Ratchet wave detection system and epilepsy detection equipment

Abstract

The invention belongs to the technical field of physiological electric signal detection, and particularly relates to a ratchet wave detection system and epilepsy detection equipment. According to the ratchet wave detection system, the ratchet wave detection unit is used for detecting signal data to obtain the ratchet wave detection result, then the ratchet wave screening unit is used for screening the ratchet wave from the ratchet wave detection result, the waveform of the ratchet wave is displayed in the waveform display area, and finally the preset value of the sensitivity is adjusted to change the displayed waveform of the ratchet wave. The detection error caused by the difference of the ratchet wave forms can be better processed, the ratchet wave forms are combined with the screening threshold value, the universality requirement is met, and the whole system has high adaptability.

Description

Technical Field

[0001] The present invention belongs to the technical field of physiological electrical signal detection, and in particular relates to a spike wave detection system and an epilepsy detection device. Background Art

[0002] Epilepsy spike detection is crucial in clinical diagnosis because spikes are the hallmark electrophysiological features of epileptic activity. With the advancement of artificial intelligence and signal processing technology, researchers have developed some automatic spike detection methods to reduce the workload of clinicians. These methods include algorithms based on machine learning and deep learning, which can automatically extract features from EEG and detect spikes. However, due to the significant differences in spike morphology between individuals, these methods have limited versatility among different patients and are difficult to adapt to the individual characteristics of each patient. For example, patent number CN114869301A discloses a method and device for detecting epileptic discharges, which mentions the comparison of at least one of the following waveform features with a waveform template to identify whether there is an epileptic discharge in the signal sub-segment: the amplitude difference of the rising edge of the spike or sharp wave in the signal sub-segment; the duration of the rising edge of the spike or sharp wave in the signal sub-segment; the amplitude difference of the falling edge of the spike or sharp wave in the signal sub-segment; the duration of the falling edge of the spike or sharp wave in the signal sub-segment; the amplitude difference of the rising edge of the slow wave following the falling edge of the spike or sharp wave in the signal sub-segment; the standard deviation of the amplitude before the rising edge of the spike or sharp wave in the signal sub-segment. This technical solution detects spikes based on the time-frequency similarity or frequency division energy of a fixed template. Since the spike wave morphology will change under different epileptic seizure states, the fixed template will have a template range that cannot cover more types of spike waves. If there is a difference between the spike wave waveform and the fixed template, the detection system may have a large number of missed detections or false detections. Summary of the invention

[0003] The present invention provides a technical field of physiological electrical signal detection, and specifically relates to a spike wave detection system and an epilepsy detection device, wherein a screening threshold is adjusted according to the spike wave recognition result to solve the problem of missed detection or false detection of the device.

[0004] In order to solve the above technical problems, the present invention provides a spike wave detection system, comprising: a processor and a data calculation unit respectively connected to the processor, a human-computer interaction machine, a spike wave detection unit, and a spike wave screening unit; wherein the spike wave detection unit detects signal data to obtain a spike wave detection result; the spike wave screening unit screens spike waves from the spike wave detection result to obtain a spike wave screening result; a sensitivity option and a waveform display area are provided on the operation interface of the human-computer interaction machine; the waveform display area is used to display the spike wave waveform in the spike wave screening result; the sensitivity option is used to input a preset value of sensitivity to adjust the displayed spike wave waveform.

[0005] Furthermore, the spike wave screening unit includes: a first screening module, which sets the sensitivity to the quantile corresponding to the amplitude difference threshold of the spike wave to screen the spike wave, including: in the spike wave detection result, using the data calculation unit to calculate the amplitude difference from the peak to the end point of each spike wave and its mean and standard deviation; sorting the amplitude difference of each spike wave according to size; using the data calculation unit to calculate the proportion of the amplitude difference of each spike wave in the sorted position to obtain the quantile corresponding to each amplitude difference; the processor determines whether the standard deviation of the amplitude difference is greater than the quantile at the corresponding position; if so, the amplitude difference at the sensitivity corresponding position is used as the amplitude difference threshold; if not, the difference between the mean of the amplitude difference and N times the standard deviation is used as the amplitude difference threshold.

[0006] Furthermore, when the sensitivity is set in the range of [0%, 20%), N is 0.5; when the sensitivity is set in the range of [20%, 80%), N is 1; when the sensitivity is set in the range of [80%, 100%), N is 2.

[0007] Furthermore, the spike wave screening unit also includes: a second screening module, which sets the sensitivity to the quantile corresponding to the slope threshold of the spike wave to screen the spike waves, including: in the spike wave detection results, using the data calculation unit to calculate the slope of each spike wave; sorting the slope of each spike wave according to size; and using the data calculation unit to calculate the proportion of the slope of each spike wave in the sorting position as the sensitivity.

[0008] Furthermore, the spike detection unit includes: a data slicing module, which performs sliding window slicing processing on the signal data to obtain data segments and their corresponding starting times; a spike detection model, whose input end is the data segment and the output end is the spike detection result of each data segment, the output end result is 1, indicating that there is a spike signal, and the output end is 0, indicating that there is no spike signal.

[0009] Furthermore, it also includes: a data correction unit; the data correction unit is used to delete resting data and high-amplitude artifact data in the spike wave detection results; mark the data segments with a standard deviation greater than a set threshold as spike wave signals; mark the data segments with a standard deviation less than a set threshold as normal signals, i.e., resting data; calculate the difference between the maximum and minimum values ​​in each data segment, as well as its mean and standard deviation, and mark the data segments with a difference greater than the mean and standard deviation as artifact signals, i.e., high-amplitude artifact data.

[0010] Furthermore, it also includes: a peak positioning unit, which marks the peak position of the spike wave signal based on the spike wave detection result, that is, using the data calculation unit to calculate the difference between the maximum value and the minimum value in the spike wave signal, sorting the difference from large to small, taking the difference at the preset position multiplied by the peak threshold coefficient as the peak threshold; searching for the peak of the spike wave signal based on the peak threshold and the half-wave width threshold, and taking the sum of the time corresponding to the peak and the starting time as the peak position of the spike wave; merging the peaks whose time difference is less than the set time threshold into one peak, and calculating the starting position and end position of each peak.

[0011] Furthermore, it also includes: a multi-data integration unit; the multi-data integration unit includes: a preprocessing module, which preprocesses the signal data to obtain the first data; a data reconstruction module, which decomposes and reconstructs the signal data into independent components to obtain the second data; and an integration module, which integrates the spike wave detection results after the spike wave detection unit detects the first data and the second data respectively.

[0012] Furthermore, the integrated spike detection results include: counting channels in the first data where the number of spikes detected is not less than a set value as merged channels, and channels where the number of spikes detected is less than the set value as other channels; in the merged channel, taking the union of the detection results corresponding to the first data and the second data, and in the union, merging the spikes whose spike position intervals are less than a threshold as the same spikes; in other channels, retaining the detection results corresponding to the first data.

[0013] In a second aspect, the present invention provides an epilepsy detection device, comprising: the spike wave detection system described above; a data display area is also provided on the operating interface of the human-computer interaction machine to display the quantitative statistical results of the epilepsy waveform; the quantitative statistical results of the epilepsy waveform include: the number and peak size of spike wave peaks of the EEG signal in the spike wave screening results.

[0014] The beneficial effect of the present invention is that the spike wave detection system of the present invention first uses the spike wave detection unit to detect signal data to obtain a spike wave detection result, then uses the spike wave screening unit to screen the spike waves from the spike wave detection result, and displays the spike wave waveform in the spike wave screening result in the waveform display area, and finally sets the preset value of the sensitivity to adjust the spike wave screening result, which can not only better handle the detection error caused by the difference in spike wave morphology, but also combine the spike wave detection result with the screening threshold to meet the universality requirements, so that the entire system has higher adaptability.

[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. In order to make the above-mentioned objects, features and advantages of the present invention more obvious and understandable, the following is a detailed description of the preferred embodiments in conjunction with the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] 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.

[0017] Figure 1 It is a workflow diagram of the spike wave detection system using two types of data detection.

[0018] Figure 2 It is a schematic diagram of data sliding window slicing.

[0019] Figure 3 It is an iterative curve chart of the spike detection model’s accuracy after training and verification.

[0020] Figure 4 It is an iterative graph of the training validation loss of the spike detection model.

[0021] Figure 5 is the confusion matrix of the detection results.

[0022] Figure 6 It is a schematic diagram of the process of two spike screening modules.

[0023] Figure 7 It is a schematic diagram of the amplitude difference in the waveform diagram.

[0024] Figure 8 This is a schematic diagram of the selection of the trial channel.

[0025] Fig. 9 It is a schematic diagram of the steepness in the waveform graph.

[0026] Fig.10 It is a schematic diagram that uses the quantile corresponding to the spike wave slope in the trial channel as the sensitivity.

[0027] Fig.11 It is a flowchart of obtaining the waveform parameters of spike waves.

[0028] Fig.12 It is a schematic diagram of the division of the first half wave and the second half wave in the waveform diagram.

[0029] Fig.13 This is a schematic diagram of the calculation process of the end point of the second half wave.

[0030] Fig.14 This is a schematic diagram of the calculation process of the starting point of the first half wave.

[0031] Fig.15 This is the detection result of the C4 channel of epileptic patients using a 1s window with no overlapping sliding window.

[0032] Fig.16 This is the detection result of a 1s window with 80% overlap of the C4 channel in epileptic patients.

[0033] Fig.17 This is the detection result of epileptic patients with multiple channels set to 0% sensitivity.

[0034] Fig.18 This is the detection result of epilepsy patients with multiple channels set to 25% sensitivity. DETAILED DESCRIPTION

[0035] 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.

[0036] See Figure 1-Figure 14 This embodiment provides a spike wave detection system, including: a processor and a data calculation unit connected to the processor, a human-computer interaction machine, a spike wave detection unit, and a spike wave screening unit; wherein the spike wave detection unit detects signal data to obtain a spike wave detection result; the spike wave screening unit screens spike waves from the spike wave detection result to obtain a spike wave screening result; a sensitivity option and a waveform display area are set on the operation interface of the human-computer interaction machine; the waveform display area is used to display the spike wave waveform in the spike wave screening result; the sensitivity option is used to input a preset value of the sensitivity to adjust the displayed spike wave waveform. The spike wave detection unit includes but is not limited to: a data slicing module, a spike wave detection model and other functional modules. Specifically, the data slicing module is used to perform sliding window slicing processing on the data, the spike wave detection model is used to detect whether each data slice contains a spike wave signal and mark the spike wave slice, the data correction unit is used to eliminate the spike wave slices that are misidentified, and the peak positioning unit calculates the waveform parameters and spike wave positions of the spike wave based on the spike wave slices, and finally is used to display the spike wave waveform in the waveform display area or calculate the screening threshold (i.e., sensitivity) of the spike wave. Generally, the corresponding function can be selected on the operation interface of the human-computer interaction machine. When the preset value of sensitivity (i.e., the screening threshold of spike wave) is set, the data passes through the spike wave detection unit (data slicing module, spike wave detection model, data correction unit, peak location unit), spike wave screening unit in sequence, and finally the spike wave waveform in the spike wave screening result is displayed in the waveform display area (such as Fig.17 or Fig.18 At this time, the spike wave screening results can be manually observed (usually by professional technicians or operators familiar with the corresponding spike wave waveforms) and the preset sensitivity value can be adjusted, for example Fig.17If there are many obvious false detections of spike waves, increase the preset sensitivity value; conversely, if there are many missed detections of spike waves, decrease the preset sensitivity value.

[0037] In some embodiments, the spike wave detection system can complete online detection and data calculation with the help of a computer and its corresponding software and hardware, and its working method can be stored in a memory or a readable storage medium to form a computer program or a computer program product, and the steps of the working method are implemented when the computer program is executed by a processor. Figure 1 , its working method comprises the following steps: Step S1, sliding window slicing processing. The data slicing module in the spike wave detection unit is used to perform sliding window slicing processing on the online multi-channel EEG data, and the data segments and the starting time of each slice are output through the operation interface or the display interface. Figure 2 As shown in the figure, to ensure that the spike wave can be covered by at least one complete window, the data is sliced ​​by overlapping sliding windows. Assuming that the data duration of a single channel is L seconds, the data is sliced ​​by sliding windows, the window length is T, and the step length is ΔT=200ms; then the starting time of the i-th slice can also be determined For example, each 1.5s spike wave data is subjected to multiple 1s data truncation processing with a random non-repeating starting point to ensure that the spike wave slice contains at least one complete spike wave. Similarly, the background EEG data is subjected to multiple 1s data truncation processing with a random non-repeating starting point.

[0038] Step S2: Data normalization: Perform zscore normalization on each 1s slice data.

[0039] Step S3, spike detection. Input the data slices into the trained spike detection model, and output the spike detection results of each slice. Calculate the standard deviation of the slice data, mark the slices with a standard deviation less than the preset value th1 as normal EEG and update the corresponding labels; calculate the difference between the maximum value and the minimum value of each slice data channel by channel, and calculate the mean and standard deviation of all the maximum and minimum value differences of each channel, and mark the slices in each channel that are greater than the mean plus the standard deviation of each channel as artifact slices and update the corresponding labels. That is, a label of 1 indicates that it contains spike signals, and a label of 0 indicates that it does not contain spike signals.

[0040] Optionally, the spike detection model includes, but is not limited to, selecting an EEGNet deep EEG model, setting the convolution kernel to k=(1, 32), the feature map F=16, using the Adam optimizer, the learning rate lr=0.01, using offline data or historical data to detect signal data, and selecting the training group model with the smallest verification loss as the final solidified model to train the spike detection model. For example, Figure 3The iterative curve of the spike wave detection model after training and verification is shown. It can be seen that the accuracy of model training and verification gradually increases with the increase of training progress, indicating that the judgment of spike waves is accurate in actual use. Figure 4 The iterative curve of the spike detection model after training and verification is shown. It can be seen that the loss of model training and verification gradually decreases with the increase of training progress, indicating that the training of the model is stable and reliable. Of course, in order to ensure the robustness of the model, the offline data or historical data can be annotated with the positions of spikes in some epilepsy data by a semi-automatic spike expert system model approved by clinicians. The data of 1.5s is intercepted with the peak of the spike as the center as the spike type training data, and the same amount of 1.5s EEG data without spikes is taken as background EEG; some additional artifact data can also be added as background. In practice, it is found that the spike detection model has been trained and verified, and the test effect of spike and background EEG data is greatly improved compared with the spike detection model that has not participated in the training. Figure 5 As shown, its test accuracy can reach over 90%, with true positives and true negatives of 99% and 89% respectively.

[0041] Step S4, data correction. The slices with spike detection results marked as 1 (i.e., spike slices) are input into the data correction unit to eliminate the misidentified spike slices; the correction module eliminates the two types of misidentified spike data labeled as resting data and high-amplitude artifact data. Specifically, the data segments with a standard deviation greater than the set threshold are marked as spike signals; the data segments with a standard deviation less than the set threshold are marked as normal signals, i.e., resting data; the difference between the maximum and minimum values ​​in each data segment, as well as its mean and standard deviation, are calculated, and the data segments with a difference greater than the mean and standard deviation are recorded as artifact signals, i.e., high-amplitude artifact data.

[0042] Step S5: peak location. For details, see Fig.11 , calculate the difference between the maximum value and the minimum value of the spike signal (i.e., the spike slice with label 1) for each channel, sort the differences from large to small, take the difference at the percentile X1 or X2 (e.g., 20%~30% position) and multiply it by the peak threshold coefficient as the peak threshold of the channel; use the calculated peak threshold and the set half-wave width threshold to search the peak position of the spike slices of each channel, and add the corresponding slice start time to the peak position of each slice, which is the real time of each peak; merge the peaks with close real time in each channel into one peak, and the so-called close means that the time difference between the two peaks is less than the tolerance time limit th2; calculate the start and end positions of each peak for each channel; calculate the peak width of each peak: the width from the start to the end point, and exclude peaks that are not within the peak width threshold range [w1 w2]. Finally, the quantitative statistical results of the epileptic waveform are given based on the number and peak value of the spikes of each channel. For example, see Fig.12, the peak point position of a peak is known. Since the spike wave width is about 40ms-70ms, 100ms from the peak point is taken as the first half wave to search for the starting point, and 200ms from the peak point is taken as the second half wave to search for the end point; 200ms from the peak point is taken as the second half wave Take the search end point as an example, and record the peak position as , 200ms later is recorded as .See Fig.13 , the end point of the second half wave is obtained by the following steps: ① Make a first-order difference on the second half wave ; ② Perform symbol sign(x) judgment, , the result is recorded as ; ③Yes Do first-order differences ④ Find The positions of all points greater than zero in ⑤Calculation Zhongcong arrive The proportion of "1" in the is the end point. If r < 0.5, continue to determine the next position. arrive The proportion of "1" in r, until , if none of them are satisfied, the termination point is .See Fig.14 The calculation starting point of the first half wave is similar to the calculation ending point of the second half wave. It is only necessary to flip the first half wave left and right with the peak as the symmetric point and then calculate it. Finally, the waveform parameters of the spike wave are obtained through the peak positioning unit to calculate the screening threshold of the spike wave.

[0043] Step S6, screening spike waves. Figure 6 , the spike wave screening unit calculates the screening threshold of the spike wave according to the waveform parameters of the spike wave. Specifically, the spike wave screening unit includes: a first screening module, which sets the sensitivity to the quantile corresponding to the amplitude threshold of the spike wave to screen the spike wave; a second screening module, which sets the sensitivity to the quantile corresponding to the slope threshold of the spike wave to screen the spike wave. Obviously, the two screening modules can be used separately or simultaneously. In actual operation, the first screening module is generally used for the initial screening, and then the second screening module is used to perform a secondary screening on the spike waves after the initial screening, and finally the result is used as the spike wave screening result. But the order of the two can also be reversed.

[0044] Optionally, in the first screening module, setting the sensitivity to the quantile corresponding to the amplitude difference threshold of the spike wave includes: in the spike wave detection result, obtaining the amplitude difference from the peak to the end point of the spike wave and its mean and standard deviation; sorting the amplitude differences of each spike wave by size; using the data calculation unit to calculate the proportion of the amplitude difference of each spike wave in the sorted position to obtain the quantile corresponding to each amplitude difference; the processor determines whether the standard deviation of the amplitude difference is greater than the quantile at the corresponding position; if so, the amplitude difference at the sensitivity corresponding position is used as the amplitude difference threshold; if not, the difference between the mean of the amplitude difference and N times the standard deviation is used as the amplitude difference threshold. For details, see Figure 7 , set the relative amplitude to be defined as: record the peak position obtained in step S5 as , the peak amplitude is recorded as , end point position , the end point amplitude is ; Amplitude difference ; Calculate the amplitude difference from the peak value to the end point of each peak of each channel, recorded as the amplitude difference; re-order the amplitude differences of all peaks of each channel, take the amplitude difference at percentile X1 as the amplitude difference P1 of the channel, calculate the mean M and standard deviation S of the amplitude differences of all peaks of each channel, if the standard deviation S of a channel is greater than P1, then select P1 as the amplitude difference threshold th3 of the channel, otherwise select the standard deviation minus the mean as the amplitude difference threshold th3; judge channel by channel, remove the peaks with amplitude differences less than the amplitude difference threshold th3, and the remaining peaks are used as secondary screening spike peaks. The percentile X1 corresponding to the spike amplitude difference threshold is an adjustable parameter exposed to the user, and the user can adjust it according to his needs. If the spike amplitudes are sorted from small to large, the larger the X1 value, the higher the specificity of spike detection, and the smaller the value, the higher the sensitivity of spike detection. Generally, in order to overcome the problem of data distribution differences, if the amplitude difference distribution of spikes is very scattered, it means that the data fluctuates violently, and there are extreme values ​​or outliers. It is appropriate to use the amplitude difference at the quantile corresponding to the preset value of sensitivity as the amplitude difference threshold to eliminate outliers; if the amplitude difference distribution of spikes is relatively tight, it means that the data is relatively stable and the value is relatively stable. The difference between the mean of the amplitude difference at the quantile corresponding to the preset value of sensitivity and N times the standard deviation can be used as the amplitude difference threshold to eliminate individual outliers. Preferably, when the sensitivity is set in the range of [0%, 20%), in order to obtain a higher detection sensitivity, N is 0.5; when the sensitivity is set in the range of [20%, 80%), in order to balance sensitivity and specificity, N is 1; when the sensitivity is set in the range of [80%, 100%], in order to obtain a higher specificity, N is 2.

[0045] Optionally, in the second screening module, setting the sensitivity to the quantile corresponding to the spike wave slope includes: in the spike wave detection result, using the data calculation unit to calculate the slope of each spike wave; sorting the slope of each spike wave by size; using the data calculation unit to calculate the proportion of the slope of each spike wave in the sorted position, and obtaining the slope quantile of the spike wave as the sensitivity. The spike wave slope at the quantile corresponding to the preset value of the sensitivity is set as the slope threshold to screen the spike wave. For details, see Figure 8 , count the number of remaining candidate peaks in each channel, and take the first two channels with the largest number as trial channels. If there is only one channel, only this channel is selected as the trial channel. Fig. 9 and Fig.10 , set the steepness to be defined as: The peak position is recorded as , the peak amplitude is recorded as , end point position , the end point amplitude is ;have , The steepness calculation formula is: ; Calculate the ratio of the amplitude difference from the peak to the end point of each peak in each channel to the width of the peak to the end point, and record it as the steepness (i.e., the slope of the spike wave); re-sort the steepness of all peaks in the trial channel by size, take the steepness at the percentile X2 as the preset value of sensitivity, remove the peaks in all channels that are steeper than the sensitivity threshold, and the remaining peaks are the peaks finally identified as spike waves; the percentile X2 of the steepness is an adjustable parameter exposed to the user, and the user can adjust it according to their needs. If the slope of the spike wave is sorted from small to large, the larger the X2 value, the higher the specificity of spike wave detection, and the smaller the value, the higher the sensitivity of spike wave detection.

[0046] In addition, in order to avoid missed detections in a single recognition result, the online multi-channel EEG data can also be decomposed and reconstructed by independent components, and spike detection can be performed on the original data and the reconstructed data respectively, and finally the spike detection results can be integrated by a multi-data integration unit to complement each other. The multi-data integration unit includes: a preprocessing module that preprocesses the signal data to obtain the first data; a data reconstruction module that performs independent component decomposition and reconstruction on the signal data to obtain the second data; and an integration module that integrates the spike detection results after the spike detection unit detects the first data and the second data respectively. Specifically, the independent component decomposition and reconstruction include: performing ICA decomposition (independent component decomposition) on the preprocessed EEG data to obtain different source signals; detecting spikes in each source signal based on the deep spike detection method, and counting the number of spikes in each source signal; retaining the top 30% of the source signals in terms of the number of spikes, and reconstructing these source signals into EEG signals. The integration module counts the channels with the top 30% of the number of spikes detected in the first data, directly takes the union of the detection results of these channels in the first data and the second data, and merges the spikes whose spike position interval is less than the threshold (generally 10ms) as the same spikes, and retains the detection results of the first data for other channels. Of course, the multi-data integration unit can be selected according to needs, and it can be unselected when processing a single type of data.

[0047] In some embodiments, an epilepsy detection device is provided, comprising: the spike wave detection system described above; a data display area is also provided on the operation interface of the human-computer interaction machine to display the quantitative statistical results of the epilepsy waveform; the quantitative statistical results of the epilepsy waveform include: the number and peak size of the spike wave peaks of the EEG signal in the spike wave screening results.

[0048] In some embodiments, when the function or model of the spike detection system is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device or its processor (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0049] Test example 1.

[0050] The EEG data were tested and the performance indicators under different test conditions were compared. Data type: 50-80s EEG signal of C4 channel of epileptic patients; Performance indicators: spike wave correct recognition rate and false detection rate.

[0051] Test condition 1: 1s window test result without overlapping sliding window, from Fig.15 It can be seen that the continuous lines are EEG data waveforms, the interval raised lines are detection results, the spike wave correct detection rate is: 22 / 29≈75.8%; the false recognition rate is: 0.

[0052] Test condition 2: Overlap 80% sliding window 1s window detection results, from Fig.16 It can be seen that the continuous lines are EEG data waveforms, the interval raised lines are detection results, the spike wave correct detection rate is: 29 / 29=100%; the false recognition rate is: 0.

[0053] It can be seen that overlapping sliding windows can reduce missed detection of spike waves.

[0054] Test example 2.

[0055] The EEG data was tested and the performance indicators under different test conditions were compared. Data type: 800-820s EEG signals of multiple channels of epileptic patients; Performance indicators: spike wave correct recognition rate and false detection rate.

[0056] Test condition 3: Fig.17 The detection results at 0% sensitivity (i.e. no sensitivity, the screening threshold is set as the amplitude difference of the peaks of the spike waves at the 0% position and the steepness at the 0% position) are shown. Fig.17 The spike wave correct detection rate: 16 / 20≈80%; the false recognition rate: quite high.

[0057] Test condition 4: Fig.18 The results of the spike peak setting of 25% sensitivity (i.e., setting the screening threshold in turn through the amplitude difference at the 25th percentile and the steepness at the 25th percentile) are shown. Fig.18 The spike wave correct detection rate is 16 / 20≈80%; the false recognition rate is 0.

[0058] The comparison results show that the spike wave screening unit can effectively reduce false positives caused by misdetection and improve accuracy by adjusting the sensitivity. Of course, users can adjust the value of percentile X1 or percentile X2 after selecting different sensitivity preset values ​​as needed, thereby adjusting the sensitivity of the device to balance the problems of false detection and missed detection.

[0059] With the above-mentioned ideal embodiment of the present invention as inspiration, through the above-mentioned description content, relevant staff can make various changes and modifications without departing from the technical idea of ​​the present invention, that is, the technical scope of the present invention is not limited to the contents of the specification.

Claims

1. A spike wave detection system, characterized in that: include: A processor and a data calculation unit, a human-computer interaction machine, a spike wave detection unit, and a spike wave screening unit respectively connected to the processor; wherein The spike wave detection unit detects signal data to obtain a spike wave detection result; The spike wave screening unit screens spike waves from the spike wave detection results to obtain spike wave screening results; The operation interface of the human-computer interaction machine is provided with a sensitivity option and a waveform display area; The waveform display area is used to display the spike wave waveform in the spike wave screening result; The sensitivity option is used to input a preset value of sensitivity to adjust the displayed spike waveform.

2. The spike wave detection system according to claim 1, characterized in that: The spike wave screening unit comprises: The first screening module sets the sensitivity to the quantile corresponding to the spike amplitude difference threshold to screen the spikes, including: In the spike wave detection results, the amplitude difference from the peak value to the end point of each spike wave and its mean value and standard deviation are calculated using the data calculation unit; Sort the amplitude differences of each spike wave by size; The data calculation unit is used to calculate the ratio of the amplitude difference of each spike wave in the sorted position to obtain the quantile corresponding to each amplitude difference; The processor determines whether the standard deviation of the amplitude difference is greater than the quantile at the corresponding position; If so, the amplitude difference at the position corresponding to the sensitivity is taken as the amplitude difference threshold; If not, the difference between the mean of the amplitude difference and N times the standard deviation is used as the amplitude difference threshold.

3. The spike wave detection system according to claim 2, characterized in that: When the sensitivity is set in the range of [0%, 20%), N is 0.5; When the sensitivity is set in the range of [20%, 80%), N is 1; When the sensitivity is set in the range of [80%, 100%], N is 2.

4. The spike wave detection system according to claim 1, characterized in that: The spike wave screening unit further comprises: The second screening module sets the sensitivity to the quantile corresponding to the slope threshold of the spike wave to screen the spike wave, including: In the spike wave detection result, the slope of each spike wave is calculated using the data calculation unit; Sort the slopes of each spike wave by size; The data calculation unit is used to calculate the ratio of the slope of each spike wave in the sorted position as the sensitivity.

5. The spike wave detection system according to claim 1, characterized in that: The spike wave detection unit comprises: The data slicing module performs sliding window slicing processing on the signal data to obtain data segments and their corresponding starting times; The spike detection model has a data segment as an input and a spike detection result of each data segment as an output. The output result is 1, indicating that there is a spike signal, and the output result is 0, indicating that there is no spike signal.

6. The spike wave detection system according to claim 1, characterized in that: Also includes: Data correction unit; The data correction unit is used to delete the resting data and high-amplitude artifact data in the spike wave detection result; The data segments whose standard deviation is greater than the set threshold are marked as spike signals; The data segments with standard deviation less than the set threshold are marked as normal signals, i.e., resting data; The difference between the maximum and minimum values ​​in each data segment and its mean and standard deviation are calculated, and the data segments with differences greater than the mean and standard deviation are recorded as artifact signals, i.e., high-amplitude artifact data.

7. The spike wave detection system according to claim 1 or 6, characterized in that: Also includes: The peak location unit marks the peak position of the spike wave signal based on the spike wave detection result, that is, Calculate the difference between the maximum value and the minimum value in the spike wave signal by using the data calculation unit, sort the differences from large to small, and take the difference at the preset position multiplied by the peak threshold coefficient as the peak threshold; The peak value of the spike wave signal is searched based on the peak value threshold and the half-wave width threshold, and the sum of the time corresponding to the peak value and the starting time is taken as the peak position of the spike wave; The peaks whose time difference is less than the set time threshold are merged into one peak, and the starting position and ending position of each peak are calculated.

8. The spike wave detection system according to claim 1, characterized in that: Also includes: Multiple data integration unit; The multi-data integration unit comprises: A preprocessing module preprocesses the signal data to obtain first data; A data reconstruction module performs independent component decomposition and reconstruction on the signal data to obtain second data; The integration module integrates the spike wave detection results after the spike wave detection unit detects the first data and the second data respectively.

9. The spike wave detection system according to claim 8, The integrated spike wave detection results include: The channels with the number of spike waves detected in the first data not less than the set value are counted as the merged channels, and the channels with the number of spike waves less than the set value are counted as other channels; In the merging channel, the detection results corresponding to the first data and the second data are combined into a union set, and the spikes whose spike position interval is less than a threshold are combined as the same spikes in the union set; In other channels, the detection results corresponding to the first data are retained.

10. An epilepsy detection device, characterized in that: include: The spike wave detection system according to any one of claims 1 to 9; A data display area is also provided on the operation interface of the human-computer interaction machine to display the quantitative statistical results of epileptic waveforms; The quantitative statistical results of the epileptic waveform include: the number and peak value of spike wave of the EEG signal in the spike wave screening result.

Images