changes. Therefore, ranking based on the absolute value of the correlation coefficient, rather than the value itself, is more reasonable.

[0109] In a preferred embodiment, step 1072 specifically includes:

[0110] Step a1: Assign initial weight information to the leads in the lead sequence according to their ranking.

[0111] Step a2: Determine the final weight information of the leads based on the initial weight information of all subjects' leads.

[0112] For example, after sorting the 16 leads of each subject, a weight is assigned to each lead according to the order of the sorting. The higher the ranking, the higher the weight assigned, and the lower the ranking, the lower the weight assigned. The first-ranked lead is assigned the highest weight of 16 points, the second-ranked lead is assigned 15 points, and so on, with the sixteenth-ranked lead assigned the lowest weight of 1 point. Figure 4 This is a graph showing the initial ranking of leads and the corresponding weights. Subsequently, the average weights for each lead across all subjects are calculated.

[0113] The ranking of all leads was based on the average of the weighted scores of each lead, rather than the correlation coefficients of the leads themselves. This is because EEG signals are weak and easily interfered with, especially under load, and are more susceptible to various factors such as electromyography (EMG), electrooculography (EOG), body movement, rotation, and magnetic fields. Although the experimental design requires subjects to exert minimal force during G-load and keep their backs pressed against the chair back to maintain their original posture as much as possible to reduce the impact of body swaying, subjects often involuntarily exert some force to prevent fainting during G-load, making the generation of EMG unavoidable. Considering that speaking, blinking, and body movement also have a significant impact on the signal and can easily create artifacts, some of which are difficult to remove, the experimental design, although requiring the experimenter to avoid talking to the subjects as much as possible and requiring the subjects to minimize blinking and other facial movements during G-load, still has some influence on the EEG signal.

[0114] Therefore, EEG signals exhibit instability and specificity. If the correlation coefficient between EEG signals and load changes is used for final ranking, some leads may show strong correlations with load changes, resulting in large correlation coefficients, but inconsistent positive and negative values. In some subjects, the correlation may be opposite to that of others. Directly averaging these correlation coefficients will fail to reflect the strong correlation between a lead and load changes, i.e., it will not reflect the lead's sensitivity to load changes. However, assigning different weights to leads based on their ranking and then averaging these weights considers the relative magnitude of the correlation between that lead and load changes among all leads—that is, the relative position of the correlation between that lead and load changes. Averaging the weights better eliminates the individual instability and specificity of EEG signals and better represents the relative magnitude of the correlation coefficient between a particular lead and load changes within the entire network. Therefore, using average weights instead of average correlation coefficients for final ranking is more scientific and reasonable.

[0115] Step 1073: Based on the weight information and preset conditions, filter out leads that meet the conditions.

[0116] For example, based on the final result after lead sorting, 8 leads, 6 leads, or 4 leads can be selected as the simplified selection result according to the simplification requirements.

[0117] According to this method, following its steps and procedures, the 16 leads of EEG acquired under centrifugation + Gz conditions were sorted and simplified. The final sorting result, in order of ranking, is T5, C3, FP1, F3, T4, F4, FP2, F7, T6, O2, P3, F8, P4, T3, C4, and O1. If simplified to 6 leads, the selected 6 leads are T5, C3, FP1, F3, T4, and F4.

[0118] In the above embodiments, based on the preset conditions and sorting results, the final leads are obtained. According to the obtained autocorrelation values, which are low, medium, and high correlation, the autocorrelation of each EEG lead is relatively strong under the +Gz load, that is, it shows medium to high correlation. This indicates that the changes in EEG under the load have a certain stability, or in other words, relatively strong stability.

[0119] 1. Based on the correlation coefficient values, most are low-level correlations, indicating that the changes in EEG are affected by the load and have a certain correlation with the load, but the correlation is not strong. However, some leads show a strong correlation, with the maximum absolute value reaching 0.8620, showing the potential for lead optimization.

[0120] 2. For all subjects, most leads showed a consistent negative correlation, but some leads showed inconsistent positive and negative correlations, indicating that the regularity of their changes was not very strong. The reason for this is that EEG data is easily affected by various signals under dynamic conditions. Under centrifuge conditions (+Gz), the equipment undergoes high-speed rotation and triaxial motion, making EEG signals more susceptible to interference from electromyography (EMG), electrocardiography (ECG), electrooculography (EOG), body movement, baseline drift, electrode interference, power supply, rotation, and other factors. The amplitude of some artifact signals can be several times or even tens of times greater than that of the EEG signal, thus increasing the randomness of the EEG signal and affecting the correlation analysis of the EEG signal, reducing the correlation between EEG and loading. To reduce the impact of the randomness and instability of EEG changes, the autocorrelation value was taken as the mean over one block (1 second).

[0121] The selected leads were mostly on the left side. The reasons are as follows:

[0122] The first reason is that during the +Gz action, due to centrifugal force, the person is in a semi-reclining position with the left arm down. Affected by systemic circulation such as blood circulation, the right side of the body is more affected by ischemia than the left side. In other words, the right side of the body is more affected by G endurance than the left side. This greatly reduces the complexity and activity of the right-side EEG, making it appear weaker. Therefore, the left-side leads show relatively stronger changes, and the correlation coefficient between the EEG and the load in the left-side leads is relatively larger. That is, overall, the autocorrelation changes of the left-side EEG and the load are more strongly correlated than those on the right side.

[0123] The second reason is related to the inherent characteristics of EEG variations. Multiple studies have shown that left-side leads exhibit more pronounced variations than right-side leads. One study used the common spatial pattern algorithm (CSP) to select EEG leads and obtain classification accuracy for three types of motor imagery (left hand, right hand, and foot) from three subjects. The results showed that the selected leads were mostly concentrated in the motor cortex region. Whether using CSP-2 norm selection or manual selection, the optimized leads were predominantly left-side leads. Another study explored EEG features closely related to emotion and optimized the minimum set of leads. By correcting the F.Score algorithm and combining F.Score feature extraction with SVM, an optimized set of leads was obtained, including FT7, T7, FC4, TP10, O1, and FPl. It is evident that the optimized leads also predominated on the left side, and there was overlap between the selected leads and the top-ranked leads in this study. Other researchers have conducted studies on EEG lead selection algorithms and classification for patients with depression. They collected EEG data from 16 adolescent patients with depression and 16 healthy individuals under resting conditions with eyes closed for 4 minutes. For 64 leads from 32 participants, they extracted time-domain and frequency-domain features using Spectral Asymmetry Index (SASI) and Detrended Fluctuation Analysis (DFA) algorithms, constructing 32×64-dimensional feature matrices for SAIS and DFA respectively. A genetic algorithm was then used to select, crossover, and mutate features for each individual and each lead, ultimately selecting the lead labels corresponding to the features of the offspring. The distribution map of lead selection shows that left-side leads were the most frequently selected. The aforementioned literature studied the EEG characteristics and lead classification of different subjects, including those with motor imagery, emotional triggering, and depression. This study optimized the leads of EEG under high load, and the selected leads were mostly on the left side, which is consistent with the results of previous related literature. The similarity of the lead selection results under different conditions indicates that it is largely related to the brain's own change characteristics. At the same time, related literature has verified the results of this study.

[0124] The basic basis of this embodiment is the change in the correlation coefficient between the G-load and the EEG autocorrelation. Based on the topographic distribution of the correlation coefficients of the 16 leads, the leads under the load are sorted and simplified to select the required leads. According to the results, the leads with more significant changes are T5, C3, and FP1. The maximum value is -0.8620 at T5. The leads with significant changes are mostly distributed in the left anterior and right posterior sides of the brain, close to areas such as the frontal and temporal regions.

[0125] This application embodiment is the first to simplify and optimize the selection of leads for EEG signals under the action of +Gz.

[0126] First, the autocorrelation coefficients of the 16 EEG leads under different loads were calculated, and autocorrelation analysis was performed. Then, based on this, the correlation coefficient between load and EEG autocorrelation was calculated. By analyzing the topographic distribution of the correlation coefficients between load and EEG autocorrelation, the correlation between centrifuge G load and EEG changes was studied. Furthermore, the EEG leads under +Gz load were ranked and simplified, selecting leads with greater contribution and strong representativeness. This is the first study to rank and select leads under the special +Gz environment using a specific method, and it is also the first time that correlation coefficients have been applied to the study of the correlation between endurance and EEG changes. The research results provide a new method and basis for future studies on lead ranking and simplification under the special +Gz environment of centrifuges. The research results have significant application value for improving EEG acquisition and measurement techniques under the special +Gz environment and for studying the characteristic changes of EEG under this environment. They also provide a reference for similar studies related to EEG under +Gz load, and have significant reference value.

[0127] The selection of autocorrelation values ​​was based on the autocorrelation analysis results of each EEG lead. Different G-loads and their corresponding autocorresponding values ​​were treated as two variables, and the correlation coefficients between different loads and their corresponding autocorrelation values ​​were calculated for each of the 16 EEG leads. When selecting autocorrelation values ​​within a single block—that is, for the 1 second period during which the corresponding ear pulse showed significant changes (i.e., within a single block)—to observe overall trend changes and reduce the randomness and uncertainty of EEG under G-load, the autocorrelation values ​​were not taken from the envelope (maximum value) of the data within a single block, but rather from the mean value within that block. This reduces the impact of the randomness and instability of EEG on the correlation analysis between the load and the EEG to some extent.

[0128] The design of a secondary ranking based on the correlation coefficient between EEG and payload: When ranking and selecting leads, instead of directly ranking them based on the correlation coefficient values, a secondary ranking is performed. First, the leads are initially ranked based on the absolute values ​​of their correlation coefficients. Then, a secondary ranking is performed based on the average weight values ​​of the corresponding ranks of each lead after the initial ranking. This significantly reduces the impact of randomness, weak signals, and poor anti-interference capabilities of EEG on the correlation coefficient, thereby reducing the influence of the correlation coefficient on the final lead ranking.

[0129] This embodiment also provides an EEG lead selection device for implementing the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware implementations, or a combination of software and hardware, are also possible and contemplated.

[0130] This embodiment provides an EEG lead selection device, such as... Figure 5 As shown, it includes:

[0131] The information acquisition module 501 is used to acquire at least one ear pulse information set and a first EEG signal set of a subject during different load training processes; the first EEG signal set includes multiple EEG signal subsets, and each EEG signal subset in the first EEG signal set is acquired by multiple leads corresponding to the subject, and the number of EEG signal subsets is the same as the number of leads.

[0132] The first filtering sub-information module 502 is used to filter the first sub-ear pulse signal with the largest change rate in the first ear pulse signal from the first ear pulse signal. The first ear pulse signal is any one in the ear pulse information set.

[0133] The second screening sub-information module 503 is used to filter the relevant EEG signal subset within the corresponding time period information from the EEG signals corresponding to the first EEG signal subset based on the time period information corresponding to the first sub-ear pulse signal. The first EEG signal subset and the first ear pulse signal are the same subject.

[0134] The sub-signal division module 504 is used to divide each EEG signal in the relevant EEG signal subset into multiple sub-EEG signal sets according to the sampling frequency of the pre-acquired first EEG signal subset, and each sub-EEG signal set corresponds to a lead;

[0135] The correlation analysis module 505 is used to analyze each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the lead corresponding to the first sub-EEG signal set under the corresponding load. The first sub-EEG signal set is any one of multiple sub-EEG signal sets.

[0136] The relationship determination module 506 is used to determine the relationship information between the load and the lead based on the autocorrelation information of the leads corresponding to all test subjects under the corresponding loads and the loads corresponding to the autocorrelation information.

[0137] The lead determination module 507 is used to determine the leads that meet the conditions based on the associated information and preset conditions.

[0138] In some optional implementations, the information acquisition module specifically includes:

[0139] The signal acquisition unit is used to acquire the subject's second set of electroencephalogram (EEG) signals.

[0140] An interference processing unit is used to process the second set of EEG signals to obtain a third set of EEG signals.

[0141] The amplification processing unit is used to amplify the third set of EEG signals to obtain the fourth set of EEG signals;

[0142] The filtering unit is used to perform noise reduction filtering on the fourth EEG signal set to obtain the first EEG signal set.

[0143] In some optional implementations, the relationship determination module specifically includes:

[0144] The correlation coefficient determination unit is used to determine the correlation coefficient between the load and the lead based on the autorelated information and the load corresponding to the autorelated information.

[0145] The relationship unit is used to determine the correlation information between the load and the lead based on the correlation coefficient of each lead corresponding to different loads.

[0146] In some alternative implementations, the lead module is determined, specifically including:

[0147] The sequence unit is determined and used to sort the leads according to the correlation information to obtain the lead sequence;

[0148] The weighting determination unit is used to assign weight information to the leads in the lead sequence according to their ranking.

[0149] The filtering unit is used to filter leads that meet the criteria based on weight information and preset conditions.

[0150] In some optional implementations, the weight determination unit specifically includes:

[0151] The initial weight determination subunit is used to assign initial weight information to the leads in the lead sequence according to their ranking in the lead sequence;

[0152] The weighting subunit is used to determine the final weighting information of the leads based on the initial weighting information of all subjects.

[0153] In some alternative implementations, the leads are sorted according to the absolute value of the associated information to obtain a lead sequence.

[0154] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0155] In this embodiment, the EEG lead selection device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0156] This invention also provides a computer device having the above-described features. Figure 5 The EEG lead selection device shown.

[0157] Please see Figure 6 , Figure 6 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 6 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 6 Take a processor 10 as an example.

[0158] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0159] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0160] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0161] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0162] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0163] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0164] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for selecting electroencephalogram (EEG) leads, characterized in that, The method includes: Acquire at least one subject's auricular pulse information set and a first EEG signal set during different load training processes; the first EEG signal set includes multiple EEG signal subsets, each EEG signal subset in the first EEG signal set is acquired by multiple leads corresponding to the subject, and the number of EEG signal subsets is the same as the number of leads; In the first ear pulse signal, the first sub-ear pulse signal with the largest rate of change of the first ear pulse signal is selected, and the first ear pulse signal is any one of the ear pulse information sets; Based on the time period information corresponding to the first sub-ear pulse signal, a relevant EEG signal subset within the corresponding time period information is selected from the EEG signals corresponding to the first EEG signal subset. The first EEG signal subset and the first ear pulse signal are from the same subject. Based on the sampling frequency of the first subset of EEG signals obtained in advance, each EEG signal in the relevant subset of EEG signals is divided into multiple sub-sets of EEG signals, and each sub-set of EEG signals corresponds to a lead. Analyze each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the lead corresponding to the first sub-EEG signal set under the corresponding load. The first sub-EEG signal set is any one of multiple sub-EEG signal sets. Based on the autocorrelation information of the leads corresponding to all test subjects under the corresponding load, and the load corresponding to the autocorrelation information, the correlation information between the load and the lead is determined; Based on the associated information and preset conditions, determine the leads that meet the conditions; The step of determining the correlation information between the load and the lead based on the autocorrelation information of all subjects' leads under corresponding loads and the loads corresponding to the autocorrelation information specifically includes: Based on the autocorrelation information and the load corresponding to the autocorrelation information, determine the correlation coefficient between the load and the lead; Based on the correlation coefficient of each lead corresponding to a different load, the association information between the load and the lead is determined; The step of determining the leads that meet the conditions based on the associated information and preset conditions specifically includes: The leads are sorted according to the association information to obtain a lead sequence; Based on the ranking of the leads in the lead sequence, weight information is assigned to the leads in the lead sequence; Based on the weight information and preset conditions, leads that meet the conditions are selected.

2. The method according to claim 1, characterized in that, Acquiring the first set of EEG signals specifically includes: Obtain the second set of EEG signals from the subject; The second set of EEG signals is subjected to interference processing to obtain the third set of EEG signals; The third set of EEG signals is amplified to obtain the fourth set of EEG signals; The fourth set of EEG signals is subjected to noise reduction filtering to obtain the first set of EEG signals.

3. The method according to claim 1, characterized in that, The step of assigning weight information to the leads in the lead sequence based on their ranking specifically includes: Based on the ranking of the leads in the lead sequence, initial weight information is assigned to the leads in the lead sequence; The final weight information of the leads is determined based on the initial weight information of all subjects' leads.

4. The method according to claim 1, characterized in that, The leads are sorted according to the absolute value of the associated information to obtain a lead sequence.

5. A brainwave lead selection device, characterized in that, The device includes: An information acquisition module is used to acquire at least one subject's auricular pulse information set and a first EEG signal set during different load training processes; the first EEG signal set includes multiple EEG signal subsets, each EEG signal subset in the first EEG signal set is acquired by multiple leads corresponding to the subject, and the number of EEG signal subsets is the same as the number of leads; The first filtering sub-information module is used to filter the first sub-ear pulse signal with the largest change rate of the first ear pulse signal from the first ear pulse signal, wherein the first ear pulse signal is any one of the ear pulse information sets; The second filtering sub-information module is used to filter the relevant EEG signal subset within the corresponding time period information from the EEG signals corresponding to the first EEG signal subset based on the time period information corresponding to the first sub-ear pulse signal. The first EEG signal subset and the first ear pulse signal are from the same subject. The sub-signal segmentation module is used to divide each EEG signal in the relevant EEG signal subset into multiple sub-EEG signal sets according to the sampling frequency of the pre-acquired first EEG signal subset, with each sub-EEG signal set corresponding to a lead; The correlation analysis module is used to analyze each sub-EEG signal in the first sub-EEG signal set to obtain the autocorrelation information of the lead corresponding to the first sub-EEG signal set under the corresponding load. The first sub-EEG signal set is any one of multiple sub-EEG signal sets. The relationship determination module is used to determine the association information between the load and the lead based on the autocorrelation information of the leads corresponding to all test subjects under the corresponding loads and the loads corresponding to the autocorrelation information. The lead determination module is used to determine the leads that meet the conditions based on the association information and preset conditions; The step of determining the correlation information between the load and the lead based on the autocorrelation information of all subjects' leads under corresponding loads and the loads corresponding to the autocorrelation information specifically includes: Based on the autocorrelation information and the load corresponding to the autocorrelation information, determine the correlation coefficient between the load and the lead; Based on the correlation coefficient of each lead corresponding to a different load, the association information between the load and the lead is determined; The step of determining the leads that meet the conditions based on the associated information and preset conditions specifically includes: The leads are sorted according to the association information to obtain a lead sequence; Based on the ranking of the leads in the lead sequence, weight information is assigned to the leads in the lead sequence; Based on the weight information and preset conditions, leads that meet the conditions are selected.

6. The apparatus according to claim 5, characterized in that, The information acquisition module specifically includes: A signal acquisition unit is used to acquire a second set of electroencephalogram (EEG) signals from the subject. An interference processing unit is used to perform interference processing on the second set of EEG signals to obtain a third set of EEG signals. An amplification processing unit is used to amplify the third set of EEG signals to obtain a fourth set of EEG signals; The filtering unit is used to perform noise reduction filtering on the fourth EEG signal set to obtain the first EEG signal set.

7. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the EEG lead selection method according to any one of claims 1 to 4 by executing the computer instructions.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the EEG lead selection method according to any one of claims 1 to 4.

Images