Power frequency removal method based on phase change cycle, evoked potential mapping machine

Abstract

The present invention belongs to the field of signal detection technology, and specifically relates to a power frequency removal method based on phase change cycle and an induced potential map machine. The power frequency removal method based on phase change cycle includes: preprocessing, dividing signal data into multiple data segments and obtaining the phase of each data segment; whole cycle reorganization, that is, reorganizing each data segment with the phase change cycle as a group to obtain a target data segment matching the target phase sequence; noise reduction processing, that is, superimposing the target data segments in any group in pairs to obtain a noise reduction signal, which effectively solves the adverse effect of the slow change of the power frequency amplitude on the anti-phase cancellation of the power frequency in physiological induced electricity.

Description

Technical Field

[0001] The invention belongs to the technical field of signal detection, and in particular relates to a power frequency removal method based on a phase variation cycle and an induced potential map machine. Background Art

[0002] Power frequency noise refers to a type of interference caused by the power system, with a frequency of 50 / 60Hz, which can cause serious interference to the recognition of bioelectric signals, especially the extremely small amplitudes of some bioelectric signals, such as the amplitudes of some components of the somatosensory evoked potential (SEP) signal are less than 5μV, and these signals are easily submerged in strong power frequency noise. Because the stimulus artifact will have sinusoidal oscillations after notch filtering, these oscillations will also affect SEP measurement, so it is not recommended to use notch filtering to remove power frequency in SEP detection.

[0003] In the related art, by processing the signal, the positive and negative periodic inversion data can be used to eliminate the power frequency noise. For example, patent CN111983698B discloses a method for suppressing power frequency noise of towed transient electromagnetic data based on component phase inversion, including the following steps: S1, obtaining the power frequency noise fundamental frequency of the environment; S2, determining the transmission frequency according to the power frequency noise fundamental frequency of the environment; S3, calculating the number of cycles of valid data collection in a measuring point according to the transmission frequency determined in step S2; S4, extracting all the valid collection data of the first measuring point, extracting the positive and negative periodic inversion data, and subtracting them to obtain a set of data; S5, superimposing and averaging all the data extracted in step S4 to obtain the original data after preprocessing of the first measuring point; S6, processing other measuring points according to steps S4 and S5 to obtain all the preprocessed original data of the towed data. Obviously, the towed transient electromagnetic data can assume that the phase of the power frequency noise is unchanged, and on this basis, noise reduction can be achieved through the inversion characteristics of the signal and noise, or accurate real-time measurement of the phase is required to quickly transmit the signal of the corresponding phase. From the attached Figure 2 , attached Figure 3 It can be seen that the actual power frequency phase change is uneven, and it is impossible to achieve absolute absence of delay. The unevenness of the power frequency phase change makes the phase unpredictable, so this solution is difficult to implement when the power frequency phase changes unevenly. The amplitude of some components of bioelectric signals, especially somatosensory evoked potential (SEP) signals, is lower than 5μV, which is easily submerged in strong power frequency noise. In addition, the power frequency amplitude of these signals changes slowly, which makes it difficult to remove the power frequency noise of the somatosensory evoked potential (SEP) signal by anti-phase cancellation power frequency. Summary of the invention

[0004] The present invention provides a power frequency removal method based on phase change cycle and an induced potential map machine, which regroups data segments by using phase change cycle as groups to solve the problem of slow power frequency amplitude variation in the process of induced power using reverse phase to offset power frequency.

[0005] In order to solve the above technical problems, the present invention provides a method for removing industrial frequency based on phase change cycle, including: preprocessing, dividing signal data into multiple data segments and obtaining the phase of each data segment; whole cycle reorganization, that is, reorganizing each data segment into groups with phase change cycle to obtain target data segments matching the target phase sequence; noise reduction processing, that is, superimposing the target data segments in any group in pairs to obtain noise reduction signals; screening noise reduction signals, including: setting the initial phase of the first group of data segments and updating it with a step size, traversing the phase change cycle, and obtaining noise reduction signals corresponding to different initial phases; selecting the corresponding noise reduction signal as the noise reduction result according to the evaluation target.

[0006] Furthermore, the whole cycle reorganization includes: obtaining the phase difference of each data segment, and determining the first data segment whose phase difference exceeds a first set threshold as the next group of data segments.

[0007] Furthermore, the whole cycle reorganization also includes: dividing the phase change period [pi, -pi] of any group of data segments by a sampling phase interval to obtain 2n target phase sequences, wherein the sampling phase interval is ; Select the 2n data segments closest to the target phase sequence as the target data segments.

[0008] Furthermore, the whole cycle reorganization also includes: in any group of data segments, when multiple target data segments are detected to be the same data segment, only one target data segment is retained and the remaining target data segments are replaced by data segments of other groups; and\or in any group of data segments, when the phase change of the detected data segment exceeds a second set threshold, the data segment is judged to be an unstable data segment and deleted.

[0009] Furthermore, the evaluation target includes: any one of a first target, a second target, and a third target; wherein the first target includes: the maximum peak value of the noise reduction signal waveform at a set point; the second target includes: obtaining the absolute phase error between the true phase of the target data segment and the target phase sequence, and summing the absolute errors of each target data segment; the third target includes: obtaining the phase sum of paired superimposed target data segments, and obtaining the absolute error between the phase sum and zero.

[0010] In a second aspect, the present invention provides an evoked potential mapping machine, comprising: a processor, which executes the steps of the method; an input unit, which is used to input signal data; and an output unit, which is used to output a noise reduction signal.

[0011] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the power frequency removal method.

[0012] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the power frequency removal method when executed by a processor.

[0013] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which implements the steps of the power frequency removal method when executed by a processor.

[0014] The beneficial effect of the present invention is that the power frequency removal method based on the phase change cycle and the induced potential map machine of the present invention first slice the signal data, then reorganize the data segments by the phase change cycle as a group, obtain the target data segments matching the target phase sequence, and finally in any group of data segments, use two data segments with opposite phases to superimpose and cancel the power frequency waveform to perform noise reduction processing. Taking the phase change cycle as a group can reduce the adverse effect of the slow change of the power frequency amplitude on the noise reduction effect, and screening the target data segments matching the target phase sequence can further improve the effect of superposition and cancellation, which not only avoids the influence of notch filtering on the measurement of induced power, but also improves the noise reduction result of the anti-phase cancellation power frequency on the induced power.

[0015] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0016] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 It is a principle block diagram of the power frequency removal method of the present invention.

[0019] Figure 2 It is the phase value of each data segment of C3 channel at 50Hz.

[0020] Figure 3 It is the phase value of each data segment of C4 channel at 50Hz.

[0021] Figure 4 It is a schematic diagram of the grouping of data groups in the C3 channel.

[0022] Figure 5 It is a schematic diagram of obtaining the target phase sequence in the data group.

[0023] Figure 6 It is a schematic diagram of selecting a target data segment in a data group.

[0024] Figure 7 It is a schematic diagram of selecting a target data segment to be replaced in a data group.

[0025] Figure 8 This is a schematic diagram after replacing the target data segment in the data group.

[0026] Fig. 9 This is a schematic diagram after deleting some target data segments in the data group.

[0027] Fig.10 It is a schematic diagram of using loss function 1 as the evaluation target.

[0028] Fig.11 It is a schematic diagram of using loss function 2 as the evaluation target.

[0029] Fig.12 It is a schematic diagram of using loss function 3 as the evaluation target.

[0030] Fig.13 This is the denoising result of the C3 channel data under experimental condition 1.

[0031] Fig.14 This is the denoising result of the C4 channel data under experimental condition 1.

[0032] Fig.15 is the noise reduction result of the signal data under experimental condition 2.

[0033] Fig.16 This is the noise reduction result of the signal data under experimental condition 3.

[0034] Fig.17 is the noise reduction result of the signal data under experimental condition 4.

[0035] Fig.18 is the noise reduction result of the signal data under experimental condition 5.

[0036] Fig.19 This is the principle block diagram of another power frequency removal method. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Example 1.

[0039] To solve the problem of uneven phase change, see Figure 1-Figure 8 , this embodiment 1 provides a method for removing power frequency based on phase change cycle, and performs noise reduction processing based on data segment grouping, including the following steps: step S1, preprocessing, dividing the signal data into multiple data segments (i.e., Epochs hereinafter) and obtaining the phase of each data segment; step S2, whole cycle reorganization, i.e., reorganizing each data segment based on the phase change cycle as a group to obtain a target data segment matching the target phase sequence; step S3, noise reduction processing, i.e., superimposing the target data segments in any group in pairs to obtain a noise reduction signal.

[0040] Optionally, the signal data includes, but is not limited to, somatosensory evoked potential (SEP) signals, myoelectric evoked potential signals, cortical inter-evoked potential (CCEP) signals, motor evoked potential signals, brainstem auditory evoked potentials, visual evoked potential (VEP) signals and other evoked potential signal data or other bioelectric signals with small amplitudes (such as amplitudes less than 5μV). The evoked potential signal after electrical stimulation can be intercepted by identifying the stimulus mark (trigger), and then the signal data can be divided into multiple data segments through preprocessing and the phase of each data segment can be obtained. Of course, the preprocessing can also include processing such as removing the baseline.

[0041] As an optional implementation of the phase identification.

[0042] Obtaining the phase of each data segment includes: performing fast Fourier transform on the signal data to obtain the frequency representation value y of any data segment; shifting the data segment to move the zero-frequency component of the frequency representation value y to the center of the spectrum to obtain the coordinate value (a, b) of the data segment after translation; obtaining the complex signal representation z=a+bi corresponding to any data segment, and the phase angle represents the angle formed between the positive direction of the horizontal axis and the ray from the origin to the coordinate value (a, b).

[0043] As an optional implementation of the whole cycle reorganization.

[0044] In step S2, after setting the initial phase, the whole cycle reorganization includes: step S21, grouping the data segments to obtain multiple data groups; step S22, obtaining the target phase sequence; step S23, screening the target data segment in any group of data segments.

[0045] Optionally, in step S21, data segment grouping includes: obtaining the phase difference of each data segment, and determining the data segment whose first phase difference exceeds a first set threshold (generally set according to an empirical value, for example but not limited to 1.5 pi) as the first data segment of the next data group. In general, if the power frequency amplitude remains unchanged and the waveform in the data group is stable, then when two data segments with opposite phases are summed, the power frequencies can cancel each other out to achieve a noise reduction effect. Therefore, the larger the amplitude of the signal and the more stable the power frequency amplitude, the better the noise reduction effect of using anti-phase to cancel the power frequency. However, the power frequency amplitude of the induced signal changes slowly due to the influence of many factors, such as voltage fluctuations caused by changes in grid load; line impedance and capacitance effects during power line transmission; poor equipment grounding; signal delays and phase distortion caused by internal circuit components of the equipment; other electrical equipment as external interference sources and other factors. For details, see Figure 2 and Figure 3 , respectively intercept the evoked electrical data from 4ms to 100ms after stimulation in the C3 channel and the C4 channel, obtain its phase after fast Fourier transform, observe the phase value of each data segment (Epoch number represents different data segments) at the power frequency (50Hz), and find that the phase value range is -π to π. The jump from -π to π indicates that the phase offset of the power frequency has gone through a phase change cycle. Therefore, see Figure 4 , multiple mutually opposite data segments can be selected and superimposed in pairs within a short period of time (for example, a single phase change cycle is a group) to eliminate the influence of the slow change of the power frequency amplitude, and their power frequencies are close to canceling each other out. In this case, the C3 channel and the C4 channel are different channel data collected at the same time. Figure 2 and Figure 3 It shows that there is a phenomenon of power frequency phase offset in multiple channels. The actual power frequency phase change is uneven, and it is impossible to absolutely eliminate delay. The uneven power frequency phase change makes the phase unpredictable. Therefore, when physiological electrical signals use positive and negative cycle inversion data to eliminate power frequency noise, it is necessary to deal with the uneven power frequency phase change first.

[0046] Optionally, in step S22, obtaining the target phase sequence includes: dividing the phase change period [pi, -pi] of any group of data segments by the sampling phase interval to obtain 2n target phase sequences, wherein the sampling phase interval is pi / n; selecting the 2n data segments closest to the target phase sequence as the target data segments. For example, the phase of any data group changes from π to -π, the initial phase is set to π, the sampling phase interval is π / 25, and the target phase sequence (π, π, π, …, - π, -π), a total of 51 (such as Figure 5), delete the middle target phase π, 50 target phases remain. In the current data group, select the 50 data segments whose phase values ​​are closest to the target phase (such as Figure 6 As shown), this can make the phase of the selected data segment closer to uniform change, and the phase sum of the first paired data segment closer to zero, that is, the two power frequencies are closer to canceling each other.

[0047] Optionally, in step S23, screening the target data segment includes: in any group of data segments, when multiple target data segments are detected to be the same data segment, only one target data segment is retained and the remaining target data segments are replaced by data segments of other groups. In a single data group, adjacent target phases may select the same data segment, which will cause the phase sum of the paired data segments to deviate greatly from zero, that is, the sum of the two data segments will still have more power frequency components. Figure 7 As shown in the figure, the black box (the data segment with Epoch number between 40-50) has the phenomenon that the phase value of the two data segments before and after does not change for many times (i.e. Figure 7 The phase value is displayed as a horizontal line in the middle. When this happens, it means that the adjacent target phases have selected the same data segment. Therefore, it is necessary to select the data segment closest to the target phase from other data groups to replace the repeated data segment selected by the current group (such as Figure 8 As shown, the phase value in the black box does not have a horizontal line), so that the phase of the selected data segment is closer to a linear change. And\or in any group of data segments, when the phase change of the detected data segment exceeds the second set threshold (generally set according to an empirical value, such as but not limited to 0.3 pi), the data segment is judged as an unstable data segment and deleted ( Fig. 9 The figure shows the grouping effect after deleting some unstable data segments in the C3 channel).

[0048] The whole cycle reorganization in this embodiment uses the phase change cycle as a group to reorganize each data segment, and selects the target data segment that matches the target phase sequence in each data group, which effectively solves the problem of slow changes in the power frequency amplitude. The power frequency can be removed without using a notch filter, the power frequency interference can be suppressed, and the signal component can be highlighted; no notch filter is used to avoid the generation of filter artifacts; at the same time, the amount of calculation is small, the computational burden is small, and the time consumption is short. No more processing is done on the original data before superposition, so a large number of calculation errors will not be introduced.

[0049] Example 2.

[0050] See Fig.19Based on Example 1, this Example 2 provides another power frequency removal method, which is based on traversing the noise reduction signal to optimize the noise reduction result and its corresponding initial phase and target phase sequence, including: screening the noise reduction signal; the screening of the noise reduction signal includes: setting the initial phase of the first group of data segments and updating it with a step size, traversing the phase change cycle, and obtaining the noise reduction signal corresponding to different initial phases; selecting the corresponding noise reduction signal as the noise reduction result according to the evaluation target. Optionally, the evaluation target includes: any one of the first target, the second target, and the third target.

[0051] Optionally, the step size is configured as, for example but not limited to, 1 / 10 of the sampling phase interval pi / n. For example, the first initial phase is pi, the second is pi*249 / 250, and so on, to obtain different noise reduction signals, and then select the noise reduction signal according to the evaluation target (i.e., noise reduction expectation), and use the corresponding initial phase and target phase sequence as the preferred solution, which can be used preferentially when the next segment data is denoised.

[0052] Optionally, the first target includes: the maximum peak value of the noise reduction signal waveform at the setting. Specifically, since the SEP signal waveform exists within 40ms, the maximum peak value of the waveform at 40-100ms can be considered as the maximum intensity of the power frequency, which is used as the loss function 1. The smaller the value, the greater the suppression of the power frequency by the noise reduction signal obtained after superposition, and the more suitable the corresponding initial phase and target phase sequence are for suppressing the power frequency. Fig.10 , blue line: all epochs in a single phase cycle (-pi~pi) are superimposed and averaged, and there is still a large fluctuation, that is, the power frequency. Red line: epochs in a single phase cycle (-pi~pi) are first screened to make the phases of each epoch evenly distributed after screening, and then the screened epochs are superimposed and averaged, and the fluctuation is significantly reduced, that is, the power frequency is significantly suppressed. Therefore, the physiological component N20 can be seen.

[0053] The second goal includes: obtaining the absolute phase error between the real phase of the target data segment and the target phase sequence, summing the absolute errors of each target data segment, and using the minimum value as the loss function 2, which represents the minimum power frequency removal error, and the corresponding initial phase and target phase sequence are more suitable for reducing signal distortion in the noise reduction process. Fig.11 , and the result is Fig.10 The processing process is similar to that of , the difference is that loss function 2 is used to screen epochs during the processing.

[0054] The third goal includes: obtaining the phase sum of the target data segments superimposed in pairs, and obtaining the absolute error between the phase sum and zero. This is used as the loss function 3. The smaller the value, the greater the suppression of the power frequency by the noise reduction signal obtained after superposition, and the more suitable the corresponding initial phase and target phase sequence are for suppressing the power frequency. Fig.12 , and the result is Fig.10 The processing process is similar to that of , except that loss function 3 is used to filter epochs.

[0055] In summary, in the application scenario of physiological electrical signals, the signal data is not phase-locked with the power frequency, but with the trigger (stimulation start time), and the epoch is intercepted according to the trigger, so all signals remain in phase and are located at a fixed time point in the data segment, which means that we do not need to pay attention to the signal phase, and only select the data segment with the opposite power frequency phase, and the power frequency amplitude is also gradually changing. Therefore, in this case, with different loss functions as reference, the initial phase and target phase sequence with better effect are selected from multiple groups of data segments with opposite phases.

[0056] Example 3.

[0057] Based on Example 1 or 2, this Example 3 provides an evoked potential imaging machine, including: a processor, which runs the steps of the method; an input unit, which is used to input signal data, such as but not limited to an acquisition module for acquiring evoked electric signals or a connector for transmitting signals; an output unit, which is used to output noise reduction signals, such as but not limited to an oscilloscope or a display; a stimulation unit, which is used to input stimulation signals (such as but not limited to electrical stimulation, visual stimulation, sound stimulation, etc.) to enable the subject to generate evoked electric signals.

[0058] Example 4.

[0059] Based on Embodiment 1 or 2, this Embodiment 4 provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the power frequency removal method.

[0060] Example 5.

[0061] On the basis of Embodiment 1 or 2, this Embodiment 5 provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the power frequency removal method are implemented.

[0062] Example 6.

[0063] On the basis of Embodiment 1 or 2, this Embodiment 6 provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the power frequency removal method are implemented.

[0064] Test example.

[0065] Test condition 1: The noise reduction signal obtained by superimposing and averaging the original data and the noise reduction signal obtained by superimposing and averaging the original data after notch filtering. Fig.13 and Fig.14 When the stimulation artifacts of the C3 channel data and the C4 channel data are filtered by notch filtering, not only the current moment data is needed, but also the data of the past period of time. If the data used contains stimulation artifacts, due to its suddenness and high-frequency characteristics, it will cause the output of the filter to produce instantaneous and drastic fluctuations. This drastic fluctuation will submerge the signal waveform, that is, it is easy to produce filtering artifacts, which in turn affects SEP recognition (N20 component, that is, the signal corresponding to 20ms after stimulation).

[0066] Experimental condition 2: The original data is subjected to denoising processing with a single group and a single initial phase (i.e., no grouping and no traversal), and then the denoised signal is obtained by superimposing and averaging. Fig.15 , a group of all epochs within the phase-pi to pi change period (one cycle) in the original data, a single group only uses the original data of one cycle, sets an initial phase, screens to obtain an epoch sequence, the phases of each epoch in the sequence are evenly distributed, and then the screened epochs are superimposed and averaged. The blue line indicates that all epochs in a single cycle are directly averaged without screening to obtain the denoised result, and the red line indicates the denoised result after screening and averaging.

[0067] Experimental condition 3: The original data is subjected to denoising processing with a single group of multiple initial phases (i.e., not grouped but traversed) and then the denoised signal is obtained by superimposing and averaging. Fig.16 ,and Fig.15 The difference is that more initial phases are set, each initial phase corresponds to an epoch sequence, the phases in the sequence are evenly distributed, and then the filtered epochs are superimposed and averaged. The blue line indicates that all epochs in a single cycle are directly averaged without screening to obtain the denoised result, and the red line indicates the denoised result after screening and averaging.

[0068] Experimental condition 4: The original data is subjected to noise reduction processing of multiple groups of single initial phases (i.e., grouped but not traversed) and then the noise reduction signal is obtained by superposition and averaging. Fig.17 ,and Fig.15 The difference is that multiple raw data from -pi to pi cycles are selected, but only one initial phase is set. The initial phase corresponds to an epoch sequence, and the phases in the sequence are evenly distributed. Then the filtered epochs are superimposed and averaged. The blue line indicates that all epochs in multiple cycles are directly averaged without screening to obtain the denoised result, and the red line indicates the denoised result after screening and averaging.

[0069] Experimental condition 5: The original data is subjected to denoising processing with multiple groups and multiple initial phases (i.e. grouped and traversed), and then the denoised signal is superimposed and averaged. Fig.18 ,and Fig.15 Compared with the previous one, the difference is that multiple raw data from -pi to pi cycles are selected, and multiple initial phases are set. Each initial phase corresponds to an epoch sequence, and the phases in the sequence are evenly distributed. Then the filtered epochs are superimposed and averaged. The blue line indicates the denoised result obtained by directly averaging all epochs in multiple cycles without screening (there is still a large fluctuation between 0.04-0.1s), and the red line is filtered and averaged to obtain the denoised result (as the number of epochs selected for superposition increases, the data fluctuation becomes less and less obvious, indicating that the power frequency is suppressed).

[0070] Test conclusion: The noise reduction signals under various test conditions are compared, as shown in Table 1 below.

[0071] Table 1 Noise reduction results under various test conditions.

[0072] Test conditions Test results Test condition 1 Obvious filtering artifacts Test condition 2 Power frequency suppression is not obvious Test condition 3 Power frequency suppression is not obvious Test condition 4 Power frequency suppression is obvious Test condition 5 Power frequency suppression is obvious

[0073] From Table 1, we can see that:

[0074] (1) In multiple phase change cycles, the target data segments selected according to the phase change cycles are matched with the target phase sequence and then superimposed, which can further suppress the power frequency component.

[0075] (2) While ensuring that the total number of data segments is the same, selecting multiple data segment sequences with initial phase selections can improve the power frequency suppression, but the improvement is not obvious.

[0076] In the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0077] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0078] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0079] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for removing power frequency based on phase change cycle, characterized in that: include: Preprocessing, dividing the signal data into multiple data segments and obtaining the phase of each data segment; Whole cycle reorganization, that is, reorganizing each data segment based on the phase change cycle to obtain the target data segment that matches the target phase sequence; Noise reduction processing, i.e., superimposing the target data segments in any group in pairs to obtain a noise-reduced signal; Screening the noise reduction signal includes: setting the initial phase of the first group of data segments and updating it with a step size, traversing the phase change cycle, and obtaining the noise reduction signals corresponding to different initial phases; According to the evaluation target, the corresponding noise reduction signal is selected as the noise reduction result; the whole cycle reorganization also includes: The phase change period [pi, -pi] of any group of data segments is divided by the sampling phase interval to obtain 2n target phase sequences, where the sampling phase interval is ; The 2n data segments closest to the target phase sequence are selected as the target data segments.

2. The power frequency removal method according to claim 1, characterized in that: The whole cycle reorganization includes: The phase difference of each data segment is obtained, and the data segment whose first phase difference exceeds a first set threshold is determined as the next group of data segments.

3. The power frequency removal method according to claim 1, characterized in that: The whole cycle reorganization also includes: In any group of data segments, when multiple target data segments are detected to be the same data segment, only one target data segment is retained and the remaining target data segments are replaced by data segments of other groups; and\or In any group of data segments, when the phase change of the detected data segment exceeds the second set threshold, the data segment is determined to be an unstable data segment and deleted.

4. The power frequency removal method according to claim 1, characterized in that: The evaluation target includes: any one of the first target, the second target, and the third target; wherein The first target includes: the maximum peak value of the noise reduction signal waveform at the set point; The second objective includes: obtaining the absolute phase error between the real phase of the target data segment and the target phase sequence, and summing the absolute errors of each target data segment; The third objective includes: obtaining a phase sum of target data segments superimposed in pairs, and obtaining an absolute error between the phase sum and zero.

5. An evoked potential image machine, characterized in that: include: A processor, executing the steps of the power frequency removal method as claimed in claim 1 to obtain a noise reduction signal; An input unit, used for inputting signal data; The output unit is used to output the noise reduction signal.

6. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the power frequency removal method as claimed in claim 1.

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 steps of the power frequency removal method as claimed in claim 1 are implemented.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the power frequency removal method as claimed in claim 1 are implemented.

Images