Brain function state detection method, device and equipment based on electroencephalogram

Abstract

The present invention discloses a method, device and equipment for detecting brain function state based on electroencephalogram, and relates to the technical field of electroencephalogram signal data processing. The idea of ​​the present invention is not to directly process electroencephalogram signals, but to slice the selected lead electroencephalogram data, and use a machine learning algorithm to predict the data state, and then perform differential and dynamic root mean square calculations on the target data segments, thereby removing the interference of low-frequency disturbances, and there is no need to standardize the signal, so as to realize automatic and accurate identification of different data states in the electroencephalogram, making the indicator calculation process more intuitive, simple, low in computing power requirements and high in accuracy. The present invention can automatically interpret electroencephalogram data of any length, filling the gap in the automatic interpretation of electroencephalogram data state in this scenario, and realizing the estimation of burst suppression ratio, so as to assist in the diagnosis and treatment of doctors, thereby greatly reducing the workload of clinicians and improving their work efficiency and diagnosis and treatment accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of electroencephalogram (EEG) signal data processing, and in particular to a method, device and equipment for detecting brain function status based on an electroencephalogram (EEG). Background Art

[0002] Electroencephalogram (EEG) is a time domain signal containing multiple leads. The amplitude unit of the time domain signal is or , the time domain signals of different leads correspond to the cerebral cortical electrical signals collected from different brain regions. The term burst suppression describes an electroencephalogram (EEG) pattern that consists of a continuous alternation between high-voltage slow waves (occasionally sharp waves) and low-voltage (or suppressed) EEG activity.

[0003] At present, the interpretation of burst suppression in EEG mainly relies on the clinical experience of clinicians. When faced with a large amount of EEG data, this manual interpretation based on experience is difficult to be comprehensive and accurate, and it is even more difficult to accurately provide specific indicator data such as burst suppression time, burst suppression ratio, and burst peak. In particular, Figure 1 The five burst suppression EEG characteristics shown in ①~⑤ are schematic diagrams. Different patients exhibit different burst suppression EEG characteristics at different times and in different states. The superposition of irregular high-frequency and low-frequency interference makes it extremely difficult to standardize the EEG signals if they are directly analyzed in the time-frequency domain.

[0004] Therefore, there is no effective solution in the industry for automatically detecting burst suppression in EEG signal data. Summary of the invention

[0005] In view of the above, the present invention aims to provide a method, device and equipment for detecting brain function status based on electroencephalogram to solve the specific problems mentioned above.

[0006] The technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention provides a method for detecting brain function status based on electroencephalogram, comprising:

[0008] Corresponding to the established assessment requirements, different numbers of lead sites are selected in combination with the spatial distribution of the leads;

[0009] Acquire corresponding EEG data based on the selected lead site, and slice the acquired EEG data to obtain a number of EEG signal data segments;

[0010] Combine the EEG signal data segments with the pre-trained EEG burst suppression state classification model for prediction processing;

[0011] The target EEG signal data predicted to be in a burst suppression state is differentiated and the dynamic root mean square is calculated, the burst segment and the suppression segment are determined using the calculation results, and the burst suppression ratio is obtained based on the burst segment and the suppression segment.

[0012] In at least one possible implementation, slicing the acquired EEG data includes: slicing the EEG data corresponding to the lead site in the time dimension according to a preset capture time length and overlap time length.

[0013] In at least one possible implementation manner, the process of obtaining the burst suppression ratio specifically includes:

[0014] Obtaining a differential signal of the target EEG signal data, and calculating a dynamic root mean square signal of the differential signal;

[0015] Using the mean value of the differential signal and the preset standard deviation, the state demarcation threshold is obtained;

[0016] Determine a portion of the target EEG signal data segment where the dynamic root mean square signal is higher than the state demarcation threshold as a burst segment, and determine a portion where the dynamic root mean square signal is lower than the state demarcation threshold as a suppression segment;

[0017] The burst suppression ratio is obtained by accumulating the burst and suppression times respectively and calculating the ratio between them.

[0018] In at least one possible implementation, the target EEG signal data is obtained by:

[0019] The EEG data of each selected lead in the EEG signal data segment predicted to be in the burst suppression state are spliced ​​in the time dimension to obtain the target EEG signal data.

[0020] In at least one possible implementation, the training method of the EEG burst inhibition state classification model includes:

[0021] By marking the states and lead positions of EEG signal data segments, a data set is constructed;

[0022] Based on the machine learning model, the time domain information of the EEG signal data segment and the corresponding lead position label are used as the model input, and the state label representing whether the burst is suppressed or not is used as the model output to build the architecture of the EEG state classification model;

[0023] Use the constructed model architecture and dataset for training.

[0024] In at least one possible implementation, selecting different numbers of lead sites includes:

[0025] If it is a preset primary rapid assessment, the information of some leads in the frontal area, temporal area, parietal area, occipital area and central axis area of ​​the head are selected;

[0026] If it is a preset intermediate assessment, the information of all leads in the frontal, temporal, parietal, occipital and axial regions of the head are selected;

[0027] If it is the preset advanced comprehensive assessment, the full-time information of all the leads in the head will be selected.

[0028] In a second aspect, the present invention provides a device for detecting brain function status based on electroencephalogram, comprising:

[0029] The lead selection module is used to select different numbers of lead sites according to the established evaluation requirements and the spatial distribution of the leads;

[0030] The EEG data slicing module is used to obtain the corresponding EEG data based on the selected lead site, and slice the obtained EEG data to obtain a number of EEG signal data segments;

[0031] A data state prediction module is used to perform prediction processing on the EEG signal data segment in combination with a pre-trained EEG burst suppression state classification model;

[0032] The EEG data characteristic index calculation module is used to perform differential and dynamic root mean square calculation on the target EEG signal data predicted to be in a burst suppression state, determine the burst segment and the suppression segment using the calculation results, and obtain the burst suppression ratio based on the burst segment and the suppression segment.

[0033] In a third aspect, the present invention provides an electronic device, comprising: one or more processors, a memory and one or more computer programs, wherein the memory may adopt a non-volatile storage medium, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, and when the instructions are executed by the device, the electronic device performs the method as described in the first aspect or any possible implementation manner of the first aspect.

[0034] In a fourth aspect, the present invention provides a computer-readable storage medium, which stores a computer program, which, when executed on a computer, enables the computer to execute the method as described in the first aspect or any possible implementation of the first aspect.

[0035] In a fifth aspect, the present invention further provides a computer program product, which, when executed by a computer, is used to execute the method in the first aspect or any possible implementation of the first aspect. In a possible design of the fifth aspect, the relevant programs involved in the product may be stored in whole or in part on a memory packaged with the processor, or may be stored in whole or in part on a storage medium not packaged with the processor.

[0036] Compared with the prior art, the main design concept of the present invention is that it does not directly process EEG signals, but processes the selected lead EEG data in pieces, then uses the trained machine learning algorithm to predict whether each data segment is in a burst suppression state, then performs differential calculation on the data segments predicted to be in a burst suppression state, amplifies the EEG data features of different states, and proposes to describe the two data states of burst and suppression with a dynamic root mean square index, thereby removing the interference of low-frequency disturbances on the expression of EEG signal burst suppression, and there is no need to standardize the signal, so as to realize automatic and accurate identification of different data states in the EEG, making the burst suppression ratio calculation process more intuitive, simple, low in computing power and high in accuracy. The present invention can effectively automatically interpret EEG data of any length, evaluate the burst suppression state data segment and the time and lead position in the EEG data, fill the gap in the automatic interpretation of EEG data state in this scenario, and realize the estimation of burst suppression ratio, so as to assist in the diagnosis and treatment of doctors, thereby greatly reducing the workload of clinicians and improving their work efficiency and diagnosis and treatment accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:

[0038] Figure 1 A schematic diagram of the time domain waveforms of different burst suppression EEG signals provided by an embodiment of the present invention;

[0039] Figure 2 A schematic diagram of a data processing flow of a method for detecting brain function status based on electroencephalogram provided in an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of the spatial distribution of leads provided by an embodiment of the present invention;

[0041] Figure 4 A schematic diagram of multi-lead EEG signal data slices provided by an embodiment of the present invention;

[0042] Figure 5 A schematic diagram of the architecture of the EEG burst suppression state classification model provided by an embodiment of the present invention;

[0043] Figure 6 A schematic diagram of a flow chart of a burst suppression ratio estimation method provided by an embodiment of the present invention;

[0044] Figure 7 A schematic diagram of the burst suppression ratio estimation process provided by an embodiment of the present invention;

[0045] Figure 8 A block diagram of a brain function state detection device based on electroencephalogram provided by an embodiment of the present invention;

[0046] Fig. 9 A schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0047] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be interpreted as limiting the present invention.

[0048] The present invention proposes an embodiment of a method for detecting brain function status based on electroencephalogram. Specifically, Figure 2 As shown, it includes: the offline preparation stage mainly used for model training and the online interpretation stage after the trained model is deployed, which can also be understood as the real-time processing stage.

[0049] In the overall implementation framework of the offline preparation stage and the online interpretation stage, steps 1 and 2 are similar, so a unified explanation is given here:

[0050] Step 1. Lead information selection (lead optimization): In order to match the assessment requirements of different levels of burst suppression, the burst suppression assessment requirements can be divided into three situations in advance: primary rapid assessment, intermediate assessment and advanced comprehensive assessment. These three burst suppression assessment requirements correspond to different lead positions, namely:

[0051] (1) Primary rapid assessment. In order to reduce the amount of calculation and improve the efficiency of interpretation, this type of assessment requires only the selection of some leads in the frontal, temporal, parietal, occipital and central axis areas of the head, as reference Figure 3 Indicatively, the leads at positions F5, F6, T7, T8, C3, C4, PO7, and PO8 may be mainly included as the objects to be evaluated;

[0052] (2) Intermediate assessment. This type of assessment requires a balance between analysis efficiency and result accuracy. Therefore, all leads in the frontal, temporal, parietal, occipital, and central axis areas of the head can be selected. Figure 3Schematically, the evaluation objects may mainly include AF7, F7, F5, F3, AF8, F4, F6, F8, FT7, T7, TP7, FT8, T8, TP8, C3, C1, CP3, CP1, C2, C4, CP2, CP4, PO3, PO7, O1, PO4, PO8, O2, Fpz, Fz, FCz, Cz, CPz, Pz, POz, and Oz position leads;

[0053] (3) Advanced comprehensive assessment: This type of assessment requires comprehensiveness and accuracy, and strives to conduct an assessment analysis such as Figure 3 The full-time information of all leads is shown.

[0054] Step 2, data slicing: For different patients and different situations, the length of EEG time required for interpretation is different each time. In order to normalize the scale of EEG digital signal processing and meet the unified requirements of subsequent algorithms for data length, the present invention believes that it is necessary to slice the EEG data in the time dimension. The specific slicing method includes but is not limited to the following method: Figure 4 Illustration: The EEG data of the selected lead site is sliced ​​in the time dimension with 10s (optional) as the unit length and 2s (optional) as the overlap length, and the starting time of the first segment of EEG data is obtained, such as 0 to 10s, and so on. The starting time of the second segment of EEG data is 8 to 18s, and the starting time of the third segment of EEG data is 16 to 26s, etc. Among them, slicing the data with 2s as the overlap length can effectively improve the integrity of the boundary features of the sliced ​​data.

[0055] Therefore, the subsequent steps of the offline preparation phase are first described in detail here.

[0056] Step 3, establish a data set: for several segments of EEG data collected in advance, associate the slice data with whether the slice data is in a burst suppression state. In actual operation, it is possible to judge whether the slice time of the EEG data corresponding to each lead is in a burst suppression state based on expert knowledge. If a certain EEG signal data segment is judged to be in a burst suppression state, the burst suppression label corresponding to the EEG data of the slice time is marked as 1, otherwise the burst suppression label corresponding to the EEG data of the slice time is marked as 0; at the same time, the lead position label can also be marked, such as lead 1 to lead 8 can correspond to the left frontal area, right frontal area, left temporal area, right temporal area, left parietal area, right parietal area, left occipital area, and right occipital area of ​​the brain respectively.

[0057] Step 4, model construction: Build a machine learning network model, such as but not limited to CNN, LSTM, SVM and other machine learning models. The input of the model is the i-th data slice and the position label corresponding to each lead in the i-th data slice, and the output is the label representing the burst suppression state. Figure 5For reference: The machine learning network model can include two parts. The first part is a nonlinear operator, which can be a Gaussian operator, a sigmoid operator, etc. Each nonlinear operator receives the time domain information of a single lead in the i-th period and the position label of the lead, and outputs the calculation result to the multi-layer neural network in the second part; the input of the multi-layer neural network in the second part is the final calculation result of the nonlinear operator, and the output is the burst suppression label. Figure 5 The model architecture and construction method shown are not only simple in structure and reduce the computing power requirements of the hardware, but also take into account the time domain characteristics of the EEG signal data and the spatial distribution characteristics of the selected leads.

[0058] Step 5, model training: Use the machine learning model established in step 4 and the data set established in step 3 to train according to mature machine model training methods to obtain a trained EEG state classification model and deploy it.

[0059] Refer to Figure 2 , following the above steps 1 and 2, continue to explain the specific implementation process of the online interpretation stage:

[0060] Step 6, perform model prediction on slice data: the EEG data that needs to be automatically interpreted is processed through steps 1 and 2 to obtain the selected lead EEG data after slicing, and then the data is input into the machine learning model trained in step 5 to obtain the prediction results. The number of prediction results is equal to the number of slice segments. The prediction results are the burst inhibition state prediction labels corresponding to the selected lead EEG data after each slice. 1 means that the EEG data at the slice time is in the burst inhibition state, and 0 means that the EEG data at the slice time is in the non-burst inhibition state.

[0061] Step 7, burst suppression ratio estimation: reference Figure 6 As shown in the figure, if after step 6 prediction If the EEG data of the slice time is in the burst suppression state, then The EEG data of each selected lead in the segment EEG data are spliced ​​in the time dimension to obtain spliced ​​EEG data It should be noted that the present invention proposes splicing the EEG data of each selected lead, which can reduce the impact of abnormal data of individual leads on the overall result. In addition, the data segments in the non-burst suppression state are skipped.

[0062] Continuing from the previous text, then Differential to get differential signal , specific The method to obtain is:

[0063]

[0064] Use the following formula to find Dynamic RMS signal :

[0065]

[0066] in, For the The time dynamic RMS signal of the segment slice data, d is the cumulative step length of the dynamic RMS signal.

[0067] Then, the differential signal can be calculated by The mean + standard deviation of the burst suppression cutoff value is obtained ;

[0068] Combination Figure 7 The schematic diagram of the effect is as follows. and For comparison, Lower than The part is the inhibition segment , Higher than The part is the burst segment , and The unit is seconds. The burst and suppression time are accumulated and compared to get the target index. :

[0069]

[0070] Step 8: until the above processing of all slice data (i.e. the aforementioned EEG signal data segments) is traversed and completed.

[0071] Finally, it should be pointed out that at present, there are two main types of burst suppression ratio calculations: first, interpretable numerical calculations and calibration of burst time and suppression time; second, the use of deep learning to perform black-box "brute force calculations" on EEG signals to obtain the burst suppression ratio. However, in the preliminary analysis, the present invention believes that these two methods have obvious shortcomings: although the first method has a small amount of calculation, the accuracy is very low; although the second method has a high accuracy, the amount of calculation is huge and the computing power requirements are high, resulting in not only high data processing costs, but also long processing time, which is especially not suitable for long-term EEG signal data processing.

[0072] Compared with the existing solutions in the industry, the present invention takes a different approach and does not directly process the EEG signal. Instead, it differentiates the EEG signal in the aforementioned step 7, and then uses the differential method to amplify the difference between the burst feature and the inhibition feature of the EEG data in the time domain, so that the root mean square of the discharge part increases rapidly. At the same time, the design indicator "dynamic root mean square" describes the difference in the burst inhibition feature of the EEG signal in the time domain. The characteristic shown is that when the dynamic root mean square is higher than a predetermined threshold (burst inhibition boundary value), it indicates that the current state is a burst state, and when the dynamic root mean square is lower than a certain threshold, it indicates that the current state is an inhibition state.

[0073] The method proposed by the present invention can effectively remove the interference of low-frequency disturbances on the expression of burst suppression of EEG signals, and at the same time adjust the comparison benchmark from the floating time domain signal to the 0 axis, so that the signal does not need to be standardized, making the burst suppression ratio calculation process more intuitive, simple, low in computing power requirements and high in accuracy, and has significant advantages that cannot be matched by existing technologies in the analysis of long-term EEG signals. Although the scheme of the present invention is not directly used for diagnosis and treatment, the scheme of the present invention establishes a combined association between data slices and burst suppression ratios during the analysis and processing of long-term EEG signal data, so it can assist in presenting the development trend of burst suppression ratios at different stages, and thus play an objective and powerful supporting and auxiliary role for the subsequent actual diagnosis and treatment of doctors, such as the estimation of the attack period and the evaluation of the effect of medication.

[0074] In summary, the main design concept of the present invention is that it does not directly process EEG signals, but processes the selected lead EEG data in pieces, then uses the trained machine learning algorithm to predict whether each data segment is in a burst suppression state, then performs differential calculation on the data segments predicted to be in a burst suppression state, amplifies the EEG data features of different states, and proposes to describe the two data states of burst and suppression with a dynamic root mean square index, thereby removing the interference of low-frequency disturbances on the expression of EEG signal burst suppression, and there is no need to standardize the signal, so as to realize automatic and accurate identification of different data states in the EEG, making the burst suppression ratio calculation process more intuitive, simple, low in computing power and high in accuracy. The present invention can effectively automatically interpret EEG data of any length, evaluate the burst suppression state data segment and the time and lead position in the EEG data, fill the gap in the automatic interpretation of EEG data state in this scenario, and realize the estimation of the burst suppression ratio, so as to assist in the diagnosis and treatment of doctors, thereby greatly reducing the workload of clinicians and improving their work efficiency and diagnosis and treatment accuracy.

[0075] Corresponding to the above-mentioned embodiments and preferred solutions, the present invention also provides an embodiment of a brain function state detection device 800 based on electroencephalogram, such as Figure 8As shown, it may specifically include the following components:

[0076] The lead selection module 801 is used to select different numbers of lead sites in accordance with the established evaluation requirements and in combination with the spatial distribution of the leads;

[0077] The EEG data slicing module 802 is used to obtain corresponding EEG data based on the selected lead site, and slice the obtained EEG data to obtain a plurality of EEG signal data segments;

[0078] The data state prediction module 803 is used to perform prediction processing on the EEG signal data segment in combination with the pre-trained EEG burst suppression state classification model;

[0079] The EEG data characteristic index calculation module 804 is used to perform differential and dynamic root mean square calculation on the target EEG signal data predicted to be in a burst suppression state, determine the burst segment and the suppression segment using the calculation results, and obtain the burst suppression ratio based on the burst segment and the suppression segment.

[0080] It should be understood that the above Figure 8 The division of the various components in the EEG-based brain function state detection device shown is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these components can all be implemented in the form of software calling through processing elements; some components can also be implemented in the form of software calling through processing elements, and some components can be implemented in the form of hardware. For example, one of the above modules can be a separately established processing element, or it can be integrated in a chip of an electronic device. The implementation of other components is similar. In addition, all or part of these components can be integrated together, or they can be implemented independently. In the implementation process, each step of the above method or the above components can be completed by the hardware integrated logic circuit in the processor element or the instructions in the form of software.

[0081] For example, the above components may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more microprocessors (DSP), or one or more field programmable gate arrays (FPGA). For another example, these components may be integrated together to be implemented in the form of a system-on-a-chip (SOC).

[0082] Based on the above embodiments and their preferred solutions, those skilled in the art can understand that, in actual operation, the technical concept involved in the present invention can be applied to a variety of implementations. The present invention uses the following carrier as a schematic illustration:

[0083] (1) An electronic device. The device may specifically include: one or more processors, a memory, and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, and when the instructions are executed by the device, the device performs the steps / functions of the aforementioned embodiment or equivalent implementation.

[0084] The electronic device may specifically be electronic device related to computers, such as but not limited to various computing terminals and electronic products.

[0085] Specifically, the processor, communication interface, and memory can all communicate with each other through a communication bus. Among them, the processor may be a central processing unit CPU, DSP, microcontroller or digital signal processor, and may also include GPU, embedded neural network processor (Neural-network Process Units; hereinafter referred to as: NPU) and image signal processor (Image Signal Processor; hereinafter referred to as: ISP). The processor may also include a specific integrated circuit ASIC, or one or more integrated circuits configured to implement the embodiment of the present invention, etc. In addition, the processor may have the function of operating one or more software programs, and the software programs may be stored in a storage medium such as a memory; and the aforementioned memory / storage medium may include: non-volatile memory (Non-Volatile Memory), such as a non-removable disk, U disk, mobile hard disk, optical disk, etc., as well as read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (RandomAccess Memory; hereinafter referred to as: RAM), etc.

[0086] Although the present invention does not limit the specific form of the electronic device, for the purpose of illustrative introduction, Fig. 9As shown, the present invention provides a schematic diagram of the structure of an embodiment of a brain function state detection device based on an electroencephalogram. Specifically, the brain function state detection device 900 based on an electroencephalogram includes a processor 910 and a memory 930. Among them, the processor 910 and the memory 930 can communicate with each other through an internal connection path to transmit control and / or data signals, the memory 930 is used to store a computer program, and the processor 910 is used to call and run the computer program from the memory 930. The above-mentioned processor 910 and the memory 930 can be combined into a processing device, and more commonly, they are independent components from each other. The processor 910 is used to execute the program code stored in the memory 930 to implement the above-mentioned functions. In specific implementation, the memory 930 can also be integrated in the processor 910, or, independent of the processor 910.

[0087] In addition, in order to make the function of the EEG-based brain function state detection device 900 more complete, the EEG-based brain function state detection device 900 may also include one or more of an input unit 960, a display unit 970, an audio circuit 980, a camera 990, and a sensor 901, and the audio circuit may also include a speaker 982, a microphone 984, etc. Among them, the display unit 970 may include a display screen.

[0088] Furthermore, the above-mentioned EEG-based brain function state detection device 900 may also include a power supply 950 for providing electrical energy to various devices or circuits in the EEG-based brain function state detection device 900 .

[0089] It should be understood that Fig. 9 The EEG-based brain function state detection device 900 shown can implement each process of the method provided in the aforementioned embodiment. The operation and / or function of each component in the EEG-based brain function state detection device 900 can be respectively used to implement the corresponding process in the aforementioned method embodiment. For details, please refer to the description of the embodiments of the method, device, etc. in the previous text. To avoid repetition, the detailed description is appropriately omitted here.

[0090] It should be understood that Fig. 9 The processor 910 in the EEG-based brain function state detection device 900 shown can be a system-on-chip SOC, and the processor 910 can include a central processing unit (Central Processing Unit; hereinafter referred to as: CPU), and can further include other types of processors, such as: a graphics processor (Graphics Processing Unit; hereinafter referred to as: GPU), etc., which will be specifically introduced below.

[0091] In summary, the various processors or processing units within the processor 910 can work together to implement the previous method flow, and the corresponding software programs of the various processors or processing units can be stored in the memory 930.

[0092] (2) A computer data storage medium on which a computer program or the above-mentioned device is stored, which, when the computer program or the above-mentioned device is executed, enables a computer to execute the steps / functions of the aforementioned embodiment or equivalent implementation.

[0093] In several embodiments provided by the present invention, any function, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer data storage medium. Based on this understanding, some technical solutions of the present invention are essentially or part of the technical solution that contributes to the prior art or part of the technical solution can be embodied in the form of the software product described below.

[0094] It is particularly important to point out that the storage medium may refer to a server or a similar computer device, specifically, a storage device in a server or a similar computer device storing the aforementioned computer program or the aforementioned apparatus.

[0095] (3) A computer program product (which may include the above-mentioned apparatus), which, when executed on a terminal device, enables the terminal device to execute the brain function state detection method based on electroencephalogram of the above-mentioned embodiment or an equivalent implementation.

[0096] It can be seen from the description of the above implementation methods that those skilled in the art can clearly understand that all or part of the steps in the above implementation method can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the above computer program product may include but is not limited to APP.

[0097] In the embodiments of the present invention, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. A and B may be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c may represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c may be single or multiple.

[0098] Those skilled in the art will appreciate that the various modules, units, and method steps described in the embodiments disclosed in this specification can be implemented by a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0099] In addition, each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. In particular, for embodiments such as devices and equipment, since they are basically similar to method embodiments, the relevant parts can refer to the partial description of the method embodiments. The embodiments of the devices and equipment described above are merely schematic, wherein the modules, units, etc. described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed in multiple places, such as nodes of a system network. Specifically, some or all of the modules and units may be selected according to actual needs to achieve the purpose of the above-mentioned embodiment scheme. Those skilled in the art can understand and implement it without paying creative labor.

[0100] The above describes in detail the structure, features and effects of the present invention based on the embodiments shown in the drawings, but the above is only a preferred embodiment of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred methods can be reasonably combined and matched into a variety of equivalent schemes by those skilled in the art without departing from or changing the design ideas and technical effects of the present invention; therefore, the present invention is not limited to the scope of implementation shown in the drawings, and all changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which still do not exceed the spirit covered by the specification and drawings, should be within the protection scope of the present invention.

Claims

1. A method for detecting brain function status based on electroencephalogram, characterized in that: include: Corresponding to the established assessment needs, combined with the spatial distribution of the leads, different numbers of lead sites are selected, including: if it is a preset primary rapid assessment, the information of some leads in the frontal area, temporal area, parietal area, occipital area and central axis area of ​​the head is selected; if it is a preset intermediate assessment, the information of all leads in the frontal area, temporal area, parietal area, occipital area and central axis area of ​​the head is selected; if it is a preset advanced comprehensive assessment, the full-time information of all leads of the head is selected; Acquire corresponding EEG data based on the selected lead site, and slice the acquired EEG data to obtain a number of EEG signal data segments; Combine the EEG signal data segments with the pre-trained EEG burst suppression state classification model for prediction processing; The target EEG signal data predicted to be in a burst suppression state are differentiated and the dynamic root mean square is calculated, the burst segment and the suppression segment are determined using the calculation results, and the burst suppression ratio is obtained based on the burst segment and the suppression segment, including: obtaining the differential signal of the target EEG signal data, and calculating the dynamic root mean square signal of the differential signal; obtaining the state boundary threshold using the mean value of the differential signal and a preset standard deviation; determining the portion of the target EEG signal data segment where the dynamic root mean square signal is higher than the state boundary threshold as the burst segment, and determining the portion where the dynamic root mean square signal is lower than the state boundary threshold as the suppression segment; respectively accumulating the time of the burst segment and the suppression segment and obtaining the ratio between the two to obtain the burst suppression ratio.

2. The method for detecting brain function status based on electroencephalogram according to claim 1, characterized in that: The slicing of the acquired EEG data includes: slicing the EEG data corresponding to the lead site in the time dimension according to a preset interception time length and overlap time length.

3. The method for detecting brain function status based on electroencephalogram according to claim 1, characterized in that: The target EEG signal data is obtained by: The EEG data of each selected lead in the EEG signal data segment predicted to be in the burst suppression state are spliced ​​in the time dimension to obtain the target EEG signal data.

4. The method for detecting brain function status based on electroencephalogram according to claim 1, characterized in that: The training method of the EEG burst inhibition state classification model includes: By marking the states and lead positions of EEG signal data segments, a data set is constructed; Based on the machine learning model, the time domain information of the EEG signal data segment and the corresponding lead position label are used as the model input, and the state label representing whether the burst is suppressed or not is used as the model output to build the architecture of the EEG state classification model; Use the constructed model architecture and dataset for training.

5. A brain function state detection device based on electroencephalogram using the detection method according to any one of claims 1 to 4, characterized in that: include: The lead selection module is used to select different numbers of lead sites according to the established evaluation requirements and the spatial distribution of the leads; The EEG data slicing module is used to obtain the corresponding EEG data based on the selected lead site, and slice the obtained EEG data to obtain a number of EEG signal data segments; A data state prediction module is used to perform prediction processing on the EEG signal data segment in combination with a pre-trained EEG burst suppression state classification model; The EEG data characteristic index calculation module is used to perform differential and dynamic root mean square calculation on the target EEG signal data predicted to be in a burst suppression state, determine the burst segment and the suppression segment using the calculation results, and obtain the burst suppression ratio based on the burst segment and the suppression segment.

6. An electronic device, characterized in that: include: One or more processors, a memory and one or more computer programs, wherein the one or more computer programs are stored in the memory, and the one or more computer programs include instructions, which, when executed by the electronic device, enable the electronic device to execute the brain function state detection method based on electroencephalogram according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for detecting brain function status based on electroencephalogram (EEG) as described in any one of claims 1 to 4 is implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for detecting brain function status based on electroencephalogram (EEG) as described in any one of claims 1 to 4 is implemented.

Images