Deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method and its application

Abstract

The present application proposes a deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method and its application, which includes the following steps: including the following steps: S00, collecting EEG data during sleep; S10, filtering the EEG data during sleep; S20, constructing a deterministic sinusoidal signal, and performing Hilbert transform on the filtered signal; S30, performing time-frequency spectrum enhancement on the filtered signal through synchronous compression transform of deterministic sinusoidal intervention; S40, obtaining the instantaneous frequency change curve of the frequency band where the spindle wave is located, and obtaining the time-frequency coefficient; S50, reconstructing the time domain signal according to the time-frequency coefficient and filtering out the deterministic sinusoidal signal; S60, setting a first threshold to divide the instantaneous frequency change curve into a first spindle wave interval, obtaining a correlation coefficient spectrum according to the reconstructed time domain signal and the filtered signal, and dividing the correlation coefficient spectrum into a second spindle wave interval by setting a second threshold; S70, determining the final spindle wave interval according to the first spindle wave interval and the second spindle wave interval.

Description

Technical Field

[0001] The present application relates to EEG signal analysis technology, and in particular to a deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method and its application. Background Art

[0002] Electroencephalography (EEG) is a fundamental tool in daily clinical practice and plays a crucial role in sleep health analysis. Currently, research on signal processing, sleep staging, and sleep health based on multi-lead EEG data from polysomnography (PSG) is relatively mature. However, research on single-lead EEG data for daily health monitoring, sleep staging, and proactive health interventions for consumers is still gaining momentum.

[0003] In the existing technology, the monitoring and separation of key sleep event waveforms (such as spindle waves) are all explored on PSG data, and there are few solutions that can be studied on EEG data.

[0004] Therefore, the current spindle wave schemes are mainly the following two:

[0005] 1. Use purely data-driven methods (such as deep learning) to monitor key event waveforms and use neural networks to divide the intervals containing spindles;

[0006] Second, based on time-frequency analysis algorithms, such as Fourier transform, wavelet decomposition, empirical mode decomposition and other classic algorithms, the components containing spindles are separated through filtering or decomposition algorithms, and the area where the spindles are located is determined by setting thresholds and other methods.

[0007] Some algorithms also combine the above two schemes, using time-frequency analysis algorithms to separate the spindle components, and then using neural networks instead of threshold decisions to delineate the area where the spindles are located.

[0008] However, these algorithms all have limitations. For example, data-driven approaches monitor a limited range of events and are still based on PSG signals, resulting in limited algorithmic performance. While classic algorithms based on Fourier transforms, wavelet decomposition, and time-series correlation can be directly applied to single-lead waveform monitoring, they suffer from significant drawbacks. These classical methods require numerous parameters, making it difficult to determine thresholds and wavelet functions. This makes parameter adjustment difficult for single-lead EEG signals, which exhibit significant individual variability and low signal-to-noise ratio. Real-time monitoring requires prior information, as fixed information is less robust to time-varying signal parameters, resulting in performance far below expectations. Even combined approaches can mitigate the decision-making errors introduced by individual algorithm decision intervals, but the excessive number of threshold parameters introduced makes real-time monitoring less effective. The algorithms in these combined approaches are all based on the high signal-to-noise ratio of PSG data, resulting in poor performance on the low signal-to-noise ratio of single-lead signals.

[0009] Therefore, there is an urgent need for a spindle wave extraction method based on deterministic sinusoidal intervention synchronous compression transform and its application, which can effectively and accurately extract spindle waves from sleep EEG signals through deterministic sinusoidal intervention synchronous compression transform, thereby solving the problems existing in the existing technology. Summary of the Invention

[0010] The embodiments of the present application provide a deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method and its application, which address the many defects existing in current technologies.

[0011] The core technology of this invention is mainly a spindle wave extraction algorithm based on deterministic sinusoidal intervention synchronous compression transform. Through deterministic sinusoidal intervention synchronous compression transform, spindle waves in sleep EEG signals can be effectively and accurately extracted.

[0012] In a first aspect, the present application provides a method for extracting spindle waves using a deterministic sinusoidal intervention synchronous compression transformation, the method comprising the following steps:

[0013] S00, collecting EEG data during sleep containing characteristic spindle waves;

[0014] S10, filtering the EEG data during sleep to obtain a filtered signal;

[0015] S20, constructing a deterministic sinusoidal signal and performing Hilbert transform on the filtered signal;

[0016] S30, performing time-frequency spectrum enhancement on the filtered signal by using a synchronous compression transform with deterministic sine intervention according to the deterministic sine signal;

[0017] S40, obtaining an instantaneous frequency change curve of the frequency band where the spindle wave is located, and obtaining a time-frequency coefficient corresponding to the region where the instantaneous frequency change curve is located;

[0018] S50, reconstructing the time domain signal according to the time-frequency coefficient and filtering out the deterministic sinusoidal signal;

[0019] S60, setting a first threshold to divide the instantaneous frequency change curve into a first spindle interval, simultaneously obtaining a correlation coefficient spectrum based on the reconstructed time domain signal and the filtered signal, and dividing the correlation coefficient spectrum into a second spindle interval by setting a second threshold;

[0020] S70 , determining a final spindle interval according to the first spindle interval and the second spindle interval.

[0021] Furthermore, in step S20, when performing Hilbert transform on the filtered signal, the frequency of the deterministic sinusoidal signal is set according to the average value of the instantaneous frequency of the frequency band where the spindle waves are located, and the power of the filtered signal is multiplied by the coefficient as the amplitude of the deterministic sinusoidal signal.

[0022] Furthermore, in step S30, the specific steps are:

[0023] Add a deterministic sinusoidal signal to the filtered signal to construct a new composite signal;

[0024] Performing wavelet transform on the composite signal to obtain a first signal;

[0025] A two-dimensional instantaneous frequency estimate is obtained by estimating the first signal through synchronous compression transformation, and the energy in the time-frequency spectrum of the first signal is compressed to align it with the two-dimensional instantaneous frequency estimate, so as to achieve time-frequency spectrum enhancement.

[0026] Furthermore, in step S40, the specific steps are:

[0027] Adaptively obtain the specific area through the edge extraction algorithm under the preset optimization iterative framework, and obtain the minimum optimization target of the preset optimization iterative framework;

[0028] Calculate the average frequency of each instantaneous frequency in the minimum optimization target and select the average frequency within the set interval;

[0029] According to the instantaneous frequency corresponding to the average frequency after screening, the first spindle interval of the area where spindle waves may appear is obtained, and the energy is compressed and redistributed to obtain the time-frequency coefficient near the specific area.

[0030] Furthermore, in step S50, the specific steps are:

[0031] Reconstruct the specific area in time domain according to the time-frequency coefficient and the spectrum of the spindle wave, obtain the time domain component of the spindle wave specific area, and use the time domain component as the reconstructed time domain signal;

[0032] A corresponding notch filter is constructed according to the frequency of the deterministic sinusoidal signal, and the reconstructed time domain signal is notched filtered by the notch filter to remove the deterministic sinusoidal signal.

[0033] Furthermore, in step S60, the specific steps of obtaining the second spindle interval are:

[0034] Set the window length of the rectangular window so that it changes with a preset step size;

[0035] By sliding the rectangular window through the filtered signal and the time domain component, the correlation coefficient of each window is calculated to obtain a correlation coefficient sequence that changes with the window length;

[0036] Interpolate the correlation coefficient sequence to obtain the correlation coefficient spectrum that changes with time;

[0037] A second threshold is set to divide the correlation coefficient spectrum to obtain correlation coefficient intervals as second spindle intervals.

[0038] Furthermore, in step S70, the specific steps are:

[0039] Binarizing the first spindle interval and the second spindle interval, and adding the binarized first spindle interval and the second spindle interval to obtain a spindle function;

[0040] Set the window length of the preset window function, and convolve the spindle wave function with the window length of the window function to obtain the spindle wave area;

[0041] The third threshold is set, and the spindle wave area is divided to obtain the final spindle wave interval.

[0042] In a second aspect, the present application provides a deterministic sinusoidal intervention synchronous compression transform spindle wave extraction device, comprising:

[0043] An acquisition module, used for acquiring EEG data during sleep containing characteristic spindle waves;

[0044] A filtering module, used for filtering EEG data during sleep to obtain a filtered signal;

[0045] A processing module is configured to construct a deterministic sinusoidal signal and perform a Hilbert transform on the filtered signal; based on the deterministic sinusoidal signal, perform time-frequency spectrum enhancement on the filtered signal through a synchronous compression transform with deterministic sinusoidal intervention; obtain an instantaneous frequency variation curve of the frequency band where the spindle waves are located, and obtain time-frequency coefficients corresponding to the region where the instantaneous frequency variation curve is located; reconstruct the time domain signal based on the time-frequency coefficients and filter out the deterministic sinusoidal signal;

[0046] a dividing module, configured to divide the instantaneous frequency change curve into a first spindle interval by setting a first threshold, obtain a correlation coefficient spectrum based on the reconstructed time domain signal and the filtered signal, and divide the correlation coefficient spectrum into a second spindle interval by setting a second threshold;

[0047] An output module is used to determine and output a final spindle interval based on the first spindle interval and the second spindle interval.

[0048] In a third aspect, the present application provides an electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the above-mentioned deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method.

[0049] In a fourth aspect, the present application provides a readable storage medium, in which a computer program is stored. The computer program includes a program code for controlling a process to execute a process, and the process includes the above-mentioned deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method.

[0050] The main contributions and innovations of the present invention are as follows: 1. Compared with the existing technology, this application proposes a method for extracting spindle-specific sequences, innovatively introduces the idea of ​​deterministic sinusoidal intervention, and intervenes in the synchronous compression transform, making the time-frequency spectrum enhancement more accurate and purposeful (the existing technology mainly focuses on how to improve accuracy). When the edge extraction algorithm adaptively extracts specific regions, it reduces the influence of noise unrelated to the frequency band, and then screens the specific regions and reconstructs them back into the time domain sequence. The notch filter is used to effectively remove the introduced deterministic sinusoidal signal, greatly reducing the noise interference of the separated signal. At the same time, because it is not based on PSG signals but on EEG signals, there is no problem of limited algorithm performance. It is also not based on classic algorithms such as Fourier transform, wavelet decomposition, and time series correlation, and there are no problems such as too many parameters and difficulty in determining thresholds, wavelet functions, etc.

[0051] 2. Compared with existing technologies, this application is not only applicable to synchronous compression transforms, but also to more effective time-frequency transform schemes, eliminating the need for complex filter design and demonstrating excellent noise robustness and universality. Furthermore, this application is also applicable to other different EEG waveforms, such as high-frequency epileptic oscillations. This method can effectively extract these characteristic waveforms, significantly improving the signal-to-noise ratio of these waveform sequences and facilitating epilepsy identification and early warning.

[0052] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0054] Figure 1 This is a process of a deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to an embodiment of the present application;

[0055] Figure 2 is a spindle wave characteristic diagram according to an embodiment of the present application;

[0056] Figure 3 This is a time-frequency diagram of spindle waves under sinusoidal intervention according to an embodiment of the present application;

[0057] Figure 4 is a time domain diagram of separated spindle waves according to an embodiment of the present application;

[0058] Figure 5 is a simplified flow chart according to an embodiment of the present application;

[0059] Figure 6 Schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0060] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The implementations described in the following exemplary embodiments are not intended to represent all implementations consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of one or more embodiments of this specification, as detailed in the appended claims.

[0061] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than those described in this specification. In addition, a single step described in this specification may be broken down into multiple steps for description in other embodiments, and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0062] Current algorithms all have their own flaws. Data-driven approaches monitor a limited range of events and are still based on PSG signals, resulting in limited algorithmic performance. While classic algorithms based on Fourier transforms, wavelet decomposition, and time-series correlation can be directly applied to single-lead signal waveform monitoring, they suffer from significant drawbacks. These classical methods require numerous parameters, making it difficult to determine thresholds and wavelet functions. This makes parameter adjustment difficult for single-lead EEG signals, which exhibit significant individual variability and low signal-to-noise ratio. Real-time monitoring requires prior information, as signal parameters vary over time. Fixed priors are less robust to these parameters, resulting in performance far below expectations. Even combined approaches can mitigate the decision-making errors introduced by individual algorithm judgment intervals, but the excessive number of threshold parameters introduced makes real-time monitoring less effective. The algorithms in these combined approaches are all based on the high signal-to-noise ratio of PSG data, resulting in poor performance on the low signal-to-noise ratio of single-lead signals.

[0063] Based on this, the present invention solves the above problems by providing a solution that can effectively monitor and separate spindles based on EEG data.

[0064] Example 1

[0065] This application aims to propose a spindle wave extraction algorithm based on deterministic sinusoidal intervening synchronous compression transform.

[0066] Specifically, the embodiment of the present application provides a deterministic sinusoidal intervention synchronous compression transformation spindle wave extraction method, which can effectively solve the problems existing in the prior art. Specifically, referring to Figure 1 and Figure 5 As shown, the method includes the following steps:

[0067] S00, collecting EEG data during sleep containing characteristic spindle waves;

[0068] In this step, specifically: using a single-channel EEG acquisition device to collect EEG data during sleep containing characteristic spindle waves according to a preset sampling frequency and sampling duration.

[0069] S10, filtering the EEG data during sleep to obtain a filtered signal;

[0070] In this step, specifically: according to the frequency band of the spindle waves, the collected EEG data is band-pass filtered at 11-16 Hz to obtain a filtered signal.

[0071] S20, constructing a deterministic sinusoidal signal, performing Hilbert transform on the filtered signal, setting the frequency of the deterministic sinusoidal signal according to the average value of the instantaneous frequency of the frequency band where the spindle waves are located, and multiplying the power of the filtered signal by a coefficient as the amplitude of the deterministic sinusoidal signal;

[0072] In this embodiment, the specific steps of step S20 are:

[0073] S21. Assume that the filtered signal is x(t). Perform Hilbert transform on the filtered signal and calculate the imaginary part signal y(t) corresponding to x(t). Specifically, it is as shown in Formula 1:

[0074] (1)

[0075] Where P represents the Cauchy principal value of the complex integral and t represents time.

[0076] S22. Construct the analytical signal z(t) corresponding to x(t) based on the imaginary signal, as shown in Formula 2:

[0077] (2)

[0078] The amplitude , phase , Expressing imaginary units;

[0079] S23. Derivative the phase function θ(t) of the analytical signal z(t) to obtain the frequency function ω(t), as shown in Formula 3:

[0080] (3)

[0081] Where A, x, and y represent A(t), x(t), and y(t), respectively;

[0082] S24, calculating the average value of the frequency function minus the fixed parameter, and using the average value as the frequency ωconst of the deterministic sinusoidal signal;

[0083] S25. Calculate the average power of the filtered signal and multiply it by a fixed coefficient, and use this as the amplitude A of the deterministic sinusoidal signal. const ;

[0084] S26. Constructing a deterministic sinusoidal signal A const sin(ω const t).

[0085] S30, such as Figure 3 As shown, according to the deterministic sinusoidal signal, the time-spectrum enhancement of the filtered signal is performed by the synchronous compression transformation with deterministic sinusoidal intervention;

[0086] In this embodiment, the specific steps of step S30 are:

[0087] S31. Add a deterministic sinusoidal signal to the filtered signal to construct a new composite signal x'(t), as shown in Formula 4:

[0088] x'(t)=x(t)+A const sin(ω const t)(4)

[0089] S32, select Gabor wavelet to perform wavelet transform on the composite signal to obtain a first signal, as shown in Formula 5:

[0090] (5)

[0091] Where α is the scale parameter of wavelet transform, b is the time shift parameter of wavelet transform, * Express the conjugate function of the wavelet function.

[0092] S33. Estimating the first signal by synchronous compression transform to obtain a two-dimensional instantaneous frequency estimate, and compressing the energy in the time-frequency spectrum of the first signal to align it with the two-dimensional instantaneous frequency estimate, so as to achieve time-frequency spectrum enhancement, as shown in Formula 6:

[0093] (6)

[0094] Among them, ω, δ(.) represent frequency and impulse function respectively, A(b)={a,W x (a,b)≠0}, i represents a unit imaginary number.

[0095] S40, obtaining an instantaneous frequency change curve of the frequency band where the spindle wave is located, and obtaining a time-frequency coefficient corresponding to the region where the instantaneous frequency change curve is located;

[0096] S41. Adaptively obtain a specific region through an edge extraction algorithm under a preset optimization iterative framework, and obtain a minimum optimization target of the preset optimization iterative framework, as shown in Formula 7 of the minimum optimization target:

[0097] (7)

[0098] Among them, φ k represents the instantaneous frequency of each energy distribution, K represents the total number of energy distributions, λ and β represent the optimization coefficients of the first and second orders. Preferably, in the actual process, λ is fixed to 0.8, β is fixed to 0.05, ω, d, b represent the frequency and integral;

[0099] S42. The minimum optimization target includes the instantaneous frequency of each energy distribution in the time-frequency energy spectrum, calculates the average frequency of each instantaneous frequency, and selects the average frequency within the set interval, as shown in Formula 8:

[0100] (8)

[0101] in, Indicates the average frequency, N indicates the total number of b, since spindle waves are mainly concentrated in 11~16hz, if If it is in this area, the instantaneous frequency corresponding to the average frequency is retained as the first threshold, thereby obtaining {φ m (b)|m=1,...,M}, thereby completing the screening of instantaneous frequency;

[0102] S43. Based on the instantaneous frequency corresponding to the average frequency after screening, the first spindle interval of the area where spindles may appear is obtained, and energy is compressed and redistributed to obtain the time-frequency coefficient near the specific area, as shown in Formula 9:

[0103] T m (b)=T x (φ m (b),b)(9)

[0104] Among them, T m (b) represents the time-frequency coefficient, φ m (b) is the instantaneous frequency after screening.

[0105] S50, reconstructing the time domain signal according to the time-frequency coefficients (obtained for the specific region) and filtering out the deterministic sinusoidal signal;

[0106] S51. Reconstruct the specific region in the time domain according to the time-frequency coefficient and the spectrum of the spindle wave, obtain the time domain component of the spindle wave specific region, and use the time domain component as the reconstructed time domain signal, as shown in Formula 10:

[0107] (10)

[0108] Among them, P() represents the cubic interpolation function, T m (b) represents the time-frequency coefficient, T sp (t) represents the spindle wave spectrum, T sp (b) represents the time-frequency coefficient of spindle waves.

[0109] S52, according to the frequency ω of the deterministic sinusoidal signal const Construct a corresponding notch filter, and use the notch filter to perform notch filtering on the reconstructed time domain signal to remove the deterministic sinusoidal signal.

[0110] S60, such as Figure 4 As shown, a first threshold is set to divide the instantaneous frequency change curve into a first spindle interval, and a correlation coefficient spectrum is obtained based on the reconstructed time domain signal and the filtered signal. The correlation coefficient spectrum is divided into a second spindle interval by setting a second threshold;

[0111] S61, setting the window length of the rectangular window so that it changes with a preset step size;

[0112] S62, calculating the correlation coefficient for each window through the rectangular window filtered signal and the time domain component sliding window, and obtaining a correlation coefficient sequence that varies with the window length;

[0113] S63, interpolating the correlation coefficient sequence to obtain a correlation coefficient spectrum that changes over time;

[0114] S64. Set a second threshold to divide the correlation coefficient spectrum to obtain correlation coefficient intervals as second spindle intervals.

[0115] S70, such as Figure 1 As shown, the final spindle interval is determined based on the first spindle interval and the second spindle interval (joint decision making).

[0116] S71, binarizing the first spindle interval and the second spindle interval, and adding the binarized first spindle interval and the second spindle interval to obtain a spindle function;

[0117] S72, setting a window length of a preset window function, and convolving the spindle wave function with the window length of the window function to obtain a spindle wave region;

[0118] S73: Set a third threshold value, and divide the spindle wave area to obtain a final spindle wave interval.

[0119] The spindle wave feature map represents the spindle wave interval after the joint decision of the first and second thresholds. The time-frequency map of the spindle wave signal under sinusoidal intervention represents the time-frequency map of the signal converted to the time-frequency domain through synchronous compression transform after introducing the deterministic sine wave. Finally, the separated spindle wave time domain map represents the spindle wave time domain map separated according to the proposed method.

[0120] Example 2

[0121] Based on the same concept, this application also proposes a deterministic sinusoidal intervention synchronous compression transformation spindle wave extraction device, comprising:

[0122] An acquisition module, used for acquiring EEG data during sleep containing characteristic spindle waves;

[0123] A filtering module, used for filtering EEG data during sleep to obtain a filtered signal;

[0124] A processing module is configured to construct a deterministic sinusoidal signal and perform a Hilbert transform on the filtered signal; based on the deterministic sinusoidal signal, perform time-frequency spectrum enhancement on the filtered signal through a synchronous compression transform with deterministic sinusoidal intervention; obtain an instantaneous frequency variation curve of the frequency band where the spindle waves are located, and obtain time-frequency coefficients corresponding to the region where the instantaneous frequency variation curve is located; reconstruct the time domain signal based on the time-frequency coefficients and filter out the deterministic sinusoidal signal;

[0125] a dividing module, configured to divide the instantaneous frequency change curve into a first spindle interval by setting a first threshold, obtain a correlation coefficient spectrum based on the reconstructed time domain signal and the filtered signal, and divide the correlation coefficient spectrum into a second spindle interval by setting a second threshold;

[0126] An output module is used to determine and output a final spindle interval based on the first spindle interval and the second spindle interval.

[0127] Example 3

[0128] This embodiment also provides an electronic device, referring to Figure 6 , includes a memory 404 and a processor 402, wherein the memory 404 stores a computer program, and the processor 402 is configured to run the computer program to perform the steps in any of the above method embodiments.

[0129] Specifically, the processor 402 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0130] Memory 404 may include a large-capacity memory 404 for data or instructions. By way of example, and not limitation, memory 404 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 404 may include removable or non-removable (or fixed) media. Where appropriate, memory 404 may be internal or external to the data processing device. In certain embodiments, memory 404 is non-volatile memory. In certain embodiments, memory 404 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. In appropriate circumstances, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM may be a fast page mode dynamic random access memory 404 (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0131] The memory 404 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 402 .

[0132] The processor 402 reads and executes computer program instructions stored in the memory 404 to implement the arbitrary deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method in the above embodiments.

[0133] Optionally, the electronic device may further include a transmission device 406 and an input / output device 408 , wherein the transmission device 406 is connected to the processor 402 , and the input / output device 408 is connected to the processor 402 .

[0134] Transmission device 406 can be used to receive or transmit data via a network. Specific examples of such networks may include wired or wireless networks provided by the electronic device's communications provider. In one embodiment, the transmission device includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, transmission device 406 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0135] The input / output device 408 is used to input or output information. In this embodiment, the input information may be EEG data, and the output information may be the final spindle interval.

[0136] Example 4

[0137] This embodiment also provides a readable storage medium, in which a computer program is stored. The computer program includes a program code for controlling a process to execute a process. The process includes the deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to the first embodiment.

[0138] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be repeated here.

[0139] In general, various embodiments may be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects of the invention may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device, but the invention is not limited thereto. Although various aspects of the invention may be shown and described as block diagrams, flow charts, or using some other graphical representation, it should be understood that, as non-limiting examples, the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or a controller or other computing device, or some combination thereof.

[0140] Embodiments of the present invention can be implemented by computer software, which is executable by the data processor of the mobile device, such as in the processor entity, or is implemented by hardware, or is implemented by a combination of software and hardware. Computer software or programs (also referred to as program products) including software routines, applets and / or macros can be stored in any device-readable data storage medium, and they include program instructions for performing specific tasks. The computer program product can include one or more computer executable components configured to perform the embodiment when the program is running. One or more computer executable components can be at least one software code or a part thereof. In addition, at this point, it should be noted that any box of the logic flow in the figure can represent a program step, or interconnected logical circuits, boxes and functions, or a combination of program steps and logical circuits, boxes and functions. The software can be stored in physical media such as memory chips or storage blocks implemented in the processor, magnetic media such as hard disks or floppy disks, and optical media such as, for example, DVDs and their data variants, CDs. Physical media is non-transient media.

[0141] Those skilled in the art should understand that the technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0142] The above embodiments merely illustrate several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. Deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method, characterized in that: The following steps are involved: S00, collecting EEG data during sleep containing characteristic spindle waves; S10, filtering the EEG data during sleep to obtain a filtered signal; S20, constructing a deterministic sinusoidal signal, and when performing Hilbert transform on the filtered signal, setting the frequency of the deterministic sinusoidal signal according to the average value of the instantaneous frequency of the frequency band where the spindle waves are located, and multiplying the power of the filtered signal by a coefficient as the amplitude of the deterministic sinusoidal signal; S30. Performing time-frequency spectrum enhancement on the filtered signal by using a synchronous compression transform with deterministic sine intervention according to the deterministic sine signal; S40, obtaining an instantaneous frequency change curve of the frequency band where the spindle wave is located, and obtaining a time-frequency coefficient corresponding to a specific region where the instantaneous frequency change curve is located; S50, performing time domain reconstruction on the specific region according to the time-frequency coefficient and the spectrum of the spindle wave, obtaining a time domain component of the spindle wave specific region, and using the time domain component as a reconstructed time domain signal; Constructing a corresponding notch filter according to the frequency of the deterministic sinusoidal signal, and performing notch filtering on the reconstructed time domain signal through the notch filter to remove the deterministic sinusoidal signal; S60, setting a first threshold to divide the instantaneous frequency change curve into a first spindle interval, simultaneously obtaining a correlation coefficient spectrum based on the reconstructed time domain signal and the filtered signal, and dividing the correlation coefficient spectrum into a second spindle interval by setting a second threshold; S70: Determine a final spindle interval according to the first spindle interval and the second spindle interval.

2. The deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to claim 1, characterized in that: In step S30, the specific steps are: Adding the deterministic sinusoidal signal to the filtered signal to construct a new composite signal; Performing wavelet transform on the composite signal to obtain a first signal; A two-dimensional instantaneous frequency estimate is obtained from the first signal estimate through synchronous compression transformation, and the energy in the time-frequency spectrum of the first signal is compressed to align it with the two-dimensional instantaneous frequency estimate, so as to achieve time-frequency spectrum enhancement.

3. The deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to claim 1, characterized in that: In step S40, the specific steps are: Adaptively obtaining a specific region through an edge extraction algorithm under a preset optimization iterative framework, and obtaining a minimum optimization target of the preset optimization iterative framework; Calculating the average frequency of each instantaneous frequency in the minimum optimization target and screening the average frequency within a set interval; According to the instantaneous frequency corresponding to the average frequency after screening, the first spindle interval of the area where spindle waves may appear is obtained, and energy compression and redistribution are performed to obtain the time-frequency coefficient near the specific area.

4. The deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to claim 1, characterized in that: In step S60, the specific steps of obtaining the second spindle interval are: Set the window length of the rectangular window so that it changes with a preset step size; Calculate the correlation coefficient of each window through the filtered signal and the time domain component sliding window of the rectangular window to obtain a correlation coefficient sequence that changes with the window length; interpolating the correlation coefficient sequence to obtain a correlation coefficient spectrum that varies with time; A second threshold is set to divide the correlation coefficient spectrum to obtain correlation coefficient intervals as second spindle intervals.

5. The deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to claim 1, characterized in that: In step S70, the specific steps are: Binarizing the first spindle interval and the second spindle interval, and adding the binarized first spindle interval and the second spindle interval to obtain a spindle function; Setting a window length of a preset window function, and convolving the spindle wave function with the window length of the window function to obtain a spindle wave region; The third threshold is set, and the spindle wave area is divided to obtain the final spindle wave interval.

6. A deterministic sinusoidal interferometric synchronous compression transform spindle wave extraction device, characterized in that: include: An acquisition module, used for acquiring EEG data during sleep containing characteristic spindle waves; A filtering module, used for filtering EEG data during sleep to obtain a filtered signal; A processing module is used to construct a deterministic sinusoidal signal, and when performing Hilbert transform on the filtered signal, the frequency of the deterministic sinusoidal signal is set according to the average value of the instantaneous frequency of the frequency band where the spindle waves are located, and the power of the filtered signal is multiplied by a coefficient as the amplitude of the deterministic sinusoidal signal; based on the deterministic sinusoidal signal, the filtered signal is time-frequency enhanced by a synchronous compression transform with deterministic sinusoidal intervention; the instantaneous frequency change curve of the frequency band where the spindle waves are located is obtained, and the time-frequency coefficient corresponding to the specific region where the instantaneous frequency change curve is located is obtained; the specific region is reconstructed in the time domain according to the time-frequency coefficient and the spectrum of the spindle waves, the time domain component of the spindle wave-specific region is obtained, and the time domain component is used as the reconstructed time domain signal; a corresponding notch filter is constructed according to the frequency of the deterministic sinusoidal signal, and the reconstructed time domain signal is notch filtered by the notch filter to remove the deterministic sinusoidal signal; a dividing module, configured to divide the instantaneous frequency change curve into a first spindle interval by setting a first threshold, obtain a correlation coefficient spectrum based on the reconstructed time domain signal and the filtered signal, and divide the correlation coefficient spectrum into a second spindle interval by setting a second threshold; An output module is used to determine and output a final spindle interval based on the first spindle interval and the second spindle interval.

7. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to any one of claims 1 to 5.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which includes a program code for controlling a process to execute a process, wherein the process includes the deterministic sinusoidal intervention synchronous compression transform spindle wave extraction method according to any one of claims 1 to 5.

Images