An epilepsy prediction system based on simultaneous acquisition of eeg-gastrointestinal signals

Abstract

The application relates to the field of disease prediction, and particularly relates to an epilepsy prediction system based on electroencephalogram-electrogastrogram signals collected synchronously, which comprises the following modules: a data acquisition module, which is used for acquiring data, wherein the data comprises electroencephalogram data and electrogastrogram data; a first analysis module, which is used for obtaining average electrode inconsistency indexes of first channels and second channels based on a first model according to time sequence data; and a prediction module, which is used for comparing the average electrode inconsistency indexes of a subject with preset evaluation thresholds, so as to predict epilepsy patients. The system provided by the application can more accurately and efficiently distinguish epilepsy patients from healthy people, and can be applied to the preliminary screening of epilepsy for a large population (for example, a community or a physical examination center) as an auxiliary diagnosis tool.

Description

[0001] This application is based on the Chinese invention patent application with the application number 202411311041.2 and the application date of September 19, 2024, and the invention name of "A seizure prediction system based on synchronous acquisition of electroencephalogram-gastrointestinal electrical signals". TECHNICAL FIELD

[0002] The present application relates to the field of disease prediction, in particular to a seizure prediction system based on synchronous acquisition of electroencephalogram-gastrointestinal electrical signals. BACKGROUND

[0003] Seizure is a chronic neurological disease, which is one of the five major neurological and mental diseases listed by the World Health Organization for global prevention and treatment, and is also a key public relations area of the global brain science plan. Seizure is caused by abnormal discharge of a group of brain cells, and different parts of the brain can become the site of abnormal discharge. Its characteristics are repeated seizures, sudden onset and sudden stop, and brain electrical activity is difficult to capture during the seizure period. During a seizure, a part of the body or the whole body often appears a short period of involuntary convulsions, sometimes accompanied by loss of consciousness and incontinence of urine and feces. The seizure ranges from very short loss of consciousness or muscle reflex to severe and persistent convulsions. The frequency of seizures can also vary from less than once a year to several times a day.

[0004] Currently, the diagnosis of epilepsy is based on clinical seizures and epileptiform discharges on electroencephalogram. On the one hand, the assessment of electroencephalogram abnormalities can be subjective and not very sensitive. On the other hand, a large proportion of patients with epilepsy have very short seizures, and it is difficult to capture their abnormal discharges even if they are subjected to electroencephalogram examination. Therefore, the existing technology based on electroencephalogram can only detect the epileptiform discharges of the subject, and is thus suitable for the diagnosis of seizures, but not for the screening of epilepsy.

[0005] The long-term misunderstanding, fear and prejudice of epilepsy patients towards epilepsy lead to a considerable number of people in China who are reluctant to go to medical institutions for further examination even if they have seizures; and the existing diagnosis of epilepsy is based on the detection of epileptiform discharges of the subject, which leads to the fact that patients with epilepsy who do not have epileptiform discharges during electroencephalogram examination are missed by the existing technology. Another part of the patients may have non-epileptic seizure disorders (such as neurotic seizures, transient ischemic attacks), but they are often worried that they have epilepsy, and it is difficult to distinguish epilepsy from them only by clinical manifestations. Correct diagnosis is the premise of effective treatment, and the existing technology is difficult to distinguish between non-epileptic patients and epileptic patients only by the patient's oral report and electroencephalogram without abnormal discharges, especially in a large population, such as community and medical centers.

[0006] Chinese patent application CN117481666A discloses a resting state stomach-brain electrical signal coupling method, which reflects the degree of bidirectional influence between the stomach electrical raw signal and the brain electrical raw signal by extracting the phase amplitude coupling, phase phase coupling and amplitude amplitude coupling indicators of the stomach electrical raw signal and the brain electrical raw signal in different frequency bands, and the transfer entropy. However, its application scenario is limited to regulating stomach function, and it is difficult to realize the diagnosis of diseases. SUMMARY

[0007] The present application provides an epilepsy prediction system based on synchronous acquisition of electroencephalogram-electrogastrogram signals, characterized by comprising:

[0008] a data acquisition module for acquiring data, the data including electroencephalogram data and electrogastrogram data, the electroencephalogram data being collected from a first channel, and the electrogastrogram data being collected from a second channel; the data including sample data from a sample population and subject data from a subject;

[0009] a data preprocessing module for preprocessing the data to obtain corresponding time series data;

[0010] a first analysis module for obtaining average electrode inconsistency indices of the first channel and the second channel based on a first model according to the time series data; the first model comprising: ; wherein JSD represents the JS divergence score function, t represents time, represents the time series data of the corresponding channel of the first channel and the second channel recorded at time point t seconds of the subject, m represents the total number of channels, and n represents the total duration of the time series data of the corresponding channel; ICI represents the inconsistency index of the corresponding channel;

[0011] a prediction module for comparing the average electrode inconsistency index of the subject with a preset evaluation threshold, and when the average electrode inconsistency index of the subject is not lower than the preset evaluation threshold, the subject is predicted as an epilepsy patient;

[0012] wherein the first channel includes F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz and POz; and the second channel includes the body of the stomach, the lesser curvature of the stomach, the antrum of the stomach, the ascending colon, the transverse colon, the descending colon and the rectal channel.

[0013] In some embodiments, the system further comprises a second analysis module for converting the time series data into probability density distribution data.

[0014] In some embodiments, the second analysis module is further configured to obtain an inconsistency index for the corresponding channel based on the probability density distribution data and a second model, wherein the second model includes: Where ICI represents the inconsistency exponential function, Let t represent the KL divergence function, and t represent time. This represents the time-series data of the corresponding channel recorded at time point t seconds for the subject. This represents the time-series data of the corresponding channel recorded at time point t for healthy individuals, where n is the total duration of the time-series data for the corresponding channel. Represents the sample distribution. Indicates the reference distribution.

[0015] In some embodiments, the system further includes a database for storing data obtained by the data acquisition module, the data preprocessing module, the first analysis module, the second analysis module, and the prediction module, as well as for generating corresponding sample datasets.

[0016] In some embodiments, the system further includes an evaluation threshold setting module, which randomly obtains the mean electrode inconsistency index of 50-70% of the epilepsy patient samples in the sample dataset and calculates the critical value for the occurrence of epilepsy, and sets the mean electrode inconsistency index corresponding to the minimum critical value as the evaluation threshold.

[0017] In some embodiments, the electroencephalogram (EEG) data and the gastrointestinal electroencephalogram (GEG) data are acquired simultaneously.

[0018] In some embodiments, the acquisition time for the electroencephalogram (EEG) data and the gastrointestinal electrical data is at least 5 minutes.

[0019] In some embodiments, the preprocessing includes filtering the EEG data to remove EEG signals below 1 Hz and above 30 Hz, and normalizing the gastrointestinal EEG data and the filtered EEG data to obtain the time series data.

[0020] In some embodiments, the data is collected before or after a meal.

[0021] In some embodiments, the data is collected from subjects in a normal state. As used herein, "normal state" means that the subject's brain electrical activity is normal, i.e., there are no detectable epileptiform discharges on the subject's electroencephalogram (EEG). In some embodiments, when the subject is a healthy individual (i.e., a non-epilepsy patient), the normal state is any state of the subject. In some embodiments, when the subject has epilepsy, the normal state is during the subject's non-seizure period (or interictal period).

[0022] In some embodiments, the subject is an epileptic patient.

[0023] In some embodiments, the epilepsy patient has a focal or generalized origin.

[0024] Compared with the prior art, the beneficial effects of the present invention include at least the following aspects:

[0025] This invention provides an epilepsy prediction system based on synchronously acquired EEG-gastrointestinal electrical signals. Existing technologies process EEG and gastrointestinal signals by performing phase-amplitude coupling, phase-phase coupling, and amplitude-amplitude coupling on EEG and gastrointestinal signals from different frequency bands. There are currently no reports on using EEG and gastrointestinal signals for epilepsy prediction and diagnosis.

[0026] Unlike conventional approaches that extract frequency bands from electroencephalogram (EEG) signals, this invention achieves accurate prediction of epilepsy by collecting simplified EEG and gastrointestinal EEG data (i.e., the first and second channels of this invention) and combining them with processing techniques such as inconsistency indices. It is important to emphasize that, on the one hand, this invention breaks away from the general approach of simultaneously selecting highly correlated channels from both EEG and gastrointestinal EEG channels. By simplifying the EEG channels that are highly correlated with the gastrointestinal EEG channels and replacing them with specific gastrointestinal EEG channels, it achieves channel simplification, reducing the practical operational difficulty while improving the accuracy and sensitivity of epilepsy diagnosis. On the other hand, this invention does not detect epileptic seizures by detecting abnormal discharges in the EEG, but rather diagnoses whether a subject has epilepsy based on the subject's normal EEG (and gastrointestinal) readings.

[0027] In large-scale screening, even experienced clinicians often struggle to determine whether a subject has epilepsy based solely on verbal descriptions of "epileptic" seizure symptoms, especially under time constraints and with limited diagnostic evidence. The simplified channel and specific algorithm of this invention provide a predictive model with high accuracy and sensitivity. It can capture and amplify the differences between epilepsy patients and healthy individuals who exhibit nearly identical brain electrical activity under conventional methods (e.g., visual observation), thus differentiating them. This model is suitable for assisting in the diagnosis of subjects in large-scale screening who may not show abnormal discharges on their electroencephalograms (e.g., healthy individuals exhibiting seizure-like symptoms but not epilepsy, and epilepsy patients with infrequent seizures).

[0028] Based on this, the system provided by the present invention not only requires less time to collect data, but also can more accurately and efficiently distinguish between epilepsy patients and healthy people. It can be used as an auxiliary diagnostic tool for the initial screening of epilepsy patients in a larger population (such as communities and health check centers). Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. The elements or parts in the drawings are not necessarily drawn to scale. Obviously, the drawings described below are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0030] Figure 1 This is a simplified schematic diagram showing the positions of the first and second channels of the present invention.

[0031] Figure 2 The graph shows the performance results of different prediction models obtained in this invention.

[0032] Figure 3 A flowchart illustrating the method for analyzing synchronously acquired EEG-gastrointestinal electrical signals provided by this invention;

[0033] Figure 4 This is a schematic diagram of the epilepsy prediction system based on synchronously acquired EEG-gastrointestinal electrical signals provided by the present invention.

[0034] 100 is the prediction system, 102 is the data acquisition module, 104 is the data preprocessing module, 106 is the first analysis module, 108 is the second analysis module, 110 is the prediction module, 112 is the database, and 114 is the evaluation threshold setting module. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0036] In this document, suffixes such as "module," "part," or "unit" used to denote elements are used only for the purpose of illustrative purposes and have no specific meaning in themselves. Therefore, "module," "part," or "unit" may be used interchangeably.

[0037] In this document, the terms "upper," "lower," "inner," "outer," "front," "rear," "one end," and "the other end," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the present invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0038] In this document, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, a direct connection, or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0039] In this document, "and / or" includes any and all combinations of one or more of the listed related items.

[0040] In this article, "multiple" means two or more, that is, it includes two, three, four, five, etc.

[0041] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0042] As used in this specification, the term "about" typically means + / -5% of the value, more typically + / -4%, more typically + / -3%, more typically + / -2%, even more typically + / -1%, even more typically + / -0.5%.

[0043] In this specification, certain embodiments may be disclosed in a range-bound format. It should be understood that this "range-bound" description is merely for convenience and brevity and should not be construed as a rigid limitation on the disclosed range. Therefore, the description of a range should be considered as having specifically disclosed all possible subranges and the individual numerical values ​​within those ranges. For example, a description of the range 1-6 should be considered as having specifically disclosed subranges such as from 1 to 3, from 1 to 4, from 1 to 5, from 2 to 4, from 2 to 6, from 3 to 6, etc., and the individual numbers within those ranges, such as 1, 2, 3, 4, 5, and 6. This rule applies regardless of the breadth of the range.

[0044] Example 1

[0045] 1.1 Dataset

[0046] Subjects: This study included 102 patients with focal or generalized epilepsy (57 males, age 33.82±12.65) and 46 healthy controls (26 males, age 31.00±8.45 years).

[0047] 1.1.1 Electroencephalogram (EEG) recording

[0048] Subjects sat in a quiet room to prepare for the test. Five minutes of resting-state EEG data were acquired from all subjects using 64 Ag / AgCl electrodes (Brain Products, Munich, Germany), placed after a 10 / 20 system with an impedance <10 kΩ. EEG data were amplified and collected using an actiCAmp amplifier (BrainVision system amplifier; Brain Products, Munich, Germany) at a sampling rate of 1000 Hz. Subjects were instructed to keep their gaze fixed on the center of the screen and to avoid body movement as much as possible during EEG recording.

[0049] 1.1.2 Gastrointestinal electrical signal (EGG) recording

[0050] Gastrointestinal electromyography (EMG) activity was measured using an 8-channel gastrointestinal EMG system (XDJ-S8, Hefei Kaili Co., Hefei, China). Subjects were instructed to avoid alcohol and spicy or irritating foods for at least three days and to fast for at least six hours prior to the examination. Measurements were performed in a supine position. Four gastric electrodes (reflecting the gastric body, antrum, lesser curvature, and greater curvature) and four intestinal electrodes (reflecting the ascending colon, transverse colon, descending colon, and rectum) were placed on the abdominal skin (Hanjie Co., Shanghai, China). The specific electrode placement locations were as follows: Gastric body: 3-5 cm to the left and 1 cm above the midpoint of the line connecting the xiphoid process and the umbilicus; Gastric antrum: 2-4 cm to the right; Lesser curvature: 1 / 2 of the distance above the midpoint of the line connecting the xiphoid process and the umbilicus; Greater curvature: 1 / 2 of the distance below the midpoint of the line connecting the xiphoid process and the umbilicus. Ascending colon: 2-4 cm to the right, level with the umbilicus; Transverse colon: 1 cm below the umbilicus; Descending colon: 2-4 cm to the left, level with the umbilicus; Rectum: Below the coccyx on the back. The ground electrode was placed on the medial ankle of the right calf, and the reference electrode was placed on the medial wrist of the right hand. Subjects were instructed to remain quiet and not to move or speak during the test. Additionally, a mealtime functional load test was performed after EGG recording 6 minutes before the meal, followed by a meal of approximately 200 kcal.

[0051] 1.2 Prediction Algorithm Based on JS Divergence (Jensen–Shannon divergence, JSD)

[0052] 1.2.1 Data Preprocessing

[0053] The dataset was divided into training and test sets in a 7:3 ratio. For independent training, only EEG or EGG signals were used. For EEG-EGG multimodal fusion training, both EEG and EGG signals were used.

[0054] First, the raw EEG signal is filtered using EEGLAB in MATLAB to maintain data integrity by eliminating signals below 1 Hz and above 30 Hz. Second, the filtered EEG data and the raw EGG signal are normalized. The feature matrices of the preprocessed EEG time series or preprocessed EGG time series are shown below.

[0055]

[0056] Where m and n represent the dimensions of the feature matrix of the EEG signal or EGG signal, respectively. m is the total number of electrodes, therefore the i-th row represents the number of electrodes. The data is recorded per second. n is the total duration of the EEG or EGG time-series signal, meaning the entire EEG or EGG data contains n seconds. Therefore, column t represents the EEG or EGG data measured at each electrode at time point t. Thus, The electrode was recorded at time point t seconds. The EEG signal or EGG signal.

[0057] 1.2.2 Convert the preprocessed EEG time series or preprocessed EGG time series into probability density distribution data.

[0058] The data for each electrode is converted into probability density distribution data. In the training set, time-series data for each electrode are used. To fit the sample distribution , where the mean and standard deviation As a sample feature. Furthermore, for the healthy control data in the training set, the time series data of each electrode... Fit to reference distribution In, the mean and standard deviation Reference features for differentiation

[0059]

[0060] in, Represents the sample distribution. This represents the time-series data for each electrode of the sample. Represents the sample mean. This represents the standard deviation of the sample.

[0061]

[0062] in, Indicates the reference distribution. This represents the time series data for each reference electrode. Indicates the reference mean. This represents the standard deviation for reference.

[0063] When the difference between sample features and reference features is small, it means there is no substantial difference between the sample distribution and the reference distribution, thus indicating the absence of epilepsy. Conversely, when the difference is quite large, it means there is a significant difference between the sample distribution and the reference distribution, indicating the presence of epilepsy.

[0064] 1.2.3 Constructing Inconsistency Metrics Based on JSD

[0065] JS divergence was used to quantify the deviation of the sample distribution from the reference distribution to measure the differences between different states (i.e., epilepsy and healthy controls); the inconsistency index (ICI) of each electrode was calculated to represent the differences between time points.

[0066]

[0067] in, Denotes the KL divergence function. Represents the sample distribution. Indicates the reference distribution. This represents the time series data of the corresponding channel of the sample recorded at time point t seconds. This represents the time-series data of the corresponding channel of the reference recorded at time point t, where t represents time and T represents the total time.

[0068]

[0069] ICI stands for Inconsistency Index Function. Represents the sample distribution. Indicates the reference distribution. This represents the time series data of the corresponding channel of the sample recorded at time point t seconds. This represents the reference time series data of the corresponding channel recorded at time point t, where t represents time and n is the total duration of the time series data of the corresponding channel.

[0070] By calculating the ICI (Integrated Chance Index), the fluctuations at all time points caused by the reference distribution are quantified. The average ICI score for each electrode can be used as a quantitative measure to assess the degree of distributional difference over time. A small average ICI score indicates that there is no substantial difference between the sample distribution and the reference distribution of the electrode. Conversely, a large average ICI score indicates that there is a significant difference between the sample distribution and the reference distribution of the electrode.

[0071] 1.2.4 Identification of DNB Electrodes

[0072] Based on the DNB (Dynamic Network Biomarker) theory, this paper assumes that a set of known molecules emit pre-disease state signals for complex diseases before significant transitions, and calculates the average ICI score for each electrode. These scores are then ranked according to the relative importance of key moments. The fluctuations caused by the reference distribution across all time points are obtained using the above equation, and the ICI scores of each electrode are then ranked.

[0073] In this embodiment, the selection of DNB electrodes follows two different strategies. For EEG training alone, electrodes ranking in the top 50% based on their highest average scores are classified as DNB electrodes.

[0074] For EEG-EGG multimodal fusion training, a total of 78 channels were included (62 EEG channels and 16 EGG channels). First, the top 30 channels with the highest average scores for each individual were selected. Then, if a certain channel appeared in more than half of the individuals in the training set, it was used as the DNB electrode. Since different electrodes may be associated with different regions of the brain or stomach, a higher score for a specific electrode across the entire array indicates a significant relationship with epileptic activity.

[0075] 1.2.5 Calculating JSD for Global Time Series

[0076] The average electrode inconsistency index for global time series is defined as follows:

[0077]

[0078] Where JSD represents the JS divergence scoring function, and t represents time. This represents the time series data of the corresponding channel recorded at time point t seconds, where m represents the total number of channels, n represents the total duration of the time series data of the corresponding channel, and ICI represents the inconsistency index of the corresponding channel.

[0079] The JSD score is used to quantify the distributional differences among all electrodes, thus helping to identify any significant deviations between distributions. A higher score indicates a significant difference between distributions; when this difference is exceeded, it indicates that the difference has escalated to a critical level, thus predicting epilepsy. To find an appropriate threshold level for each patient, the following training scenario was employed: using 70% of the epilepsy data as training data to obtain the threshold. For each epilepsy case in the training dataset, a critical value in the DNB theory for each seizure was obtained based on the JSD score, and the minimum of these critical values ​​was used as the threshold. When the score at a certain time point exceeds this threshold, the subject is diagnosed with epilepsy.

[0080] 1.3 Correlation analysis of EEG and EGG channels

[0081] The probability distribution of each channel is evaluated by analyzing 62 channels of EEG data and 16 channels of EGG data for each individual, based on the following equation:

[0082]

[0083] Density function P(obtained using EEG and EGG channels) ) and P ( The p-value was calculated using the mean and standard deviation of the p-values. The correlation between EEG and EGG channels was determined by comparing their p-values. A p-value less than 0.05 indicates no significant difference between the two, thus supporting the hypothesis that a relationship exists between them. To identify the EEG channels with the strongest correlation to the EGG channels, the three EEG channels with the smallest statistical deviation from the distribution of the 16 EGG channels in each sample were selected.

[0084] Example 2

[0085] It should be noted that in this invention, the selection of DNB electrodes is performed in two different ways. For standalone EEG training, the top 50% of electrodes with the highest average ICI scores are selected as DNB electrodes based on the total number of electrodes used. For EEG-EGG multimodal fusion training, the top 30 channels with the highest average ICI scores for each individual in the training set are selected. If a channel appears in more than half of the training set, it is designated as a common DNB electrode for predicting epilepsy. The results are as follows: Figure 1 As shown, the 30 channels obtained by this invention include: 23 EEG channels (first channel): F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz and POz; and 7 gastrointestinal EEG channels (second channel): gastric body channel (i.e., electrode 1), lesser curvature channel (i.e., electrode 2), antrum (i.e., electrode 4), ascending colon channel (i.e., electrode 5), transverse colon channel (i.e., electrode 6), descending colon channel (i.e., electrode 7) and rectal channel (i.e., electrode 8).

[0086] Figure 2 In this context, EEG refers to the prediction model trained using the method of this invention and data collected from 62 EEG channels; EEG * The first part, "EEG+EGG," represents a prediction model trained using the method of this invention, employing data from the top 50% of electrodes with the highest average ICI scores across 62 EEG channels. The second part, "EEG+EGG," represents a prediction model trained using the method of this invention, employing data from 78 channels (i.e., 62 EEG channels and 16 EGG channels). * This refers to the prediction model obtained by using the method of the present invention and training it with data collected from the first and second channels.

[0087] like Figure 2As shown, the prediction model obtained by processing EEG or EEG+EGG data using the method of this invention has already shown good predictive ability. However, while it requires collecting and calculating more data, its accuracy is difficult to reach 95%. In contrast, the prediction model constructed based on the simplified electrodes obtained by this invention not only requires fewer electrodes to be placed, but also improves the performance of predicting epilepsy, with an observed accuracy of 98.92% and a sensitivity of 98.82%.

[0088] Correlation analysis between EEG channels and EGG channels.

[0089] This embodiment further explored the correlation between different EEG channels and EGG channels in predicting epilepsy using probability distribution. Specifically, this embodiment analyzed the probability distribution of 62 EEG channels and 16 EGG channels for each subject. In 148 samples, the frequency of occurrence of the 62 EEG channels was statistically similar to that of the 16 EGG channels. Therefore, this embodiment extracted the top three EEG channels associated with each EGG channel in the subjects, and these channels were defined as key correlation factors for specific EGG channels.

[0090] The distribution of all gastrointestinal and electroencephalogram (EEG) channels was compared for similarity. The frequency of eight gastrointestinal leads with no significant difference in distribution from the EEG channels was calculated among 148 subjects. A higher frequency indicates a stronger correlation between the EEG lead and the gastrointestinal signal. In this embodiment, the top 20% of EEG channels, which did not show statistically significant distribution differences compared to the 16 EEG channels, were designated as highly correlated EEG channels, as shown in Table 1.

[0091] Table 1: Statistical analysis of the frequency of the top 20% of EEG channels related to gastrointestinal electrical activity in 148 sample results

[0092]

[0093] The results showed that most of the EEG channels shown in Table 1 were not included in the first channel. Based on this, the present invention believes that the predictive role of most of the EEG channels shown in Table 1 for epilepsy can be replaced by the relevant gastrointestinal EEG channels (i.e., the second channel mentioned above).

[0094] Example 3

[0095] See Figure 3 This invention provides a method for analyzing synchronously acquired EEG-gastrointestinal electrical signals, comprising the following steps:

[0096] S101 Input the synchronously collected EEG and gastrointestinal EEG data of the subject;

[0097] In some embodiments, the electroencephalogram (EEG) data is acquired from a first channel, and the gastrointestinal EEG data is acquired from a second channel.

[0098] In this invention, "channel", "electrode" and "lead" have the same meaning unless otherwise stated.

[0099] In some embodiments, the first channel includes F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz, and POz; the second channel includes the gastric body, lesser curvature, antrum, ascending colon, transverse colon, descending colon, and rectum.

[0100] It should be noted that the first channel and the second channel are the simplified EEG channel and gastrointestinal EEG channel of the present invention, respectively, and the first channel and the second channel have low correlation in epilepsy.

[0101] Unlike existing techniques that couple gastrointestinal electrical signals and brain signals, this invention, after identifying brain channels in epilepsy patients that are strongly correlated with gastrointestinal electrical signals, excludes these brain channels from the actual brain channels to be acquired, thus obtaining the aforementioned first channel. This invention discovers that the predictive function of these brain channels can be replaced by related gastrointestinal electrical channels (i.e., the aforementioned second channel). By using the aforementioned first and second channels, which have lower correlation in predicting epilepsy, not only is the acquisition operation more difficult, but the accuracy of prediction for epilepsy patients is also improved, making it easier for practical applications. Using a lower number of brain channels (e.g., conventional 16-lead or 32-lead channels) is insufficient to cover the entire brain, making it difficult to predict epilepsy.

[0102] In some embodiments, the acquisition time for the electroencephalogram (EEG) data and the gastrointestinal electrical data is at least 5 minutes.

[0103] In some embodiments, the data is collected before or after a meal.

[0104] In some embodiments, the data is collected from subjects in a normal state.

[0105] In some embodiments, the subject is an epileptic patient.

[0106] In some embodiments, the epilepsy patient has a focal or generalized origin.

[0107] S102 Preprocesses the synchronously acquired EEG data and gastrointestinal electroencephalogram data to obtain corresponding time series data;

[0108] In some embodiments, the preprocessing includes filtering the EEG data to remove EEG signals below 1 Hz and above 30 Hz, and normalizing the gastrointestinal EEG data and the filtered EEG data to obtain the time series data.

[0109] S103 Based on the first model, the average electrode inconsistency index of the first channel and the second channel is obtained;

[0110] In some embodiments, the first model includes: Where JSD represents the JS divergence scoring function, and t represents time. This represents the time-series data of the first and second channels recorded at time point t, where m represents the total number of channels and n represents the total duration of the time-series data of the corresponding channel; ICI represents the inconsistency index of the corresponding channel.

[0111] In some embodiments, the corresponding channels refer to F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz, POz, gastric body channel, lesser curvature channel, antral channel, ascending colon channel, transverse colon channel, descending colon, and rectal channel. In some embodiments, the total duration refers to the total duration of the entire synchronously acquired EEG and gastrointestinal electrical data; for example, when the acquisition time is 5 minutes, the total duration is 5 minutes.

[0112] Based on the above and the first and second channels, the present invention not only reduces the difficulty of acquisition operations, but also reduces the amount of data that needs to be analyzed and calculated, improving the efficiency of calculation while maintaining a high level of accuracy and sensitivity.

[0113] In some embodiments, the method further includes S201 converting the time series data into probability density distribution data.

[0114] In some embodiments, the method further includes S202 obtaining an inconsistency index for the corresponding channel based on the probability density distribution data and a second model, wherein the second model includes: Where ICI represents the inconsistency exponential function, Let t represent the KL divergence function, and t represent time. This represents the time-series data of the corresponding channel recorded at time point t seconds for the subject. This represents the time-series data of the corresponding channel recorded at time point t for healthy individuals, where n is the total duration of the time-series data for the corresponding channel. Represents the sample distribution. Indicates the reference distribution.

[0115] S104 Compare the subject's mean electrode inconsistency index with a preset assessment threshold. When the subject's mean electrode inconsistency index is not lower than the preset assessment threshold, the subject is predicted to be an epilepsy patient.

[0116] In some embodiments, the method for obtaining the preset evaluation threshold includes S203 randomly obtaining the mean electrode inconsistency index of 50-70% of the cognitive impairment patient samples in the sample dataset and calculating the critical value for the occurrence of epilepsy (based on DNB theory), and setting the mean electrode inconsistency index corresponding to the minimum critical value as the evaluation threshold.

[0117] Example 4

[0118] See Figure 4 This invention provides an epilepsy prediction system 100 based on synchronously acquired EEG-gastrointestinal electrical signals, characterized in that it comprises:

[0119] Data acquisition module 102 is used to acquire data, including electroencephalogram (EEG) data and gastrointestinal electroencephalogram (GEG) data;

[0120] In some embodiments, the electroencephalogram (EEG) data and the gastrointestinal electroencephalogram (GEG) data are acquired simultaneously.

[0121] In some embodiments, the acquisition time for the electroencephalogram (EEG) data and the gastrointestinal electrical data is at least 5 minutes.

[0122] In some embodiments, the data is collected before or after a meal.

[0123] In some embodiments, the electroencephalogram (EEG) data is acquired from a first channel, and the gastrointestinal EEG data is acquired from a second channel.

[0124] In some embodiments, the data includes sample data from the sample population and subject data from the subjects.

[0125] In some embodiments, the sample population includes healthy individuals and patients with epilepsy.

[0126] In some embodiments, the data is collected from subjects in a normal state.

[0127] In some embodiments, the subject is an epileptic patient.

[0128] In some embodiments, the epilepsy patient has a focal or generalized origin.

[0129] In some embodiments, the first channel includes F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz, and POz; the second channel includes the gastric body, lesser curvature, antrum, ascending colon, transverse colon, descending colon, and rectum.

[0130] Data preprocessing module 104 is used to preprocess the data to obtain corresponding time series data;

[0131] In some embodiments, the preprocessing includes filtering the EEG data to remove EEG signals below 1 Hz and above 30 Hz, and normalizing the gastrointestinal EEG data and the filtered EEG data to obtain the time series data.

[0132] The first analysis module 106 is used to obtain the average electrode inconsistency index of the first channel and the second channel based on the time series data and a first model.

[0133] In some embodiments, the first model includes: Where JSD represents the JS divergence scoring function, and t represents time. This represents the time-series data of the first and second channels recorded at time point t, where m represents the total number of channels and n represents the total duration of the time-series data of the corresponding channel; ICI represents the inconsistency index of the corresponding channel.

[0134] The second analysis module 108 is used to convert the time series data into probability density distribution data.

[0135] In some embodiments, the second analysis module 108 is further configured to obtain the inconsistency index of each sub-channel based on the probability density distribution data and a second model.

[0136] In some embodiments, the second model includes: Where ICI represents the inconsistency exponential function, Let t represent the KL divergence function, and t represent time. This represents the time-series data of the corresponding channel recorded at time point t seconds for the subject. This represents the time-series data of the corresponding channel recorded at time point t for healthy individuals, where n is the total duration of the time-series data for the corresponding channel. Represents the sample distribution. Indicates the reference distribution.

[0137] In some embodiments, the first analysis module 106 and the second analysis module 108 may be combined into a single analysis module.

[0138] The prediction module 110 is used to compare the subject's mean electrode inconsistency index with a preset assessment threshold. When the subject's mean electrode inconsistency index is not lower than the preset assessment threshold, the subject is predicted to be an epilepsy patient.

[0139] Database 112 is used to store the data obtained by the data acquisition module 102, the data preprocessing module 104, the first analysis module 106, the second analysis module 108 and the prediction module 110, and to generate corresponding sample datasets.

[0140] In some embodiments, the sample dataset includes a dataset of healthy individuals and a dataset of patients with epilepsy.

[0141] The evaluation threshold setting module 114 randomly obtains the average electrode inconsistency index of 50-70% of the epilepsy patient samples in the epilepsy sample dataset and calculates the critical value for the occurrence of epilepsy, and sets the average electrode inconsistency index corresponding to the smallest critical value as the evaluation threshold.

[0142] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a computer terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0143] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. An epilepsy prediction system based on synchronously acquired EEG-gastrointestinal electrical signals, characterized in that, include: The data acquisition module is used to acquire data, including electroencephalogram (EEG) data and gastrointestinal electroencephalogram (GEG) data. The EEG data is acquired from a first channel, and the GEG data is acquired from a second channel. The data includes sample data from the sample population and subject data from the subjects. The data preprocessing module is used to preprocess the data to obtain corresponding time series data; The first analysis module is used to obtain the average electrode inconsistency index of the first channel and the second channel based on the time series data and a first model. The first model includes: Where JSD represents the JS divergence scoring function, and t represents time. This represents the time-series data of the first and second channels recorded at time point t, where m represents the total number of channels, n represents the total duration of the time-series data of the corresponding channel, and ICI represents the inconsistency index of the corresponding channel. The prediction module is used to compare the subject's mean electrode inconsistency index with a preset assessment threshold. When the subject's mean electrode inconsistency index is not lower than the preset assessment threshold, the subject is predicted to be an epilepsy patient. The first channel includes F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz, and POz; the second channel includes the gastric body, lesser curvature, antrum, ascending colon, transverse colon, descending colon, and rectum.

2. The system as described in claim 1, characterized in that, The system also includes a second analysis module for converting the time series data into probability density distribution data.

3. The system as described in claim 2, characterized in that, The second analysis module is further configured to obtain the inconsistency index of the corresponding channel based on the probability density distribution data and a second model, wherein the second model includes: Where ICI represents the inconsistency exponential function, Let t represent the KL divergence function, and t represent time. This represents the time-series data of the corresponding channel recorded at time point t seconds for the subject. This represents the time-series data of the corresponding channel recorded at time point t for healthy individuals, where n is the total duration of the time-series data for the corresponding channel. Represents the sample distribution. Indicates the reference distribution.

4. The system as described in claim 1, characterized in that, The electroencephalogram (EEG) data and the gastrointestinal electroencephalogram (GEG) data were collected simultaneously.

5. The system as described in claim 1, characterized in that, The acquisition time for the electroencephalogram (EEG) data and the gastrointestinal electroencephalogram (GEG) data is at least 5 minutes.

6. A method for analyzing synchronously acquired EEG-gastrointestinal electrical signals, characterized in that, Includes the following steps: S101 Inputs synchronously collected EEG and gastrointestinal electrical data of the subject; the EEG data is collected from a first channel, and the gastrointestinal electrical data is collected from a second channel; the first channel includes F3, C4, P4, O2, T7, FC1, CP1, CP6, TP9, TP10, F2, C1, C2, P2, FC3, FC4, CP3, CP4, PO4, P6, PO8, CPz, and POz; the second channel includes the gastric body, lesser curvature, antrum, ascending colon, transverse colon, descending colon, and rectum. S102 Preprocesses the synchronously acquired EEG data and gastrointestinal electroencephalogram data to obtain corresponding time series data; S103 Based on the first model, the average electrode inconsistency index of the first channel and the second channel is obtained; The first model includes: Where JSD represents the JS divergence scoring function, and t represents time. This represents the time-series data of the first and second channels recorded at time point t, where m represents the total number of channels, n represents the total duration of the time-series data of the corresponding channel, and ICI represents the inconsistency index of the corresponding channel. S104 Compare the subject's mean electrode inconsistency index with a preset assessment threshold. When the subject's mean electrode inconsistency index is not lower than the preset assessment threshold, the subject is predicted to be an epilepsy patient.

7. The method as described in claim 6, characterized in that, The method further includes step S201, which converts the time series data into probability density distribution data.

8. The method as described in claim 7, characterized in that, The method further includes step S202: Based on the probability density distribution data, obtaining the inconsistency index of the corresponding channel using a second model, wherein the second model includes: Where ICI represents the inconsistency exponential function, Let t represent the KL divergence function, and t represent time. This represents the time-series data of the corresponding channel recorded at time point t seconds for the subject. This represents the time-series data of the corresponding channel recorded at time point t for healthy individuals, where n is the total duration of the time-series data for the corresponding channel. Represents the sample distribution. Indicates the reference distribution.

9. The method as described in claim 6, characterized in that, The data was collected before or after meals.

10. The method as described in claim 6, characterized in that, The data was collected from subjects in a normal state.

Images