Epilepsy detection method based on emotion recognition model and related device thereof

Abstract

The invention relates to an epilepsy detection method based on an emotion recognition model and a related device thereof.The epilepsy detection method comprises the steps that on the basis of a plurality of electrodes arranged on a target object, electroencephalogram signals of the target object are collected, and the electroencephalogram signals at least comprise a forehead cortex signal, an edge system signal and a temporal cortex signal; acquiring physiological feature data of the target object, wherein the physiological feature data at least comprises blood flow rate and heartbeat rate; inputting the electroencephalogram signal and the physiological feature data into an emotion recognition model to obtain an emotion recognition state which at least comprises excitation, anxiety and sadness; based on a preset epilepsy trigger feature, analyzing the electroencephalogram signal to obtain a signal analysis result; and determining whether the target object is in an epileptic state or not according to the emotion recognition state and the signal analysis result. According to the scheme provided by the invention, the accuracy and timeliness of epilepsy detection can be improved by identifying the emotional state of the user.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an epilepsy detection method based on an emotion recognition model and a related device thereof. Background Art

[0002] In recent years, with the improvement of people's quality of life and the improvement of medical level, the awareness and attention of brain diseases such as epilepsy have gradually increased, and the number of EEG monitoring has also increased. EEG analysis is an important detection method for early prevention of EEG diseases.

[0003] In related technologies, traditional epilepsy detection methods have difficulty accurately identifying emotional triggers before epileptic seizures, and cannot accurately identify abnormal feature combinations of signals from different brain regions under specific emotional states, resulting in poor accuracy of epilepsy detection. Summary of the Invention

[0004] In order to solve or partially solve the problems existing in the related art, the present application provides an epilepsy detection method based on an emotion recognition model and a related device, which can improve the accuracy and timeliness of epilepsy detection by identifying the user's emotional state.

[0005] In a first aspect, the present application provides an epilepsy detection method based on an emotion recognition model, comprising: collecting EEG signals of the target object based on a plurality of electrodes arranged on the target object, wherein the EEG signals include at least prefrontal cortex signals, limbic system signals, and temporal cortex signals; obtaining physiological characteristic data of the target object, wherein the physiological characteristic data include at least blood flow rate and heart rate; inputting the EEG signals and the physiological characteristic data into an emotion recognition model to obtain an emotion recognition state, wherein the emotion recognition state includes at least excitement, anxiety, and sadness; parsing the EEG signals based on preset epilepsy triggering features to obtain signal analysis results; and determining whether the target object is in an epileptic state based on the emotion recognition state and the signal analysis results.

[0006] In combination with the first aspect, in a possible implementation of the first aspect, before collecting the EEG signals of several electrodes on the target object, it includes: collecting the EEG detection signals of the electrodes; performing time domain detection and frequency domain detection on the EEG detection signals based on preset detection conditions to obtain detection results; if the detection results meet the preset detection threshold, it is determined that the electrodes are normal.

[0007] In combination with the first aspect, in a possible implementation of the first aspect, the inputting the EEG signal into the emotion recognition model to obtain the emotion recognition state includes: preprocessing the EEG signal to obtain a preprocessed EEG signal; performing feature extraction on the preprocessed EEG signal to obtain EEG signal features; and inputting the EEG signal features into the emotion recognition model to obtain the emotion recognition state.

[0008] In combination with the first aspect, in a possible implementation of the first aspect, a first EEG signal is input into the emotion recognition model to obtain a first emotion recognition state; a second EEG signal is input into the emotion recognition model to obtain a second emotion recognition state; if the first emotion recognition state is different from the second emotion recognition state, preliminary epilepsy warning information is generated.

[0009] In combination with the first aspect, in a possible implementation of the first aspect, the EEG signal is analyzed based on the preset epilepsy triggering characteristics to obtain a signal analysis result, including: respectively analyzing the waveforms of the prefrontal cortex signal, the limbic system signal, the temporal cortex signal, and the parietal cortex signal to obtain the signal analysis result; wherein, the signal analysis result includes whether the EEG signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform.

[0010] In conjunction with the first aspect, in a possible implementation of the first aspect, determining whether the target object is in an epileptic state based on the emotion recognition state and the signal analysis result includes:

[0011] If the target object's emotion recognition state is anxiety, and the temporal cortex signal has a sharp-slow wave complex waveform and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state; if the target object's emotion recognition state is sadness, and the prefrontal cortex signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state; if the target object's emotion recognition state is excitement, and the prefrontal cortex signal has a high-frequency oscillation waveform, the limbic system signal has a high-frequency oscillation waveform, and the temporal cortex signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state.

[0012] In combination with the first aspect, in a possible implementation of the first aspect, it also includes: when the target object is in an epileptic state, generating an epileptic state report; transmitting the epileptic state report to a cloud server, and generating a warning message.

[0013] The second aspect of the present application provides an epilepsy detection device based on an emotion recognition model, including an acquisition module for collecting EEG signals from several electrodes on a target object, wherein the EEG signals include at least prefrontal cortex signals, limbic system signals, and temporal cortex signals; an acquisition module for acquiring physiological characteristic data of the target object, wherein the physiological characteristic data include at least blood flow rate and heart rate; an input module for inputting the EEG signals into an emotion recognition model to obtain an emotion recognition state, wherein the emotion recognition state includes at least excitement, anxiety, and sadness; an analysis module for analyzing the EEG signals based on preset epilepsy triggering features to obtain signal analysis results; and a determination module for determining whether the target object is in an epileptic state based on the emotion recognition state and the signal analysis results.

[0014] In combination with the second aspect, in a possible implementation of the second aspect, the acquisition module is also used to acquire the EEG detection signal of the electrode; based on preset detection conditions, time domain detection and frequency domain detection are performed on the EEG detection signal to obtain a detection result; if the detection result meets a preset detection threshold, it is determined that the electrode is normal.

[0015] In combination with the second aspect, in a possible implementation of the second aspect, the input module is also used to preprocess the EEG signal to obtain a preprocessed EEG signal; perform feature extraction on the preprocessed EEG signal to obtain EEG signal features; and input the EEG signal features into the emotion recognition model to obtain the emotion recognition state.

[0016] In combination with the second aspect, in a possible implementation of the second aspect, the input module is also used to input the first EEG signal into the emotion recognition model to obtain a first emotion recognition state; input the second EEG signal into the emotion recognition model to obtain a second emotion recognition state; if the first emotion recognition state is different from the second emotion recognition state, a preliminary epilepsy warning message is generated.

[0017] In combination with the second aspect, in a possible implementation of the second aspect, the analysis module is also used to analyze the waveforms of the prefrontal cortex signal, the limbic system signal, the temporal cortex signal and the parietal cortex signal respectively to obtain the signal analysis results; wherein, the signal analysis results include whether the EEG signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform.

[0018] In combination with the second aspect, in a possible implementation of the second aspect, the analysis module is further used to determine that the target object is in an epileptic state if the emotion recognition state of the target object is anxiety, and the temporal lobe cortex signal has a sharp-slow wave complex waveform, and the limbic system signal has a high-frequency oscillation waveform; if the emotion recognition state of the target object is sadness, and the prefrontal cortex signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform, and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state; if the emotion recognition state of the target object is excitement, and the prefrontal cortex signal has a high-frequency oscillation waveform, the limbic system signal has a high-frequency oscillation waveform, and the temporal lobe cortex signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state.

[0019] In combination with the second aspect, in a possible implementation of the second aspect, the determination module is further used to generate an epilepsy status report when the target object is in an epileptic state; transmit the epilepsy status report to the cloud server, and generate a warning message.

[0020] A third aspect of the present application provides an electronic device, including:

[0021] processor; and

[0022] The memory stores executable codes thereon, and when the executable codes are executed by the processor, the processor is caused to execute the method described above.

[0023] A fourth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method described above.

[0024] A fifth aspect of the present application provides a computer program product, comprising a computer program / instruction, which implements the method described above when executed by a processor.

[0025] The technical solution provided by this application may have the following beneficial effects:

[0026] The present application discloses an epilepsy detection method based on an emotion recognition model and a related device, comprising collecting an EEG signal of the target object based on a plurality of electrodes arranged on the target object, the EEG signal including at least a prefrontal cortex signal, a limbic system signal, and a temporal cortex signal; obtaining physiological characteristic data of the target object, the physiological characteristic data including at least a blood flow rate and a heart rate; inputting the EEG signal and the physiological characteristic data into an emotion recognition model to obtain an emotion recognition state, the emotion recognition state including at least excitement, anxiety, and sadness; parsing the EEG signal based on a preset epilepsy triggering feature to obtain a signal parsing result; and determining whether the target object is in an epileptic state based on the emotion recognition state and the signal parsing result. This method can improve the accuracy and timeliness of epilepsy detection by identifying the user's emotional state.

[0027] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0029] Figure 1 is a flow chart of an epilepsy detection method based on an emotion recognition model shown in an embodiment of the present application;

[0030] Figure 2 is a structural diagram of an epilepsy detection device based on an emotion recognition model shown in an embodiment of the present application;

[0031] Figure 3 It is a structural diagram of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0032] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although the accompanying drawings illustrate embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0033] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0034] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0035] In recent years, with the improvement of people's quality of life and the improvement of medical level, the awareness and attention of brain diseases such as epilepsy have gradually increased, and the number of EEG monitoring has also increased. EEG analysis is an important detection method for early prevention of EEG diseases.

[0036] In related technologies, epilepsy is a common neurological disease, and its attacks are often closely related to emotional fluctuations. Traditional epilepsy detection methods have difficulty accurately identifying the emotional triggers before epileptic attacks, and cannot accurately identify the abnormal characteristic combinations of signals in different brain regions under specific emotional states, resulting in poor accuracy of epilepsy detection.

[0037] In response to the above problems, an embodiment of the present application provides an epilepsy detection method based on an emotion recognition model, which can improve the accuracy and timeliness of epilepsy detection by identifying the user's emotional state.

[0038] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0039] Figure 1 3 is a flow chart of an epilepsy detection method based on an emotion recognition model shown in an embodiment of the present application.

[0040] See also Figure 1 , an epilepsy detection method based on an emotion recognition model, comprising:

[0041] S110: Based on a plurality of electrodes set on the target object, the EEG signal of the target object is collected, where the EEG signal includes at least a prefrontal cortex signal, a limbic system signal, and a temporal cortex signal.

[0042] Specifically, several electrodes are placed on the target subject's head to collect EEG signals. Prefrontal cortex signals, which refer to electrical activity in brain regions responsible for higher-level cognitive functions and emotional regulation, are collected via electrodes placed on leads Fp1 and Fp2. Limbic system signals, involving core emotion processing areas like the amygdala and hippocampus, can be monitored using electrodes in the anterior temporal lobe. The temporal cortex, located on the side of the brain, processes hearing, language, memory, emotion, and higher-level visual information. Temporal cortex signals can be collected using temporal electrodes from a standard EEG electrode placement system.

[0043] S120: Acquire physiological characteristic data of the target object, where the physiological characteristic data at least includes blood flow velocity and heart rate.

[0044] Specifically, the physiological characteristic data of the target object can be obtained through a variety of instruments. For example, the blood flow rate can be measured by near-infrared spectroscopy technology to measure the changes in blood oxygen in the frontal lobe, and the heart rate can be obtained by a chest strap sensor, which can be used to assist in judging the emotional recognition status of the target object.

[0045] S130: Inputting the EEG signal and the physiological characteristic data into an emotion recognition model to obtain an emotion recognition state, which includes at least excitement, anxiety, and sadness.

[0046] Specifically, the emotion recognition model can use a convolutional neural network to extract the time-frequency features of EEG signals, combined with a long-short-term memory network to process the temporal changes of physiological parameters. By inputting the EEG signals and physiological feature data collected by electrodes into the emotion recognition model, for example, the emotion recognition model can identify the excitement state by analyzing the power changes of the γ band in the prefrontal cortex signal, detect the ratio of the θ wave to the α wave in the limbic system to determine the anxiety level, and judge the emotion recognition state of the target object.

[0047] S140: Analyze the EEG signal based on the preset epilepsy triggering characteristics to obtain a signal analysis result.

[0048] Specifically, EEG signals can be analyzed based on preset epilepsy triggering features, for example, detecting whether high-frequency oscillation waveforms lasting more than 100 milliseconds appear in each brain region, and obtaining signal analysis results of different EEG signals. The preset epilepsy triggering features can be used to analyze different EEG signals.

[0049] S150: Determine whether the target object is in an epileptic state based on the emotion recognition state and the signal analysis result.

[0050] Specifically, the current state of the target object can be judged based on the emotion recognition state and signal analysis results. For example, when a sharp-slow wave complex waveform appears in the temporal lobe cortex and is accompanied by anxiety emotion recognition, it can be determined that the target object has epilepsy. By combining the target object's emotional state with the EEG signals of different brain areas, the accuracy and timeliness of epilepsy detection can be improved.

[0051] The present application discloses an epilepsy detection method based on an emotion recognition model, comprising collecting EEG signals of the target object based on a plurality of electrodes arranged on the target object, the EEG signals including at least prefrontal cortex signals, limbic system signals, and temporal cortex signals; obtaining physiological characteristic data of the target object, the physiological characteristic data including at least blood flow velocity and heart rate; inputting the EEG signals and physiological characteristic data into an emotion recognition model to obtain an emotion recognition state, the emotion recognition state including at least excitement, anxiety, and sadness; parsing the EEG signals based on preset epilepsy triggering features to obtain signal analysis results; determining whether the target object is in an epileptic state based on the emotion recognition state and the signal analysis results, thereby improving the accuracy and timeliness of epilepsy detection by identifying the user's emotional state.

[0052] In one possible embodiment, before collecting EEG signals from a plurality of electrodes on a target object, the process includes: collecting EEG detection signals from the electrodes; performing time domain detection and frequency domain detection on the EEG detection signals based on preset detection conditions to obtain detection results; if the detection results meet a preset detection threshold, determining that the electrodes are normal.

[0053] Specifically, the EEG detection signal can be pre-detected in the frequency domain or in the time domain based on preset detection conditions. Time domain detection refers to the quantitative analysis of the amplitude, waveform continuity and noise level of the EEG detection signal, which can be used to determine whether the electrode signal is distorted due to poor contact or environmental interference. Frequency domain detection refers to the analysis of the frequency distribution characteristics of the EEG detection signal. For example, the frequency band energy ratio can be extracted by fast Fourier transform, which can be used to identify whether the electrode signal frequency is abnormal due to hardware failure, and obtain the detection result. The detection result is compared with the preset detection threshold to determine whether the electrode is normal. The preset detection threshold is the qualified range of the pre-set time domain and frequency domain detection indicators. For example, the peak value needs to be between 50μV and 200μV, and the main frequency energy ratio in the 0.5Hz to 40Hz frequency band exceeds 80%. The threshold judgment can ensure that the electrode signal meets the physiological signal acquisition requirements.

[0054] Specifically, the electrode status can be preset as normal, poor contact / high impedance, environmental / physiological noise interference, hardware failure, etc. Before the electrode is officially used to collect EEG signals, the electrode status can be pre-judged. When the electrode is performing time domain detection, two detection indicators can be set: signal peak-to-peak value (Vpp) and whether the signal is saturated or flat. The normal range of signal peak-to-peak value (Vpp) is set to 5μV <Vpp<200μV、信号饱和或平线的条件为不存在持续超过1秒的饱和或平线,当电极的Vpp<5μV时,可以认为电极的状态为接触不良 / 高阻抗,当电极的Vpp> When the voltage is 200μV, there may be electromyographic or motion artifact interference, and the electrode state can be considered to be subject to physiological noise interference; when the electrode is subjected to frequency domain detection, two detection indicators can also be set: the proportion of power frequency noise energy and the proportion of energy in the main physiological frequency band. The normal range of the power frequency noise energy proportion should be that the energy peak of the 50Hz / 60Hz frequency point is significantly lower than the energy of the main frequency band, and the normal range of the main physiological frequency band energy proportion should be that the energy proportion of the 0.5Hz-80Hz frequency band is >80%. When there is a significant power frequency spike in the spectrum of the EEG signal obtained by the electrode, the electrode state can be considered to be subject to environmental noise interference; when the energy in the spectrum of the EEG signal obtained by the electrode is mainly concentrated in the non-physiological frequency band, the electrode state can be considered to be poor contact or hardware failure. Only when all the detection indicators of the electrode meet the conditions of the normal range can the electrode be finally judged to be in a normal state and can the electrode be used for subsequent signal acquisition.

[0055] Further, for abnormal electrodes, different means can be used to adjust the electrodes. For surface electrodes (such as EEG cap electrodes), if it is detected that the state of the electrode is "poor contact / high impedance" or "environmental noise interference", real-time feedback and prompt information can be generated to suggest the operator to check and reposition the corresponding electrode, clean the scalp area to reduce the contact impedance, or check the equipment grounding to exclude environmental interference. After adjustment, the pre-detection process can be re-executed until the electrode state returns to normal. If it is determined that the state of the electrode is "hardware failure", the operator is prompted to replace the electrode. For implanted electrodes (such as stereotactic EEG SEEG or cortical EEG ECoG electrodes): Since their positions are fixed and cannot be physically repositioned or cleaned, in this case, if a contact is determined to be "hardware failure" or persistent "high impedance", the contact can be marked as unusable and the data from the contact can be automatically ignored in subsequent analysis to ensure the accuracy of the brain region signal as much as possible. Specifically, before the electrode starts to collect the formal EEG signal, a short time (for example, 10 seconds) of EEG detection signal collection is performed to detect the signal in time domain and frequency domain. In time domain detection, if the peak-to-peak value of the signal is less than 50 μV, it may indicate that the electrode contact impedance is too high, and if it exceeds 200 μV, it may be caused by electromyographic interference; in frequency domain detection, if the energy of the main frequency in the range of 0.5 Hz to 40 Hz accounts for less than 80%, it may indicate that the electrode has hardware filtering abnormalities or external electromagnetic interference. When both detection results meet the preset threshold, it is determined that the electrode is in a normal working state, and at this time the formal EEG signal collection process is started. If either detection does not meet the standard, an electrode state abnormality prompt is triggered to avoid distorted results of subsequent emotion recognition and epilepsy detection due to electrode failure, to avoid invalid data collection due to electrode failure or interference, to improve the reliability of the input data of the emotion recognition model, to reduce the probability of false positives or false negatives in epilepsy detection, and to enhance the stability and practicality of the detection system.

[0056] In a possible implementation, the EEG signal is input into the emotion recognition model to obtain an emotion recognition state, including: pre-processing the EEG signal to obtain a pre-processed EEG signal; extracting features from the pre-processed EEG signal to obtain EEG signal features; and inputting the EEG signal features into the emotion recognition model to obtain the emotion recognition state.

[0057] Specifically, pre-processing refers to noise filtering and signal enhancement operations on the original EEG signal to eliminate electromyographic interference and power frequency noise, thereby improving signal quality. Feature extraction refers to extracting key waveform features related to emotions from the pre-processed EEG signal, which can capture energy distribution and time domain waveform changes in different frequency bands to obtain the pre-processed EEG signal.

[0058] Specifically, an emotion recognition model refers to a model that classifies EEG signal features through a machine learning algorithm, which can be a convolutional neural network or a support vector machine. It is used to map features to emotion categories such as excitement, anxiety, and sadness. The EEG signal first undergoes a preprocessing stage. Subsequently, the preprocessed signal is decomposed into multiple sub-bands, such as extracting the time-frequency energy features of delta waves, theta waves, alpha waves, and beta waves through wavelet transform. These features are input into a pre-trained emotion recognition model, such as using a convolutional neural network to perform nonlinear mapping on the features, and finally output the emotion recognition state. By preprocessing the EEG signal, the EEG signal can be made more accurate. Through the synergistic effect of preprocessing and feature extraction, noise is effectively suppressed and features that are highly correlated with epilepsy-triggered emotions are extracted, which significantly improves the input data quality of the emotion recognition model and strengthens the reliability of the classification results.

[0059] In a possible embodiment, it also includes: inputting the first EEG signal into the emotion recognition model to obtain a first emotion recognition state; inputting the second EEG signal into the emotion recognition model to obtain a second emotion recognition state; if the first emotion recognition state is different from the second emotion recognition state, generating preliminary epilepsy warning information.

[0060] Specifically, the first EEG signal and the second EEG signal refer to EEG signals collected at different time periods. These signals can be collected continuously by dividing the time periods into 5-second or 10-second intervals. During the period when the target subject wears or implants the electrodes, the EEG signals are divided into multiple continuous time periods. For example, the first EEG signal can be data collected within the current time window, and the second EEG signal can be historical data from the previous adjacent time window. After the two signals are input into the emotion recognition model, if a sudden change in the emotional state from anxiety to excitement, or from calm to sadness, is detected, it is determined to be an abnormal emotional fluctuation. At this point, a preliminary warning message can be generated, but the final epilepsy determination is not triggered yet. Instead, it provides an auxiliary judgment basis for subsequent signal analysis. By comparing the emotion recognition results of adjacent time periods, the correlation between abnormal emotional fluctuations and epileptic precursors can be discovered earlier, and potential risks can be identified in advance when the EEG signal waveform characteristics do not yet fully meet the epilepsy triggering conditions, thereby improving the timeliness of epilepsy detection.

[0061] In one possible embodiment, based on preset epilepsy triggering characteristics, the EEG signal is analyzed to obtain a signal analysis result, including: analyzing the waveforms of the prefrontal cortex signal, the limbic system signal, the temporal cortex signal, and the parietal cortex signal respectively to obtain a signal analysis result; wherein, the signal analysis result includes whether the EEG signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform.

[0062] Specifically, the high-frequency oscillation waveform can be an EEG activity waveform with a frequency higher than 80 Hz, which often manifests as abnormal synchronous discharge of local neuronal clusters before an epileptic seizure. The spike-slow wave composite waveform refers to a composite waveform composed of a spike wave and a subsequent slow wave, which is usually directly related to the abnormal discharge of neurons during an epileptic seizure. By analyzing different EEG information, the signal analysis results of each EEG signal are obtained.

[0063] Specifically, during the signal analysis process, independent waveform analysis is first performed on the signals from the four brain regions. For example, wavelet transform is used to extract energy in the 80-500Hz frequency band from the prefrontal cortex signal. High-frequency oscillations are determined when the energy exceeds three standard deviations from the baseline level. By identifying the waveforms of signals from different brain regions and determining whether high-frequency oscillation waveforms or spike-slow wave complex waveforms are present, abnormal discharge conduction characteristics between different brain regions before an epileptic seizure can be captured, improving the accuracy of early warning. For example, if the emotion recognition model outputs negative or high-arousal emotions such as "anxiety" or "sadness," but the EEG signal analysis results do not find significant epilepsy triggering features in the relevant brain regions (i.e., no high-frequency oscillation waveforms or spike-slow wave complex waveforms), the patient will not be diagnosed with an epileptic state. In this case, the patient may be marked as a "high-risk monitoring" or "emotional warning" stage. If the emotional state is identified as a low-arousal emotion such as "calmness," but the EEG signal analysis detects clear waveforms that meet the definition of classic epileptiform discharges, it can be identified as a potential epileptic event and marked as a "non-emotion-induced abnormality." When the target object is in an "excited" or "happy" emotional state, its EEG signal may also show high-frequency activity (such as physiological Gamma waves, about 30-80Hz). The specific characteristics of the high-frequency activity can be further analyzed. For example, energy analysis can be performed on the frequency band above 80Hz through a characteristic algorithm, and matched with a morphological template to filter out the interference of physiological Gamma waves, and determine whether high-frequency oscillation waveforms are also detected in the limbic system and temporal lobe cortex. In this specific emotional state, the "co-occurrence pattern" or "conduction pattern" of abnormal discharges in multiple key brain regions is the key basis for distinguishing physiological excitement from multifocal or rapidly conducting pathological discharges. The high-frequency activity of a single brain region is likely to be physiological, while the synchronous abnormalities of multiple brain regions that meet the preset rules can determine that the patient has an epileptic event.

[0064] In a possible implementation, the determining whether the target object is in the epilepsy state according to the emotion recognition state and the signal analysis result comprises: if the emotion recognition state of the target object is anxiety, the temporal lobe cortex signal has a sharp-slow wave complex waveform, and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target object is in the epilepsy state; if the emotion recognition state of the target object is sadness, the prefrontal cortex signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform, and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target object is in the epilepsy state; if the emotion recognition state of the target object is excitement, the prefrontal cortex signal has a high-frequency oscillation waveform, the limbic system signal has a high-frequency oscillation waveform, and the temporal lobe cortex signal has a high-frequency oscillation waveform, it is determined that the target object is in the epilepsy state.

[0065] Specifically, in the epilepsy state determination process, first, the temporal lobe cortex signal is detected for a sharp-slow wave complex waveform, when a waveform meeting the sharp-slow wave characteristics is detected, it is synchronously checked whether the limbic system signal has a high-frequency oscillation lasting more than 100 milliseconds, for the anxiety emotional state, the signal correlation of the temporal lobe and the limbic system is monitored, when both of them have abnormal waveforms at the same time, the epilepsy determination is triggered, in the sadness emotional scene, the combination of the high-frequency oscillation or the sharp-slow wave complex waveform of the prefrontal cortex and the high-frequency oscillation of the limbic system is set as a determination condition, for example, by setting double thresholds to detect the two waveform characteristics respectively; in the excitement emotional state, high-frequency oscillations are required in the prefrontal cortex, the limbic system and the temporal lobe, for example, a multi-channel signal synchronous analysis method is used to ensure that the abnormal discharges of the three regions have time correlation, so as to determine whether the target object is in the epilepsy state, by setting the multi-region waveform combination condition, the epilepsy seizure mode triggered by different emotional states can be accurately identified, specific discharge combinations of the temporal lobe-limbic system under the anxiety emotion, abnormal waveform characteristics of the prefrontal cortex under the sadness emotion, and high-frequency oscillations of the whole brain region under the excitement emotion are established respectively, the determination rules corresponding to the three are established respectively, which can improve the specificity of epilepsy detection.

[0066] In a possible implementation, the method further comprises: when the target object is in the epilepsy state, generating an epilepsy state report; transmitting the epilepsy state report to a cloud server and generating a warning information.

[0067] Specifically, the epilepsy state report refers to a standardized data set containing the electroencephalogram signal characteristics of the target object, the emotion recognition state and the corresponding time stamp, which can be obtained through a structured data format such as JSON or XML, and can be used to integrate the electroencephalogram signal waveform characteristics and the emotion recognition result to obtain the epilepsy state, the epilepsy state report is transmitted to the cloud database through an encrypted network protocol for storage, the cloud server can perform real-time remote access and long-term storage on the epilepsy state report, and at the same time, a warning information such as a buzzer alarm can be generated, which can be used to transmit an emergency state signal to medical personnel or guardians.

[0068] Corresponding to the aforementioned embodiment of the method for realizing the application function, the present application also provides an epilepsy detection device, an electronic device and corresponding embodiments based on an emotion recognition model.

[0069] Figure 2 Schematic diagram of the structure of an epilepsy detection device based on an emotion recognition model shown in an embodiment of the present application.

[0070] See also Figure 2 , an epilepsy detection device 200 based on an emotion recognition model, comprising:

[0071] The acquisition module 210 is used to acquire EEG signals from a plurality of electrodes on the target object, where the EEG signals include at least prefrontal cortex signals, limbic system signals, and temporal cortex signals.

[0072] The acquisition module 220 is used to acquire physiological characteristic data of the target object, where the physiological characteristic data at least includes blood flow velocity and heart rate.

[0073] The input module 230 is used to input the EEG signal into the emotion recognition model to obtain the emotion recognition state, which includes at least excitement, anxiety, and sadness.

[0074] The analysis module 240 is used to analyze the EEG signal based on the preset epilepsy triggering characteristics to obtain a signal analysis result.

[0075] The determination module 250 is configured to determine whether the target object is in an epileptic state according to the emotion recognition state and the signal analysis result.

[0076] In one possible embodiment, the acquisition module 210 is also used to acquire the EEG detection signal of the electrode; based on the preset detection conditions, the EEG detection signal is subjected to time domain detection and frequency domain detection to obtain a detection result; if the detection result meets the preset detection threshold, it is determined that the electrode is normal.

[0077] In a possible embodiment, the input module 230 is also used to preprocess the EEG signal to obtain a preprocessed EEG signal; perform feature extraction on the preprocessed EEG signal to obtain EEG signal features; and input the EEG signal features into the emotion recognition model to obtain the emotion recognition state.

[0078] In one possible embodiment, the input module 230 is also used to input the first EEG signal into the emotion recognition model to obtain a first emotion recognition state; input the second EEG signal into the emotion recognition model to obtain a second emotion recognition state; if the first emotion recognition state is different from the second emotion recognition state, a preliminary epilepsy warning message is generated.

[0079] In one possible embodiment, the analysis module 240 is also used to analyze the waveforms of the prefrontal cortex signal, the limbic system signal, the temporal cortex signal, and the parietal cortex signal respectively to obtain signal analysis results; wherein, the signal analysis results include whether the EEG signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform.

[0080] In one possible embodiment, the analysis module 240 is also used to determine that the target object is in an epileptic state if the target object's emotion recognition state is anxiety, and the temporal lobe cortex signal has a sharp-slow wave complex waveform and the limbic system signal has a high-frequency oscillation waveform; if the target object's emotion recognition state is sadness, and the prefrontal cortex signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state; if the target object's emotion recognition state is excitement, and the prefrontal cortex signal has a high-frequency oscillation waveform, the limbic system signal has a high-frequency oscillation waveform, and the temporal lobe cortex signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state.

[0081] In a possible implementation, the determination module 250 is further configured to generate an epileptic state report when the target object is in an epileptic state; transmit the epileptic state report to a cloud server, and generate a warning message.

[0082] The present application discloses an epilepsy detection device based on an emotion recognition model, comprising collecting EEG signals of the target object based on a plurality of electrodes arranged on the target object, the EEG signals including at least prefrontal cortex signals, limbic system signals, and temporal cortex signals; obtaining physiological characteristic data of the target object, the physiological characteristic data including at least blood flow velocity and heart rate; inputting the EEG signals and physiological characteristic data into an emotion recognition model to obtain an emotion recognition state, the emotion recognition state including at least excitement, anxiety, and sadness; parsing the EEG signals based on preset epilepsy triggering features to obtain signal parsing results; determining whether the target object is in an epileptic state based on the emotion recognition state and the signal parsing results, and being able to improve the accuracy and timeliness of epilepsy detection by identifying the user's emotional state.

[0083] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0084] An embodiment of the present application also provides an electronic device. Figure 3This is a hardware structure diagram of an embodiment of an electronic device of the present application. The electronic device includes a memory 320 and at least one processor 310. The memory 320 is electrically connected to the at least one processor 310. The memory 320 stores instructions, and the at least one processor 310 invokes the instructions in the memory 320, causing the electronic device to execute the epilepsy detection method based on the emotion recognition model according to any of the aforementioned embodiments of the present application.

[0085] Specifically, the processor 310 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0086] The memory 320 may include a large capacity memory 320 for data or instructions. By way of example and not limitation, the memory 320 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 320 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 320 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 320 is a non-volatile solid-state memory. In a specific embodiment, the memory 320 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0087] In one example, the control device may further include a communication interface 330 and a bus 340. The processor 310, the memory 320, and the communication interface 330 are connected via the bus 340 and communicate with each other.

[0088] The communication interface 330 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0089] Bus 340 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus can include accelerated graphics port (AGP) or other graphics bus, enhanced industry standard architecture (EISA) bus, front side bus (FSB), hypertransport (HT) interconnection, industry standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 340 can include one or more buses. Although the present application embodiment describes and shows specific bus, the application considers any suitable bus or interconnection.

[0090] In addition, in conjunction with the above-mentioned epilepsy detection method based on the emotion recognition model, the present application can provide a computer-readable storage medium for implementation. The computer-readable storage medium stores instructions that, when executed by a processor, implement any of the above-mentioned epilepsy detection methods based on the emotion recognition model.

[0091] The present application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications and additions, or change the order of the steps after understanding the spirit of the present application.

[0092] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. Programs or code segments can be stored in machine-readable media, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable media" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memories, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0093] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0094] Alternatively, the present application also provides a computer program product that can implement part or all of the steps of the method in the above embodiments. The computer program product includes a computer program / instructions, which implement part or all of the steps of the method in the above embodiments when executed by a processor.

[0095] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. An epilepsy detection method based on an emotion recognition model, characterized in that: include: Based on a plurality of electrodes set on the target object, the target object's electroencephalogram (EEG) signal is collected, where the EEG signal includes at least a prefrontal cortex signal, a limbic system signal, and a temporal cortex signal; Acquiring physiological characteristic data of the target object, wherein the physiological characteristic data at least includes blood flow velocity and heart rate; Inputting the EEG signal and the physiological characteristic data into an emotion recognition model to obtain an emotion recognition state, wherein the emotion recognition state includes at least excitement, anxiety, and sadness; Analyzing the EEG signal based on the preset epilepsy triggering characteristics to obtain a signal analysis result; Determine whether the target object is in an epileptic state based on the emotion recognition state and the signal analysis result.

2. The method according to claim 1, characterized in that Before collecting EEG signals from a plurality of electrodes on the target object, the method includes: collecting electroencephalogram (EEG) detection signals from the electrodes; Based on the preset detection conditions, the EEG detection signal is subjected to time domain detection and frequency domain detection to obtain a detection result; If the detection result meets the preset detection threshold, it is determined that the electrode is normal.

3. The method according to claim 1, characterized in that Inputting the EEG signal into an emotion recognition model to obtain an emotion recognition state includes: Preprocessing the EEG signal to obtain a preprocessed EEG signal; Performing feature extraction on the preprocessed EEG signal to obtain EEG signal features; The EEG signal feature is input into the emotion recognition model to obtain the emotion recognition state.

4. The method according to claim 1, wherein Also includes: Inputting the first EEG signal into the emotion recognition model to obtain a first emotion recognition state; Inputting the second EEG signal into the emotion recognition model to obtain a second emotion recognition state; If the first emotion recognition state is different from the second emotion recognition state, preliminary epilepsy warning information is generated.

5. The method according to claim 1, wherein The EEG signal is analyzed based on the preset epilepsy triggering characteristics to obtain a signal analysis result, including: respectively analyzing the waveforms of the prefrontal cortex signal, the limbic system signal, the temporal cortex signal, and the parietal cortex signal to obtain the signal analysis results; The signal analysis result includes whether the EEG signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform.

6. The method according to claim 5, characterized in that The determining whether the target object is in an epileptic state according to the emotion recognition state and the signal analysis result includes: If the target subject's emotion recognition state is anxiety, and the temporal lobe cortex signal has a sharp-slow wave complex waveform and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target subject is in an epileptic state; If the target subject's emotion recognition state is sadness, and the prefrontal cortex signal has a high-frequency oscillation waveform or a sharp-slow wave complex waveform, and the limbic system signal has a high-frequency oscillation waveform, it is determined that the target subject is in an epileptic state; If the target object's emotion recognition state is excitement, and the prefrontal cortex signal has a high-frequency oscillation waveform, the limbic system signal has a high-frequency oscillation waveform, and the temporal cortex signal has a high-frequency oscillation waveform, it is determined that the target object is in an epileptic state.

7. The method according to claim 1, characterized in that Also includes: When the target subject is in an epileptic state, generating an epileptic state report; The epilepsy status report is transmitted to a cloud server, and a warning message is generated.

8. An epilepsy detection device based on an emotion recognition model, characterized in that: include: An acquisition module, configured to acquire EEG signals from a plurality of electrodes on a target subject, wherein the EEG signals include at least prefrontal cortex signals, limbic system signals, and temporal cortex signals; An acquisition module, configured to acquire physiological characteristic data of the target object, wherein the physiological characteristic data at least includes blood flow velocity and heart rate; An input module, configured to input the EEG signal into an emotion recognition model to obtain an emotion recognition state, wherein the emotion recognition state includes at least excitement, anxiety, and sadness; An analysis module, configured to analyze the EEG signal based on preset epilepsy triggering features to obtain a signal analysis result; A determination module is used to determine whether the target object is in an epileptic state according to the emotion recognition state and the signal analysis result.

9. An electronic device, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: An executable code is stored thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the method according to any one of claims 1 to 7.

Images