Non-stationary signal cross-frequency coupling frequency domain feature acquisition method and system

Abstract

The invention discloses a non-stationary signal cross-frequency coupling frequency domain feature acquisition method and system, and relates to the technical field of biomedical signal processing.The method comprises the steps that multi-channel electroencephalogram signals before and after an event occur are acquired, and the electroencephalogram signals are cut into multiple segments according to preset duration; the electroencephalogram signal is a non-stationary signal; the electroencephalogram signals of each segment are decomposed by adopting an intrinsic mode function (IMF), and electroencephalogram signals of different frequency bands are determined; according to the electroencephalogram signals of the different frequency bands, determining coupling values among the different frequency bands, matching the coupling values among the different frequency bands with the occurrence of the event, and obtaining cross-frequency coupling values before and after the occurrence of the event; the cross-frequency coupling values before and after the event occur are compared, and the cross-frequency coupling values exceeding a preset baseline value are used as final frequency domain features.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of biomedical signal processing, and particularly relates to a non-stationary signal cross-frequency coupling frequency domain feature acquisition method and system. BACKGROUND

[0002] Event-Related Potentials (ERP) as the brain electrical signals matching specific event occurrence, its analysis is crucial to understand the information encoding mechanism of the brain in the event processing process.

[0003] Although the electroencephalogram signal is a time series signal, its frequency information is closely related to specific physiological functions, especially the cross-frequency coupling phenomenon between different frequency bands, which has been proven to be closely related to the cognitive function of animals.

[0004] However, the traditional time-frequency conversion method is limited by the Heisenberg uncertainty principle, and cannot accurately obtain the time domain and frequency domain information at the same time, which seriously affects the accuracy of cross-frequency coupling analysis. Especially when dealing with short duration and non-stationary event-related potential signals, high-resolution time domain and frequency domain analysis is particularly important. SUMMARY

[0005] The present application provides a non-stationary signal cross-frequency coupling frequency domain feature acquisition method and system to solve the above problems existing in the prior art, that is, how to improve the accuracy of cross-frequency coupling analysis in the prior art. The present application provides a non-stationary signal cross-frequency coupling frequency domain feature acquisition method, which comprises: Obtaining the multi-channel electroencephalogram signal before and after the event, and cutting the electroencephalogram signal into multiple segments according to a preset time length; the electroencephalogram signal is a non-stationary signal; Decomposing the electroencephalogram signal of each segment by using intrinsic mode function (IMF), to determine the electroencephalogram signal of different frequency bands; According to the electroencephalogram signal of different frequency bands, determining the coupling values between different frequency bands, matching the coupling values between different frequency bands with the event occurrence, and obtaining the cross-frequency coupling values before and after the event occurrence; By comparing the cross-frequency coupling values before and after the event occurrence, the cross-frequency coupling values exceeding the preset baseline value are taken as the final frequency domain features.

[0006] Optionally, obtaining the multi-channel electroencephalogram signal before and after the event, and cutting the electroencephalogram signal into multiple segments according to a preset time length; the electroencephalogram signal is a non-stationary signal; Decomposing the electroencephalogram signal of each segment by using intrinsic mode function (IMF), to determine the electroencephalogram signal of different frequency bands; According to the brain electrical signals of different frequency bands, coupling values between different frequency bands are determined, the coupling values between different frequency bands are matched with event occurrence, and cross-frequency coupling values before and after the event occurrence are obtained; By comparing the cross-frequency coupling values before and after the event occurrence, the cross-frequency coupling values exceeding a preset baseline value are taken as the final frequency domain features.

[0007] Optionally, the brain electrical signals of different frequency bands are determined by decomposing the brain electrical signals of each segment by using intrinsic mode functions (IMFs), and specifically comprising: The projection direction of the original brain electrical signals of each segment of the multi-channel is obtained, the projection signals are determined according to the projection direction, and the local extreme points corresponding to each projection signal are determined; The local extreme points are mapped back to the multi-dimensional channel to generate multi-dimensional envelope lines in each direction; The mean values of the multi-dimensional envelope lines in each direction are obtained, the mean values of the multi-dimensional envelope lines are subtracted from the original brain electrical signals, and the brain electrical signals of different frequency bands corresponding to each channel are determined.

[0008] Optionally, the multi-dimensional interpolation method is used to generate the multi-dimensional envelope lines in each direction.

[0009] Optionally, the brain electrical signals are brain electrical signals of a human or an experimental animal.

[0010] Optionally, the brain electrical signals of each segment are brain electrical signal segments with equal time lengths.

[0011] The application provides a non-stationary signal cross-frequency coupling frequency domain feature acquisition system, comprising: An acquisition module is configured to acquire brain electrical signals of a multi-channel before and after an event occurrence, and cut the brain electrical signals into a plurality of segments according to a preset time length; the brain electrical signals are non-stationary signals; A decomposition module is configured to decompose the brain electrical signals of each segment by using intrinsic mode functions (IMFs) to determine brain electrical signals of different frequency bands; A coupling analysis module is configured to determine coupling values between different frequency bands according to the brain electrical signals of different frequency bands, match the coupling values between different frequency bands with event occurrence, and obtain cross-frequency coupling values before and after the event occurrence; An extraction module is configured to compare the cross-frequency coupling values before and after the event occurrence, and take the cross-frequency coupling values exceeding a preset baseline value as the final frequency domain features.

[0012] The application provides a computer readable storage medium, the storage medium stores a computer program, and the computer program is executed by a processor to realize the non-stationary signal cross-frequency coupling frequency domain feature acquisition method.

[0013] The application provides a computer device, comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, and the processor implements the above-mentioned non-stationary signal cross-frequency coupling frequency domain feature acquisition method when executing the program.

[0014] Compared with the prior art, the application has the following beneficial effects: the application provides a non-stationary signal cross-frequency coupling frequency domain feature acquisition method, which decomposes the electroencephalogram signal into a series of modal functions, i.e., waves of different frequencies, through the adaptive signal decomposition method proposed in the application, so that the high-resolution accurate analysis of the event-related potential time domain and frequency domain can be realized, the limitation of the time-frequency conversion in the prior art is effectively avoided, and the accuracy of the cross-frequency coupling analysis is improved; meanwhile, the application can realize real-time cross-frequency coupling analysis and frequency feature extraction through seamless connection with the existing electroencephalogram recording device. BRIEF DESCRIPTION OF DRAWINGS

[0015] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the application and, together with the specification, serve to explain the principles of the application.

[0016] Figure 1 A flowchart of a non-stationary signal cross-frequency coupling frequency domain feature acquisition method provided for an embodiment of the application; Figure 2 A schematic diagram of the adaptive signal decomposition method for decomposing the electroencephalogram signal into different frequency bands provided for an embodiment of the application; Figure 3 A structural schematic diagram of a non-stationary signal cross-frequency coupling frequency domain feature acquisition device provided for an embodiment of the application; Figure 4 A computer device schematic diagram of a non-stationary signal cross-frequency coupling frequency domain feature acquisition method provided for an embodiment of the application. DETAILED DESCRIPTION

[0017] To make the objects, technical solutions and advantages of the application clearer, the technical solutions in the application will be described clearly and completely below with reference to the drawings in the application. Obviously, the described embodiments are only a part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0018] The technical solutions of the application and how the technical solutions of the application solve the above-mentioned technical problems will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described again in some embodiments. The embodiments of the application will be described below with reference to the drawings.

[0019] Embodiment 1 As Figure 1 And Figure 2 The non-stationary signal cross-frequency coupling frequency domain feature acquisition method shown in the embodiment includes: S1: acquiring multi-channel electroencephalogram signals before and after an event, and cutting the electroencephalogram signals into multiple segments according to a preset time length; the electroencephalogram signals are non-stationary signals.

[0020] Exemplarily, human or experimental animal electroencephalogram data can be acquired, the electroencephalogram signals are cut into multiple equal-length segments in chronological order by setting a cutting segment time length (involving time resolution), and finally the part of less than the time length is discarded, and the acquired original electroencephalogram signals are complex non-stationary signals.

[0021] S2: decomposing the electroencephalogram signals of each segment by using intrinsic mode function (IMF), to determine the electroencephalogram signals of different frequency bands.

[0022] Generally, as non-stationary signals, the electroencephalogram signals cannot be simultaneously accurately determined in time domain and frequency domain according to existing time-frequency conversion methods due to the limitation of Heisenberg uncertainty principle, that is, the time domain resolution is increased and the frequency domain resolution (high frequency part) is decreased, and vice versa. For the event-related potential signals with short duration and non-stationary, it is particularly important to simultaneously ensure the time domain resolution and the frequency domain resolution for cross-frequency coupling analysis. The Heisenberg uncertainty principle limitation on time domain resolution and frequency domain resolution can be effectively avoided by decomposing the electroencephalogram signals into a series of non-sinusoidal waves through an adaptive signal decomposition method.

[0023] Exemplarily, the electroencephalogram signals are decomposed into a group of intrinsic mode functions for representing intrinsic vibration of the signals by setting a frequency point step (involving frequency resolution), which can specifically include the following steps: The multi-channel electroencephalogram signals are decomposed as a whole, so as to ensure the mode correspondence and consistency of the decomposition results among the channels. The number of projection directions n is defined as the number of channels n, the signals are projected along each projection direction and the local extreme points of the projected signals are calculated; then, the extreme points are mapped back to the original n-dimensional space, and finally a multi-dimensional envelope curve is generated through multi-dimensional interpolation, the mean value of all direction envelope curves is calculated, the intrinsic mode function is obtained by subtracting the envelope curve mean value from the original signal, and the series of intrinsic mode functions are obtained by subtracting the intrinsic mode function from the original signal and repeating the above steps.

[0024] Exemplarily, the multi-brain region electroencephalogram signals can be decomposed into intrinsic mode functions by using the following formula: Wherein, v(t) is the original electroencephalogram signal, c i(t) is a multi-channel intrinsic mode function, where each channel corresponds to an intrinsic mode function component, r(t) is a multi-channel residual signal, and K is the number of intrinsic mode functions obtained by decomposition. Among them, the intrinsic mode function is used to represent the intrinsic vibration of the signal.

[0025] S3: According to the electroencephalogram of different frequency bands, the coupling values between different frequency bands are determined, and the coupling values between different frequency bands are matched with the event occurrence to obtain the cross-frequency coupling values before and after the event occurrence.

[0026] For example, the coupling calculation between the specified frequency bands can be set, and the characteristic frequency closely related to the event can be marked and extracted by summarizing the continuous cross-frequency coupling calculation results.

[0027] For example, each IMF corresponds to a different frequency band of the electroencephalogram, and the instantaneous phase and amplitude information of each IMF can be calculated respectively; then the coupling values between different frequency bands of IMF are evaluated, and the modulation index (Modulation Index, MI) is used to quantify the coupling strength between the phase and amplitude; finally, a coupling matrix is constructed according to the coupling values, which reveals the interaction and dynamic change characteristics between different frequency bands, so as to quantitatively reflect the coupling relationship between different frequency bands, that is, the coupling values between different frequency bands. Generally, specific cues will be given to induce changes in electroencephalogram during electroencephalogram recording, such as showing a cigarette photo to a person who wants to quit smoking. The time point of the cigarette photo (i.e. the event occurrence) will be recorded in the recording system, and the electroencephalogram change induced by the cigarette photo is called event-related potential.

[0028] S4: By comparing the cross-frequency coupling values before and after the event occurrence, the cross-frequency coupling values exceeding the preset baseline value are taken as the final frequency domain features.

[0029] For example, after extracting the characteristic frequency (coupling value), it is paired with the event occurrence. Specifically, the cross-frequency coupling values before and after the event occurrence can be compared according to the set time before and after the event occurrence (such as 5 seconds before the event occurrence and 3 seconds after the event occurrence), and the frequency band with significant change in value (such as a preset baseline value of 20%) is found.

[0030] For example, the analysis and calculation and feature extraction program can be opened and run in the computer through the USB interface, and real-time cross-frequency coupling analysis and frequency feature extraction can be performed.

[0031] Embodiment 2 This embodiment analyzes the emotion-related electroencephalogram cross-frequency coupling of volunteers, which can specifically include: (1) Acquisition of emotion-related potential of subjects: in the electroencephalogram recording environment, image, sound, odor or tactile stimulation and other means are used to induce the subjects to produce related emotional changes, and the emotion-related electroencephalogram produced by the subjects is recorded.

[0032] (2) connecting the data recording computer of the present application to read the emotion-related potential electroencephalogram.

[0033] (3) using the adaptive signal decomposition method of the present application to perform signal cutting and waveform adaptive decomposition on the emotion-related potential of the volunteer.

[0034] (4) performing cross-frequency coupling calculation on each cutting signal and performing caching.

[0035] (5) extracting the cross-frequency coupling characteristic frequency band information of the emotion-related electroencephalogram through comparison.

[0036] (6) storing the characteristic frequency band information.

[0037] Embodiment 3 This embodiment analyzes the cross-frequency coupling of the related electroencephalogram of the animal when it is frightened, which can specifically include the following steps: (1) implanting an in-vivo electroencephalogram recording electrode in the skull of the experimental animal and performing postoperative recovery.

[0038] (2) placing the animal in a training box, playing a specific frequency and decibel sound stimulus, and accompanying a punishment event (such as foot shock), and training the animal to learn the fear response to the specific sound.

[0039] (2) connecting the data recording computer of the present application to read the emotion-related potential electroencephalogram.

[0040] (3) recording the animal fear-related electroencephalogram.

[0041] (4) using the present application to perform signal cutting and waveform adaptive decomposition on the animal fear response-related potential.

[0042] (5) performing cross-frequency coupling calculation on each cutting signal and performing caching.

[0043] (6) extracting the cross-frequency coupling characteristic frequency band information of the emotion-related electroencephalogram through comparison.

[0044] (7) storing the characteristic frequency band information.

[0045] The above is a non-stationary signal cross-frequency coupling frequency domain feature acquisition method provided by one or more embodiments of the present application, and based on the same idea, the present application also provides a corresponding non-stationary signal cross-frequency coupling frequency domain feature acquisition system, which includes: The acquisition module is configured to acquire a plurality of electroencephalogram signals before and after an event, and cut the electroencephalogram signals into a plurality of segments according to a preset time length; the electroencephalogram signals are non-stationary signals. The decomposition module is configured to decompose the electroencephalogram signal of each segment by using intrinsic mode function (IMF) to determine the electroencephalogram signal in different frequency bands. The coupling analysis module is configured to determine the coupling values between different frequency bands according to the electroencephalogram signal in different frequency bands, match the coupling values between different frequency bands with the event occurrence, and obtain the cross-frequency coupling values before and after the event occurrence. The extraction module is configured to compare the cross-frequency coupling values before and after the event occurrence, and take the cross-frequency coupling values exceeding the preset baseline value as the final frequency domain features.

[0046] As shown in Figure 3 The present application is connected with the electroencephalogram recording system through the USB interface and reads the electroencephalogram data, and the actual calculation and storage are realized by the operation chip and the storage chip, respectively. The operation chip realizes signal cutting, waveform adaptive decomposition, cross-frequency coupling calculation, and feature frequency band information comparison and extraction. The storage chip mainly realizes cross-frequency coupling calculation result caching and feature frequency band information storage.

[0047] The specific limitations of the non-stationary signal cross-frequency coupling frequency domain feature acquisition system can be referred to the limitations of the non-stationary signal cross-frequency coupling frequency domain feature acquisition method in the above, which will not be repeated here. Each module in the above non-stationary signal cross-frequency coupling frequency domain feature acquisition system can be realized by software, hardware, and combinations thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0048] The present application also provides a computer readable storage medium, which stores a computer program, and the computer program can be used to execute the non-stationary signal cross-frequency coupling frequency domain feature acquisition method provided above.

[0049] The present application also provides a computer readable storage medium, which stores a computer program, and the computer program can be used to execute the non-stationary signal cross-frequency coupling frequency domain feature acquisition method provided above. Figure 4 As shown in Figure 4 As shown in, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and of course can also include other hardware required by the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs to realize the non-stationary signal cross-frequency coupling frequency domain feature acquisition method provided by the above embodiments.

[0050] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments of the methods. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0051] The technical features of the above embodiments can be combined in any way. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

Claims

1. A method for obtaining the frequency domain characteristics of cross-frequency coupling of non-stationary signals, characterized in that, include: Acquire multi-channel EEG signals before and after the event, and segment the EEG signals into multiple segments according to a preset duration; The electroencephalogram (EEG) signal is a non-stationary signal; By using the Intrinsic Mode Function (IMF) to decompose the EEG signal of each segment, the EEG signals of different frequency bands are determined. Based on EEG signals of different frequency bands, the coupling values ​​between different frequency bands are determined, and the coupling values ​​between different frequency bands are matched with the occurrence of the event to obtain the cross-frequency coupling values ​​before and after the event. By comparing the cross-frequency coupling values ​​before and after the event, the cross-frequency coupling values ​​exceeding the preset baseline value are used as the final frequency domain features.

2. The method for obtaining the frequency domain characteristics of non-stationary signal cross-frequency coupling as described in claim 1, characterized in that, The method of decomposing the EEG signal of each segment using the Intrinsic Mode Function (IMF) to determine the EEG signal of different frequency bands specifically includes: The projection direction of the raw EEG signal for each segment in the multi-channel array is obtained, the projection signal is determined based on the projection direction, and the local extremum point corresponding to each projection signal is determined. Map local extreme points back to multidimensional channels to generate multidimensional envelopes in various directions; The mean value of the multidimensional envelope in each direction is obtained. The mean value of the multidimensional envelope is subtracted from the original EEG signal to determine the EEG signal of different frequency bands corresponding to each channel.

3. The method for obtaining the frequency domain characteristics of non-stationary signal cross-frequency coupling as described in claim 2, characterized in that, Multidimensional envelopes in each direction are generated using multidimensional interpolation.

4. The method for obtaining the frequency domain characteristics of non-stationary signal cross-frequency coupling as described in claim 1, characterized in that, The EEG signals are those from humans or experimental animals.

5. The method for obtaining the frequency domain characteristics of non-stationary signal cross-frequency coupling as described in claim 1, characterized in that, Each segment of the EEG signal is a segment of EEG signal of equal length over a given time period.

6. A system for acquiring the frequency domain characteristics of a non-stationary signal through cross-frequency coupling, characterized in that, include: The acquisition module is used to acquire multi-channel EEG signals before and after the event, and to cut the EEG signals into multiple segments according to a preset duration. The electroencephalogram (EEG) signal is a non-stationary signal; The decomposition module is used to decompose the EEG signal of each segment by using the Intrinsic Mode Function (IMF) to determine the EEG signal of different frequency bands. The coupling analysis module is used to determine the coupling values ​​between different frequency bands based on EEG signals of different frequency bands, match the coupling values ​​between different frequency bands with the occurrence of the event, and obtain the cross-frequency coupling values ​​before and after the event. The extraction module is used to compare the cross-frequency coupling values ​​before and after the event, and use the cross-frequency coupling values ​​that exceed the preset baseline value as the final frequency domain features.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method for obtaining the cross-frequency domain characteristics of non-stationary signals as described in any one of claims 1-5.

8. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the method for obtaining cross-frequency domain features of non-stationary signals as described in any one of claims 1-5.

Images