A brain state analysis device based on electroencephalogram parameters

Abstract

The application provides a brain state analysis device based on electroencephalogram parameters, comprising: a preprocessing module for obtaining electroencephalogram parameters, screening available parameters and obtaining corresponding characteristic parameters; a parameter purification module for matching a corresponding communication channel for each characteristic parameter and performing parameter purification to obtain electroencephalogram typical parameters; a parameter conversion module for obtaining a power spectrum diagram based on the electroencephalogram typical parameters; a power matching module for obtaining corresponding specific microstates based on the time-space distribution fluctuations of the power spectrum diagram; and a state analysis module for comprehensively analyzing all specific microstates to obtain a brain state. Through screening and feature extraction of electroencephalogram parameters, characteristic parameters are obtained, corresponding communication channels are matched, electroencephalogram typical parameters are analyzed, all microstates are comprehensively analyzed, detailed analysis of the brain state is obtained, brain disorders are better mastered, and the accuracy of brain disorder diagnosis is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a brain state analysis device based on electroencephalogram parameters. BACKGROUND

[0002] At present, with the development of science and technology, science and technology have armed medicine and created instruments, which not only subvert the traditional concept of medicine, but also completely change the practice mode of medicine. With the deep cross-fusion of medicine and modern information technology, material technology and other technologies, the development of medical technology has been greatly promoted, and the achievements of modern medical technology are applied to the clinic more and more quickly, which is revolutionizing the disease diagnosis and treatment mode. The analysis of brain diseases in modern medical technology is the most important, and the analysis and processing of electroencephalogram information is the key link of brain treatment.

[0003] Therefore, the present application provides a brain state analysis device based on electroencephalogram parameters. SUMMARY

[0004] The present application provides a brain state analysis device based on electroencephalogram parameters, which obtains characteristic parameters by screening and feature extraction of electroencephalogram parameters, matches corresponding communication channels, obtains electroencephalogram typical parameters and performs power spectrum analysis, comprehensively analyzes all micro-states, and obtains detailed analysis of brain state, further grasps brain diseases, and improves the accuracy of electroencephalogram parameter analysis and the accuracy of brain disease diagnosis.

[0005] The present application provides a brain state analysis device based on electroencephalogram parameters, which comprises:

[0006] The preprocessing module acquires electroencephalogram parameters, screens available parameters and obtains corresponding characteristic parameters;

[0007] The parameter purification module matches corresponding communication channels for each characteristic parameter and performs parameter purification to obtain electroencephalogram typical parameters;

[0008] The parameter conversion module obtains a power spectrum diagram based on the electroencephalogram typical parameters;

[0009] The power matching module obtains corresponding specific micro-states based on the temporal and spatial distribution fluctuations of the power spectrum diagram;

[0010] The state analysis module comprehensively analyzes all specific micro-states to obtain brain state.

[0011] Preferably, a brain state analysis device based on electroencephalogram parameters comprises:

[0012] The initial diagnosis unit is used to determine an electroencephalogram measurement scheme for the patient according to the patient's disease description;

[0013] An electroencephalogram measurement unit is configured to place the matched measurement devices at the target positions of the patient according to the electroencephalogram measurement scheme to obtain the electroencephalogram measurement results of each measurement device;

[0014] A region determination unit is configured to divide the electroencephalogram measurement results into brain regions and obtain available brain regions by comparing and analyzing the standard parameters of the electroencephalogram regions.

[0015] A screening parameter unit is configured to screen all the electroencephalogram parameters based on the available brain regions to obtain available parameters.

[0016] Preferably, the brain state analysis device based on electroencephalogram parameters further comprises:

[0017] A feature information analysis unit is configured to obtain a plurality of feature information based on all the available parameters.

[0018] A sub-feature encoding unit is configured to decompose each feature information to obtain a plurality of sub-feature information and encode each sub-feature information to obtain a sub-feature code.

[0019] A sub-feature vector conversion unit is configured to map each sub-feature code to obtain a sub-feature vector, and further obtain a sub-feature vector set corresponding to each feature information.

[0020] A feature degree calculation unit is configured to calculate the current feature degree of the corresponding feature information based on the sub-feature vector set.

[0021] A feature degree reservation unit is configured to reserve the corresponding feature information if the current feature degree is greater than a preset feature degree.

[0022] A feature parameter construction unit is configured to construct a feature parameter corresponding to the available parameters based on all the reserved feature information.

[0023] Preferably, the feature degree calculation unit comprises:

[0024] ; wherein, represents the current feature degree of the corresponding feature information; represents the number of sub-feature vectors in the sub-feature vector set; represents the norm of the th sub-feature vector in the sub-feature vector set; represents the average value of the norms of all the sub-feature vectors in the sub-feature vector set; represents the feature weight of the th sub-feature vector; represents a preset relative coefficient corresponding to feature information; represents an average vector of all sub-feature vectors in a sub-feature vector set; represents a sub-feature vector in a sub-feature vector set; represents a sub-feature vector in a sub-feature vector set; represents a module of a vector represents a larger one in represents obtaining represents obtaining represents a difference absolute value of represents a difference absolute value of represents a difference absolute value of

[0025] Preferably, a brain state analysis device based on electroencephalogram parameters, the parameter purification module comprises:

[0026] A detection signal acquisition unit is configured to input feature parameters corresponding to available parameters to a detection communication channel to obtain a detection signal.

[0027] A detection signal spectrum acquisition unit is configured to obtain a detection signal spectrum of the detection signal based on a filter display.

[0028] A detection frequency band acquisition unit is configured to obtain corresponding feature frequency band information based on the detection signal spectrum.

[0029] A parameter-characteristics conversion unit is configured to obtain loss characteristics corresponding to the feature parameters based on a parameter-characteristics analysis model.

[0030] A channel matching unit is configured to match a corresponding first communication channel from a channel database based on the loss characteristics and the feature frequency band information.

[0031] A feature signal acquisition unit is configured to input feature parameters corresponding to available parameters to the first communication channel to obtain a feature signal.

[0032] A feature signal spectrum acquisition unit is configured to obtain a feature signal spectrum of the feature signal.

[0033] An effective wave impurity removal unit is configured to obtain effective waves in the feature signal spectrum and remove aperiodic signals in the effective waves.

[0034] An effective wave region division unit is configured to divide the effective waves into a high-intensity signal region, a medium-intensity signal region, and a low-intensity signal region based on signal intensity in the effective waves.

[0035] A signal set acquisition unit is configured to obtain corresponding high-frequency signal sets, medium-frequency signal sets, and low-frequency signal sets based on the high-intensity signal region, the medium-intensity signal region, and the low-intensity signal region.

[0036] a low-frequency signal attenuation analysis unit configured to obtain a low-frequency signal attenuation trend based on the set of low-frequency signals;

[0037] a non-ideal attenuation point acquisition unit configured to obtain a non-ideal attenuation point with the largest attenuation trend based on the low-frequency signal attenuation trend;

[0038] an attenuation point delay unit configured to introduce a preset modulation parameter at the non-ideal attenuation point to delay the non-ideal attenuation point to an ideal attenuation point, and obtain a set of ideal low-frequency signals;

[0039] a signal adjustment unit configured to adjust an effective wave based on the set of ideal low-frequency signals to obtain a set of ideal high-frequency signals and a set of ideal intermediate-frequency signals;

[0040] a quantity difference calculation unit configured to sequentially calculate a first signal quantity difference between the set of ideal high-frequency signals and the set of high-frequency signals and a second signal quantity difference between the set of ideal intermediate-frequency signals and the set of intermediate-frequency signals;

[0041] an electroencephalogram typical parameter acquisition unit configured to decode and parameterize a characteristic signal obtained based on the first communication channel to obtain electroencephalogram typical parameters if the first signal quantity difference and the second signal quantity difference are within a preset adjustment difference.

[0042] Preferably, a brain state analysis device based on electroencephalogram parameters, the parameter conversion module comprises:

[0043] an analog power conversion unit configured to input the electroencephalogram typical parameters into a parameter-power conversion model to obtain an electroencephalogram analog power;

[0044] a power spectrum diagram acquisition unit configured to obtain a power spectrum diagram based on the electroencephalogram analog power.

[0045] Preferably, a brain state analysis device based on electroencephalogram parameters, the analog power conversion unit comprises:

[0046] a training sample set construction block configured to obtain all electroencephalogram typical parameters and corresponding feature parameter names to construct a training sample set;

[0047] a model construction block configured to construct a parameter-power conversion model, wherein the parameter-power conversion model comprises a name classification layer, an electroencephalogram analog layer, and an output power layer;

[0048] a model training block configured to input the training sample set into the parameter-power conversion model to obtain the electroencephalogram analog power.

[0049] Preferably, a brain state analysis device based on electroencephalogram parameters, the power matching module comprises:

[0050] The category acquisition unit is used to obtain the corresponding microstate category based on the feature category of the feature parameter corresponding to the power spectrum diagram;

[0051] A power spectrum analysis unit is used to obtain the spatiotemporal distribution fluctuations based on the power spectrum diagram;

[0052] The fluctuation analysis unit is used to obtain the fluctuation trend, the fluctuation difference between the maximum and minimum peak values ​​of the spatiotemporal distribution fluctuations, the average peak value and the average valley value of the spatiotemporal distribution fluctuations based on the spatiotemporal distribution fluctuations.

[0053] The index acquisition unit is used to acquire a micro-state parameter lookup table to obtain a micro-state index corresponding to the fluctuation trend, fluctuation difference, average peak value, and average valley value.

[0054] Preferably, a brain state analysis device based on electroencephalogram (EEG) parameters includes a state analysis module comprising:

[0055] The symptom analysis unit is used to extract symptom features from symptom descriptions and determine the probability of each symptom feature appearing in association with all symptom features, as well as the symptom uniqueness.

[0056] ;

[0057] ;in, Indicates the first Individual disease characteristics With all remaining symptoms The association function between them; Indicate the characteristics of each disease With all remaining symptoms The summation function after association; Intersection symbol; The union symbol represents the set of sets. Indicates the first The probability of association between characteristics of a disease; Indicates the first The uniqueness of each disease characteristic; Indicates the first The history of each symptom characteristic is given a unique value;

[0058] The weight acquisition unit is used to set a disease identifier for each disease feature based on the probability of association occurrence and the uniqueness of the disease, and to obtain the weight of each category micro-state from the weight setting database. The disease identifier is related to the probability of association occurrence of the disease, the uniqueness of the disease, and the category matched by the disease.

[0059] The brain state acquisition unit is configured to input all microstate categories and corresponding microstate indexes and weights of microstates into a microstate synthesis model to obtain a brain state.

[0060] Additional features and advantages of the application will be set forth in the description that follows, and in part will be apparent from the description, or can be learned by practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

[0061] The technical solutions of the present application are described in further detail below with reference to the accompanying drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0062] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0063] Figure 1 Fig. 1 is a structural block diagram of a brain state analysis device based on electroencephalogram parameters according to an embodiment of the present application. DETAILED DESCRIPTION

[0064] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, and it should be understood that the preferred embodiments described herein are merely intended to illustrate and explain the present application, and are not intended to limit the present application. EMBODIMENT

[0065] The embodiment of the present application provides a brain state analysis device based on electroencephalogram parameters, as shown in Fig. 1, which comprises: Figure 1

[0066] The preprocessing module is configured to acquire electroencephalogram parameters, screen available parameters, and obtain corresponding feature parameters.

[0067] The parameter purification module is configured to match each feature parameter to a corresponding communication channel, and purify the parameters to obtain electroencephalogram typical parameters.

[0068] The parameter conversion module is configured to obtain a power spectrum graph based on the electroencephalogram typical parameters.

[0069] The power matching module is configured to obtain corresponding specific microstates based on the temporal and spatial distribution fluctuations of the power spectrum graph.

[0070] The state analysis module is configured to comprehensively analyze all specific microstates to obtain a brain state.

[0071] In this embodiment, the electroencephalogram parameters refer to information parameters obtained after the overall activity information of brain nerve cells is converted into electrical signals by a medical device, so as to master the state of the brain, which includes ion exchange and metabolism of the overall activity of brain nerve cells.​

[0072] In this embodiment, the available parameter refers to a parameter selected from all electroencephalogram parameters according to a description of a patient's disease, so as to achieve the purpose of narrowing the analysis range.

[0073] In this embodiment, the characteristic parameter refers to a parameter capable of representing the characteristics of the available parameter obtained by analyzing each sub-characteristic information in the characteristic information obtained by decomposing the available parameter.

[0074] In this embodiment, the electroencephalogram typical parameter refers to a parameter capable of representing the electroencephalogram parameter obtained by impurity removal through the decomposition and analysis of the characteristic parameter.

[0075] In this embodiment, the power spectrum diagram refers to a diagram of the relationship between the power frequency and the energy of each electroencephalogram typical parameter corresponding to the simulation of the electroencephalogram state obtained by simulating the electroencephalogram typical parameter.

[0076] In this embodiment, the spatiotemporal distribution fluctuation refers to a fluctuation diagram of the power frequency and the energy corresponding to each electroencephalogram typical parameter of the simulation of the electroencephalogram state in time and space, so as to achieve the purpose of analyzing the current brain state.

[0077] In this embodiment, the specific microstate refers to a state of a small part of the brain obtained by matching the fluctuation trend of the spatiotemporal distribution fluctuation, the fluctuation difference between the maximum peak value and the minimum peak value of the spatiotemporal distribution fluctuation, the average peak value and the average valley value of the spatiotemporal distribution fluctuation, and the microstate parameter table.

[0078] The working principle and beneficial effects of the above technical solution are as follows: through the screening and feature extraction of the electroencephalogram parameters, the characteristic parameters are obtained, the corresponding communication channels are matched, the electroencephalogram typical parameters are obtained, the power spectrum analysis is performed, all microstates are comprehensively analyzed, the detailed analysis of the brain state is obtained, the brain disease is further mastered, the accuracy of the electroencephalogram parameter analysis is improved, and the accuracy of the brain disease diagnosis is enhanced. Embodiment

[0079] The embodiment provides a brain state analysis device based on electroencephalogram parameters.

[0080] The initial diagnosis unit is configured to determine an electroencephalogram measurement scheme of the patient according to a description of a disease of the patient.

[0081] The electroencephalogram measurement unit is configured to place the matched measurement device at a target part of the patient to perform electroencephalogram measurement according to the electroencephalogram measurement scheme, and obtain an electroencephalogram measurement result of each measurement device.

[0082] A region determining unit is configured to divide the brain region according to the EEG measurement result, and obtain available brain regions by comparing and analyzing the EEG region standard parameters.

[0083] A screening parameter unit is configured to screen all EEG parameters based on the available brain regions, and obtain available parameters.

[0084] In this embodiment, the EEG measurement scheme refers to a scheme obtained by initially performing EEG measurement according to the patient's disease description.

[0085] In this embodiment, the brain region division refers to a region divided according to the physiological characteristics of the brain, and the EEG parameters of each region can independently or in combination represent a physiological state of the brain.

[0086] In this embodiment, the EEG region standard parameter refers to a parameter of the brain region obtained according to the brain region division, which can represent the parameter of the brain region in the lowest standard state of the brain physiology. When the parameter in the EEG measurement result does not conform to the EEG region standard parameter, it is determined that the measurement is incorrect, and the parameter is not available.

[0087] In this embodiment, the available brain region refers to a brain region corresponding to the EEG measurement result conforming to the EEG region standard parameter, and the parameter available brain region.

[0088] The working principle and beneficial effects of the above technical solution are as follows: the EEG measurement scheme is given according to the patient's disease description, and the available parameters of the available brain region are obtained by dividing the brain region and verifying the result, thereby narrowing the range of EEG measurement, reducing the interference of the unavailable parameters, and improving the accuracy and efficiency of the EEG analysis. Embodiment

[0089] The embodiment provides a brain state analysis device based on EEG parameters, and the preprocessing module further comprises:

[0090] A feature information analysis unit is configured to obtain a plurality of feature information corresponding to all available parameters.

[0091] A sub-feature encoding unit is configured to decompose each feature information, obtain a plurality of sub-feature information, and encode each sub-feature information to obtain a sub-feature code.

[0092] A sub-feature vector conversion unit is configured to map each sub-feature code to obtain a sub-feature vector, and further obtain a sub-feature vector set corresponding to each feature information.

[0093] A feature degree calculation unit is configured to calculate the current feature degree of the corresponding feature information based on the sub-feature vector set.

[0094] a feature degree reservation unit, configured to reserve the corresponding feature information if the current feature degree is greater than a preset feature degree;

[0095] a feature parameter construction unit, configured to construct a feature parameter corresponding to the available parameter based on all the reserved feature information.

[0096] In this embodiment, the feature information refers to detailed information capable of representing the characteristics of the available parameter obtained by analyzing the available parameter, such as the conduction speed of the neural impulse in the loop, the length of the loop, the delay loop, the existence of the neural lateral path, etc.

[0097] In this embodiment, the sub-feature information refers to the sub-feature information obtained by disassembling the feature information according to the sub-feature category.

[0098] In this embodiment, the current feature degree refers to the degree of feature manifestation of the current feature information obtained by calculating the sub-feature vector in the current feature information.

[0099] In this embodiment, the preset feature degree refers to a threshold feature degree that can distinguish between the reserved feature information and the non-reserved feature information. If the current feature degree is less than the preset feature degree, the feature manifestation of the current feature information does not reach the degree that needs to be analyzed, and is not reserved.

[0100] The working principle and beneficial effects of the above technical solution are: by disassembling and analyzing the feature information, the sub-feature vector is obtained and calculated to obtain the current feature degree corresponding to the feature information, and the feature parameter is screened,

[0101] The interference of non-feature parameters is reduced, and the accuracy and efficiency of the electroencephalogram analysis are improved. Embodiment

[0102] The embodiment provides a brain state analysis device based on electroencephalogram parameters, and the feature degree calculation unit comprises:

[0103] Among them, indicates the current feature degree of the corresponding feature information; indicates the number of sub-feature vectors in the sub-feature vector set; indicates the norm of the th sub-feature vector in the sub-feature vector set; indicates the average value of the norms of all sub-feature vectors in the sub-feature vector set; indicates the feature weight of the th sub-feature vector; indicates the preset relative coefficient corresponding to the feature information; An average vector representing all sub-feature vectors in the sub-feature vector set; A sub-feature vector in the sub-feature vector set; A sub-feature vector in the sub-feature vector set; A module of the vector An acquisition A larger one in An acquisition An absolute value of the difference. The working principle and beneficial effects of the above technical solutions are: through the calculation of the current feature degree, the feature parameters that meet the standard are screened, the interference of non-feature parameters is reduced, and the accuracy and efficiency of the electroencephalogram analysis are improved.

[0104] The working principle and beneficial effects of the above technical solutions are: through the calculation of the current feature degree, the feature parameters that meet the standard are screened, the interference of non-feature parameters is reduced, and the accuracy and efficiency of the electroencephalogram analysis are improved.

[0105] The working principle and beneficial effects of the above technical solutions are: through the calculation of the current feature degree, the feature parameters that meet the standard are screened, the interference of non-feature parameters is reduced, and the accuracy and efficiency of the electroencephalogram analysis are improved. Embodiment

[0106] The embodiment provides a brain state analysis device based on electroencephalogram parameters, and the parameter purification module comprises:

[0107] A detection signal acquisition unit is configured to input the feature parameters corresponding to the available parameters into a detection communication channel to obtain a detection signal;

[0108] A detection signal spectrum acquisition unit is configured to obtain a detection signal spectrum of the detection signal based on a filter display;

[0109] A detection frequency band acquisition unit is configured to obtain corresponding feature frequency band information based on the detection signal spectrum;

[0110] A parameter-characteristic conversion unit is configured to obtain loss characteristics corresponding to the feature parameters based on a parameter-characteristic analysis model;

[0111] A channel matching unit is configured to match a corresponding first communication channel from a channel database based on the loss characteristics and the feature frequency band information;

[0112] A feature signal acquisition unit is configured to input the feature parameters corresponding to the available parameters into the first communication channel to obtain a feature signal;

[0113] A feature signal spectrum acquisition unit is configured to obtain a feature signal spectrum of the feature signal;

[0114] An effective wave impurity removal unit is configured to obtain effective waves in the feature signal spectrum and remove aperiodic signals in the effective waves; An effective wave impurity removal unit is configured to obtain effective waves in the feature signal spectrum and remove aperiodic signals in the effective waves;

[0115] an effective wave region division unit, configured to divide the effective wave into a high-intensity signal region, a medium-intensity signal region and a low-intensity signal region based on signal intensity in the effective wave;

[0116] a signal set obtaining unit, configured to obtain a high-frequency signal set, a medium-frequency signal set and a low-frequency signal set respectively based on the high-intensity signal region, the medium-intensity signal region and the low-intensity signal region;

[0117] a low-frequency signal attenuation analysis unit, configured to obtain a low-frequency signal attenuation trend based on the low-frequency signal set;

[0118] a non-ideal attenuation point obtaining unit, configured to obtain a non-ideal attenuation point with the largest attenuation trend based on the low-frequency signal attenuation trend;

[0119] an attenuation point delay unit, configured to introduce a preset modulation parameter at the non-ideal attenuation point, delay the non-ideal attenuation point to an ideal attenuation point, and obtain an ideal low-frequency signal set;

[0120] a signal adjusting unit, configured to adjust the effective wave based on the ideal low-frequency signal set, and obtain an ideal high-frequency signal set and an ideal medium-frequency signal set;

[0121] a quantity difference calculating unit, configured to sequentially calculate a first signal quantity difference between the ideal high-frequency signal set and the high-frequency signal set, and a second signal quantity difference between the ideal medium-frequency signal set and the medium-frequency signal set;

[0122] an electroencephalogram typical parameter obtaining unit, configured to, if the first signal quantity difference and the second signal quantity difference are within a preset adjustment difference, decode and parameterize a characteristic signal obtained based on the first communication channel, and obtain an electroencephalogram typical parameter.

[0123] In this embodiment, detecting the communication channel refers to setting the parameters of the communication channel as normal operation parameters, so as to achieve the purpose of detecting the signal in the communication channel.

[0124] In this embodiment, the filter display refers to filtering out a specific waveband frequency that is clutter in the signal, and displaying a signal spectrum curve of the detection signal.

[0125] In this embodiment, the detection signal spectrum refers to a frequency distribution curve that can display the frequency density of the detection signal strength.

[0126] In this embodiment, the characteristic frequency band information refers to the maximum frequency bandwidth of the signal that can effectively pass through the detection communication channel, i.e., the frequency bandwidth occupied by the detection signal.

[0127] In this embodiment, the parameter-characteristic analysis model refers to a model trained by parameters and corresponding loss characteristics, which can be based on the loss characteristics of the feature parameters corresponding to the feature parameters.

[0128] In this embodiment, the loss characteristic refers to the characteristic performance of the feature parameter itself normally lost in the transmission process based on the analysis of the feature parameter, so as to achieve the purpose of eliminating the reasonable loss in the transmission of the feature parameter, wherein the loss characteristic includes: characteristics susceptible to polarization of medium, crosstalk characteristics and external radiation characteristics.

[0129] In this embodiment, the channel database refers to a database composed of all kinds of communication channels of the current device.

[0130] In this embodiment, the characteristic signal spectrum refers to a frequency distribution curve that can show the frequency density of the detected signal strength.

[0131] In this embodiment, the aperiodic signal refers to a signal without periodicity.

[0132] In this embodiment, the high-frequency signal set refers to a set of signals in the high-intensity signal area representing high signal strength.

[0133] In this embodiment, the medium-frequency signal set refers to a set of signals in the medium-intensity signal area representing medium signal strength.

[0134] In this embodiment, the low-frequency signal set refers to a set of signals in the low-intensity signal area representing low signal strength.

[0135] In this embodiment, the non-ideal attenuation point refers to the actual attenuation point with the largest attenuation trend obtained by analyzing the low-frequency signal attenuation trend, and the actual attenuation point is before the ideal attenuation point.

[0136] In this embodiment, the modulation parameter refers to a process of mixing the signal in the low-frequency signal set with the high-energy carrier signal to generate a new high-energy signal, so as to achieve the purpose of delaying the attenuation of the low-frequency signal.

[0137] In this embodiment, the ideal attenuation point refers to the attenuation point of the low-frequency signal attenuation trend under the condition that the low-frequency signal utilization is the largest within the range that the modulation parameter can affect.

[0138] In this embodiment, the ideal low-frequency signal set refers to a set of signals in the low-intensity signal area constructed after the modulation of the modulation parameter.

[0139] In this embodiment, the ideal high-frequency signal set refers to a set of signals in the high-intensity signal area constructed after the modulation of the modulation parameter.

[0140] In this embodiment, the ideal intermediate frequency signal set refers to a set of signals in the medium intensity signal region after modulation of the modulation parameter.

[0141] In this embodiment, the preset adjustment difference refers to a preset difference that can represent the high frequency signal set and the intermediate frequency signal set within the reasonable influence before and after the modulation of the modulation parameter. If the first signal quantity difference is greater than the preset adjustment difference, the modulation fails. If the second signal quantity difference is greater than the preset adjustment difference, the modulation fails.

[0142] The working principle and beneficial effects of the above technical solution are: by detecting the feature parameters, the corresponding feature frequency band information is obtained, and by analyzing the loss characteristics of the feature parameters, the loss features are obtained, and the corresponding communication channel is obtained, and by analyzing the feature signal spectrum of the purified feature signal and performing effective wave division analysis, the low frequency signal is modulated, so that the feature parameters are purified in transmission, and the accuracy and efficiency of the electroencephalogram analysis are improved. Embodiment

[0143] The embodiment provides a brain state analysis device based on electroencephalogram parameters, and the parameter conversion module comprises:

[0144] The analog power conversion unit is configured to input the electroencephalogram typical parameters into a parameter-power conversion model to obtain electroencephalogram analog power.

[0145] The power spectrum diagram acquisition unit is configured to obtain a power spectrum diagram based on the electroencephalogram analog power.

[0146] In this embodiment, the parameter-power conversion model refers to a model trained by the electroencephalogram typical parameters and the corresponding power, which can convert the electroencephalogram typical parameters into power.

[0147] The working principle and beneficial effects of the above technical solution are: by simulating the electroencephalogram typical parameters, the corresponding power spectrum diagram is obtained, which is beneficial to the analysis of the electroencephalogram typical parameters, thereby achieving the purpose of obtaining specific micro-state information, and is beneficial to the analysis of the brain state. Embodiment

[0148] The embodiment provides a brain state analysis device based on electroencephalogram parameters, and the analog power conversion unit comprises:

[0149] The training sample set construction block is configured to obtain all electroencephalogram typical parameters and corresponding feature parameter names, and construct a training sample set.

[0150] The model construction block is configured to construct a parameter-power conversion model, wherein the parameter-power conversion model comprises a name classification layer, an electroencephalogram simulation layer, and an output power layer.

[0151] The model training block is configured to input the training sample set into the parameter-power conversion model to obtain the electroencephalogram simulation power.

[0152] The technical scheme has the advantages that: all the electroencephalogram typical parameters and the corresponding feature parameter names are used to construct a training sample set, the parameter-power conversion model is trained, and the electroencephalogram typical parameters are simulated, so that the specific microstate information is obtained, and the brain state is analyzed. Embodiment

[0153] The embodiment provides a brain state analysis device based on electroencephalogram parameters, and the power matching module comprises:

[0154] The category acquisition unit is configured to obtain the corresponding microstate category based on the feature category of the feature parameter corresponding to the power spectrum graph.

[0155] The power spectrum analysis unit is configured to obtain the space-time distribution fluctuation based on the power spectrum graph.

[0156] The fluctuation analysis unit is configured to obtain the fluctuation trend of the space-time distribution fluctuation, the fluctuation difference between the maximum peak value and the minimum peak value of the space-time distribution fluctuation, the average peak value and the average valley value of the space-time distribution fluctuation based on the space-time distribution fluctuation.

[0157] The index acquisition unit is configured to obtain the microstate parameter table to obtain the microstate index corresponding to the fluctuation trend, the fluctuation difference, the average peak value and the average valley value.

[0158] In the embodiment, the feature category refers to the category obtained by classifying the feature information according to the brain function, including the neural category, the blood category and the cell category.

[0159] In the embodiment, the microstate category refers to the category of the corresponding microstate obtained according to the category of the feature information, including the neural microstate, the blood microstate and the cell microstate.

[0160] In the embodiment, the fluctuation trend of the space-time distribution fluctuation refers to the trend of the power frequency and energy of each electroencephalogram typical parameter corresponding to the space-time distribution fluctuation graph at each time, so that the microstate is analyzed.

[0161] In the embodiment, the average valley value of the space-time distribution fluctuation refers to the minimum value of the time variable of the power frequency and energy of each electroencephalogram typical parameter corresponding to the space-time distribution fluctuation graph in every two unit time.

[0162] In this embodiment, the microstate parameter reference table refers to a reference table constructed by the fluctuation trend of the spatiotemporal distribution fluctuation, the fluctuation difference between the maximum peak value and the minimum peak value of the spatiotemporal distribution fluctuation, the average peak value and the average valley value of the spatiotemporal distribution fluctuation, and the corresponding microstate index.

[0163] In this embodiment, the microstate index refers to the index of the microstate obtained by jointly referring to the fluctuation trend of the spatiotemporal distribution fluctuation, the fluctuation difference between the maximum peak value and the minimum peak value of the spatiotemporal distribution fluctuation, the average peak value and the average valley value of the spatiotemporal distribution fluctuation on the microstate parameter reference table.

[0164] The working principle and beneficial effects of the above technical solution are as follows: by analyzing the power spectrum diagram, the spatiotemporal distribution fluctuation is obtained and analyzed to obtain the fluctuation trend of the spatiotemporal distribution fluctuation, the fluctuation difference between the maximum peak value and the minimum peak value of the spatiotemporal distribution fluctuation, the average peak value and the average valley value of the spatiotemporal distribution fluctuation, and the microstate index of each microstate category on the microstate reference table, which is beneficial to the analysis of the brain state and improves the accuracy and efficiency of the brain analysis. Embodiment

[0165] The embodiment provides a brain state analysis device based on an electroencephalogram parameter, and the state analysis module comprises:

[0166] A disease analysis unit is configured to extract disease characteristics in disease descriptions and determine an associated occurrence probability and disease uniqueness of each disease characteristic based on all disease characteristics.

[0167] ;

[0168] ; wherein, represents the correlation function between the i th disease characteristic and all the remaining disease characteristics. represents the accumulated sum function after the correlation of each disease characteristic and all the remaining disease characteristics. represents the intersection symbol. represents the union symbol. represents the associated occurrence probability of the i th disease characteristic. represents the disease uniqueness of the i th disease characteristic. represents the history of the i th disease characteristic being assigned a unique value.

[0169] ​​​​​​​​The weight acquisition unit is configured to set a disease identifier to each disease feature based on the associated occurrence probability and the disease uniqueness, and acquire the weight of each category of microstate from a weight setting database, wherein the disease identifier is related to the associated occurrence probability of the disease, the disease uniqueness, and the category matched by the disease.

[0170] The brain state acquisition unit is configured to input all microstate categories and corresponding microstate indexes and microstate weights to a microstate comprehensive model to obtain a brain state.

[0171] In this embodiment, the disease feature refers to a feature extracted from a disease description of a patient and capable of representing the essential characteristics of the disease.

[0172] In this embodiment, the associated occurrence probability refers to the probability that each disease is associated with other disease features and occurs together.

[0173] In this embodiment, the disease uniqueness refers to the distinctive feature of a disease corresponding to a disease feature, which is different from other diseases corresponding to other disease features.

[0174] In this embodiment, the weight setting database refers to a database composed of disease identifiers and corresponding microstate weights.

[0175] In this embodiment, the microstate weight refers to the importance of a category of microstate to the brain state.

[0176] In this embodiment, the microstate comprehensive model refers to a model trained by microstate categories and corresponding microstate indexes and microstate weights, and capable of comprehensively analyzing the microstate categories and corresponding microstate indexes and microstate weights to convert them into a brain state.

[0177] The working principle and beneficial effects of the above technical solution are as follows: by analyzing the associated occurrence probability and the disease uniqueness of the disease features based on all disease features, the corresponding microstate weights are obtained, and all microstate categories and corresponding microstate indexes and microstate weights are analyzed to obtain a brain state analysis result, thereby improving the accuracy of electroencephalogram parameter analysis and enhancing the diagnosis ability of brain diseases.

[0178] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application.

Claims

1. A brain state analysis device based on electroencephalogram (EEG) parameters, characterized in that, include: Preprocessing module: Acquires EEG parameters, filters available parameters, and obtains corresponding feature parameters; Parameter purification module: Matches a corresponding communication channel for each feature parameter and purifies the parameter to obtain typical EEG parameters; Parameter conversion module: Based on typical EEG parameters, a power spectrum diagram is obtained; Power matching module: Based on the spatiotemporal distribution fluctuations of the power spectrum, it obtains the corresponding specific microstates; State Analysis Module: Comprehensively analyzes all specific microstates to obtain the brain state; The power matching module includes: The category acquisition unit is used to obtain the corresponding microstate category based on the feature category of the feature parameter corresponding to the power spectrum diagram; A power spectrum analysis unit is used to obtain the spatiotemporal distribution fluctuations based on the power spectrum diagram; The fluctuation analysis unit is used to obtain the fluctuation trend, the fluctuation difference between the maximum and minimum peak values ​​of the spatiotemporal distribution fluctuations, the average peak value and the average valley value of the spatiotemporal distribution fluctuations based on the spatiotemporal distribution fluctuations. The index acquisition unit is used to acquire a micro-state parameter lookup table and obtain a micro-state index corresponding to the fluctuation trend, fluctuation difference, average peak value and average valley value. The status analysis module includes: The symptom analysis unit is used to extract symptom features from symptom descriptions and determine the probability of each symptom feature appearing in association with all symptom features, as well as the symptom uniqueness. ； ;in, Indicates the first Individual disease characteristics With all remaining symptoms The association function between them; Indicate the characteristics of each disease With all remaining symptoms The summation function after association; Intersection symbol; The union symbol represents the set of sets. Indicates the first The probability of association between characteristics of a disease; Indicates the first The uniqueness of each disease characteristic; Indicates the first The history of each symptom characteristic is given a unique value; The weight acquisition unit is used to set a disease identifier for each disease feature based on the probability of association occurrence and the uniqueness of the disease, and to obtain the weight of each category micro-state from the weight setting database. The disease identifier is related to the probability of association occurrence of the disease, the uniqueness of the disease, and the category matched by the disease. The brain state acquisition unit is used to input all microstate categories, their corresponding microstate indices, and microstate weights into the microstate synthesis model to obtain the brain state.

2. The brain state analysis device based on electroencephalogram (EEG) parameters according to claim 1, characterized in that, The preprocessing module includes: The initial diagnostic unit is used to determine the electroencephalogram (EEG) measurement plan for the patient based on the patient's description of symptoms. The electroencephalogram (EEG) measurement unit is used to place the matching measuring device on the target site of the patient to perform EEG measurement according to the EEG measurement scheme, and to obtain the EEG measurement results of each measuring device. The region determination unit is used to divide the brain regions based on the EEG measurement results and obtain usable brain regions through comparative analysis of standard parameters of the EEG regions; The parameter filtering unit is used to filter all EEG parameters based on available brain regions to obtain usable parameters.

3. The brain state analysis device based on electroencephalogram (EEG) parameters according to claim 2, characterized in that, The preprocessing module further includes: The feature information parsing unit is used to obtain several corresponding feature information based on all available parameters; The sub-feature coding unit is used to decompose each feature information to obtain several sub-feature information and encode each sub-feature information to obtain the sub-feature code; The sub-feature vector transformation unit is used to map each sub-feature code to obtain a sub-feature vector, and then obtain the set of sub-feature vectors corresponding to each feature information; The feature degree calculation unit is used to calculate the current feature degree of the corresponding feature information based on the set of sub-feature vectors; The feature retention unit is used to retain the corresponding feature information if the current feature value is greater than the preset feature value. The feature parameter construction unit is used to construct feature parameters corresponding to the available parameters based on all retained feature information.

4. The brain state analysis device based on EEG parameters according to claim 3, characterized in that, The feature degree calculation unit includes: ;in, This indicates the current feature degree of the corresponding feature information; This indicates the number of sub-feature vectors in the sub-feature vector set; The first sub-feature vector in the set of sub-feature vectors Sub-feature vectors The model; Represents all sub-feature vectors in the sub-feature vector set. The average value of the modulus; Indicates the first Feature weights of each sub-feature vector; This represents a preset relative coefficient representing the corresponding feature information; This represents the average vector of all sub-feature vectors in the sub-feature vector set; The first sub-feature vector in the set of sub-feature vectors Sub-feature vectors; Representing vectors The model; Indicates obtaining and The larger of the two; Indicates obtaining The absolute value of the difference.

5. The brain state analysis device based on electroencephalogram (EEG) parameters according to claim 1, characterized in that, The parameter purification module includes: The detection signal acquisition unit is used to input the feature parameters corresponding to the available parameters into the detection communication channel to obtain the detection signal; The detection signal spectrum acquisition unit is used to obtain the detection signal spectrum based on the filter display. The detection frequency band acquisition unit is used to obtain corresponding characteristic frequency band information based on the spectrum of the detection signal; The parameter-feature conversion unit is used to obtain the loss characteristics corresponding to the feature parameters based on the parameter-feature analysis model. The channel matching unit is used to match the corresponding first communication channel from the channel database based on the loss characteristics and characteristic frequency band information; The feature signal acquisition unit is used to input the feature parameters corresponding to the available parameters into the first communication channel to obtain the feature signal; A feature signal spectrum acquisition unit is used to acquire the feature signal spectrum of the feature signal; An effective wave de-noising unit is used to acquire the effective wave in the spectrum of the characteristic signal and remove the aperiodic signal from the effective wave; The effective wave region division unit is used to divide the effective wave into high-intensity signal region, medium-intensity signal region and low-intensity signal region based on the signal intensity in the effective wave; The signal set acquisition unit is used to obtain the corresponding high-frequency signal set, medium-frequency signal set, and low-frequency signal set based on the high-intensity signal region, medium-intensity signal region, and low-intensity signal region, respectively. The low-frequency signal attenuation analysis unit is used to obtain the low-frequency signal attenuation trend based on a set of low-frequency signals. The non-ideal attenuation point acquisition unit obtains the non-ideal attenuation point with the largest attenuation trend based on the attenuation trend of the low-frequency signal. The attenuation point delay unit is used to introduce a preset modulation parameter at the non-ideal attenuation point, delay the non-ideal attenuation point to the ideal attenuation point, and obtain an ideal low-frequency signal set. The signal adjustment unit is used to adjust the effective wave based on the ideal low-frequency signal set to obtain the ideal high-frequency signal set and the ideal intermediate-frequency signal set; The quantity difference calculation unit is used to sequentially calculate the first signal quantity difference between the ideal high-frequency signal set and the high-frequency signal set, and the second signal quantity difference between the ideal intermediate frequency signal set and the intermediate frequency signal set. The typical EEG parameter acquisition unit is used to decode and purify the feature signals obtained based on the first communication channel to obtain typical EEG parameters if both the first signal quantity difference and the second signal quantity difference are within a preset adjustment difference value.

6. The brain state analysis device based on electroencephalogram (EEG) parameters according to claim 1, characterized in that, The parameter conversion module includes: The analog power conversion unit is used to input typical EEG parameters into the parameter-power conversion model to obtain the simulated EEG power. The power spectrum acquisition unit is used to obtain a power spectrum based on the brainwave simulation power.

7. The brain state analysis device based on electroencephalogram (EEG) parameters according to claim 6, characterized in that, The analog power conversion unit includes: The training sample set building block is used to obtain all typical EEG parameters and their corresponding feature parameter names to construct the training sample set. A model building block is used to construct a parameter-power conversion model, wherein the parameter-power conversion model includes a name classification layer, an EEG simulation layer, and an output power layer; The model training block is used to input the training sample set into the parameter-power conversion model to obtain the EEG simulated power.

Images