Method and device for detecting epileptiform discharges

Abstract

The present application proposes a method for detecting epileptic discharges, comprising: obtaining an electroencephalogram (EEG) signal of a patient during an epileptic seizure, wherein the EEG signal includes one or more signal segments on at least one lead channel, each signal segment includes one or more signal sub-segments; performing feature matching on the signal sub-segments to identify the signal sub-segments with epileptic discharges; and determining the statistical data of epileptic discharges. The method can effectively reduce the workload of medical staff and the probability of false detection and improve diagnostic efficiency by using an automated detection method.

Description

Background Art

[0002] Electroencephalogram (EEG) is a signal data form commonly used to record the electrical activity generated by the human cerebral cortex. The high temporal resolution, non-invasiveness and low cost of EEG signals make them widely used in the preoperative evaluation of patients with epileptic lesions. In long-term EEG signal recordings, the epileptic seizure period only occupies a very small part of the EEG signal in time. In fact, patients are in a state of no epileptic seizure (i.e., interictal period) more during the monitoring process.

[0003] The detection of epileptic discharges between attacks plays an important role in the diagnosis of epilepsy syndromes and the localization of lesions. At present, the analysis and judgment of whether long-term EEG signals contain epileptic discharge information is completed by doctors observing and giving diagnostic opinions based on experience. This manual method is time-consuming and labor-intensive.

[0004] In addition, the traditional spike signal detection scheme has low sensitivity, high false detection rate, and cannot be accurately detected on a specific lead channel. The traditional scheme also does not specifically exclude physiological waveforms in the signal.

[0005] Therefore, there exists a need for improved schemes for detecting epileptiform discharges from interictal EEG signals. Summary of the invention

[0006] In order to solve at least one of the above problems, an embodiment of the present application proposes an automated solution for detecting epileptic discharges in EEG signals.

[0007] According to one aspect of the present application, a method for detecting epileptic discharges is proposed, comprising: acquiring an electroencephalogram (EEG) signal of a patient during an epileptic seizure, wherein the EEG signal includes one or more signal segments on at least one lead channel, each signal segment including one or more signal sub-segments; performing feature matching on the signal sub-segments to identify signal sub-segments in which epileptic discharges exist; and determining statistical data of epileptic discharges.

[0008] According to another aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored. The computer program includes executable instructions. When the executable instructions are executed by a processor, the method described above is executed.

[0009] According to another aspect of the present application, an electronic device is provided, comprising a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the method as described above.

[0010] According to the scheme for detecting epileptic discharges in EEG signals of the embodiments of the present application, compared with the commonly used scheme of diagnosing the patient's epileptic seizure through manual observation of EEG signals by doctors, an automated EEG signal detection method is introduced, which can be used as a computer-aided detection tool to output the detection results and indicators related to whether there are epileptic discharges in the EEG signals monitored by the patient during the interictal period, and quantitatively calculate the statistics and use visualization and other presentation methods for the doctor's reference, effectively reducing the workload of medical staff and the probability of false detection and improving diagnostic efficiency. At the same time, the automated detection scheme of the present application can be accurate to the specific lead channel of the EEG signal, and optimize and update the detection standards and templates based on the accumulated historical data, further improving the accuracy and speed of detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other features and advantages of the present application will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.

[0012] Figure 1 The figure is a schematic flow chart of a method for detecting epileptic discharges according to an embodiment of the present application.

[0013] Figure 2 3 is a comparison diagram of signal waveform processing for identifying epileptic discharges in electroencephalogram signals according to an embodiment of the present application.

[0014] Figure 3 exemplarily a time-frequency template for time-frequency matching according to an embodiment of the present application.

[0015] Figure 4 An exemplary visual presentation of statistical data of epileptiform discharge-based recognition results according to an embodiment of the present application.

[0016] Figure 5 It is a schematic structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] The exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided to make the contents of this application comprehensive and complete and to fully convey the concepts of the exemplary embodiments to those skilled in the art. In the drawings, the size of some elements may be exaggerated or deformed for clarity. The same reference numerals in the drawings represent the same or similar structures, and thus their detailed descriptions will be omitted.

[0018] In addition, described features, structures or characteristics can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without having to include one or more of the specific details, or other methods, elements, etc. can be adopted to practice the technical scheme proposed by the present application. In other cases, known structures, methods or operations are not shown or described in detail to avoid blurring the various aspects of the present application.

[0019] In clinical practice, electroencephalogram (EEG) equipment (such as electroencephalogram (EEG) amplifiers) are used to collect patients' electroencephalogram (EEG) signals. The basic rhythm of human electroencephalogram (EEG) signals consists of signal parts such as delta waves and theta waves, alpha waves and beta waves corresponding to different frequencies. Usually, there are no sharp waves or spikes, or the sharp waves or spikes are not obvious. The spikes in the EEG signal are a sudden, transient fluctuation waveform that generally lasts for 20-70ms. The main component of the spike is the negative phase, and the amplitude is variable. The typical spike rising edge waveform is steep, and the falling edge may have a slope. Spikes are mostly pathological EEG waveforms, which are common in localized epilepsy, grand mal epileptic seizures, myoclonic seizures, diencephalic epilepsy, etc. Therefore, the spike waveform in the EEG signal can be used to detect information related to epileptic lesions. Sharp waves are also a sudden fluctuation waveform, but compared with spikes with a shorter duration (for example, during a transient time period of 20-70ms), the fluctuation duration of the spike is relatively long. During the onset and interictal period of epileptic diseases, the pathological EEG waveform will not only produce corresponding changes in spikes or sharp waves, but also produce corresponding changes in slow wave waveforms such as spike-slow waves and sharp-slow waves. Therefore, epileptic discharges related to epileptic lesions can be detected by spike waveforms, sharp wave waveforms, and waveforms of slow waves such as spike-slow waves and sharp-slow waves in the electroencephalogram signal. In the actual reading diagnosis of electroencephalogram signals, these four epileptic discharge-related waveforms account for most of the cases of epileptic discharges. The waveforms related to epileptic discharges in electroencephalogram signals are referred to herein as abnormal discharges (waveforms) related to epileptic symptoms or lesions.

[0020] When using an EEG amplifier that meets clinical standards to collect EEG signals, electrodes are placed on the surface of a person's head. The placement of the electrodes on the human body surface (head) and the connection method between the electrodes and the EEG signal amplifier are called EEG leads. EEG leads include multiple lead channels. For example, a 10-20 lead system has 19 lead channels, of which the Fp1 and Fp2 lead channels detect the EEG signals of the left and right foreheads, the Cz lead channel detects the EEG signals of the top of the head, and there are also F7, F8 lead channels, etc.

[0021] Combine the following Figure 1An example process of a method for detecting epileptic discharges according to an embodiment of the present application is described.

[0022] The method first obtains the collected electroencephalogram (EEG) signal in step S110. The sampling rate of the EEG amplifier ranges from 200 to 1000 Hz. The number of lead channels of the EEG amplifier is between 16 and 32. In an embodiment of the present application, some lead channels are reference lead channels without detection signals, and there are actually 19 lead channels with detection signals of EEG signals. Clinically, the monitoring time of EEG signals is usually around 2 hours or longer, and its purpose is to detect EEG signals during epileptic seizures. Automated monitoring of EEG signals between epileptic seizures does not need to wait for the occurrence of epileptic seizures, so a shorter monitoring time can be used, such as 30 minutes, 15 minutes, 10 minutes or shorter time periods.

[0023] For ease of processing, in the scheme described in the embodiment of the present application, the electroencephalogram (EEG) signal is divided into signal segments with the same time length according to each lead channel. For example, the time length of the signal segment can be selected as 30 seconds, 1 minute, 2 minutes or longer. It is also possible to select a time length shorter than 30 seconds, such as 20 seconds, 10 seconds, etc. In this way, the electroencephalogram (EEG) signal on each lead channel includes one or more signal segments. It should be understood by those skilled in the art that the purpose of the operation of dividing the electroencephalogram (EEG) signal into multiple signal segments during the signal processing process is to facilitate signal analysis and processing (such as determination of relevant thresholds). In actual operation, it is also possible not to segment the electroencephalogram (EEG) signal but to directly select the corresponding part in the continuous electroencephalogram (EEG) signal for calculation and analysis according to the duration length of the signal segment of interest.

[0024] After obtaining the patient's electroencephalogram signal, the original waveform data of the electroencephalogram signal may be preprocessed in step S120. The preprocessing operation is mainly used to reduce noise, reduce the amount of processed data, thereby increasing the detection speed and improving the detection accuracy.

[0025] The preprocessing step S120 may include a sub-step S121 of reducing the sampling frequency, a sub-step S122 of removing power frequency interference, a sub-step S123 of removing baseline drift interference, a sub-step S124 of removing electromyographic interference, a sub-step S125 of shielding the faulty lead channel, a re-reference processing sub-step S126, and a sub-step S127 of removing artifact data. One or more of the sub-steps S121 to S127 may be selected for preprocessing, and the order of these sub-steps may be adjusted as required during the execution process.

[0026] The sub-step S121 of reducing the sampling frequency can, for example, reduce the sampling rate of the EEG amplifier from more than 500 Hz to about 200 Hz. In the process of detecting epileptic discharges, reducing the sampling frequency of the collected electroencephalogram (EEG) signal can speed up the calculation. The sampling frequency of epileptic discharge detection can generally be reduced to 1 / 2 or less of the sampling frequency of the EEG amplifier. However, according to the requirements of detection accuracy, the detection frequency should also be high enough. In an embodiment of the present application, the sampling frequency of epileptic discharge detection is set to more than 150 Hz, for example, 200 Hz.

[0027] The sub-step S122 of removing power frequency interference is used to eliminate the interference from the power system in the EEG signal. The frequency of the power system is generally 50 Hz or 60 Hz, so a notch filter with an operating frequency of 50 Hz and integer multiples of 50 Hz can be designed to remove power frequency noise.

[0028] Baseline drift interference comes from the patient's slow movements, such as electrode displacement caused by the patient moving his body. Based on experience, the frequency of baseline drift generally does not exceed 0.5 Hz, so a high-pass filter with an operating frequency of 0.5 Hz can be used in sub-step S123 to remove baseline drift interference.

[0029] Myoelectric interference, for example, includes interference caused by muscle force when a patient bites his teeth, which must exist in the EEG signal. The electromyographic signal (EMG) is the superposition of motor unit action potentials (MUAPs) in many muscle fibers in time and space. The surface electromyographic signal (SEMG) is the combined effect of the superficial muscle EMG and the electrical activity on the nerve trunk on the surface of the skin, which can reflect the activity of the nerves and muscles to a certain extent. Therefore, the EEG amplifier will also collect the electromyographic signal when detecting the EEG signal. The electromyographic interference signal is usually a higher frequency signal. Combined with the spike or sharp wave frequency associated with epileptic discharges in the EEG, a low-pass filter with an operating frequency of 70 Hz can be designed in sub-step S124 to remove the electromyographic interference.

[0030] Sub-step S125 involves fault detection and processing of the device. When the EEG device is in use, some lead channels may fail. When it is found that some lead channels fail and there is no EEG signal waveform, the failed lead channels can be shielded or replaced in sub-step S125.

[0031] The re-reference processing in sub-step S126 involves calibration of the electroencephalogram (EEG) signal based on the reference signal. The EEG signals detected on multiple or all lead channels can be averaged as a reference signal. The detection signals on all lead channels minus the calculated reference signal can eliminate the common errors existing on all lead channels. For example, there may be epileptic discharges on the lead channel of the left forehead of the patient, while there are no epileptic discharges on the lead channel of the right forehead. The re-reference processing can remove the error signals that exist in common on the two lead channels without affecting the detection results of the epileptic discharges on the respective lead channels.

[0032] The operation of removing artifact data in sub-step S127 is used to eliminate artifacts with obvious noise in EEG signals. Artifacts are deviation components in the detected EEG brain wave signal that are significantly different from the real electroencephalogram signal, and are obvious errors that exceed the preset threshold. Artifacts may exist on one or more lead channels of the EEG signal. For example, signal segments of the electroencephalogram with excessive amplitude are generally considered to be signal interference caused by electrode detachment, etc., or physiological artifacts caused by the patient's large movements during electroencephalogram monitoring. According to experience, electroencephalogram signal segments with signal amplitudes above 300uV can be identified as signal segments with artifacts and are not detected. The removal of artifact data can be completed by discarding the signal segments with artifacts.

[0033] After preprocessing, the method performs feature matching on the signal segments in step S140 to identify signal segments containing epileptic discharges.

[0034] Before performing feature matching, a step S130 of pre-screening the signal segments may be included to reduce the computational complexity of feature matching. According to the characteristic that the spikes or sharp waves in the EEG signal have high-frequency signal energy exceeding the background signal, the normal EEG signal may be used as the background signal to pre-screen the signal segments. According to an embodiment of the present application, the pre-processed EEG signal is filtered to obtain a high-frequency EEG signal S with a frequency of 10-40 Hz. H For adults, epileptiform discharges have high-frequency spike or sharp wave waveforms that exceed the threshold between 14-40Hz, while the frequency of epileptiform discharges in children is slightly lower, so EEG waveforms in the frequency range of 10-40Hz are selected for filtering and initial screening.

[0035] Based on high-frequency EEG signal S H Calculate the corresponding high frequency energy E H For example, high frequency energy E H It can be calculated as the high-frequency EEG signal S H The square of . Generally speaking, high frequency energy E HSignal segments with energy exceeding the threshold may include discharges other than epileptiform discharges, but the high-frequency energy of signal segments without epileptiform discharges will necessarily not exceed the threshold. For a certain signal segment of an electroencephalogram (EEG) signal, this signal segment includes n signal sub-segments. Assume that the high-frequency energy of signal sub-segment i (where i is an integer and 0 < i ≤ n) is E Hi , then the high-frequency energy threshold of the signal segment to which signal sub-segment i belongs can be calculated by the following formula:

[0036] Threshold = mean(E Hi ) + std(E Hi )

[0037] where Threshold is the high-frequency energy threshold of the signal segment to which signal sub-segment i belongs, mean(E Hi ) is the average value of the high-frequency energy E Hi of all signal sub-segments i within the signal segment, and std(E Hi ) is the standard deviation of the high-frequency energy E Hi of all signal sub-segments i within the signal segment.

[0038] If there is at least one signal sub-segment among the signal sub-segments included in a signal segment such that the high-frequency energy of this signal sub-segment exceeds the corresponding threshold of the signal segment to which it belongs, then the signal sub-segments with high-frequency energy exceeding the corresponding threshold in this signal segment are retained for further processing. At this time, all the signal sub-segments included in the signal segment after preliminary screening are signal sub-segments with high-frequency energy exceeding the corresponding threshold. If the high-frequency energy of all the signal sub-segments included in a signal segment does not exceed the threshold of this signal segment, then these signal sub-segments do not have epileptiform discharges and are screened out, that is, this signal segment is screened out. It should be understood that for each signal segment, the high-frequency energy threshold of the signal segment can be calculated using the high-frequency energy values of all the signal sub-segments within the signal segment according to the formula in the above text, and the high-frequency energy of each signal sub-segment within this signal segment can be compared with the high-frequency energy threshold of this signal segment to complete the filtering and preliminary screening of the signal sub-segments within this signal segment. According to the embodiments of the present application, the signals after preliminary screening in step 130 are used for subsequent feature matching in units of signal sub-segments.

[0039] The step S130 of preliminary screening signal segments is used to reduce the number of signal sub-segments for feature matching. In subsequent processing steps, the original EEG signals based on all frequencies or the pre-screened signal sub-segments of the pre-processed EEG signals are still used to detect epileptiform discharges.

[0040] Next, the method proceeds to step S140 to perform feature matching on the screened signal sub-segments that may have epileptiform discharges.

[0041] Feature matching can be completed based on the comparison of the signal features of the signal sub-segment and the feature template. Signal features include waveform features and time-frequency features, so feature matching can specifically include waveform matching of waveform features with corresponding waveform templates and / or time-frequency matching of time-frequency features with corresponding time-frequency templates. As shown in the figure, feature matching includes sub-step S141 of waveform matching based on the waveform features of the signal sub-segment and sub-step S142 of time-frequency matching based on the time-frequency features of the signal sub-segment.

[0042] In sub-step S141, according to the morphology of spike / sharp waves and slow waves (including spike-slow waves and sharp-slow waves), the following waveform features of the EEG signal in the signal sub-segment are mainly examined:

[0043] 1) The amplitude difference A of the rising edge of the spike or sharp wave 1 and duration D 1 ;

[0044] 2) The amplitude difference A of the falling edge of the spike or sharp wave 2 and duration D 2 ;

[0045] 3) The amplitude difference A between the falling edge of a spike or sharp wave and the rising edge of a slow wave following it 3 ;

[0046] 4) Standard deviation of the amplitude before the rising edge of the spike or sharp wave Std pre .

[0047] The horizontal axis of the EEG signal is time, and the vertical axis is the amplitude of the signal waveform. First, find the moment when the waveform signal in the signal sub-segment is at the maximum amplitude (maximum value, here is the absolute value) in the signal sub-segment, and then examine the waveform characteristics of the EEG signal in the time period before and after this moment. Among them, the amplitude difference A of the rising edge of the spike or sharp wave is 1 It is the amplitude difference (vertical height difference) between the minimum value (minimum value of the rising process) and the maximum value (maximum value of the rising process) of the rising spike / spike wave waveform during the continuous rising process before the waveform of the EEG signal reaches the maximum value of the signal sub-segment. The continuous rising process includes continuous rising, and may also include the process in which the amplitude of the waveform remains unchanged for a short period of time after continuous rising and then continues to rise again. The duration D of the rising edge of the spike or sharp wave 1 The length of the time period (horizontal axis length) between the times corresponding to the minimum and maximum values ​​of the rising process of the rising spike / spike wave waveform. 2It is the amplitude difference (vertical height difference) between the maximum value of the descending process and the minimum value of the descending process during the time period defined by the start and end of the descending process in the continuous descending process after the waveform of the EEG signal reaches the maximum value of the signal sub-segment. The continuous descending process is similar to the situation covered by the continuous ascending process, which includes both continuous decline and the process in which the amplitude of the waveform remains unchanged for a short period of time after a continuous decline and then continues to decline again. The duration D of the descending edge of the spike or sharp wave 2 The amplitude difference A is the length of the time period between the time points corresponding to the maximum and minimum values ​​of the descending process of the descending spike / spike wave waveform (horizontal axis length). 3 It means that in the falling edge of the spike or sharp wave, the waveform reaches the minimum of the falling process and then there is a new continuous rising process. In this new continuous rising process, the amplitude difference (vertical axis height difference) between the minimum value (new rising process minimum value) and the maximum value (new rising process maximum value) of the rising spike / spike wave waveform during the time period defined by the start time and the end time of the rising process is the amplitude difference (vertical axis height difference). The maximum value of the new rising process is generally smaller than the maximum value of the signal sub-segment, that is, smaller than the maximum value of the rising process of the previous spike or sharp wave rising process, and the new rising process involves a slow wave component that may exist after the spike or sharp wave, so the new rising process is called the "slow wave rising edge" following the falling edge of the spike or sharp wave. The amplitude standard deviation Std before the rising edge of the spike or sharp wave pre Refers to the standard deviation of the amplitude of the signal waveform in a period of time before the rising edge of the spike or sharp wave of the signal sub-segment. For example, the signal amplitude standard deviation can be calculated in a period of about 100ms before the moment corresponding to the minimum value of the rising edge of the spike or sharp wave of the signal sub-segment.

[0048] The corresponding waveform feature threshold can be set for each waveform morphology to form a waveform feature threshold set, that is, a waveform template 14A. The above-mentioned waveform features of the signal sub-segment are compared with the waveform template 14A to determine whether there is an epileptic discharge in the candidate signal sub-segment. For example, when the measured value of the waveform feature exceeds the corresponding threshold in the waveform template 14A, it is considered that the threshold requirement is met in the waveform feature. The importance of the above-mentioned waveform features can be the same. According to an embodiment of the present application, when at least one waveform feature meets the threshold requirement, it can be determined that there is an epileptic discharge in the signal sub-segment. The greater the number of waveform features that meet the threshold requirement, the higher the possibility of the presence of epileptic discharge. The possibility of more than 4 waveform features meeting the threshold requirement is greater than that of less than 3 waveform features. According to the requirements, signal sub-segments in which 5 or all 6 waveform features meet the threshold requirements in the waveform template 14A can be selected as candidate signal sub-segments. Signal sub-segments whose number of waveform features that meet the threshold requirement is less than the set number (for example, 4, 5) are identified as waveform mismatches and are screened out.

[0049] In sub-step S142, the signal sub-segment that may contain epileptic discharges is screened based on another matching rule. Sub-step S142 further includes sub-step S1421 of respectively calculating the energy of the signal sub-segment in the low-frequency, medium-frequency and high-frequency bands to form the time-frequency features of the signal sub-segment, sub-step S1422 of determining the time-frequency similarity of the signal sub-segment based on the time-frequency features and the time-frequency template 14B, and sub-step S1423 of identifying whether there is epileptic discharge in the signal sub-segment based on the comparison of the time-frequency similarity with the corresponding threshold.

[0050] According to the commonly used division standard of EEG signal analysis, the low-frequency energy E of the signal sub-segment is calculated according to the definition of low-frequency δ, θ (1.6-8 Hz), medium-frequency α (8-15 Hz) and high-frequency β (15-40 Hz) bands. L , intermediate frequency energy E M And high frequency energy E H . The energy combination of each frequency band in the time period of 120ms before and after the moment corresponding to the maximum amplitude value of the signal sub-segment determined in sub-step S141 can be selected to form the time-frequency characteristics of the signal sub-segment. The time length of the time period of 120ms is selected based on the theory or experience of covering the spike / spike wave waveform of the EEG signal. Generally speaking, the length of the time period of 120ms before and after the moment corresponding to the maximum amplitude value (a total of 240ms) is sufficient to cover the spike / spike wave waveform characteristics.

[0051] In the 240ms time sub-segment, according to the detection sampling frequency mentioned above, it can be known that there are 48 sampling moments (sampling points) included therein. Accordingly, the sub-band energy combination can obtain a 48*3-dimensional (48 sampling moments of energy data, each sampling moment has 3 energy values ​​corresponding to the frequency bands) time-frequency energy matrix. Correspondingly, the preset time-frequency template 14B is also a time-frequency energy matrix template based on the sub-band energy at the sampling moment, with a dimension of 48*3. The time-frequency template 14B is a sub-band energy template that characterizes the reference time-frequency characteristics of epileptic discharges, and can be presented in the form of a visual image, such as Figure 3 As shown in Figure 3 In the time-frequency energy matrix template, the horizontal axis is time, and the vertical axis is the sub-band energy value. The darker the color, the higher the sub-band energy. On the contrary, the lighter the color, the lower the sub-band energy. In the time-frequency energy matrix template, the value of each unit is the energy weight of the corresponding sub-band at that moment.

[0052] In sub-step S1422, the time-frequency similarity of the signal sub-segment can be determined based on the calculated time-frequency energy matrix. The time-frequency similarity characterizes the degree of matching between the signal sub-segment and the reference signal sub-segment with epileptic discharge in terms of time-frequency characteristics. When calculating the unit time-frequency similarity of each unit of the time-frequency energy matrix, the sub-band energy measurement value represented by each unit in the time-frequency energy matrix can be binarized. The unit time-frequency similarity at each unit is calculated by comparing the sub-band energy measurement value (the value of the corresponding unit of the time-frequency energy matrix) at each sampling moment with the sub-band energy threshold (the sub-band energy threshold of the corresponding unit of the time-frequency energy matrix template). A corresponding energy threshold can be set for each of the multiple frequency bands, and a corresponding sub-band energy threshold can be further set for each unit. The energy threshold can be calculated by smoothing or averaging the sub-band energy of the EEG waveform when epileptic discharge is present, or the energy threshold parameters of each unit in the time-frequency energy matrix template can be set based on experience. If the measured value of the sub-band energy of each unit of the matrix exceeds the corresponding sub-band energy threshold, the unit time-frequency similarity indicates that there is a high possibility of epileptic discharge, otherwise the possibility is low within the energy threshold. Different unit time-frequency similarities can be set according to the comparison result of the measured value and the energy threshold, for example, the unit exceeding the energy threshold has a larger unit time-frequency similarity value, and the unit not exceeding has a smaller unit time-frequency similarity value, and it can be further set that the greater the degree of exceeding the threshold, the greater the unit time-frequency similarity value, etc. The binarization process described above uses two unit time-frequency similarity values ​​such as 0 and 1 to characterize the possibility that the measured value of the sub-band energy involves epileptic discharge, wherein 1 indicates that the unit (i.e., the sampling time corresponding to the measured value of the sub-band energy of the unit) involves epileptic discharge or the epileptic discharge process includes the sampling time corresponding to the measured value of the unit, and 0 indicates that the unit (i.e., the sampling time corresponding to the measured value of the sub-band energy of the unit) does not involve epileptic discharge or the epileptic discharge process does not include the sampling time corresponding to the measured value of the unit. The unit time-frequency similarity of the unit whose measurement value exceeds the energy threshold is 1, and the unit time-frequency similarity of the unit that does not exceed the energy threshold is 0. Then, the corresponding unit time-frequency similarity of each unit in the time-frequency energy matrix is ​​a value in [0,1]. The time-frequency energy matrix template 14B sets different weights for each unit based on the importance of detection for the sub-band energy at different times (the weight value is, for example, a real value in [0,1], or further in [-0.5,1]). Then, the time-frequency energy matrix is ​​multiplied by the time-frequency energy matrix template 14B to obtain a weighted time-frequency energy matrix. Among them, the unit time-frequency similarity value of each unit is multiplied by the weight value of the corresponding unit in the matrix template 14B to obtain the weighted unit time-frequency similarity of each unit.

[0053] The time-frequency similarity of the signal sub-segment is calculated by summing the weighted unit time-frequency similarities of each unit in the time-frequency energy matrix. Then, in sub-step S1423, the time-frequency similarity of the signal sub-segment is compared with the corresponding time-frequency similarity threshold to identify whether there is an epileptic discharge in the signal sub-segment. For example, if the time-frequency similarity exceeds the time-frequency similarity threshold, it indicates that there is an epileptic discharge in the signal sub-segment, otherwise it is considered that there is no epileptic discharge. Whether the time-frequency similarity threshold standard is met is equivalent to judging the degree of fit between the signal sub-segment and the reference signal sub-segment template representing the interictal period with epileptic discharges. The signal sub-segment that is determined to meet the requirements through time-frequency matching is retained.

[0054] like Figure 2 As shown, for the original EEG signal waveform in the upper part, it is found in the preliminary screening of step S130 that the high-frequency energy of the filtered signal sub-segments 201 to 205 exceeds the high-frequency energy threshold, and the relevant preliminary screening results are presented in the middle part. Therefore, these signal sub-segments are retained for feature matching in step S140. The lower part specifically shows the time-frequency feature matching of multi-band energy in sub-step S142. Figure 3 In the time-frequency template shown, only the time-frequency similarity of the signal sub-segment 203 meets the corresponding threshold requirement, so the signal sub-segment 203 is considered to have epileptic discharges and is retained, while the signal sub-segments 201, 202, 204 and 205 are screened out. Those skilled in the art will understand that Figure 2 The epileptic discharge detection process shown is only an example and not a limitation, and sub-steps S141 and S142 in step S140 related to feature matching can be performed separately or in combination.

[0055] After feature matching of the signal sub-segments, the recognition results can be further corrected in step S150. The correction step can filter out signal sub-segments that are incorrectly identified as having epileptic discharges and originally belong to normal EEG, and can also retrieve missed signal sub-segments for re-detection.

[0056] Step S150 includes sub-step S151 of filtering out physiological waveforms, sub-step S152 of filtering out non-stereotyped repetitive waveforms, and / or sub-step S153 of updating feature templates.

[0057] In sub-step S151, signal sub-segments of physiological waveform morphology that may be mistakenly identified as having epileptic discharges are excluded. The morphological characteristics and time-frequency characteristics of some physiological waveforms are very similar to abnormal discharges of epileptic discharges, but physiological waveforms usually have specific occurrence locations. Therefore, when more abnormal signal sub-segments are detected at the corresponding specific locations, these signal sub-segments can be excluded through the physiological waveform exclusion sub-step S152 to reduce the impact on the detection results.

[0058] Physiological waveforms may specifically include:

[0059] 1) Blinking and eye movement physiological waveforms: Blinking may have similar physiological waveforms in the symmetrical left and right forehead lead channels Fp1 and Fp2, while eye movement may have similar physiological waveforms in the four lead channels Fp1, Fp2, F7, and F8. Since the waveforms of the contralateral lead channels will have high similarity during blinking and eye movement, they can be eliminated through the waveform correlation between the lead channels.

[0060] 2) Top wave physiological waveform: The top wave is related to sleep and is not an abnormal discharge, but the top wave morphology is similar to abnormal discharge. The top wave physiological waveform usually appears in lead channel Cz and nearby lead channels, and can be excluded by the correlation of the lead channel waveform near the center of the head.

[0061] 3) Physiological waveform of alpha activity in the occipital area: The alpha rhythm in the occipital area is a normal physiological waveform, but its morphology also has certain similarities with abnormal discharges, which can be excluded by the frequency and duration of the EEG signal sub-segments detected on the occipital lead channel.

[0062] Sub-step S152 screens out the signal sub-segments where non-stereotyped repetitive waveforms that do not have stereotyped repetitive waveform features are located. Abnormal discharges such as epileptic discharges usually have stereotyped repetitive waveform features. Stereotyped repetition is the continuous / multiple repetitions of similar spikes or sharp waves in the EEG signal waveform of the same lead channel. The spike or sharp wave features in the EEG waveforms of different patients are generally different, so the screening of non-stereotyped repetitive waveforms cannot be completed between the EEG signals of different patients, and usually only the EEG signals on the same lead channel of the same patient are compared. Non-stereotyped repetitive waveforms can be excluded by calculating the correlation between the detected signal sub-segments that are considered to have abnormal discharges, while retaining stereotyped repetitive waveforms that are determined to have stereotyped repetitive features. For example, correlation data between different signal sub-segments on the same lead channel, especially signal sub-segments with abnormal waveforms, can be calculated, and the correlation data can be compared with the corresponding threshold to determine whether the signal sub-segment has stereotyped repetitive waveform features.

[0063] The correlation between signal sub-fragments can be calculated using the Pearson correlation value, and the formula is as follows:

[0064]

[0065] Among them, ρ X,Yis the Pearson correlation value. X and Y represent different abnormal discharge signal sub-segments on the same lead channel, respectively. A corresponding threshold is set for the Pearson correlation value. When the average value of the correlation between a signal sub-segment with abnormal discharge and other signal sub-segments with abnormal discharge on the same lead channel is less than the corresponding correlation threshold, the signal sub-segment is considered to have no stereotyped repetitive waveform. On the contrary, if the average value of the correlation of the signal sub-segment is greater than the corresponding correlation threshold, it is considered that the signal sub-segment has a stereotyped repetitive waveform and is an abnormal waveform, which is retained for the detection of epileptic discharges. Other algorithms for calculating the correlation between signal sub-segments and / or average algorithms can be used to calculate the overall correlation between the signal sub-segment and multiple other signal sub-segments on the same lead channel, thereby screening out non-stereotyped repetitive waveforms.

[0066] Sub-step S153 detects information on the waveform of abnormal discharges (especially epileptic discharges) in the key lead channels to update feature matching templates such as waveform templates and time-frequency templates, so that the templates in the feature matching process can be more accurately used as references for monitoring epileptic discharges with individual patient characteristics, and the signal sub-segments that were previously wrongly screened out are re-added to the candidate signal sub-segment set. Sub-step S153 further includes sub-step S1531 for determining the key lead channels and sub-step S1532 for determining the abnormal discharge feature templates.

[0067] In sub-step S1531, the lead channel with more signal sub-segments of abnormal waveform detected is determined as the key lead channel. For example, the lead channel with the total time length of the signal sub-segments of abnormal waveform exceeding 20 seconds (greater than 20s / h) per hour is regarded as the key lead channel. Calculated based on the length of each abnormal discharge signal sub-segment being 480ms (approximately 500ms), it is necessary to have at least 40 signal sub-segments with abnormal discharge waveforms per hour on the lead channel to be identified as the key lead channel.

[0068] Since the abnormal discharge waveform on the same lead channel usually has a stereotyped and repetitive signal feature, which further includes a waveform feature and / or a time-frequency energy feature, the abnormal discharge signal sub-segment detected on the same key lead channel can be waveform averaged to obtain an abnormal discharge waveform template corresponding to the patient, and at least one of the abnormal discharge time-frequency templates of the abnormal discharge signal sub-segment can be extracted. On each lead channel, a waveform template and a time-frequency template of the patient's abnormal discharge on the lead channel can be generated, so as to find the common waveform features and time-frequency features of the epileptic-like discharge on the lead channel. The feature template of the abnormal discharge waveform can be a waveform template or a time-frequency template, so the generated abnormal discharge feature template can update the waveform template in step S141 and / or the time-frequency template in step S142. Further, the similarity between the previously screened signal sub-segments on the key lead channel and the abnormal discharge waveform template and / or the time-frequency template can be calculated, or the waveform matching sub-step S141 and / or the time-frequency matching sub-step S142 can be performed again, and those signal sub-segments with a high degree of similarity and / or meeting the waveform feature threshold requirements and / or the time-frequency feature threshold requirements can be re-included in the subsequent abnormal discharge detection and statistics. Therefore, the sub-step S153 of updating the feature template can be regarded as a remedial measure for missed detection, which is different from the step of screening out wrong detection in sub-steps S151 and S152.

[0069] Next, the method determines the statistics of epileptic discharges and / or generates a detection report in step S160. The statistics may include statistically summing the duration lengths of the signal sub-segments in which epileptic discharges exist on each lead channel. The statistical data may be combined with the position coordinates of each lead channel to generate a visual report graph. Figure 4 In the top view of the patient's head shown, the duration of epileptic discharges at various positions (these positions are related to the electrode positions of the lead channels) is represented by different colors and / or depths (grayscale), wherein the darker the position, the longer the duration of abnormal discharges, indicating that there are more cases of epileptic discharges at this position, and the greater the possibility of the presence of epileptic lesions. For example, the duration of epileptic discharges at position 301 on the right side of the patient's head is longer (about 20 seconds / hour) than at position 302 on the center of the head. This visual report diagram can provide users (doctors) with a two-dimensional distribution / marking of abnormal discharge conditions and / or probabilities, and doctors can also further mark or make diagnostic descriptions on the two-dimensional distribution report diagram of abnormal discharges.

[0070] According to the scheme for detecting epileptic discharges in EEG signals as described above, compared with the commonly used scheme of diagnosing the patient's epileptic seizure through manual observation of EEG signals by doctors, the introduction of an automated EEG signal detection method can be used as a computer-aided detection tool to output the detection results and indicators related to whether there are epileptic discharges in the EEG signals monitored by the patient during the interictal period, and quantitative statistical data and visualization and other presentation methods are used for the doctor's reference, which effectively reduces the workload of medical staff and the probability of false detection and improves the efficiency of diagnosis, especially for the targeted elimination of physiological waveforms that are prone to false detection. At the same time, the automated detection scheme of the present application can be accurate to the specific lead channel of the EEG signal, and optimize and update the detection standards and templates based on the accumulated historical data, further improving the accuracy and speed of detection.

[0071] In an exemplary embodiment of the present application, a computer-readable storage medium is also provided, on which a computer program is stored, the program including executable instructions, which can implement the steps of the method for detecting epileptic discharges described in any of the above embodiments when executed by, for example, a processor. In some possible implementations, various aspects of the present application can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps described in the method for detecting epileptic discharges in this specification according to various exemplary embodiments of the present application.

[0072] The program product for implementing the above method according to the embodiment of the present application can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present application is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0073] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0074] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0075] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0076] In an exemplary embodiment of the present application, an electronic device is further provided, which may include a processor and a memory for storing executable instructions of the processor, wherein the processor is configured to execute the steps of the method for detecting epileptic discharges in any of the above embodiments by executing the executable instructions.

[0077] Those skilled in the art will appreciate that various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to as "circuit", "module" or "system" herein.

[0078] Refer to the following Figure 5 The electronic device 500 according to this embodiment of the present application is described. Figure 5 The electronic device 500 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0079] like Figure 5As shown, the electronic device 500 is in the form of a general computing device. The components of the electronic device 500 may include, but are not limited to: at least one processing unit 510, at least one storage unit 520, a bus 530 connecting different system components (including the storage unit 520 and the processing unit 510), a display unit 540, etc.

[0080] The storage unit stores a program code, which can be executed by the processing unit 510, so that the processing unit 510 performs the steps according to various exemplary embodiments of the present application described in the method for detecting epileptic discharges in this specification. For example, the processing unit 510 can perform the following steps: Figure 1 Follow the steps shown in .

[0081] The storage unit 520 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 5201 and / or a cache memory unit 5202 , and may further include a read-only memory unit (ROM) 5203 .

[0082] The storage unit 520 may also include a program / utility 5204 having a set (at least one) of program modules 5205, such program modules 5205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0083] Bus 530 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0084] The electronic device 500 may also communicate with one or more external devices 600 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 500, and / or communicate with any device that enables the electronic device 500 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 550. Furthermore, the electronic device 500 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 560. The network adapter 560 may communicate with other modules of the electronic device 500 via the bus 530. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0085] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the method for detecting epileptic discharges according to the embodiment of the present application.

[0086] Those skilled in the art will readily appreciate other embodiments of the present application after considering the description and practicing the contents disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The description and embodiments are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the appended claims.

Claims

1. A method for detecting epileptic discharges, comprising: Acquiring an electroencephalogram (EEG) signal of a patient during an epileptic seizure, wherein the EEG signal comprises one or more signal segments on at least one lead channel, and each of the signal segments comprises one or more signal sub-segments; Performing feature matching on the signal sub-segments to identify the signal sub-segments with epileptic discharges, wherein the feature matching at least includes performing time-frequency matching based on the time-frequency features of the signal sub-segments, wherein the time-frequency matching includes: respectively calculating the energy of the signal sub-segment in the low frequency, medium frequency and high frequency bands in the time period with the peak amplitude to form the time-frequency characteristics of the signal sub-segment; Determining the time-frequency similarity of the signal sub-segments based on a comparison between an energy measurement value in the time-frequency feature and a corresponding energy threshold in the time-frequency template; Identify whether there is an epileptic discharge in the signal sub-segment based on the time-frequency similarity; and Determine the statistics of epileptiform discharges.

2. The method according to claim 1, characterized in that Before performing feature matching on the signal sub-fragments, the method further comprises: Based on the comparison between the energy of the signal sub-segment in the high-frequency band and the energy threshold of the signal segment to which the signal sub-segment belongs in the high-frequency band, the signal sub-segment without epileptic discharge is screened out.

3. The method according to claim 1, characterized in that The feature matching further includes performing waveform matching based on the waveform features of the signal sub-segment, wherein the waveform matching includes identifying whether there is an epileptic discharge in the signal sub-segment based on a comparison between at least one of the following waveform features and a waveform template: The amplitude difference of the rising edge of the spike or sharp wave in the signal sub-segment; 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 between the falling edge of the spike or sharp wave and 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.

4. The method according to claim 1, characterized in that: The time-frequency template is a time-frequency energy matrix template of sub-frequency band energy based on time, and determining the time-frequency similarity of the signal sub-segment based on the comparison between the energy measurement value in the time-frequency feature and the corresponding energy threshold in the time-frequency template includes: generating a time-frequency energy matrix as the time-frequency feature based on the energy of the signal sub-segment in the low-frequency, medium-frequency and high-frequency bands calculated at the sampling time in the time period; The time-frequency energy matrix is ​​multiplied by the time-frequency energy matrix template to calculate the time-frequency similarity.

5. The method according to claim 4, characterized in that Multiplying the time-frequency energy matrix by the time-frequency energy matrix template comprises: Performing binarization processing based on the value of the unit of the time-frequency energy matrix and the corresponding unit threshold of the time-frequency energy matrix; The unit time-frequency similarity of each unit of the time-frequency energy matrix is ​​calculated by multiplying the value of the unit of the binarized time-frequency energy matrix by the value of the corresponding unit of the time-frequency energy matrix template.

6. The method according to claim 5, characterized in that The unit time-frequency similarity of the unit of the time-frequency energy matrix template is a value weighted by the weight value set based on the importance of detection for the sub-frequency band energy at different moments.

7. The method according to any one of claims 4 to 6, characterized in that The time-frequency similarity is calculated by summing the unit time-frequency similarities of all units of the time-frequency energy matrix.

8. The method according to claim 1 or 2, characterized in that: Before the feature matching, the acquired electroencephalogram signal is preprocessed, and the preprocessing includes at least one of the following: Reduce the sampling frequency; Remove power frequency interference; Remove baseline drift interference; Remove EMG interference; Shield the faulty lead channel; Re-reference processing; Remove artifact data.

9. The method according to claim 1, characterized in that: The method further includes correcting the recognition result of the epileptic discharge after performing feature matching on the signal sub-segments, wherein the correction includes at least one of the following: Screen out signal sub-segments belonging to physiological waveforms; Signal subsegments that do not have stereotyped repetitive waveforms are filtered out.

10. The method according to claim 9, characterized in that Based on the comparison between the correlation between different signal sub-segments on the same lead channel and the corresponding threshold, it is determined whether the signal sub-segment has a stereotyped and repetitive waveform.

11. The method according to claim 9, characterized in that The correction further includes: At least one of the waveform template and the time-frequency template is updated based on the waveform features and time-frequency features of the signal sub-segments on the lead channels having a plurality of signal sub-segments with epileptic discharges.

12. The method according to claim 1, characterized in that Statistics that identify epileptiform discharges include: The duration of epileptiform discharges of each lead channel is counted to generate a detection report.

13. A computer-readable storage medium having a computer program stored thereon, the computer program comprising executable instructions, and when the executable instructions are executed by a processor, the method according to any one of claims 1 to 12 is executed.

14. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the executable instructions to implement the method according to any one of claims 1 to 12.

Images