Data processing method and device based on infant electroencephalogram

Abstract

The invention relates to the technical field of data processing, and provides a data processing method and device based on an infant electroencephalogram. The method comprises the steps of obtaining a corresponding target emotion basic model according to the gender and the age group of a to-be-tested infant; generating a target emotion actual model of the to-be-detected infant according to the electroencephalogram data of the to-be-detected infant under the target emotion in the first N adjacent monitoring periods; according to the target emotion basic model and the target emotion actual model, determining a classification category to which the baby to be tested belongs; obtaining a health assessment function corresponding to the classification category; according to the actual electroencephalogram data of the to-be-detected infant under the target emotion in the next monitoring period and the health assessment function, the health assessment value of the to-be-detected infant is determined and used for assessing the health degree of the to-be-detected infant, the disease of the infant can be estimated more accurately, and therefore the problems in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a data processing method and device based on infant electroencephalogram (EEG) data. Background Art

[0002] Infants, especially those within 1 year old, are usually in the peak period of illness due to their relatively weak physical constitution. In addition, infants at this age stage have poor mobility, usually lie in bed or in swaddling clothes all day long, and do not have the ability to express themselves verbally. In case of illness, it is easy to cause delays in the treatment of the disease.

[0003] Currently, the analysis of infant EEG data is usually used for disease monitoring. For example, in Patent CN20211021807.4, a scalp EEG seizure high-frequency oscillation model for infantile spasms is disclosed, which includes collecting sample data of scalp EEG high-frequency oscillation when an infant with a confirmed infantile spasm has a seizure, and constructing a high-frequency oscillation model by training a convolutional neural network, so as to be able to identify infantile spasms.

[0004] However, in actual applications, there are certain differences among individual infants. The high-frequency oscillation model directly constructed from the sample data of scalp EEG high-frequency oscillation when an infant with a confirmed infantile spasm has a seizure usually has a low accuracy rate. Summary of the Invention

[0005] The present invention provides a data processing method and device based on infant EEG data, which can be used to solve the problem of low accuracy rate of predicting infant diseases in the prior art.

[0006] On the one hand, the present invention provides a data processing method based on infant EEG data, including:

[0007] Obtaining a corresponding target emotion base model according to the gender and age of the infant to be tested, wherein the target emotion base model is constructed based on EEG data of multiple infants of the same gender and age in a target emotion;

[0008] Generating a target emotion actual model of the infant to be tested according to the EEG data of the infant to be tested in the target emotion in the previous N adjacent monitoring periods, where N is a positive integer greater than or equal to 1;

[0009] Determining the classification category to which the infant to be tested belongs according to the target emotion base model and the target emotion actual model;

[0010] Obtaining a health assessment function corresponding to the classification category;

[0011] Determine the health assessment value of the baby to be tested based on the actual electroencephalogram data and the health assessment function under the target emotion in the next monitoring period of the baby to be tested, so as to evaluate the health degree of the baby to be tested.

[0012] Preferably, the method further includes:

[0013] Obtain the electroencephalogram data of multiple babies of the same gender and age group under the target emotion;

[0014] Based on each electroencephalogram data, calculate the average data amplitude, the average duration, and the average attack frequency under the target emotion as the basic model of the target emotion; and,

[0015] Generate the actual model of the target emotion of the baby to be tested according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring periods under the target emotion, specifically including:

[0016] According to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring periods under the target emotion, calculate the second average data amplitude, the second average duration, and the second average attack frequency under the target emotion as the actual model of the target emotion.

[0017] Preferably, determine the classification category to which the baby to be tested belongs according to the basic model of the target emotion and the actual model of the target emotion, specifically including:

[0018] Determine the attack characteristics of the target emotion of the baby to be tested according to the magnitude of the second average data amplitude and the average data amplitude, the magnitude of the second average duration and the average duration, and the magnitude of the second average attack frequency and the average attack frequency. The attack characteristics are used to characterize whether there are obvious deviations in the emotional fluctuation range, duration, and attack frequency of the target emotion of the baby to be tested;

[0019] Determine the classification category to which the baby to be tested belongs according to the attack characteristics of the target emotion of the baby to be tested.

[0020] Preferably, the attack characteristics are specifically the first ratio of the second average data amplitude to the average data amplitude, the second ratio of the second average duration to the average duration, and the third ratio of the second average attack frequency to the average attack frequency; and,

[0021] Determine the classification category to which the baby to be tested belongs according to the attack characteristics of the target emotion of the baby to be tested, specifically including:

[0022] Determine the classification category to which the baby to be tested belongs according to the numerical intervals into which the first ratio, the second ratio, and the third ratio respectively fall.

[0023] Preferably, the health assessment function is specifically: Q = a×A / A E +b×T / T E +c×F / F E ;

[0024] Wherein, Q is the health assessment value calculated by the health assessment function; A is the actual amplitude of the data of the baby to be tested in the next monitoring cycle under the target emotion; A E is the predicted amplitude of the data of the baby to be tested in the next monitoring cycle under the target emotion; T is the actual duration of the target emotion in the next monitoring cycle of the baby to be tested; T E is the predicted duration of the target emotion in the next monitoring cycle of the baby to be tested; F is the actual attack frequency of the target emotion in the next monitoring cycle of the baby to be tested; F E is the predicted attack frequency of the target emotion in the next monitoring cycle of the baby to be tested; a, b, and c are all undetermined weights;

[0025] Preferably, obtaining the health assessment function corresponding to the classification category specifically includes: obtaining the values of a, b, and c in the health assessment function corresponding to the classification category to obtain the health assessment function.

[0026] Preferably, determining the health assessment value of the baby to be tested according to the actual electroencephalogram data of the baby to be tested in the next monitoring cycle under the target emotion and the health assessment function specifically includes:

[0027] Determine the actual amplitude, the actual duration, and the actual attack frequency of the data according to the actual electroencephalogram data of the baby to be tested in the next monitoring cycle under the target emotion;

[0028] Determine the predicted amplitude, the predicted duration, and the predicted attack frequency of the data according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion;

[0029] Substitute the actual amplitude, the actual duration, and the actual attack frequency of the data, as well as the predicted amplitude, the predicted duration, and the predicted attack frequency of the data into the health assessment function Q = a×A / A E +b×T / T E +c×F / F E , to calculate the health assessment value of the baby to be tested.

[0030] Preferably, according to the actual electroencephalogram data of the baby to be tested in the next monitoring cycle under the target emotion, determine the actual amplitude, the actual duration, and the actual attack frequency of the data. Specifically, it includes:

[0031] Obtain the vibration range of each data in the actual electroencephalogram data as the actual amplitude A of the data;

[0032] Obtain the duration of each target emotion attack in the actual electroencephalogram data, and calculate the average value of each duration as the actual duration T;

[0033] Determine the number of target emotion attacks in the next monitoring cycle of the baby to be tested through the actual electroencephalogram data as the actual attack frequency F.

[0034] Preferably, according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion, determine the estimated amplitude, the estimated duration, and the estimated attack frequency of the data. Specifically, it includes:

[0035] For each of the previous N adjacent monitoring cycles, as the current adjacent monitoring cycle, obtain the electroencephalogram data of the current adjacent monitoring cycle under the target emotion as the current electroencephalogram data;

[0036] Obtain the amplitude A' of the current adjacent monitoring cycle from the current electroencephalogram data, so as to obtain the amplitudes A' corresponding to each of the previous N adjacent monitoring cycles, denoted as A1', A2'... A N ';

[0037] Perform curve fitting on A1', A2'... A N ' to obtain an amplitude trend function, and use the amplitude trend function to calculate A E ;

[0038] Obtain the target emotion duration T' of the current adjacent monitoring cycle from the current electroencephalogram data, so as to obtain the target emotion durations T' corresponding to each of the previous N adjacent monitoring cycles, denoted as T1', T2'... T N ';

[0039] Perform curve fitting on T1', T2'... T N ' to obtain a duration trend function, and use the duration trend function to calculate T E ;

[0040] The estimated attack frequency F' of the target emotion in the current near monitoring cycle is obtained from the current EEG data to obtain the estimated attack frequency F' corresponding to each near monitoring cycle in the first N near monitoring cycles, which are recorded as F1', F2'...F N ';

[0041] For F1', F2'...F N 'Perform curve fitting to obtain a frequency trend function, and use the frequency trend function to calculate F E .

[0042] In another aspect, the present invention provides a data processing device based on an infant electroencephalogram, comprising:

[0043] An acquisition unit, used to acquire a corresponding target emotion basic model according to the gender and age group of the infant to be tested, wherein the target emotion basic model is constructed according to the electroencephalogram data of multiple infants of the same gender and age group under the target emotion;

[0044] A generating unit, configured to generate an actual model of the target emotion of the infant to be tested according to the electroencephalogram data of the infant to be tested under the target emotion in the previous N adjacent monitoring cycles, wherein N is a positive integer greater than or equal to 1;

[0045] A determination unit, used for determining the classification category to which the infant to be tested belongs according to the target emotion basic model and the target emotion actual model;

[0046] A second acquisition unit, used to acquire the health assessment function corresponding to the classification category;

[0047] The second determination unit is used to determine the health assessment value of the baby to be tested according to the actual electroencephalogram data of the baby to be tested under the target emotion in the next monitoring period and the health assessment function, so as to evaluate the health level of the baby to be tested.

[0048] The data processing method based on infant electroencephalogram provided by the present invention includes first obtaining a corresponding target emotion base model according to the gender and age of the infant to be tested. The target emotion base model is constructed based on the electroencephalogram data of multiple infants of the same gender and age under the target emotion. Then, according to the electroencephalogram data of the infant to be tested in the target emotion in the previous N adjacent monitoring cycles, a target emotion actual model of the infant to be tested is generated, where N is a positive integer greater than or equal to 1. Then, according to the target emotion base model and the target emotion actual model, the classification category to which the infant to be tested belongs is determined. Then, the health assessment function corresponding to this classification category is obtained. Then, according to the actual electroencephalogram data of the infant to be tested in the target emotion in the next monitoring cycle and the health assessment function, the health assessment value of the infant to be tested is determined, which is used to evaluate the health level of the infant to be tested. In this method, by combining the target emotion actual model generated from the electroencephalogram data of the infant to be tested in the previous N adjacent monitoring cycles under the target emotion, the classification category to which the infant to be tested belongs is determined, and then the corresponding health assessment function is determined, and its health assessment value is calculated. This method can more accurately predict the diseases of infants compared with the general model method, thus solving the problems in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0050] Figure 1 It is a schematic flow chart of the data processing method based on infant electroencephalogram provided by the present invention;

[0051] Figure 2 It is a schematic structural diagram of the data processing device based on infant electroencephalogram provided by the present invention;

[0052] Figure 3 It is a schematic structural diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0054] As described above, in actual applications, there are certain differences among individual infants. Currently, the high-frequency oscillation model constructed directly from the sample data of scalp electroencephalogram high-frequency oscillations during symptom attacks in infants with diagnosed infantile spasms usually has a low accuracy rate.

[0055] In view of this, the embodiments of the present application provide a data processing method and device based on infant electroencephalogram, which can be used to more accurately evaluate the health status of infants. For the sake of easy understanding, the embodiments of the present application can be described as a whole here. For example, the technical solutions provided by the embodiments of the present application can be applied to software products. For example, the method provided by the embodiments of the present application can be designed as an application program (i.e., software product), and then installed on an electronic device such as a user terminal or a server to apply the method provided by the embodiments of the present application.

[0056] For example, the software product can be installed on an electronic device of the user terminal, including electronic devices such as mobile phones, tablets, and computers, and then the method provided by the embodiments of the present application can be executed; the software product can also be installed on an electronic device of the server, including a server or a server cluster, and then the method provided by the embodiments of the present application can be executed.

[0057] In actual applications, this method can also be applied as a hardware product. For example, a dedicated hardware device can be designed to implement the method provided by the embodiments of the present application.

[0058] As Figure 1 shown, it is a schematic flowchart of the data processing method based on infant electroencephalogram provided by the embodiments of the present application. In this example, the method is described by taking its application to a software product as an example.

[0059] Among them, the method may include:

[0060] Step S11: Obtain a corresponding target emotion base model according to the gender and age of the infant to be tested.

[0061] Among them, the infant to be tested is an infant who needs to be tested for spasms. For example, since infants within 1 year old are usually in the peak period of onset, infants within 1 year old are usually required to be tested for spasms. At this time, infants within 1 year old can be used as the infant to be tested. Of course, in actual applications, according to actual needs, other infants can also be used as the infant to be tested.

[0062] It should be noted that the electroencephalogram (EEG) data of infants will change with age. For example, for infants within 1 year old, the younger the month (age range at birth), the more the EEG data shows the characteristics of low-voltage slow-wave activity. Therefore, the EEG data of infants is related to their age range. In addition, the EEG data of infants is also related to the gender of the infants. Especially for infants over 8 months old, the correlation between their EEG data and gender becomes more and more obvious. Therefore, the EEG data of infants is also related to gender.

[0063] In this step S11, the target emotion base model is constructed based on the EEG data of multiple infants of the same gender and age range under the target emotion. Since the EEG data of infants is related to the age range and gender of the infants, the target emotion base model is also related to the gender and age range of the infant to be tested. Thus, before this step S11, the gender and age range of the infant to be tested can be obtained first. For example, a user operation interface can be set up so that the user can operate through this user operation interface. For example, the user can input or select a certain gender and a certain age range through the user operation interface, so that the electronic device can receive the gender and age range of the infant to be tested, and then obtain the corresponding target emotion base model from the base model library.

[0064] Among them, the base model library stores base models corresponding to various different age ranges, genders, and emotions. For example, the age range can be divided into 1 month, 2 months, 3 months, etc., and the gender is divided into male and female. It should be further noted that for infants within 1 year old, their daily life states are relatively simple, mainly including sleep state, feeding state, quiet state, normal activity state, and crying state. These different states correspond to different emotions respectively. However, through specific analysis, it is found that the EEG data in the crying state has the highest correlation with spasm. Therefore, in this application, the emotion in the crying state of the infant can be selected as the target emotion.

[0065] Therefore, in practical applications, after obtaining the gender and age range of the infant to be tested, the base models of various emotions of the gender and age range of the infant to be tested can be queried from the base model library, and then the base model of the emotion in the crying state among them can be further obtained, that is, the target emotion base model.

[0066] In practical applications, the following method can be used to generate the actual model of the target emotion. Specifically, multiple infants of the same gender and age group can be selected first, and electroencephalogram (EEG) data of these infants in the target emotion (i.e., the crying state) can be obtained. For example, for multiple infants of the same gender and age group, when they are in the crying state, their EEG data can be collected by an EEG acquisition device. Then, based on each EEG data, the average amplitude of the data in the target emotion, the average duration of the target emotion, and the average attack frequency of the target emotion can be calculated as the basic model of the target emotion. Here, the same gender and age group refer to the same gender and age as the infant to be tested.

[0067] Among them, the amplitude of the EEG data of each infant in the target emotion can be calculated, and then the average value of the amplitudes of the EEG data of each infant can be calculated as the average amplitude of the data. Obviously, the average amplitude of the data reflects the average situation of the fluctuation range of the EEG data of a large number of infants of the same gender and age group in the target emotion.

[0068] The calculation method of the average duration of the target emotion, for example, can calculate the duration of each crying state of each infant for the EEG data of each infant respectively, and then calculate the average value of the durations of each crying state of each infant as the average duration of the target emotion. Obviously, the average duration of the target emotion also reflects the average duration of each attack of the target emotion of a large number of infants of the same gender and age group, that is, the average duration of the crying state.

[0069] For the calculation method of the average attack frequency of the target emotion, for example, for each infant, the number of attacks of the target emotion of the infant in each statistical period can be collected. For example, the number of times the infant is in the crying state within 24 hours can be counted as the attack frequency of the target emotion of the infant. Here, the statistical period can be 24 hours (i.e., 1 day), or 72 hours or other statistical periods. In this way, the average value of the attack frequencies of the target emotion of each infant can be further calculated as the average attack frequency of the target emotion. Obviously, the average attack frequency of the target emotion reflects the average frequency of the average attack of the target emotion of a large number of infants in each statistical period.

[0070] Of course, based on the same implementation principle, multiple different basic models can be pre-constructed in advance for infants of multiple different age groups and genders, as well as multiple different emotions, and then a basic model library can be generated. In this way, in step S11, the corresponding basic model of the target emotion can be obtained according to the gender and age of the infant to be tested.

[0071] Step S12: Generate the actual model of the target emotion of the infant to be tested according to the EEG data of the infant to be tested in the target emotion in the previous N adjacent monitoring periods.

[0072] Among them, N is a positive integer greater than or equal to 1. For example, N can be 1, 3, 5, 10 or other values.

[0073] Specifically, the above-mentioned target emotion base model obviously reflects the average situation of electroencephalogram data of a large number of infants of the same gender and age group under the target emotion. However, when this average situation of a large amount of electroencephalogram data is used to detect the infant to be tested, there may be a problem of insufficient accuracy. Therefore, in step S12 of this application, the electroencephalogram data of the infant to be tested under the target emotion in the previous N adjacent monitoring periods can be further obtained. For example, the electroencephalogram data of the infant to be tested in the crying state in the previous 3 days can be obtained.

[0074] Then, based on the electroencephalogram data of the infant to be tested under the target emotion in the previous N adjacent monitoring periods, the actual target emotion model of the infant to be tested is generated. Specifically, based on the electroencephalogram data of the infant to be tested under the target emotion in the previous N adjacent monitoring periods, the average amplitude of the data under the target emotion (referred to as the second average amplitude), the average duration of the target emotion (referred to as the second average duration), and the average attack frequency of the target emotion (referred to as the second average attack frequency) can be calculated as the actual target emotion model.

[0075] For example, for the calculation method of the second average amplitude, the average value of the amplitudes of the electroencephalogram data of the infant to be tested under the target emotion in the previous N adjacent monitoring periods can be calculated as the second average amplitude. Obviously, the magnitude of the second average amplitude can reflect the actual fluctuation range of the electroencephalogram data when the infant cries.

[0076] For the calculation method of the second average duration, the duration of each crying state of the infant to be tested in the previous N adjacent monitoring periods can be determined, and then the average value of these durations can be calculated as the second average duration. Obviously, the second average duration reflects the magnitude of the actual fluctuation duration of the electroencephalogram data when the infant to be tested cries each time.

[0077] For the calculation method of the second average attack frequency, the attack frequency of the target emotion of the infant to be tested in each adjacent monitoring period in the previous N adjacent monitoring periods can be determined. For example, the number of times of crying in each adjacent monitoring period, and then the average value of these attack frequencies can be calculated as the second average attack frequency. Obviously, the second average attack frequency also reflects the actual number of attacks of the target emotion of the infant to be tested in the adjacent monitoring periods.

[0078] Step S13: Determine the classification category to which the infant to be tested belongs according to the target emotion base model and the actual target emotion model.

[0079] Among them, the classification categories usually can include duration deviation category, attack frequency deviation category, fluctuation range deviation category, no obvious tendency category, and comprehensive deviation category. Among them, for the infants in the duration deviation category, the duration of each target emotion attack is significantly deviated from that of infants of the same gender and age group; for the infants in the attack frequency deviation category, the attack frequency of the target emotion is significantly deviated from that of infants of the same gender and age group; for the infants in the fluctuation range deviation category, the fluctuation range of the emotion during each target emotion attack is significantly deviated from that of infants of the same gender and age group; for the infants in the no obvious tendency category, the attack of the target emotion has no obvious tendency, that is, there is no obvious tendency in the three dimensions of duration, attack frequency, and emotion fluctuation range; for the infants in the comprehensive deviation category, the target emotion has obvious tendencies in at least two of the duration, attack frequency, and emotion fluctuation range.

[0080] In this application, since the target emotion basic model reflects the average situation of a large number of infants of the same gender and age group, and the target emotion actual model reflects the actual situation of the to-be-tested infant, the classification category to which the to-be-tested infant belongs can be determined according to the target emotion basic model and the target emotion actual model. Specifically, the attack characteristics of the target emotion of the to-be-tested infant can be determined according to the magnitude of the second data average amplitude and the data average amplitude, the magnitude of the second average duration and the average duration, and the magnitude of the second average attack frequency and the average attack frequency. The attack characteristics are used to characterize whether there are obvious deviations in the emotion fluctuation range, duration, and attack frequency of the target emotion of the to-be-tested infant; then, according to the attack characteristics of the target emotion of the to-be-tested infant, the classification category to which the to-be-tested infant belongs can be determined.

[0081] Since the attack characteristics are used to characterize whether there are obvious deviations in the emotion fluctuation range, duration, and attack frequency of the target emotion of the to-be-tested infant, in practical applications, the difference or ratio between the second average duration and the average duration is usually used to characterize whether there is an obvious deviation in the duration of the target emotion of the to-be-tested infant. For example, the larger the absolute value of the difference between the two, the greater the deviation of the target emotion of the to-be-tested infant in terms of duration, and vice versa, the smaller the deviation of the target emotion of the to-be-tested infant in terms of duration; of course, when it is a ratio, the closer the ratio of the difference between the second data average amplitude and the data average amplitude is to 1, the smaller the deviation of the target emotion of the to-be-tested infant in terms of duration, and vice versa, the greater the deviation of the target emotion of the to-be-tested infant in terms of duration.

[0082] Similarly, the difference or ratio between the second data average amplitude and the data average amplitude can be used to characterize whether there is an obvious deviation in the fluctuation range of the target emotion of the baby to be tested. For example, the larger the absolute value of the difference between the two, the greater the deviation of the baby to be tested in the fluctuation range of the target emotion, and vice versa, the smaller the deviation of the baby to be tested in the fluctuation range of the target emotion. When it is a ratio, the closer the ratio between the second data average amplitude and the data average amplitude is to 1, the smaller the deviation of the baby to be tested in the fluctuation range of the target emotion, and vice versa, the greater the deviation of the baby to be tested in the fluctuation range of the target emotion.

[0083] The difference or ratio between the second average attack frequency and the average attack frequency can be used to characterize whether there is an obvious deviation in the attack frequency of the target emotion of the baby to be tested. For example, the larger the absolute value of the difference between the two, the greater the deviation of the baby to be tested in the attack frequency of the target emotion, and vice versa, the smaller the deviation of the baby to be tested in the attack frequency of the target emotion. When it is a ratio, the closer the ratio between the second average attack frequency and the average attack frequency is to 1, the smaller the deviation of the baby to be tested in the attack frequency of the target emotion, and vice versa, the greater the deviation of the baby to be tested in the attack frequency of the target emotion.

[0084] In this application, considering the differences in units and value ranges among frequency, amplitude, and duration, for the convenience of normalization processing, a ratio method can be used for evaluation. Therefore, the attack characteristics can specifically be the ratio of the second data average amplitude to the data average amplitude (referred to as the first ratio), the ratio of the second average duration to the average duration (referred to as the second ratio), and the ratio of the second average attack frequency to the average attack frequency (referred to as the third ratio). Obviously, the first ratio, the second ratio, and the third ratio can respectively characterize whether there is an obvious deviation in the target emotion of the baby to be tested in terms of the emotion fluctuation range, the duration, and the attack frequency. Then, according to the first ratio, the second ratio, and the third ratio, the classification category to which the baby to be tested belongs is determined. For example, according to the numerical intervals into which the first ratio, the second ratio, and the third ratio respectively fall, the classification category to which the baby to be tested belongs is determined.

[0085] In practical applications, the overall numerical interval can be divided into three numerical intervals, namely (0, 0.45], (0.45, 1.3], and the interval greater than 1.3. Among them, if a certain ratio (which can be the first ratio, the second ratio, or the third ratio) falls into (0.45, 1.3], it means there is no obvious deviation. On the contrary, if the ratio falls into the interval (0, 0.45] or greater than 1.3, it means there is an obvious deviation.

[0086] For example, if the numerical interval in which the first ratio falls is (0.45, 1.3], it indicates that there is no obvious deviation in the emotional fluctuation range of the target emotion of the baby to be tested. On the contrary, if the numerical interval in which the first ratio falls is (0, 0.45] or an interval greater than 1.3, it indicates that there is an obvious deviation in the emotional fluctuation range of the target emotion of the baby to be tested.

[0087] Similarly, it is also possible to determine whether there is an obvious deviation in the duration of the target emotion of the baby to be tested according to the numerical interval in which the second ratio falls. Specifically, if the falling numerical interval is (0.45, 1.3], there is no obvious deviation; on the contrary, if the falling numerical interval is (0, 0.45] or an interval greater than 1.3, there is an obvious deviation. It is also possible to determine whether there is an obvious deviation in the frequency of occurrence of the target emotion of the baby to be tested according to the numerical interval in which the third ratio falls. Specifically, if the falling numerical interval is (0.45, 1.3], there is no obvious deviation; on the contrary, if the falling numerical interval is (0, 0.45] or an interval greater than 1.3, there is an obvious deviation.

[0088] In this way, according to the numerical intervals respectively fallen by the first ratio, the second ratio and the third ratio, it is possible to determine the aspects with obvious deviations. For example, if the baby to be tested only has an obvious deviation in the emotional fluctuation range, its classification category is the fluctuation range deviation category; if the baby to be tested only has an obvious deviation in the duration, its classification category is the duration deviation category; if the baby to be tested only has an obvious deviation in the frequency of occurrence, its classification category is the attack frequency deviation category; if the baby to be tested has no obvious deviation in the frequency of occurrence, the emotional fluctuation range and the duration, its classification category is the no obvious tendency category; if the baby to be tested has obvious deviations in at least two of the frequency of occurrence, the emotional fluctuation range and the duration, its classification category is the comprehensive deviation category.

[0089] Step S14: Obtain the health assessment function corresponding to this classification category.

[0090] Among them, this health assessment function can be used to evaluate the health level of the baby to be tested. For example, this health assessment function can be used to calculate the health assessment value of the baby to be tested, and then the health level of the baby to be tested can be evaluated according to this health assessment value. For example, if the health assessment value of the baby to be tested is less than the lowest threshold, it indicates that the health level of the baby to be tested is relatively poor and there may be a relatively high risk of disease.

[0091] In the application scenario of the embodiment of the present application, since the electroencephalogram data of the baby is used to evaluate its health status, especially the risk of diseases such as spasm, the health assessment function can specifically be the following calculation formula:

[0092] Q=a×A / A E +b×T / T E +c×F / F E

[0093] In this calculation formula, Q is the health assessment value calculated by the health assessment function; A is the actual amplitude of the data under the target emotion in the next monitoring cycle of the infant to be tested; A E is the estimated amplitude of the data under the target emotion in the next monitoring cycle of the tested baby; T is the actual duration of the target emotion in the next monitoring cycle of the tested baby; T E is the estimated duration of the target emotion in the next monitoring cycle of the infant to be tested; F is the actual frequency of the target emotion in the next monitoring cycle of the infant to be tested; F E is the estimated frequency of the target emotion in the next monitoring period of the infant to be tested; a, b and c are all weights to be determined.

[0094] In this calculation formula, the pending weights a, b, and c correspond to the classification categories. That is to say, the present application takes into account the characteristics of the target emotion of the infant to be tested, and designs the values of the pending weights a, b, and c corresponding to its classification category. For example, if the classification category to which the infant to be tested belongs is the fluctuation range deviation category, it means that the characteristics of the target emotion of the infant to be tested are that the amplitude of the target emotion fluctuation is large, but the duration and frequency are normal. Therefore, the value of a needs to be relatively high, while the values of b and c need to be relatively low. Similarly, for other classification categories, there are corresponding values of the pending weights a, b, and c.

[0095] Therefore, for the specific implementation of step S14, the values of a, b and c in the health assessment function corresponding to the classification category can be obtained and brought into the above calculation formula to obtain the health assessment function.

[0096] Step S15: Determine the health assessment value of the infant to be tested according to the actual EEG data of the infant to be tested under the target emotion in the next monitoring cycle and the health assessment function.

[0097] Obviously, the health assessment value of the infant to be tested can be used to evaluate the health status of the infant to be tested.

[0098] In the above step S14, it is mentioned that the calculation formula of the health assessment function can be Q = a × A / A E +b×T / T E +c×F / F E In this calculation formula, the values of the undetermined weights a, b and c can be determined by step S14, but the actual amplitude A, actual duration T and actual frequency F of the data, as well as the estimated amplitude A of the data need to be further determined. E, estimated duration T E and estimated attack frequency F E , and how to determine these parameters will be described here.

[0099] In this step S15, the actual amplitude A, actual duration T, and actual attack frequency F of the data can be determined according to the actual electroencephalogram data of the baby to be tested in the next monitoring cycle under the target emotion. For example, for the actual amplitude A of the data, the actual electroencephalogram data of the baby to be tested in the next monitoring cycle under the target emotion can be collected by an electroencephalogram acquisition device, and then the actual amplitude of the data of the baby to be tested in the next monitoring cycle under the target emotion can be calculated through this actual electroencephalogram data. For example, the vibration conditions of each data in the actual electroencephalogram data can be obtained, and then its vibration range can be determined as the actual amplitude A of the data; the actual duration T of the baby to be tested in the next monitoring cycle under the target emotion can also be calculated through this actual electroencephalogram data. For example, the duration of each target emotion attack in the actual electroencephalogram data can be obtained, and then its average value can be calculated to obtain the actual duration T; similarly, the actual attack frequency F of the baby to be tested in the next monitoring cycle under the target emotion can also be calculated through this actual electroencephalogram data, that is, the number of target emotion attacks. For example, when the target emotion is the emotion in the crying state, the number of crying times can be calculated at this time, which is the actual attack frequency F.

[0100] In practical applications, the estimated amplitude A of the data can be determined according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion E , estimated duration T E and estimated attack frequency F E .

[0101] Here, the estimated amplitude A of the data E is taken as an example for illustration. For example, for each of the previous N adjacent monitoring cycles (referred to as the current adjacent monitoring cycle), the electroencephalogram data of the current adjacent monitoring cycle under the target emotion (referred to as the current electroencephalogram data) can be obtained, and the amplitude A' of the current adjacent monitoring cycle can be obtained from the current electroencephalogram data. In this way, the amplitudes A' corresponding to each of the previous N adjacent monitoring cycles, that is, A1', A2'... A N ' can be obtained, and then further according to the change trend of A1', A2'... A N ', the estimated amplitude A of the data of the baby to be tested in the next monitoring cycle under the target emotion can be calculated E , for example, A1', A2'... A NPerform curve fitting to obtain the amplitude trend function, and then use this amplitude trend function to estimate the amplitude A of the data for the next monitoring cycle of the baby to be measured under the target emotion. E 。

[0102] Based on the same principle, the duration T' of the target emotion for the current adjacent monitoring cycle can also be obtained from the current EEG data, so that the duration T' of the target emotion corresponding to each of the previous N adjacent monitoring cycles can be obtained, that is, T1', T2'... T N ', and then further according to the changing trend of T1', T2'... T N ', estimate the duration T of the target emotion for the next monitoring cycle of the baby to be measured. E For example, curve fitting can be performed on T1', T2'... T N ' to obtain the duration trend function, and then use this duration trend function to estimate the duration T of the target emotion for the next monitoring cycle of the baby to be measured. E 。

[0103] The estimated attack frequency F' of the target emotion for the current adjacent monitoring cycle can also be obtained from the current EEG data, so that the estimated attack frequency F' corresponding to each of the previous N adjacent monitoring cycles can be obtained, that is, F1', F2'... F N ', and then further according to the changing trend of F1', F2'... F N ', estimate the estimated attack frequency F of the target emotion for the next monitoring cycle of the baby to be measured. E For example, curve fitting can be performed on F1', F2'... F N ' to obtain the frequency trend function, and then use this frequency trend function to estimate the estimated attack frequency F of the target emotion for the next monitoring cycle of the baby to be measured. E 。

[0104] In this way, the actual amplitude A, actual duration T, and actual attack frequency F of the data, as well as the estimated amplitude A E 、estimated duration T E and estimated attack frequency F E can be obtained respectively. And after obtaining the actual amplitude A, actual duration T, and actual attack frequency F of the data, as well as the estimated amplitude A E 、estimated duration T E and estimated attack frequency F E through the above method, they can be substituted into the above health assessment function Q = a × A / A E+b×T / T E +c×F / F E , so as to calculate the health assessment value Q of the baby to be tested, which is used to evaluate the health level of the baby to be tested. For example, when the health assessment value of the baby to be tested is too low, it indicates that the baby has a relatively high risk of getting sick.

[0105] Adopt the data processing method based on the electroencephalogram of infants provided by the embodiment of the present application. This method includes first obtaining the corresponding target emotion basic model according to the gender and age of the baby to be tested. The target emotion basic model is constructed based on the electroencephalogram data of multiple babies of the same gender and age under the target emotion. Then, according to the electroencephalogram data of the baby to be tested in the target emotion in the first N adjacent monitoring cycles, a target emotion actual model of the baby to be tested is generated, where N is a positive integer greater than or equal to 1. Then, according to the target emotion basic model and the target emotion actual model, the classification category to which the baby to be tested belongs is determined. Then, the health assessment function corresponding to this classification category is obtained. Then, according to the actual electroencephalogram data of the baby to be tested in the target emotion in the next monitoring cycle and the health assessment function, the health assessment value of the baby to be tested is determined, which is used to evaluate the health level of the baby to be tested. In this method, the target emotion actual model generated by combining the electroencephalogram data of the baby to be tested in the target emotion in the first N adjacent monitoring cycles is used to determine the classification category to which the baby to be tested belongs, and then the corresponding health assessment function is determined, and its health assessment value is calculated. This method can improve the accuracy rate compared with the general model method, thus solving the problems in the prior art.

[0106] In practical applications, if it is detected that the health assessment value of the baby to be tested is too low, such as lower than the preset minimum value, it indicates that the baby to be tested may have a risk of getting sick. Therefore, further warnings can be given to prompt relevant personnel to conduct further health monitoring on the baby to be tested. Of course, if the health assessment value of the baby to be tested is greater than or equal to the preset minimum value, it indicates that the overall risk of the baby to be tested getting sick is not high and no treatment is required.

[0107] Based on the same inventive concept as the data processing method based on the electroencephalogram of infants provided by the embodiment of the present application, the embodiment of the present application can also provide a data processing device based on the electroencephalogram of infants, such as Figure 2 As shown in the specific structural schematic diagram of the data processing device 20. In the embodiment of this device, if there are unclear points, the content in the method embodiment can be referred to. The data processing device 20 may include: an acquisition unit 201, a generation unit 202, a determination unit 203, a second acquisition unit 204, and a second determination unit 205, where:

[0108] An acquisition unit 201, configured to obtain a corresponding target emotion base model according to the gender and age of the baby to be tested, where the target emotion base model is constructed based on electroencephalogram data of multiple babies of the same gender and age in a target emotion;

[0109] A generation unit 202, configured to generate a target emotion actual model of the baby to be tested according to the electroencephalogram data of the baby to be tested in the target emotion in the previous N adjacent monitoring periods, where N is a positive integer greater than or equal to 1;

[0110] A determination unit 203, configured to determine the classification category to which the baby to be tested belongs according to the target emotion base model and the target emotion actual model;

[0111] A second acquisition unit 204, configured to acquire a health assessment function corresponding to the classification category;

[0112] A second determination unit 205, configured to determine a health assessment value of the baby to be tested according to the actual electroencephalogram data of the baby to be tested in the target emotion in the next monitoring period and the health assessment function, so as to evaluate the health level of the baby to be tested;

[0113] Since the data processing device 20 adopts the same inventive concept as the data processing method based on the electroencephalogram of the baby provided in the embodiment of the present application, when the method can solve the technical problem, the data processing device 20 can also solve the technical problem. Specifically, the solution provided by the data processing device 20 combines the target emotion actual model generated from the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring periods in the target emotion to determine the classification category to which the baby to be tested belongs, and then determines the corresponding health assessment function and calculates its health assessment value. This method can improve the accuracy rate compared with the general model method, thereby solving the problems in the prior art.

[0114] In practical applications, by combining the data processing device 20 with other related technologies, such as adopting a distributed layout for the data processing device 20, or combining related cloud computing technologies, and deploying the data processing device 20 in the cloud, etc. are all within the protection scope of the present application. In addition, the data processing device 20 can also be combined with specific hardware devices to improve the commercial value of the hardware device, etc.

[0115] The data processing device 20 may further include a target emotion base model generation unit, configured to obtain electroencephalogram data of multiple babies of the same gender and age in the target emotion; calculate the average data amplitude, the average duration of the target emotion, and the average attack frequency of the target emotion based on each electroencephalogram data as the target emotion base model; and,

[0116] Based on the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion, generating the actual model of the target emotion of the baby to be tested may specifically include:

[0117] Based on the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion, calculating the second data average amplitude, the second average duration, and the second average attack frequency under the target emotion as the actual model of the target emotion.

[0118] Among them, based on the basic model of the target emotion and the actual model of the target emotion, determining the classification category to which the baby to be tested belongs may specifically include:

[0119] Based on the magnitudes of the second data average amplitude and the data average amplitude, the magnitudes of the second average duration and the average duration, and the magnitudes of the second average attack frequency and the average attack frequency, determining the attack characteristics of the target emotion of the baby to be tested, where the attack characteristics are used to characterize whether there are obvious deviations in the emotional fluctuation range, duration, and attack frequency of the target emotion of the baby to be tested;

[0120] Based on the attack characteristics of the target emotion of the baby to be tested, determining the classification category to which the baby to be tested belongs.

[0121] Among them, the attack characteristics are specifically the first ratio of the second data average amplitude to the data average amplitude, the second ratio of the second average duration to the average duration, and the third ratio of the second average attack frequency to the average attack frequency; and,

[0122] Based on the attack characteristics of the target emotion of the baby to be tested, determining the classification category to which the baby to be tested belongs may specifically include:

[0123] Based on the numerical intervals into which the first ratio, the second ratio, and the third ratio respectively fall, determining the classification category to which the baby to be tested belongs.

[0124] Among them, the health assessment function may specifically be: Q = a×A / A E +b×T / T E +c×F / F E ;

[0125] Among them, Q is the health assessment value calculated by the health assessment function; A is the actual amplitude of the data of the baby to be tested in the next monitoring cycle under the target emotion; A EFor the next monitoring cycle of the baby to be measured, the estimated amplitude of the data in the target emotion; T is the actual duration of the target emotion in the next monitoring cycle of the baby to be measured; T E For the next monitoring cycle of the baby to be measured, the estimated duration of the target emotion; F is the actual attack frequency of the target emotion in the next monitoring cycle of the baby to be measured; F E For the next monitoring cycle of the baby to be measured, the estimated attack frequency of the target emotion; a, b, and c are all undetermined weights;

[0126] Among them, obtaining the health assessment function corresponding to the classification category may specifically include: obtaining the values of a, b, and c in the health assessment function corresponding to the classification category to obtain the health assessment function.

[0127] Among them, determining the health assessment value of the baby to be measured according to the actual electroencephalogram data of the baby to be measured in the target emotion in the next monitoring cycle and the health assessment function may specifically include:

[0128] Determining the actual amplitude, the actual duration, and the actual attack frequency of the data according to the actual electroencephalogram data of the baby to be measured in the target emotion in the next monitoring cycle;

[0129] Determining the estimated amplitude, the estimated duration, and the estimated attack frequency of the data according to the electroencephalogram data of the baby to be measured in the target emotion in the previous N adjacent monitoring cycles;

[0130] Substituting the actual amplitude, the actual duration, and the actual attack frequency of the data, and the estimated amplitude, the estimated duration, and the estimated attack frequency of the data into the health assessment function Q = a×A / A E +b×T / T E +c×F / F E to calculate the health assessment value of the baby to be measured.

[0131] Among them, determining the actual amplitude, the actual duration, and the actual attack frequency of the data according to the actual electroencephalogram data of the baby to be measured in the target emotion in the next monitoring cycle may specifically include:

[0132] Obtaining the vibration range of each data in the actual electroencephalogram data as the actual amplitude A of the data;

[0133] Obtaining the duration of each target emotion attack in the actual electroencephalogram data and calculating the average value of each duration as the actual duration T;

[0134] The number of target emotion attacks of the infant to be tested in the next monitoring period is determined by the actual EEG data as the actual attack frequency F.

[0135] Wherein, determining the estimated amplitude of the data, the estimated duration and the estimated frequency of the attack according to the electroencephalogram data of the infant to be tested under the target emotion in the previous N adjacent monitoring cycles may specifically include:

[0136] For each of the first N proximate monitoring cycles, as the current proximate monitoring cycle, obtain the electroencephalogram data of the current proximate monitoring cycle under the target emotion as the current electroencephalogram data;

[0137] The amplitude A' of the current near monitoring cycle is obtained from the current EEG data to obtain the amplitude A' corresponding to each near monitoring cycle in the first N near monitoring cycles, which are recorded as A1', A2', ... A N ';

[0138] For A1', A2'...A N 'Perform curve fitting to obtain the amplitude trend function, and use the amplitude trend function to calculate A E ;

[0139] The target emotion duration T' of the current near monitoring cycle is obtained from the current EEG data to obtain the target emotion duration T' corresponding to each near monitoring cycle in the first N near monitoring cycles, which are recorded as T1', T2'...T N ';

[0140] For T1', T2'...T N 'Perform curve fitting to obtain the duration trend function, and use the duration trend function to calculate T E ;

[0141] The estimated attack frequency F' of the target emotion in the current near monitoring cycle is obtained from the current EEG data to obtain the estimated attack frequency F' corresponding to each near monitoring cycle in the first N near monitoring cycles, which are recorded as F1', F2'...F N ';

[0142] For F1', F2'...F N 'Perform curve fitting to obtain a frequency trend function, and use the frequency trend function to calculate F E .

[0143] Figure 3 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 3As shown in the figure, the electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communication bus 340. Among them, the processor 310, the communications interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logical instructions in the memory 330 to execute the data processing method based on the electroencephalogram of the baby provided in the embodiments of the present application. The method includes first obtaining a corresponding target emotion base model according to the gender and age of the baby to be tested. The target emotion base model is constructed based on the electroencephalogram data of multiple babies of the same gender and age in the target emotion. Then, according to the electroencephalogram data of the baby to be tested in the target emotion in the first N adjacent monitoring cycles, a target emotion actual model of the baby to be tested is generated, where N is a positive integer greater than or equal to 1. Then, according to the target emotion base model and the target emotion actual model, the classification category to which the baby to be tested belongs is determined. Then, the health assessment function corresponding to the classification category is obtained. Then, according to the actual electroencephalogram data of the baby to be tested in the target emotion in the next monitoring cycle and the health assessment function, the health assessment value of the baby to be tested is determined, which is used to evaluate the health of the baby to be tested.

[0144] Obviously, since the processor 310 can call the logical instructions in the memory 330 to execute the method provided in the embodiments of the present application. In this method, the target emotion actual model generated by combining the electroencephalogram data of the baby to be tested in the target emotion in the first N adjacent monitoring cycles is used to determine the classification category to which the baby to be tested belongs, and then the corresponding health assessment function is determined, and its health assessment value is calculated. This method can improve the accuracy compared with the general model method, thus solving the problems in the prior art.

[0145] In addition, when the logical instructions in the above-mentioned memory 330 are implemented in the form of software functional units and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0146] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the data processing method based on the electroencephalogram of infants provided in the embodiments of the present application. The method includes first obtaining a corresponding target emotion base model according to the gender and age of the infant to be tested. The target emotion base model is constructed based on the electroencephalogram data of multiple infants of the same gender and age in the target emotion. Then, according to the electroencephalogram data of the infant to be tested in the target emotion in the previous N adjacent monitoring cycles, a target emotion actual model of the infant to be tested is generated, where N is a positive integer greater than or equal to 1. Then, according to the target emotion base model and the target emotion actual model, the classification category to which the infant to be tested belongs is determined. Then, the health assessment function corresponding to the classification category is obtained. Then, according to the actual electroencephalogram data of the infant to be tested in the target emotion in the next monitoring cycle and the health assessment function, the health assessment value of the infant to be tested is determined, which is used to evaluate the health degree of the infant to be tested.

[0147] Obviously, since when the computer program is executed by a processor, the computer can execute the method provided in the embodiments of the present application. In this method, the target emotion actual model generated by combining the electroencephalogram data of the infant to be tested in the target emotion in the previous N adjacent monitoring cycles is used to determine the classification category to which the infant to be tested belongs, and then the corresponding health assessment function is determined, and its health assessment value is calculated. This method can improve the accuracy compared with the general model method, thus solving the problems in the prior art.

[0148] On the other hand, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is the measurement method of the horizontal temperature field provided in the embodiments of the present application.

[0149] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0150] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data processing method based on infant electroencephalogram, characterized in that, Including: Obtain a corresponding target emotion base model according to the gender and age range of the baby to be tested, where the target emotion base model is constructed based on the electroencephalogram (EEG) data of multiple babies of the same gender and age range under a target emotion; Generate a target emotion actual model of the baby to be tested according to the EEG data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion, where N is a positive integer greater than or equal to 1; Determine the classification category to which the baby to be tested belongs according to the target emotion base model and the target emotion actual model; Obtain the health assessment function corresponding to the classification category; Determine the health assessment value of the baby to be tested according to the actual EEG data of the baby to be tested in the next monitoring cycle under the target emotion and the health assessment function, for evaluating the health level of the baby to be tested.

2. The method according to claim 1, characterized in that The method further includes: Obtain the EEG data of multiple babies of the same gender and age range under the target emotion; Based on each EEG data, calculate the average data amplitude, the average duration, and the average attack frequency of the target emotion under the target emotion as the target emotion base model; and Generating the target emotion actual model of the baby to be tested according to the EEG data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion specifically includes: According to the EEG data of the baby to be tested in the previous N adjacent monitoring cycles under the target emotion, calculate the second average data amplitude, the second average duration, and the second average attack frequency of the target emotion as the target emotion actual model.

3. The method according to claim 2, characterized in that, Determining the classification category to which the baby to be tested belongs according to the target emotion base model and the target emotion actual model specifically includes: Determine the attack characteristics of the target emotion of the baby to be tested according to the magnitude of the second average data amplitude and the average data amplitude, the magnitude of the second average duration and the average duration, and the magnitude of the second average attack frequency and the average attack frequency. The attack characteristics are used to characterize whether there are obvious deviations in the emotional fluctuation range, duration, and attack frequency of the target emotion of the baby to be tested; Determine the classification category to which the baby to be tested belongs according to the attack characteristics of the target emotion of the baby to be tested.

4. The method according to claim 3, wherein Wherein, The attack characteristics are specifically the first ratio of the second average data amplitude to the average data amplitude, the second ratio of the second average duration to the average duration, and the third ratio of the second average attack frequency to the average attack frequency; And Determining the classification category to which the baby to be tested belongs according to the attack characteristics of the target emotion of the baby to be tested specifically includes: Determine the classification category to which the baby to be tested belongs according to the numerical intervals into which the first ratio, the second ratio, and the third ratio respectively fall.

5. The method according to claim 1, characterized in that, The specific health assessment function is: Q = a×A / A E + b×T / T E + c×F / F E ; Among them, Q is the health assessment value calculated by the health assessment function; A is the actual amplitude of the data of the to-be-tested infant in the next monitoring period under the target emotion; A E is the predicted amplitude of the data of the to-be-tested infant in the next monitoring period under the target emotion; T is the actual duration of the target emotion in the next monitoring period of the to-be-tested infant; T E is the predicted duration of the target emotion in the next monitoring period of the to-be-tested infant; F is the actual attack frequency of the target emotion in the next monitoring period of the to-be-tested infant; F E is the predicted attack frequency of the target emotion in the next monitoring period of the to-be-tested infant; a, b, and c are all undetermined weights.

6. The method according to claim 5, characterized in that Obtain the health assessment function corresponding to the classification category, specifically including: obtaining the values of a, b, and c in the health assessment function corresponding to the classification category to obtain the health assessment function.

7. The method according to claim 5, characterized in that, Determine the health assessment value of the baby to be tested according to the actual electroencephalogram data of the baby to be tested in the next monitoring period under the target emotion and the health assessment function, specifically including: Determine the actual amplitude, the actual duration, and the actual attack frequency of the data according to the actual electroencephalogram data of the baby to be tested in the next monitoring period under the target emotion; Determine the estimated amplitude, the estimated duration, and the estimated attack frequency of the data according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring periods under the target emotion; Substitute the actual amplitude, the actual duration, and the actual attack frequency of the data, as well as the predicted amplitude, the predicted duration, and the predicted attack frequency of the data into the health assessment function Q = a×A / A E +b×T / T E +c×F / F E , so as to calculate the health assessment value of the baby to be tested.

8. The method according to claim 7, wherein Determine the actual amplitude, the actual duration, and the actual attack frequency of the data according to the actual electroencephalogram data of the baby to be tested in the next monitoring period under the target emotion, specifically including: Obtain the vibration range of each data in the actual electroencephalogram data as the actual amplitude A of the data; Obtain the duration of each target emotion attack in the actual electroencephalogram data and calculate the average value of each duration as the actual duration T; Determine the number of target emotion attacks in the next monitoring period of the baby to be tested through the actual electroencephalogram data as the actual attack frequency F.

9. The method according to claim 7, wherein Determine the estimated amplitude, the estimated duration, and the estimated attack frequency of the data according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring periods under the target emotion, specifically including: For each of the previous N adjacent monitoring periods, as the current adjacent monitoring period, obtain the electroencephalogram data of the current adjacent monitoring period under the target emotion as the current electroencephalogram data; Obtain the amplitude A’ of the current adjacent monitoring period from the current electroencephalogram data, so as to obtain the amplitudes A’ corresponding to each of the previous N adjacent monitoring periods, denoted as A1’, A2’......A N ’; For A1', A2'...A N 'Perform curve fitting to obtain the amplitude trend function, and use the amplitude trend function to calculate A E ; Obtain the target emotion duration T' of the current adjacent monitoring period from the current electroencephalogram data, so as to obtain the target emotion duration T' corresponding to each of the previous N adjacent monitoring periods, denoted as T1', T2'......T N '; For T1', T2'...T N 'Perform curve fitting to obtain the duration trend function, and use the duration trend function to calculate T E ; Obtain the estimated onset frequency F' of the target emotion in the current adjacent monitoring period from the current electroencephalogram data, so as to obtain the estimated onset frequencies F' corresponding to each of the previous N adjacent monitoring periods, denoted as F1', F2'......F N '; For F1', F2'...F N 'Perform curve fitting to obtain a frequency trend function, and use the frequency trend function to calculate F E .

10. A data processing device based on infant electroencephalogram, characterized in that, Include: An acquisition unit, configured to obtain a corresponding target emotion basic model according to the gender and age of the baby to be tested, where the target emotion basic model is constructed according to the electroencephalogram data of multiple babies of the same gender and age under the target emotion; A generation unit, configured to generate a target emotion actual model of the baby to be tested according to the electroencephalogram data of the baby to be tested in the previous N adjacent monitoring periods under the target emotion, where N is a positive integer greater than or equal to 1; A determination unit, configured to determine the classification category to which the baby to be tested belongs according to the target emotion basic model and the target emotion actual model; A second acquisition unit, configured to obtain the health assessment function corresponding to the classification category; A second determination unit, configured to determine the health assessment value of the baby to be tested according to the actual electroencephalogram data of the baby to be tested in the next monitoring period under the target emotion and the health assessment function, and is used to evaluate the health degree of the baby to be tested.

Images