Pain assisted assessment system based on artificial intelligence and EEG signals

Abstract

The present application relates to the field of data processing technology, and specifically to a pain auxiliary assessment system based on artificial intelligence and electroencephalogram (EEG) signals. The system includes: collecting an EEG signal consisting of a number of potential values ​​of each electrode; dividing the EEG signal into a number of segmented signals, determining the amplitude-cycle ratio according to the amplitude and period in the segmented signal, obtaining a high-frequency evaluation according to the difference between the amplitude-cycle ratio and the mean of the amplitude-cycle ratio, and determining a high-frequency EEG signal; dividing the electrode into an influencing electrode and a normal electrode, determining a normal EEG signal and a potential abnormal EEG signal, and obtaining the degree of distortion of the potential abnormal EEG signal according to the relationship between the two; obtaining a distortion evaluation of the potential abnormal EEG signal in combination with the high-frequency evaluation and the potential difference; determining a distorted EEG signal; correcting the potential value of the distorted EEG signal and the high-frequency EEG signal, and performing pain rating on the corrected EEG signal. The corrected EEG signal of the present application can more accurately reflect the patient's pain state.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a pain-assisted assessment system based on artificial intelligence and electroencephalogram signals. Background Art

[0002] With the development of artificial intelligence technology and brain science, the method of pain assessment based on EEG signal analysis plays an irreplaceable role. EEG signals are a direct reflection of brain activity and are not affected by the patient's subjective consciousness and language expression ability. Therefore, they have high objectivity and can reflect the activity state of the brain in real time. They can also accurately capture the bioelectric signals generated by pain stimulation and achieve timely pain assessment. EEG signal acquisition usually uses non-invasive electrode patches or electrode caps, which do not require invasive surgical operations and are safer than other objective assessment methods. The pain assessment method based on EEG signal analysis has irreplaceable advantages due to its objectivity, real-time and accuracy. It can make up for the shortcomings of traditional assessment methods and provide a new and effective way for pain assessment. However, when the electrode patch measures EEG on the patient's scalp, the patient's sneezing causes the electrode patch to have poor contact, resulting in distortion of the EEG signal of the electrode; at the same time, the instantaneous pressure and reflex action of sneezing will cause the brain to quickly process these external stimuli, thereby triggering the generation of high-frequency EEG signals, affecting the accuracy of pain assessment through EEG signals. Summary of the invention

[0003] In order to solve the technical problem of inaccurate pain assessment, this application provides a pain auxiliary assessment system based on artificial intelligence and EEG signals. The technical solutions adopted are as follows:

[0004] This application proposes a pain-assisted assessment system based on artificial intelligence and EEG signals, which includes the following modules:

[0005] An acquisition module is used to acquire an EEG signal consisting of a number of potential values ​​of each electrode;

[0006] A high-frequency EEG signal screening module is used to divide the EEG signal into several segmented signals, determine the amplitude and period according to the maximum and minimum values ​​in the segmented signals; determine the amplitude-period ratio according to the amplitude and period in the segmented signals, obtain the high-frequency evaluation of the segmented signal of each electrode according to the difference between the amplitude-period ratio of the segmented signal of each electrode and the average of all amplitude-period ratios of each electrode; and determine the high-frequency EEG signal according to the high-frequency evaluation;

[0007] The distorted EEG signal screening module is used to record electrodes with high-frequency EEG signals as affected electrodes and other electrodes as normal electrodes, and determine normal EEG signals and potential abnormal EEG signals by the position of high-frequency EEG signals; obtain the degree of distortion of potential abnormal EEG signals according to the relationship between potential abnormal EEG signals and normal EEG signals; obtain the potential difference at each moment according to the difference between the potential value and the adjacent potential values; obtain the distortion evaluation of potential abnormal EEG signals according to the degree of distortion, high-frequency evaluation and potential difference of potential abnormal EEG signals; and determine the distorted EEG signals according to the distortion evaluation;

[0008] The potential value correction module corrects the potential value of the affecting electrode according to the potential value of the normal electrode closest to the affecting electrode at the same time for high-frequency EEG signals; and corrects the potential value of the same position in the distorted EEG signal of the affecting electrode according to the normal EEG signal of the affecting electrode for distorted EEG signals;

[0009] The rating module is used to output the pain rating of the corrected EEG signal through a neural network.

[0010] In the above scheme, the present application conducts in-depth analysis of the collected EEG signals, and can identify and confirm the high-frequency EEG signals based on the amplitude, period and other characteristics of the EEG signals. In addition, it also considers the poor contact between the electrodes and the skin caused by stimulation, thereby causing distortion. The distorted EEG signals are screened out by comparing the waveforms and the potential value jumps caused by the distortion, and these distorted EEG signals and high-frequency EEG signals are effectively corrected. The confirmed and corrected EEG signals can more accurately reflect the patient's pain state. This accuracy is crucial for clinical diagnosis and pain management, helping medical staff to better understand the patient's pain experience and formulate more effective treatment plans, so that an artificial intelligence-based pain auxiliary assessment system can be realized.

[0011] In one embodiment, the method for determining the amplitude and period according to the maximum and minimum values ​​in the segmented signal is:

[0012] The maximum value in the segmented signal is taken as the amplitude in the segmented signal;

[0013] The time length between two adjacent minimum values ​​in the segmented signal is regarded as a period in the segmented signal.

[0014] In one embodiment, the method for determining the amplitude-period ratio according to the amplitude and period in the segmented signal is:

[0015] The ratio of the mean of all amplitudes in the segmented signal to the mean of all periods in the segmented signal is recorded as the amplitude-to-period ratio.

[0016] In one embodiment, the method for obtaining the high frequency evaluation of the segmented signal of each electrode according to the difference between the amplitude period ratio of the segmented signal of each electrode and the average of all amplitude period ratios of each electrode is:

[0017] , Indicates The amplitude-to-period ratio of the i-th segment signal in the electrode, Indicates The number of segmented signals in an electrode, Indicates The segmented signal in the electrode is in the interval Inside The maximum value of The function is a maximum value function; Indicates The segmented signal in the electrode is in the interval Inside The minimum value of The function is the minimum function. Indicates High-frequency evaluation of the i-th segment signal in the i-th electrode.

[0018] In one embodiment, the method for determining normal EEG signals and potential abnormal EEG signals by the location of high-frequency EEG signals is:

[0019] For all segmented signals of the electrode, the signal before the high-frequency EEG signal is recorded as a normal EEG signal, and the signal after the high-frequency EEG signal is recorded as a potential abnormal EEG signal.

[0020] In one embodiment, the method for obtaining the degree of distortion of the potential abnormal EEG signal based on the relationship between the potential abnormal EEG signal and the normal EEG signal is:

[0021] The dissimilarity between the potential abnormal EEG signal and each normal EEG signal is calculated, and the mean of the dissimilarity between the potential abnormal EEG signal and all normal EEG signals is taken as the distortion degree of the potential abnormal EEG signal.

[0022] In one embodiment, the method for obtaining the potential difference at each moment according to the difference between the potential value and the adjacent potential value is:

[0023] Let the difference between each potential value and the adjacent potential value be recorded as the potential difference; if the potential value has only one adjacent potential value, then the potential difference is the absolute value of the difference between the potential value and the adjacent potential value; if the potential value has two adjacent potential values, then the potential difference is the average of the absolute values ​​of the difference between the potential value and each adjacent potential value.

[0024] In one embodiment, the method for obtaining the distortion evaluation of the potential abnormal EEG signal according to the distortion degree, high frequency evaluation and potential difference of the potential abnormal EEG signal is:

[0025] Obtaining the maximum potential difference and the minimum potential difference in the potential abnormal EEG signal, and recording the difference between the maximum potential difference and the minimum potential difference as the first potential difference;

[0026] Obtaining a distortion evaluation of a potential abnormal EEG signal based on the degree of distortion, high frequency evaluation, and first potential difference of the potential abnormal EEG signal;

[0027] The distortion evaluation was positively correlated with the distortion degree, high frequency evaluation and first potential difference of potential abnormal EEG signals.

[0028] In one embodiment, the method for correcting the potential value of the influencing electrode according to the potential value of the normal electrode closest to the influencing electrode at the same time is:

[0029] , Indicates the revised The first electrode The high-frequency EEG signal Potential value; Indicates The first electrode The high-frequency EEG signal Potential value; Indicates the The first of the two normal electrodes closest to the affected electrode The first segment signal The minimum value of a potential.

[0030] In one embodiment, the method for correcting the potential value at the same position in the distorted EEG signal affecting the electrode according to the normal EEG signal affecting the electrode is:

[0031] , Indicates the revised The first electrode The distorted EEG signal Potential value; Indicates The first electrode The distorted EEG signal Potential value; Indicates The influence of the electrode The first of the two normal EEG signals closest to the distorted EEG signal The minimum value among the potential values.

[0032] The beneficial effects of this application are:

[0033] This application conducts in-depth analysis of the collected EEG signals, and can identify and confirm high-frequency EEG signals based on the amplitude, period and other characteristics of the EEG signals. In addition, it also takes into account the distortion caused by poor contact between the electrodes and the skin due to stimulation. The distorted EEG signals are screened out by comparing the waveforms and the potential value jumps caused by the distortion, and these distorted EEG signals and high-frequency EEG signals are effectively corrected. The confirmed and corrected EEG signals can more accurately reflect the patient's pain status. This accuracy is crucial for clinical diagnosis and pain management, helping medical staff to better understand the patient's pain experience and develop more effective treatment plans, thereby realizing an artificial intelligence-based pain auxiliary assessment system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 A flowchart of a pain-assisted assessment system based on artificial intelligence and EEG signals provided in one embodiment of the present application;

[0036] Figure 2 Schematic diagram of electrode placement. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of the pain auxiliary assessment system based on artificial intelligence and EEG signals proposed in the present application are described in detail below in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0039] Embodiment of pain auxiliary assessment system based on artificial intelligence and EEG signals:

[0040] The specific scheme of the pain auxiliary assessment system based on artificial intelligence and EEG signals provided by the present application is described in detail below with reference to the accompanying drawings.

[0041] See also Figure 1 , which shows a flow chart of a pain auxiliary assessment system based on artificial intelligence and EEG signals provided by an embodiment of the present application, the system includes the following modules:

[0042] The acquisition module marks the protrusion between the patient's eyes, the left and right ears, and the top of the head. The electrode placement of the EEG device is determined according to the position of the marked points on the patient's head. The electrodes are placed according to the international 10-20 system, such as Figure 2 shown.

[0043] 21 electrodes are placed at different positions on the patient's head for monitoring, and a potential value is collected every 1 second. Each electrode collects a potential value at each moment. The time series composed of the collected potential values ​​is an EEG signal. The EEG signal of each electrode is obtained. In this embodiment, an EEG signal composed of potential values ​​for half an hour is collected. The electrodes are connected to an amplifier through a wire, and the amplifier amplifies the monitored EEG signal for recording. The amplifier is connected to a recorder, and the recorder saves the monitored EEG signal.

[0044] At this point, the EEG signals of different electrodes are obtained through the EEG device.

[0045] High-frequency EEG signal screening module, in the process of analyzing and evaluating pain based on EEG signals, EEG signals are collected by fixing electrodes on the patient's scalp through electrode patches in order to effectively measure the brain's electrical activity. However, patients may experience a physiological reaction of sneezing during the test, which is usually accompanied by rapid muscle contractions, especially in the neck and facial areas. When patients sneeze, the tension of the related muscles triggers a series of chain reactions, which are particularly evident at electrodes close to the temporal or frontal regions. These electrodes are used to monitor pain-related EEG activity, but in this case, they may be affected by muscle activity, thereby interfering with EEG signals.

[0046] Specifically, when a sneeze occurs, the instantaneous pressure and reflex action of the sneeze causes the brain to quickly process these external stimuli, thereby triggering the generation of high-frequency signals. The instantaneous contraction of the facial and neck muscles may cause the frequency of the EEG signals measured by electrodes near the temporal or frontal regions to increase, and the cycle of the EEG signals to shorten. At the same time, due to the activation of the nervous system and the strong contraction of the muscles, the amplitude of the EEG signals tends to increase significantly. This increase in amplitude may be due to the increased synchronization of the activity of the neuronal population, which leads to an increase in the amplitude of the signal.

[0047] For the EEG signal of one of the electrodes, the EEG signal is divided equally and each segment is recorded as a segmented signal; in this embodiment, the EEG signal is divided into 20 segments; the maximum value in each segmented signal is obtained, the maximum value is recorded as the amplitude, all moments between two minimum values ​​in the segmented signal are recorded as a cycle, and the mean value of the amplitude and the mean value of the cycle in each segmented signal are calculated.

[0048] The ratio of the mean amplitude to the mean period in each segmented signal is recorded as the amplitude-period ratio; the high-frequency evaluation of the segmented signal is obtained according to the difference between the amplitude-period ratio in each segmented signal and the amplitude-period ratio in all segmented signals.

[0049] The expression of high frequency evaluation of segmented signal is:

[0050] , Indicates The amplitude-to-period ratio of the i-th segment signal in the electrode, Indicates The number of segmented signals in an electrode, Indicates The segmented signal in the electrode is in the interval Inside The maximum value of The function is a maximum value function; Indicates The segmented signal in the electrode is in the interval Inside The minimum value of The function is the minimum function. Indicates High-frequency evaluation of the i-th segment signal in the i-th electrode.

[0051] Indicates The mean value of the amplitude-to-period ratio in each segment signal of the electrodes, Indicates The first The difference between the amplitude-period ratio of the first segmented signal and other segmented signals, the larger the value, the smaller the amplitude-period ratio of the first segmented signal. The first The shorter the period of a segmented signal and the larger its amplitude, the more likely it is a high-frequency EEG signal and the greater the degree of abnormality. The role here is to normalize.

[0052] Set the high-frequency evaluation threshold. The first When the high-frequency evaluation of the first segment signal is greater than the high-frequency evaluation threshold, The first The segmented signals are recorded as high-frequency EEG signals; The first When the high-frequency performance evaluation of the first segment signal is less than or equal to the high-frequency evaluation threshold, The first The segmented signals are recorded as non-high-frequency EEG signals. In this embodiment, the evaluation threshold is 0.5.

[0053] It should be noted that negative correlation means that as one variable increases, the other variable decreases accordingly, and the two variables change in opposite directions. When one variable changes from large to small or from small to large, the other variable also changes from small to large or from large to small. The specific relationship can be a ratio relationship, a subtraction relationship, etc., which is determined by actual application, and this application does not impose any special restrictions.

[0054] At this point, the segmented signals are divided into high-frequency EEG signals and non-high-frequency EEG signals.

[0055] Distorted EEG signal screening module. During the research process of using EEG signals for pain analysis and evaluation, the patient experienced a physiological reaction of sneezing during the test. At the moment of sneezing, some electrodes generated high-frequency signals. If there is a high-frequency EEG signal in the electrode, this electrode is affected by the sneeze, and this electrode is recorded as the affected electrode. The affected electrode is determined through the above steps. After the sneeze ends, the affected electrode patch may become loose due to the violent physiological reaction. This looseness will cause poor contact between the electrode and the skin, thereby generating distorted EEG signals.

[0056] Due to loose electrode patches or poor contact, the waveform of distorted EEG signals may be significantly distorted, with sharp waveforms, inconsistent amplitudes, and loss of original periodicity. Distorted EEG signals may fluctuate intermittently or suddenly, resulting in the destruction of signal continuity. Normal EEG signals usually exhibit stable waveform characteristics, with regular waveforms and a specific frequency and amplitude range. The signals are usually continuous and show a certain degree of stability, which can reflect the dynamic activities of the brain.

[0057] The influencing electrodes and the high-frequency EEG signals in the influencing electrodes are determined through the above steps. The segmented signals before the high-frequency EEG signals appear are not affected by the patient's sneezing and are recorded as normal EEG signals; the segmented signals after the high-frequency EEG signals are recorded as potential abnormal EEG signals.

[0058] For each potential abnormal EEG signal, the dissimilarity is calculated with each normal EEG signal, and the average of the dissimilarity between the potential abnormal EEG signal and all normal EEG signals is taken as the degree of distortion of the potential abnormal EEG signal. In this embodiment, the dissimilarity between the potential abnormal EEG signal and the normal EEG signal is calculated using the DTW distance. The larger the DTW distance, the greater the dissimilarity between the potential abnormal EEG signal and the normal EEG signal.

[0059] For each potential abnormal EEG signal, the difference between each potential value and the adjacent potential value is recorded as the potential difference; if the potential value has only one adjacent potential value, the potential difference is the absolute value of the difference between the potential value and the adjacent potential value; if the potential value has two adjacent potential values, the potential difference is the average of the absolute values ​​of the difference between the potential value and each adjacent potential value.

[0060] The distortion evaluation of the potential abnormal EEG signal is obtained according to the distortion degree, high frequency evaluation and the difference between the maximum potential difference and the minimum potential difference of the potential abnormal EEG signal.

[0061] The distortion evaluation was positively correlated with the distortion degree, high-frequency evaluation, maximum potential difference and minimum potential difference of potential abnormal EEG signals.

[0062] It should be noted that positive correlation means that when one variable increases, the other variable also increases, and the two variables change in the same direction. When one variable changes from large to small or from small to large, the other variable also changes from large to small or from small to large. The specific relationship is determined by actual application and this application does not impose any special restrictions.

[0063] Preferably, in this embodiment, the expression for distortion evaluation is:

[0064] , represents the high-frequency evaluation of the nth potential abnormal EEG signal, Indicates the degree of distortion of the nth potential abnormal EEG signal, Indicates Among the potential abnormal EEG signals The potential difference at each moment; Indicates Among the potential abnormal EEG signals Moments in the interval Internal Time The maximum value of The function is a maximum value function; Indicates Among the potential abnormal EEG signals Moments in the interval Internal Time The minimum value of The function is the minimum function. represents the number of moments in the nth potential abnormal EEG signal, Represents the distortion evaluation of the nth potential abnormal EEG signal.

[0065] Preferably, in another embodiment of the present application, the expression for distortion evaluation is:

[0066] , Represents the distortion evaluation of the nth potential abnormal EEG signal.

[0067] Indicates The continuity of a potential abnormal EEG signal. The larger the value, the greater the difference between adjacent potential values, which means that the potential abnormal EEG signal is more likely to be distorted. The larger the value, the greater the difference between the potential abnormal computer signal and the normal computer signal, indicating that The more potential abnormal EEG signal waveforms are, the more likely they are to be abnormal; The larger the value, the greater the high-frequency evaluation of the potential abnormal computer signal. The larger the amplitude and the smaller the period, the more likely it is that there is an abnormality.

[0068] Each potential abnormal EEG signal is normalized. In this embodiment, the normalization calculation method is linear normalization.

[0069] Set the distortion evaluation threshold. When the normalized distortion evaluation of a potential abnormal EEG signal is greater than the distortion evaluation threshold, the Potential abnormal EEG signals are recorded as distorted EEG signals; When the normalized distortion evaluation of a potential abnormal EEG signal is less than or equal to the distortion evaluation threshold, the Potential abnormal EEG signals are recorded as normal EEG signals. In this embodiment, the evaluation threshold is 0.7.

[0070] The potential value correction module determines the high-frequency EEG signals generated by the patient's sneezing and the corresponding affecting electrodes according to the above steps, as well as the distorted EEG signals generated by the loosening of the electrode patches of the affecting electrodes. In order to ensure the accuracy of the EEG signal data, it is necessary to further correct the distorted EEG signals and high-frequency EEG signals.

[0071] The normal EEG signals in the normal electrodes are used to correct the high-frequency EEG signals that affect the electrodes.

[0072] For high-frequency EEG signals, the potential value of the high-frequency EEG signal of the electrode affected by sneezing is much greater than the potential value of the segmented signal of the normal electrode.

[0073] For the influencing electrode, the two closest normal electrodes are selected, and the potential value of the high-frequency EEG signal of the influencing electrode is corrected according to the potential value of the normal electrode segmented signal at the same time. The specific correction expression is:

[0074] , Indicates the revised The first electrode The high-frequency EEG signal Potential value; Indicates The first electrode The high-frequency EEG signal Potential value; Indicates the The first of the two normal electrodes closest to the affected electrode The first segment signal The minimum value of a potential.

[0075] Then, the potential value in the distorted EEG signal is corrected by affecting the potential value in the normal EEG signal in the electrode. The specific correction expression is:

[0076] , Indicates the revised The first electrode The distorted EEG signal Potential value; Indicates The first electrode The distorted EEG signal Potential value; Indicates The influence of the electrode The first of the two normal EEG signals closest to the distorted EEG signal The minimum value among the potential values.

[0077] At this point, the correction of high-frequency EEG signals and distorted EEG signals affecting the electrodes has been completed.

[0078] The rating module corrects the distorted EEG signals and high-frequency EEG signals that affect the electrodes through the above steps, and then inputs the corrected EEG signals into the trained CNN network model for pain rating. The network structure of the CNN network model is input layer-convolution layer-pooling layer-fully connected layer-output layer; its loss function is the cross entropy loss function, and the output is the pain rating.

[0079] During training, each EEG signal pain rating is marked. In this embodiment, mark 0 represents no pain, mark 1 represents mild pain, mark 2 represents moderate pain, and mark 3 represents severe pain.

[0080] At this point, pain ratings of different EEG signals were obtained.

[0081] It should be noted that the above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

[0082] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A pain-assisted assessment system based on artificial intelligence and EEG signals, characterized in that: The system includes the following modules: An acquisition module is used to acquire an EEG signal consisting of a number of potential values ​​of each electrode; A high-frequency EEG signal screening module is used to divide the EEG signal into several segmented signals, determine the amplitude and period according to the maximum and minimum values ​​in the segmented signals; determine the amplitude-period ratio according to the amplitude and period in the segmented signals, obtain the high-frequency evaluation of the segmented signal of each electrode according to the difference between the amplitude-period ratio of the segmented signal of each electrode and the average of all amplitude-period ratios of each electrode and the ratio of the maximum and minimum value difference of all amplitude-period ratios of the electrode; determine the high-frequency EEG signal according to the high-frequency evaluation; A distorted EEG signal screening module is used to record electrodes with high-frequency EEG signals as affected electrodes and other electrodes as normal electrodes, and to determine normal EEG signals and potential abnormal EEG signals by the location of high-frequency EEG signals; Obtaining the degree of distortion of the potential abnormal EEG signal according to the relationship between the potential abnormal EEG signal and the normal EEG signal; Obtain the potential difference at each moment according to the difference between the potential value and the adjacent potential values; Obtaining distortion evaluation of potential abnormal EEG signals based on the distortion degree, high frequency evaluation and potential difference of potential abnormal EEG signals; determining a distorted EEG signal according to the distortion evaluation; The potential value correction module corrects the potential value of the affecting electrode according to the potential value of the normal electrode closest to the affecting electrode at the same time for high-frequency EEG signals; and corrects the potential value of the same position in the distorted EEG signal of the affecting electrode according to the normal EEG signal of the affecting electrode for distorted EEG signals; The rating module is used to output the pain rating of the corrected EEG signal through a neural network.

2. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for determining the amplitude and period according to the maximum and minimum values ​​in the segmented signal is: The maximum value in the segmented signal is taken as the amplitude in the segmented signal; The time length between two adjacent minimum values ​​in the segmented signal is regarded as a period in the segmented signal.

3. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for determining the amplitude-period ratio according to the amplitude and period in the segmented signal is: The ratio of the mean of all amplitudes in the segmented signal to the mean of all periods in the segmented signal is recorded as the amplitude-to-period ratio.

4. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for obtaining the high frequency evaluation of the segmented signal of each electrode according to the difference between the amplitude period ratio of the segmented signal of each electrode and the mean value of all amplitude period ratios of each electrode and the ratio of the maximum and minimum values ​​of all amplitude period ratios of the electrode is: , Indicates The amplitude-to-period ratio of the i-th segment signal in the electrode, Indicates The number of segmented signals in an electrode, Indicates The segmented signal in the electrode is in the interval Inside The maximum value of The function is a maximum value function; Indicates The segmented signal in the electrode is in the interval Inside The minimum value of The function is the minimum value function. Indicates High-frequency evaluation of the i-th segment signal in the i-th electrode.

5. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for determining normal EEG signals and potential abnormal EEG signals by the position of high-frequency EEG signals is: For all segmented signals of the electrode, the signal before the high-frequency EEG signal is recorded as a normal EEG signal, and the signal after the high-frequency EEG signal is recorded as a potential abnormal EEG signal.

6. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for obtaining the degree of distortion of the potential abnormal EEG signal according to the relationship between the potential abnormal EEG signal and the normal EEG signal is: The dissimilarity between the potential abnormal EEG signal and each normal EEG signal is calculated, and the mean of the dissimilarity between the potential abnormal EEG signal and all normal EEG signals is taken as the distortion degree of the potential abnormal EEG signal.

7. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for obtaining the potential difference at each moment according to the difference between the potential value and the adjacent potential value is: Let the difference between each potential value and the adjacent potential value be recorded as the potential difference; if the potential value has only one adjacent potential value, then the potential difference is the absolute value of the difference between the potential value and the adjacent potential value; if the potential value has two adjacent potential values, then the potential difference is the average of the absolute values ​​of the difference between the potential value and each adjacent potential value.

8. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for obtaining the distortion evaluation of the potential abnormal EEG signal according to the distortion degree, high frequency evaluation and potential difference of the potential abnormal EEG signal is: Obtaining the maximum potential difference and the minimum potential difference in the potential abnormal EEG signal, and recording the difference between the maximum potential difference and the minimum potential difference as the first potential difference; Obtaining a distortion evaluation of a potential abnormal EEG signal based on the distortion degree, high frequency evaluation, and first potential difference of the potential abnormal EEG signal; The distortion evaluation was positively correlated with the distortion degree, high frequency evaluation and first potential difference of potential abnormal EEG signals.

9. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for correcting the potential value of the influencing electrode according to the potential value of the normal electrode closest to the influencing electrode at the same time is: , Indicates the revised The first electrode The high-frequency EEG signal Potential value; Indicates The first electrode The high-frequency EEG signal Potential value; Indicates the The first of the two normal electrodes closest to the affected electrode The first segment signal The minimum value of a potential.

10. The pain auxiliary assessment system based on artificial intelligence and EEG signals as claimed in claim 1, characterized in that: The method for correcting the potential value at the same position in the distorted EEG signal of the influencing electrode according to the normal EEG signal of the influencing electrode is: , Indicates the revised The first electrode The distorted EEG signal Potential value; Indicates The first electrode The distorted EEG signal Potential value; Indicates The influence of the electrode The first of the two normal EEG signals closest to the distorted EEG signal The minimum value among the potential values.

Images