A method and device for generating magnetoencephalogram induced signals

Abstract

The present invention belongs to the field of biological signal processing technology. The present invention discloses a method and device for generating magnetoencephalography (MEG) evoked signals. The method comprises: determining the alignment cost of all signals in the MEG signal based on the minimum cost path between any two trial signals in the MEG signal, wherein the minimum cost path is used to represent the time-minimized alignment cost between the corresponding two trial signals; determining the temporal similarity weight coefficient of the signal for each trial based on the alignment cost; and generating the MEG evoked signal based on the signal for each trial and the corresponding temporal similarity weight coefficient. The method provided by the present invention effectively solves problems in the prior art, such as the inability to effectively extract signals of interest from MEG signals and the inability to effectively reduce the impact of signal jitter.

Description

Technical Field

[0001] The present invention belongs to the technical field of biological signal processing and the technical field of magnetoencephalogram signal preprocessing, and particularly relates to a method and device for generating magnetoencephalogram induced signals. Background Art

[0002] In existing technologies, arithmetic averaging, a classic method for removing noise from MEG signals, has been widely used in auditory, visual, and somatosensory experiments, but it also has obvious limitations. When an experiment is subject to significant noise or outliers (transient high-intensity noise), arithmetic averaging can only rely on increasing the number of experimental trials to minimize these noise artifacts. This is not only time-consuming, but as the experiment progresses, subject fatigue and habituation to repeated stimulation can cause changes in the evoked response waveform. Furthermore, differences between subjects and equipment can lead to blurring of the averaged waveform and shifts in peak timing and amplitude, changes known as signal jitter. Therefore, reducing the impact of high-intensity noise and jitter is a key focus of improving averaging methods. Summary of the Invention

[0003] One object of the present invention is to provide a method for generating magnetoencephalographic (MEG) induced signals, which aims to solve the problems in the prior art of being unable to effectively extract signals of interest from MEG signals and being unable to effectively reduce the influence of signal jitter.

[0004] Another object of the present invention is to provide a device for generating magnetoencephalographic (MEG) evoked signals. Yet another object of the present invention is to provide an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-described method for generating MEG evoked signals are implemented. Yet another object of the present invention is to provide a readable medium storing the computer program, and when the processor executes the computer program, the steps of the above-described method for generating MEG evoked signals are implemented.

[0005] In order to solve the technical problems in the background technology of the present invention, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides a method for generating a magnetoencephalogram induced signal, comprising:

[0007] determining the alignment cost of all signals in the MEG signal according to a minimum cost path between signals of any two trials in the MEG signal, wherein the minimum cost path is used to represent the alignment cost of minimizing time between the signals of the corresponding two trials;

[0008] determining a temporal similarity weight coefficient of the signal of each trial according to the alignment cost;

[0009] A magnetoencephalogram induced signal is generated according to the signal of each trial and the corresponding time similarity weight coefficient.

[0010] In some embodiments of the present invention, a method for generating a magnetoencephalographic induced signal further includes:

[0011] generating a cost matrix of the signals of any two trials;

[0012] determining the Euclidean distance between the signals of any two trials according to the cost matrix;

[0013] The minimum cost path is determined according to the Euclidean distance between the signals of any two trials.

[0014] In some embodiments of the present invention, determining the minimum cost path according to the Euclidean distance between the signals of any two trials includes:

[0015] determining a cumulative distance matrix between the signals of any two trials according to the Euclidean distance;

[0016] The minimum cost path is determined based on the cumulative distance matrix.

[0017] In some embodiments of the present invention, the alignment cost is the average value of the distance difference between matching points on the minimum cost path, and the overall time difference between the signals of all trials in the magnetoencephalographic signal and the magnetoencephalographic evoked signal is less than a preset value.

[0018] In a second aspect, the present invention provides a device for generating a magnetoencephalographic induced signal, the device comprising:

[0019] an alignment cost determination module, configured to determine the alignment cost of all signals in the MEG signal based on a minimum cost path between signals of any two trials in the MEG signal, wherein the minimum cost path is used to represent the alignment cost of minimizing time between the signals of the corresponding two trials;

[0020] a time weight coefficient determination module, configured to determine a time similarity weight coefficient of the signal of each trial according to the alignment cost;

[0021] The induced signal generation module is used to generate an induced magnetoencephalogram signal according to the signal of each trial and the corresponding time similarity weight coefficient.

[0022] In some embodiments of the present invention, a device for generating a magnetoencephalogram induced signal further includes:

[0023] A cost matrix generating module, configured to generate a cost matrix of the signals of any two trials;

[0024] a Euclidean distance determination module, configured to determine the Euclidean distance between the signals of any two trials according to the cost matrix;

[0025] The minimum cost path determination module is configured to determine the minimum cost path according to the Euclidean distance between the signals of any two trials.

[0026] In some embodiments of the present invention, the minimum cost path determination module includes:

[0027] a cumulative distance matrix determining unit, configured to determine a cumulative distance matrix between the signals of any two trials according to the Euclidean distance;

[0028] A minimum cost path determining unit is configured to determine the minimum cost path according to the cumulative distance matrix.

[0029] In some embodiments of the present invention, the alignment cost is the average value of the distance difference between matching points on the minimum cost path, and the overall time difference between the signals of all trials in the magnetoencephalographic signal and the magnetoencephalographic evoked signal is less than a preset value.

[0030] In a third aspect, the present invention provides a computer program product comprising a computer program / instruction, which, when executed by a processor, implements the steps of a method for generating a magnetoencephalographically induced signal.

[0031] In a fourth aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of a method for generating a magnetoencephalogram induced signal are implemented.

[0032] In a fifth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for generating magnetoencephalographic induced signals.

[0033] From the above description, it can be seen that an embodiment of the present invention provides a method and device for generating a magnetoencephalography-induced signal. The corresponding method includes: first, determining the alignment cost of all signals in the magnetoencephalography signal based on the minimum cost path between the signals of any two trials in the magnetoencephalography signal, and the minimum cost path is used to characterize the minimized time alignment cost between the signals of the corresponding two trials; then, determining the time similarity weight coefficient of the signal of each trial based on the alignment cost; finally, generating the magnetoencephalography-induced signal based on the signal of each trial and the corresponding time similarity weight coefficient.

[0034] The present invention proposes an improved robust weighted averaging method to overcome the limitations of existing averaging methods. By calculating the alignment cost between the trial signal and the average signal and using it as the temporal similarity weight coefficient, the criterion function is iteratively minimized to obtain the optimal weight for each trial, thereby reducing the negative impact of outliers and jitter. In addition, the present invention has demonstrated through simulation and real data experiments that the method provided by the present invention is superior to several existing benchmark averaging techniques in terms of signal-to-noise ratio, root mean square error, and waveform recovery. In addition, the method can also be easily extended to other biosignal measurements. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 Schematic diagram of a method for generating a magnetoencephalogram induced signal in an embodiment of the present invention Figure 1 ;

[0037] Figure 2 Schematic diagram of a method for generating a magnetoencephalogram induced signal in an embodiment of the present invention Figure 2 ;

[0038] Figure 3 A schematic flow chart of step 600 in an embodiment of the present invention;

[0039] Figure 4 1 is a flow chart of a method for generating a magnetoencephalographic induced signal in a specific embodiment of the present invention;

[0040] Figure 5 Schematic diagram of the SNR evaluation index results of various simulation experiment methods in a specific embodiment of the present invention;

[0041] Figure 6 Schematic diagram of RMSE evaluation index results of various simulation experiment methods in a specific embodiment of the present invention;

[0042] Figure 7 Schematic diagram of waveform results of different methods in simulation experiments in a specific embodiment of the present invention;

[0043] Figure 8 Schematic diagram of waveforms of different methods in a real auditory experiment in a specific embodiment of the present invention;

[0044] Figure 9A block diagram of a device for generating a magnetoencephalogram induced signal in an embodiment of the present invention Figure 1 ;

[0045] Figure 10 A block diagram of a device for generating a magnetoencephalogram induced signal in an embodiment of the present invention Figure 1 ;

[0046] Figure 11 is a block diagram of the minimum cost path determination module 60 in an embodiment of the present invention;

[0047] Figure 12 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0049] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0050] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or are inherent to these processes, methods, products, or devices. The embodiments of the present invention and the features described in the embodiments may be combined with each other unless there is a conflict. The present invention will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0051] It is understandable that the evoked response in magnetoencephalography (MEG) is of great significance to neuroscience analysis and clinical research, and is usually obtained by averaging many trials. However, the traditional arithmetic averaging method has inherent disadvantages, such as being sensitive to strong abnormal noise, which can only be reduced by increasing the number of experimental trials. In addition, signal jitter caused by factors such as the subject or the device system will also affect the amplitude and time of the average signal peak. Based on this and in order to solve at least part of the technical problems in the prior art, an embodiment of the present invention provides a specific implementation method of a method for generating a magnetoencephalography evoked signal, see Figure 1 , the method specifically includes the following contents:

[0052] Step 100: determining the alignment cost of all signals in the MEG signal according to a minimum cost path between any two trial signals in the MEG signal, wherein the minimum cost path is used to represent the time-minimized alignment cost between the corresponding two trial signals;

[0053] Step 200: Determine a temporal similarity weight coefficient of the signal of each trial according to the alignment cost;

[0054] Step 300: Generate a magnetoencephalogram evoked signal according to the signal of each trial and the corresponding temporal similarity weight coefficient.

[0055] From the above description, it can be seen that an embodiment of the present invention provides a method for generating a magnetoencephalography-induced signal, including: first, determining the alignment cost of all signals in the magnetoencephalography signal based on the minimum cost path between the signals of any two trials in the magnetoencephalography signal, the minimum cost path being used to characterize the minimized time alignment cost between the signals of the corresponding two trials; then, determining the temporal similarity weight coefficient of the signal of each trial based on the alignment cost; finally, generating a magnetoencephalography-induced signal based on the signal of each trial and the corresponding temporal similarity weight coefficient.

[0056] This method calculates the alignment cost between each trial and the average signal, converts it into a temporal similarity coefficient, and updates the weights. It then iteratively minimizes the criterion function to obtain the optimal weight for each trial. The robust weighted averaging method proposed in this paper has been tested using simulation and real-world experimental data, demonstrating superior performance compared to other averaging methods.

[0057] It can be understood that the trial in step 100 can be understood from the following aspects: a trial refers to each complete stimulus-response process in the brain magnetic experiment, which is a time slice of a single complete experimental unit under a specific experimental paradigm, including a pre-stimulus baseline period and a post-stimulus response period.

[0058] Technical parameter description: Time window: For example, "Each trial contains data from 500ms before to 1000ms after the stimulus"; Number of sampling points: For example, "Each trial contains 1500 sampling points." Baseline time: For example, "200ms before the stimulus is used as the baseline period."

[0059] Experimental design description: Number of trials: e.g., "Each subject completed 300 valid trials"; Classification: e.g., "Contains 150 target stimulus trials and 150 non-target stimulus trials"; Inter-trial interval: e.g., "Inter-trial interval randomly varied around 100ms."

[0060] Data processing description: Trial extraction: such as "trial segmentation based on trigger signals"; Trial screening: such as "removing artifact-contaminated trials and retaining valid trials"; Superposition averaging: such as "superposition averaging of all valid trials."

[0061] Regarding step 200, unlike the prior art that uses absolute values ​​as the cost function (which is insensitive to jitter between signals), the degree of temporal difference between the signals of two trials is calculated here to represent the alignment cost of all signals in the MEG signal. Specifically, dynamic programming can be used to find a nonlinear temporal mapping between the two signals to minimize the cumulative distance. First, a distance matrix (such as Euclidean distance) is constructed. Next, the minimum cumulative path from the starting point to the end point is searched. The slope of the path reflects the degree of temporal difference. Among them, the minimum cumulative distance (degree of difference) and the time-curved path.

[0062] Regarding step 300, in actual testing, it's not necessary to calculate the alignment cost for the entire trial. This not only increases computational cost but also results in some time periods dominated by uninteresting noise. Therefore, a time window of interest is set, such as the auditory P100 (approximately 100ms), and the alignment cost and corresponding temporal similarity weight coefficient are calculated only within this time window.

[0063] In some embodiments of the present invention, see Figure 2 , a method for generating a magnetoencephalogram induced signal, further comprising:

[0064] Step 400: Generate a cost matrix of the signals of any two trials;

[0065] Step 500: Determine the Euclidean distance between the signals of any two trials according to the cost matrix;

[0066] Each element in the cost matrix represents the Euclidean distance between the trial signals of the corresponding two sampling points.

[0067] Step 600: Determine the minimum cost path according to the Euclidean distance between the signals of any two trials.

[0068] In some embodiments of the present invention, see Figure 3 , step 600 includes:

[0069] Step 601: determining a cumulative distance matrix between the signals of any two trials according to the Euclidean distance;

[0070] Step 602: Determine the minimum cost path according to the cumulative distance matrix.

[0071] Specifically, the Euclidean distance between the p-th sampling point of the x-trial signal and the q-th sampling point of the v-trial signal in the cost matrix is ​​calculated by the following formula.

[0072] (1)

[0073] Then, the cumulative distance matrix is ​​obtained by recursive calculation , and use this to calculate the minimum cost path P between the two endpoints of the signal, also known as the alignment path.

[0074]

[0075] The alignment path describes the cost of minimizing the temporal alignment between two signals.

[0076] In some embodiments of the present invention, the alignment cost is the average value of the distance difference between matching points on the minimum cost path, and the overall time difference between the signals of all trials in the magnetoencephalographic signal and the magnetoencephalographic evoked signal is less than a preset value.

[0077] Preferably, the overall temporal difference between the signals of all trials in the MEG signal and the MEG-evoked signal is as small as possible.

[0078] From the above description, it can be seen that an embodiment of the present invention provides a method for generating a magnetoencephalography-induced signal, including: first, determining the alignment cost of all signals in the magnetoencephalography signal based on the minimum cost path between the signals of any two trials in the magnetoencephalography signal, the minimum cost path being used to characterize the minimized time alignment cost between the signals of the corresponding two trials; then, determining the temporal similarity weight coefficient of the signal of each trial based on the alignment cost; finally, generating a magnetoencephalography-induced signal based on the signal of each trial and the corresponding temporal similarity weight coefficient.

[0079] Averaging technology is an important step in removing noise from MEG data and obtaining a stable waveform. In view of the fact that the arithmetic averaging method in the prior art is difficult to effectively deal with high-intensity noise and signal jitter, the present invention improves the weighted averaging method of minimizing the accuracy function, and adds the time similarity weight coefficient of the alignment cost to the weight iteration process, so that each trial obtains the optimal weight. Finally, simulation and real auditory evoked experimental data were used to verify that the method proposed in the present invention is superior to a variety of existing averaging methods. Therefore, the method provided by the present invention can be regarded as an alternative to traditional averaging, and the results obtained can improve the accuracy of subsequent research.

[0080] To further illustrate the solution, the present invention also provides a specific implementation of a method for generating a magnetoencephalographic induced signal, which specifically includes the following contents.

[0081] Magnetoencephalography (MEG), a neuroimaging technique with high spatiotemporal resolution, uses non-invasive sensors to measure the weak magnetic fields generated by neural circuit activity within the brain. Similar to the event-related potential (ERP) in electroencephalography (EEG), the event-related field (ERF) in MEG represents the evoked response of the brain's magnetic field to stimulation. Therefore, ERF holds great promise in neuroscience and clinical applications. However, MEG signals are much less susceptible to noise interference. While various denoising methods are currently available, averaging trial-by-trial responses is still necessary to obtain a stable ERF waveform.

[0082] It's understandable that using the median instead of the arithmetic mean can mitigate the impact of extreme values ​​on the results, but it can also include significant high-frequency noise. Furthermore, median averaging can make the results overly robust, potentially losing some useful information. As an improvement, trimmed averaging suppresses extreme values ​​by assigning smaller weights to them by setting a cutoff threshold. However, choosing an appropriate threshold based on the data characteristics is crucial to achieve a balance between suppressing extreme values ​​and retaining useful information.

[0083] In addition to the above methods, weighted averaging is another improved method, which usually assigns a weight to each trial according to its noise level. This requires the assumption that the noise is stable within a trial, but it may not be met in actual experiments. Based on this, see Figure 4 A specific embodiment of a method for generating a magnetoencephalogram induced signal includes the following steps:

[0084] S1: scalar standard function defining the optimal weights.

[0085] Based on the idea of ​​weighted average of criterion function, weighted average can be expressed as:

[0086]

[0087] Where N is the number of trials, k is the sample sampling point, is the weight of the ith trial, is the sample of the ith trial, is the average signal. To find the optimal weight, the scalar standard function is defined as:

[0088] (6)

[0089] in is a measure of the dissimilarity of the vector parameters, is the weighted index parameter. The above formula (6) can be interpreted as V and trial X i The goal is to find the best weight vector This results in the best average signal Since V is fixed, I m The minimization of is considered as a constrained optimization problem.

[0090] S2: Determine the weight vector of the signal for each trial.

[0091] Constraining Equation (6) through Equation (5) and setting the Lagrangian gradient to zero, we finally get the vector :

[0092] (7)

[0093] By iteratively alternating between updating the weight W and averaging the signal , until W is l Iteration convergence: , is a preset small constant. Therefore (7) can be updated as:

[0094] (8)

[0095] S3: Determine the final optimal average signal.

[0096] Here we use the absolute value cost function, i.e. The final optimal average signal It can be expressed as:

[0097] (9)

[0098] S4: Determine the alignment cost of all signals in the MEG signal.

[0099] Please refer to formula (1) to formula (3) for details.

[0100] S5: Determine the temporal similarity weight coefficient of the signal of each trial based on the alignment cost.

[0101] In the calculation process, the trial x with higher temporal similarity to the average signal v is given a larger weight to reduce the impact of trials with greater temporal jitter on v. Therefore, we first use equations (1) and (3) to calculate the alignment cost: , expressed as the average of the distance costs of all matching points on the minimum cost path (the average of the distance differences between each pair of points on the optimal alignment path).

[0102] (10)

[0103] according to Determine the corresponding time similarity weight coefficient :

[0104]

[0105] It should be noted that in actual testing, it is not necessary to calculate the alignment cost of the entire trial, because this not only increases the computational cost, but also some time periods are mainly noises of no interest. Therefore, a time window of a signal of interest is set, such as the moment of auditory P100 (about 100ms), and the alignment cost and the corresponding time similarity weight coefficient are calculated only within the time window. In addition, a weight threshold c is added to the calculation process, and the weights W that are less than c after each iterative update are reset to zero, so as to further reduce the interference that may be caused by strong outliers. Finally, the process of a method for generating magnetoencephalographic evoked signals is as follows:

[0106] Step 1. Fix the weighted index parameter m, weight screening threshold c, and iterative convergence constant and the time window of interest, iterative index ,initialization is the mean or median of all trials.

[0107] Step 2. Calculate using formula (7) , calculated using formula (10) to formula (12) , the vector Multiply each sample by And filter threshold c to update vector .

[0108] Step 3. Update using formula (9) .

[0109] Step 4. If ,but And return to Step 2.

[0110] Experimental results:

[0111] The performance of the algorithm provided by the present invention was tested on simulation experiment data and real auditory evoked experiment data respectively. The signal in the simulation experiment consists of simulated brain neural signals and actual measurement noise. The simulated neural signal is used to simulate the M100 component of the auditory stimulation experiment, and is generated by a signal source placed in a real head model. The signal is a sine wave with a peak of about 500fT, a frequency of 10Hz, and a peak time of 100ms. The signal is limited to a window of 50ms. A total of 300 trials were generated, and each trial lasted for 1s. At the same time, in order to simulate signal jitter, the signal peak in each trial was randomly distributed between 90ms and 110ms. The corresponding waveform was obtained by 26 extracerebral magnetometers through the conduction model. The noise is the empty room noise measured in a shielded room without a subject using a sensor array with the same layout. It was measured for 300 seconds to correspond to the simulation signal.

[0112] The real auditory evoked experimental data used the public OPM auditory dataset, which is 85-channel OPM pure tone stimulation experimental data.

[0113] See also Figures 5 to 8 (Due to image size limitations, Figures 5 to 8 In this paper, the weighted averaging method based on criterion function minimization is abbreviated as the weighted averaging method. In experiments, the proposed algorithm (hereinafter referred to as the dtwWACFM method) was compared with several existing benchmark algorithms. These include the traditional arithmetic mean (AM), median mean (Median), trimmed mean (TM), and the weighted averaging method based on criterion function minimization (WACFM).

[0114] Experimental Setup: The data sampling rate was 1000 Hz. Both the empty room noise data and the OPM auditory data were bandpass filtered from 1 to 40 Hz in the simulation experiment. After bandpass filtering, the simulation data were denoised using the SSP algorithm. After removing corrupted trials (noise amplitude greater than 10,000 fT), the simulation experiment consisted of 250 trials and the real auditory experiment of 500 trials.

[0115] In the simulation and real auditory experiments, the threshold of trimmed average is set to 0.1. The weighted index parameter m in WACFM is 2, and the iterative convergence constant is 1e -5The weighted index parameter m in the method proposed by the present invention is 2, and the iterative convergence constant is 1e -3 , the weight screening threshold c is 1e -3 ,The time window of interest is set to 90ms-110ms.

[0116] Evaluation: To quantitatively analyze the performance of dtwWACFM versus the baseline algorithm, simulations were conducted to measure the signal-to-interference plus noise ratio (SNR) and root mean square error (RMSE) of the algorithm after averaging different numbers of trials.

[0117]

[0118] in, To simulate neural signals, for power. is the reconstructed signal after denoising, is the noise power in the reconstructed signal. N is the number of sensors. In addition, since there are no accurate neural signals in the real experiment, the peak signal-to-noise ratio of the average result is calculated.

[0119]

[0120] in, represents the peak signal during the time period of interest after stimulation, It represents the variance of the signal-noise baseline segment. In addition to the quantitative analysis, the waveform results of simulation and real experiments are also analyzed.

[0121] Figure 5 as well as Figure 6 The following simulations show the SNR and RMSE results for different averaging methods under different trial numbers. All results are averages of 20 random trials. It can be seen that the other methods all show improvement over the arithmetic mean, with the proposed method (herein referred to as dtwWACFM) achieving the best performance across all trial numbers. Compared to the second-best WACFM method, dtwWACFM achieves a maximum SNR improvement of 24% (250 trials) and an average improvement of 16.2%. The RMSE also improves by up to 15% (250 trials) and an average of 10.4%.

[0122] Figure 7Figure 2 shows a representative channel waveform result. The black line represents the simulated neural signal, and the red line represents the dtwWACFM waveform. At the 0.1s evoked signal, the peak amplitude of the dtwWACFM waveform is closest to the simulated signal compared to the other methods, indicating that dtwWACFM minimizes signal distortion. Furthermore, during the remaining time periods without simulated signal activation, the dtwWACFM waveform is also closer to zero, indicating greater noise suppression compared to the other methods.

[0123] Table 1. SNR_peak results of different methods in real listening experiments

[0124]

[0125] In the real auditory experiment, the SNR_peak results of different methods are shown in Table 1. Except for the median result being slightly lower than the arithmetic mean, the trends of other results are similar to those in the simulation experiment. The proposed dtwWACFM method achieves the best performance, improving by at least 4.1% compared to other methods.

[0126] Figure 8 This is the waveform result for one channel in the auditory experiment. It can be seen that dtwWACFM effectively preserves the M100 component (around 0.1s), with no significant distortion compared to other methods. Furthermore, the waveform is closer to zero in the remaining time periods, including the baseline and non-evoked periods, effectively suppressing noise in the signal.

[0127] In summary, MEG, EEG, and other physiological monitoring technologies all require averaging multiple trials to remove residual noise and obtain stable and distinct evoked waveforms. However, when the noise intensity is high or the evoked signal has a certain amount of jitter (not strictly aligned), the performance of existing arithmetic averaging methods will decline. Signal jitter can occur for many reasons, including variations in the evoked signal from different subjects, fatigue from repeated experiments, or adaptation to the stimulus. In addition, different experimental equipment may cause certain random systematic errors in the generation and recording of stimulation signals. Although improved averaging methods such as median, trimmed average, and weighted average have been proposed, it is still difficult to solve these two problems simultaneously. Therefore, the present invention proposes a method for generating magnetoencephalographic evoked signals. This method improves on the existing weighted averaging method by adding the calculated temporal similarity weight coefficient to the weight update process, thereby improving the algorithm's ability to cope with signal jitter.

[0128] dtwWACFM updates the weight of each trial that can minimize the criterion function during iteration, and selects the absolute value of the difference between the trial signal and the average signal as the criterion function. Therefore, this iterative process can be expressed as the trial with a greater contribution to the average signal (the more similar) can obtain a higher weight, and conversely, the trial with more noise obtains a smaller weight. However, compared with high-intensity noise, the impact of small-amplitude jitters of the induced signal between different trials on the criterion function is much smaller, but these jitters will cause changes in the time and amplitude of the induced peak of the final average result. The alignment cost can describe the temporal similarity between the two signals. The trial signal with smaller jitter has a smaller alignment cost with the average signal. Therefore, the present invention will further multiply the weight obtained by the criterion function by the temporal similarity weight coefficient, so that the trial with smaller jitter obtains a larger weight. Based on the above process, dtwWACFM achieves better performance than other averaging methods.

[0129] Simulations and real-world auditory experiments demonstrate that the proposed method outperforms other benchmark averaging methods. dtwWACFM not only achieves the best SNR and RMSE results, but also produces waveforms that are closer to theoretical or ideal values.

[0130] It is worth noting that although the method provided by the present invention is verified on OPM-MEG data, dtwWACFM can be easily extended to the application of EEG, SQUID-MEG or other physiological detection technologies.

[0131] Based on the same inventive concept, an embodiment of the present invention also provides a device for generating magnetoencephalographic induced signals, which can be used to implement the methods described in the above embodiments, such as the following embodiments. Since the principle of solving the problem by the device for generating magnetoencephalographic induced signals is similar to that of the method for generating magnetoencephalographic induced signals, the implementation of the device for generating magnetoencephalographic induced signals can refer to the implementation of the method for generating magnetoencephalographic induced signals, and the repeated parts will not be repeated. As used below, the terms "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the system described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and conceived.

[0132] The embodiment of the present invention provides a specific embodiment of a device for generating a magnetoencephalogram induced signal that can realize a method for generating a magnetoencephalogram induced signal, see Figure 9 , a device for generating magnetoencephalogram induced signals specifically includes the following contents:

[0133] an alignment cost determination module 10, configured to determine the alignment cost of all signals in the MEG signal based on a minimum cost path between any two trial signals in the MEG signal, wherein the minimum cost path is used to represent the time-minimized alignment cost between the corresponding two trial signals;

[0134] a time weight coefficient determination module 20, configured to determine a time similarity weight coefficient of the signal of each trial according to the alignment cost;

[0135] The induced signal generating module 30 is configured to generate an induced magnetoencephalogram signal according to the signal of each trial and the corresponding time similarity weight coefficient.

[0136] In some embodiments of the present invention, see Figure 10 , a device for generating a magnetoencephalogram induced signal, further comprising:

[0137] A cost matrix generating module 40 is used to generate a cost matrix of the signals of any two trials;

[0138] a Euclidean distance determination module 50, configured to determine the Euclidean distance between the signals of any two trials according to the cost matrix;

[0139] The minimum cost path determination module 60 is configured to determine the minimum cost path according to the Euclidean distance between the signals of any two trials.

[0140] In some embodiments of the present invention, see Figure 11 , the minimum cost path determination module 60 includes:

[0141] a cumulative distance matrix determining unit 60a, configured to determine a cumulative distance matrix between the signals of any two trials according to the Euclidean distance;

[0142] The minimum cost path determining unit 60b is configured to determine the minimum cost path according to the cumulative distance matrix.

[0143] In some embodiments of the present invention, the alignment cost is the average value of the distance difference between matching points on the minimum cost path, and the overall time difference between the signals of all trials in the magnetoencephalographic signal and the magnetoencephalographic evoked signal is less than a preset value.

[0144] The embodiment of the present invention also provides a specific implementation of an electronic device that can implement all the steps in the method for generating magnetoencephalogram induced signals in the above embodiment, see Figure 12 , electronic equipment specifically includes the following:

[0145] Processor 1201, memory 1202, communication interface 1203 and bus 1204;

[0146] The processor 1201, the memory 1202, and the communication interface 1203 communicate with each other via the bus 1204; the communication interface 1203 is used to implement information transmission between the server-side device and the user-side device and other related devices;

[0147] The processor 1201 is configured to call the computer program in the memory 1202. When the processor executes the computer program, all steps of the method for generating magnetoencephalographic evoked signals in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0148] Step 100: determining the alignment cost of all signals in the MEG signal according to a minimum cost path between any two trial signals in the MEG signal, wherein the minimum cost path is used to represent the time-minimized alignment cost between the corresponding two trial signals;

[0149] Step 200: Determine a temporal similarity weight coefficient of the signal of each trial according to the alignment cost;

[0150] Step 300: Generate a magnetoencephalogram evoked signal according to the signal of each trial and the corresponding temporal similarity weight coefficient.

[0151] An embodiment of the present invention further provides a computer-readable storage medium capable of implementing all steps of the method for generating a magnetoencephalographically induced signal in the above embodiment. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the computer program implements all steps of the method for generating a magnetoencephalographically induced signal in the above embodiment. For example, when the processor executes the computer program, the following steps are implemented:

[0152] Step 100: determining the alignment cost of all signals in the MEG signal according to a minimum cost path between any two trial signals in the MEG signal, wherein the minimum cost path is used to represent the time-minimized alignment cost between the corresponding two trial signals;

[0153] Step 200: Determine a temporal similarity weight coefficient of the signal of each trial according to the alignment cost;

[0154] Step 300: Generate a magnetoencephalogram evoked signal according to the signal of each trial and the corresponding temporal similarity weight coefficient.

[0155] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the hardware + program embodiments are generally similar to the method embodiments, so their description is relatively simple. For relevant portions, refer to the description of the method embodiments.

[0156] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0157] While the present invention provides method steps as shown in the embodiments or flowcharts, more or fewer steps may be included based on routine or uninventive work. The order of steps listed in the embodiments is merely one of many possible execution sequences and does not represent the only execution sequence. When implemented in a practical device or user-end product, the methods shown in the embodiments or figures may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0158] For the convenience of description, the above devices are described in terms of functions divided into various modules. Of course, when implementing the embodiments of this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the modules that implement the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0159] Those skilled in the art will also appreciate that, in addition to implementing the controller in pure computer-readable program code, it is entirely possible to implement the same functionality by logically programming the method steps in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, and the like. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be considered structures within the hardware component. Alternatively, the devices for implementing various functions can be considered both software modules implementing the method and structures within the hardware component.

[0160] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0161] Memory may include non-permanent storage in a computer-readable medium in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (RAM). Memory is an example of a computer-readable medium.

[0162] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between the various embodiments can be referenced across them. Each embodiment focuses on the differences from the other embodiments. In particular, since the system embodiments are generally similar to the method embodiments, their description is relatively simple. For relevant parts, reference can be made to the description of the method embodiments. Throughout this specification, reference to the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" means that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the embodiments in this specification. In this specification, the schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, those skilled in the art may combine and integrate the different embodiments or examples, and features of different embodiments or examples, described in this specification, without conflict.

[0163] The above description is merely an example of the embodiments of this specification and is not intended to limit the embodiments of this specification. For those skilled in the art, various modifications and variations of the embodiments of this specification are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of this specification shall be included within the scope of the claims of the embodiments of this specification.

Claims

1. A method for generating a magnetoencephalogram induced signal, characterized in that: include: determining the alignment cost of all signals in the MEG signal according to a minimum cost path between signals of any two trials in the MEG signal, wherein the minimum cost path is used to represent the alignment cost of minimizing time between the signals of the corresponding two trials; determining a temporal similarity weight coefficient of the signal of each trial according to the alignment cost; generating a magnetoencephalogram evoked signal according to the signal of each trial and the corresponding time similarity weight coefficient; Determining a temporal similarity weight coefficient of the signal of each trial according to the alignment cost includes: S1: Define the scalar standard function of the optimal weight: Based on the idea of ​​weighted average of the criterion function, the weighted average is expressed as: in N is the number of trials, k is the sample sampling point, is the weight of the ith trial, is the sample of the ith trial, is the average signal; the scalar standard function for finding the optimal weight is defined as: （6） in is a measure of the dissimilarity of the vector parameters, is the weighted index parameter; the above formula (6) is interpreted as V and trial X i The goal is to find the best weight vector This results in the best average signal ; Since V is fixed, I m The minimization of is considered as a constrained optimization problem; S2: Determine the weight vector of the signal for each trial; Constraining Equation (6) through Equation (5) and setting the Lagrangian gradient to zero, we finally get the vector : (7) By iteratively alternating between updating the weight W and averaging the signal , until W is l Iteration convergence: , is a preset constant; therefore (7) is updated as: (8) S3: Determine the final optimal average signal: use the absolute value cost function, ; Final optimal average signal Expressed as: (9) S4: Determine the alignment cost of all signals in the MEG signal; S5: Determine the temporal similarity weight coefficient of the signal of each trial based on the alignment cost; In the calculation process, the average signal v Trials with higher temporal similarity x The larger the weight obtained, the more time jitter the trial pair will have. v The impact of ; therefore, the alignment cost is calculated first , expressed as the average distance cost of all matching points on the minimum cost path; (10) according to Determine the corresponding time similarity weight coefficient : 。 2. The method for generating magnetoencephalographic induced signals according to claim 1, wherein: Also includes: generating a cost matrix of the signals of any two trials; determining the Euclidean distance between the signals of any two trials according to the cost matrix; The minimum cost path is determined according to the Euclidean distance between the signals of any two trials.

3. The method for generating magnetoencephalographic induced signals according to claim 2, wherein: Determining the minimum cost path according to the Euclidean distance between the signals of any two trials comprises: determining a cumulative distance matrix between the signals of any two trials according to the Euclidean distance; The minimum cost path is determined based on the cumulative distance matrix.

4. The method for generating magnetoencephalographic induced signals according to claim 1, wherein: The alignment cost is an average value of the distance differences between matching points on the minimum cost path, and the overall time difference between the signals of all trials in the magnetoencephalographic signal and the magnetoencephalographic evoked signal is less than a preset value.

5. A device for generating magnetoencephalogram induced signals, characterized in that: include: an alignment cost determination module, configured to determine the alignment cost of all signals in the MEG signal based on a minimum cost path between signals of any two trials in the MEG signal, wherein the minimum cost path is used to represent the alignment cost of minimizing time between the signals of the corresponding two trials; a time weight coefficient determination module, configured to determine a time similarity weight coefficient of the signal of each trial according to the alignment cost; an induced signal generating module, configured to generate an induced magnetoencephalogram signal according to the signal of each trial and the corresponding time similarity weight coefficient; Determining a temporal similarity weight coefficient of the signal of each trial according to the alignment cost includes: S1: Define the scalar standard function of the optimal weight: Based on the idea of ​​weighted average of the criterion function, the weighted average is expressed as: in N is the number of trials, k is the sample sampling point, is the weight of the ith trial, is the sample of the ith trial, is the average signal; the scalar standard function for finding the optimal weight is defined as: （6） in is a measure of the dissimilarity of the vector parameters, is the weighted index parameter; the above formula (6) is interpreted as V and trial X i The goal is to find the best weight vector This results in the best average signal ; Since V is fixed, I m The minimization of is considered as a constrained optimization problem; S2: Determine the weight vector of the signal for each trial; Constraining Equation (6) through Equation (5) and setting the Lagrangian gradient to zero, we finally get the vector : (7) By iteratively alternating between updating the weight W and averaging the signal , Until W l Iteration convergence: , is a preset constant; therefore (7) is updated as: (8) S3: Determine the final optimal average signal: use the absolute value cost function, ; Final optimal average signal Expressed as: (9) S4: Determine the alignment cost of all signals in the MEG signal; S5: Determine the temporal similarity weight coefficient of the signal of each trial based on the alignment cost; In the calculation process, the average signal v Trials with higher temporal similarity x The larger the weight obtained, the more time jitter the trial pair will have. v The impact of ; therefore, the alignment cost is calculated first , expressed as the average distance cost of all matching points on the minimum cost path; (10) according to Determine the corresponding time similarity weight coefficient : 。 6. The device for generating magnetoencephalographic induced signals according to claim 5, characterized in that: Also includes: A cost matrix generating module, configured to generate a cost matrix of the signals of any two trials; a Euclidean distance determination module, configured to determine the Euclidean distance between the signals of any two trials according to the cost matrix; The minimum cost path determination module is configured to determine the minimum cost path according to the Euclidean distance between the signals of any two trials.

7. The device for generating magnetoencephalographic induced signals according to claim 6, characterized in that: The minimum cost path determination module includes: a cumulative distance matrix determining unit, configured to determine a cumulative distance matrix between the signals of any two trials according to the Euclidean distance; A minimum cost path determining unit is configured to determine the minimum cost path according to the cumulative distance matrix.

8. The device for generating magnetoencephalographic induced signals according to claim 5, characterized in that: The alignment cost is an average value of the distance differences between matching points on the minimum cost path, and the overall time difference between the signals of all trials in the magnetoencephalographic signal and the magnetoencephalographic evoked signal is less than a preset value.

9. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method for generating magnetoencephalographic evoked signals according to any one of claims 1 to 4 are implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method for generating a magnetoencephalographic evoked signal according to any one of claims 1 to 4 are implemented.

11. 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 method for generating magnetoencephalographic evoked signals according to any one of claims 1 to 4 are implemented.

Images