A brainwave intelligent acquisition and analysis system

Abstract

This invention discloses an intelligent brainwave acquisition and analysis system, including a data acquisition module, a data analysis module, an anomaly alarm module, a data update module, and a data display interface. This invention relates to the field of biosignal processing technology. The intelligent brainwave acquisition and analysis system preprocesses the acquired data through a data preprocessing unit, extracts digital signals through a feature extraction and analysis unit to construct analysis curves, segments and amplifies the analysis curves using timestamps, and combines frequency fluctuations to segment and determine the user's brain activity state. These segments are then compared with set thresholds for different states to identify distinguishing features. A result judgment unit removes interfering features and marks abnormal results, thereby effectively improving the efficiency of acquisition and analysis, as well as the accuracy and reliability of brainwave signal analysis. This provides timely and effective feedback on the real-time status of brain signals, offering users more precise health management and decision support.

Description

Technical Field

[0001] This invention relates to the field of biosignal processing technology, specifically to an intelligent brainwave acquisition and analysis system. Background Technology

[0002] Brainwaves are bioelectrical signals generated by the electrical activity of neurons in the brain. They can reflect the functional state and cognitive processes of the brain. With the rapid development of neuroscience and artificial intelligence technology, the acquisition and analysis of brainwave signals have shown great potential in fields such as clinical medicine, neurofeedback training, and brain-computer interfaces.

[0003] The reference patent, titled "An Intelligent Brainwave Acquisition and Analysis System and Method" (Patent Publication No.: CN117708682A, Patent Publication Date: 2024-03-15), acquires the brainwave signals of a subject; extracts local neighborhood waveform features of the brainwave signals to obtain a sequence of brainwave local neighborhood waveform-related feature vectors; passes the sequence of brainwave local neighborhood waveform-related feature vectors through a global feature interaction module based on internal self-focus to obtain a sequence of globally enhanced brainwave waveform-related feature vectors; and determines the subject's emotion label based on the sequence of globally enhanced brainwave waveform-related feature vectors. This enables a more accurate description and identification of the subject's emotional state, providing more reliable technical support for the classification and identification of emotional states.

[0004] Based on the description in the above documents, existing EEG signal acquisition and analysis methods differ in their understanding of the actual changes in EEG signals under asynchronous conditions, resulting in the inability to clearly identify curves with large changes. At the same time, the operation of judging abnormal features in EEG signal acquisition is easily affected by interference caused by facial blinking or muscle changes, making the signal analysis operation slower. Therefore, this invention provides an intelligent EEG acquisition and analysis system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent brainwave acquisition and analysis system. This system solves the problems of discrepancies in the actual changes of brainwave signals under asynchronous states during the acquisition and analysis process, resulting in the inability to clearly identify curves with large changes. Furthermore, the system addresses the issue that the judgment of abnormal features during brainwave signal acquisition is easily affected by interference from facial blinking or muscle changes, which slows down the signal analysis process.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent brainwave acquisition and analysis system, comprising:

[0007] The data acquisition module generates category sequences to add user and device information, and uses an electrode array to attach to the user's head to collect brainwave signals. The data is then transmitted and stored via wireless communication technology.

[0008] The data analysis module uses the data preprocessing unit to preprocess, classify, and transform the collected data. The feature extraction and analysis unit extracts the digital signals and constructs analysis curves. The analysis curves are segmented and amplified using timestamps. The activity state of the user's brain is determined by segmenting and comparing it with the thresholds set in different states to determine distinguishing features. The result judgment unit removes interference features and marks abnormal results.

[0009] The anomaly alarm module determines the anomaly risk level based on the timestamp of the anomaly result and performs alarm notification operations based on the anomaly risk level.

[0010] The data update module updates the collected and stored data in real time for any modifications, additions, or deletions of information.

[0011] The data display interface presents the results to the user in the form of charts, curves, and text descriptions.

[0012] Preferably, the data preprocessing unit performs preprocessing, classification, and signal conversion on the collected data as follows:

[0013] First, determine the user information that needs to be tested, then associate and add the corresponding device information based on the user information, and finally extract the EEG signal data collected by the device to realize the association between user information, device information and EEG signal data.

[0014] After acquiring brainwave signals, the analog signals need to be converted into digital signals through pre-amplification, filtering, signal amplification, and analog-to-digital conversion.

[0015] Preferably, the operation of the feature extraction and analysis unit in extracting the digital signal and constructing the analysis curve is as follows:

[0016] Extract the digital signal and the corresponding timestamp at the time of acquisition;

[0017] Using the timestamp of the collected EEG signal as the horizontal axis and the frequency of the EEG signal as the vertical axis, with the starting points of the horizontal and vertical axes intersecting, the frequency values ​​of the corresponding EEG signal under the timestamp are extracted to form an analysis curve.

[0018] Preferably, the feature extraction and analysis unit performs segmented amplification of the analysis curve using timestamps as follows:

[0019] Based on the recorded timestamps, the same time period window is set to segment the analysis curve into multiple waveform segments. The bottom length of the time period window is J, the corresponding timestamp interval is △T, and the bottom length of the time period window is equal to the length of the time period on the horizontal axis, while the side length of the time period window is an integer value higher than the entire analysis curve.

[0020] Extract waveform segments where the rise and fall of the curve cannot be known, set a band magnification window to magnify the extracted waveform segments, and the bottom length of the band magnification window is equal to the length of the time period on the horizontal axis, while the side length of the band magnification window is the vertical distance from the horizontal axis to the peak of the current waveform segment.

[0021] Based on the timestamps in the waveform segment, multiple time nodes are marked. The overall waveform segment is magnified n times using a band magnification window. Then, the source data is collected and interpolation is performed on each adjacent time node to optimize the waveform segment, resulting in a waveform segment with smoother lines.

[0022] Preferably, the steps for performing interpolation on adjacent time nodes are as follows:

[0023] Tracing the timestamps corresponding to the current adjacent time nodes, and extracting data related to the time and EEG digital signals between the timestamps;

[0024] Set the number of data points to be inserted, mark the data between timestamps at equal intervals according to the number of data points, determine the timestamps according to the time division order, insert the corresponding timestamp values, and smoothly connect the data points according to the timestamp order to form a smooth waveform segment.

[0025] Preferably, the feature extraction and analysis unit combines the frequency fluctuations to segment and determine the user's brain activity state, and compares this segment with thresholds set for different states to determine distinguishing features.

[0026] The analysis of each waveform segment is performed step by step. The feature extraction window is set in the time period window. The side length of the feature extraction window is the same as that of the time period window, while the bottom length of the feature extraction window is set to the same value as the data point that is reduced by n times.

[0027] The feature extraction window moves horizontally along the horizontal axis, and the moving distance is the bottom length of the feature extraction window. Feature recognition is performed on the area that the feature extraction window is fixed in each time. The recognized feature content has features with both rising and falling curves. Specifically, the results include the following:

[0028] Result 1: If the amplitude increases first and then decreases, there is a high amplitude point, and the value of the high amplitude point P1 and the corresponding timestamp t1 are recorded.

[0029] Result 2: If the amplitude decreases first and then increases, there is a low amplitude point, and the value of the low amplitude point P2 and the corresponding timestamp t2 are recorded.

[0030] Result 3: If multiple high-amplitude points, multiple low-amplitude points, or both high-amplitude and low-amplitude points are present, then the waveform segments within the current time period exhibit distinct characteristics. Record the values ​​of each low-amplitude or high-amplitude point as Q. m Or R m and the corresponding timestamp T k ;

[0031] The range of the current waveform segment is determined based on the values ​​of the high and low amplitude points in Result 1 and Result 2. Normal EEG signal images under different states are extracted and matched with the range of the waveform segment to obtain the current brain activity state. The difference between the timestamps between amplitude points is determined to be interference features or abnormal features based on the distance between the timestamps on the horizontal axis.

[0032] Preferably, the operation of extracting normal EEG signal maps and matching the range of waveform segments under different states to obtain the current brain activity state is as follows:

[0033] The analysis curves of normal EEG signals were extracted as comparison data, and the comparison data were divided into four states: delta wave [a, b], theta wave [b, c], alpha wave [c, d], and beta wave [d, e]. The brain signal states were gradually excited from delta wave to beta wave, and 0 < a < b < c < d < e.

[0034] The values ​​of the maximum high amplitude point and the minimum low amplitude point of the current waveform segment are recorded as [P1, P2]. The values ​​of [P1, P2] are matched with the comparison data, and the interval in which the value is located corresponds to the current brain activity state.

[0035] Preferably, the operation of removing interference features and marking abnormal results by the result judgment unit is as follows:

[0036] If there are multiple high amplitude points, multiple low amplitude points, or both high amplitude points and low amplitude points in result three, then the current wave segment has distinctive characteristics.

[0037] The interference feature removal operation is as follows: if the current waveform segment is in the θ wave, α wave, or β wave state, then Q... m The waveform segments ∈ [a, b] are interference features. If the current waveform segment is in the delta wave state, and Q m The waveform segment ∈ [a, b] is determined to be a normal waveform segment by the distance between the timestamps of the amplitude points on the horizontal axis;

[0038] Similarly, if the current waveform segment is in the delta, theta, alpha, or beta state, then R m The waveform segment >e is an interference characteristic;

[0039] Furthermore, once identified as a interference feature, the analysis curve of the interference feature is removed;

[0040] The abnormal result marking operation is as follows: after determining the brain activity state corresponding to the current wave segment, determine the low amplitude point Q. m Or high point R m If a value exists that differs from the range of brain activity states, then the timestamp of the abnormal result is marked.

[0041] Preferably, the operation of determining whether a waveform segment is normal based on the distance between the timestamps of the amplitude points on the horizontal axis is as follows:

[0042] The timestamps of adjacent low-amplitude points are extracted sequentially, and the corresponding distances on the horizontal axis are calculated based on the timestamps. The calculation formula is as follows:

[0043] L=(J / △T)×(T k -T k-1 );

[0044] L is the calculated distance on the horizontal axis from the current adjacent low-amplitude point, and T is the distance from the horizontal axis. k-1 The timestamp is located at the adjacent, lower amplitude point on the front side.

[0045] Set the spacing threshold of adjacent low-amplitude points in a normal waveform segment that satisfies the delta wave as g. Compare the calculated distance with the spacing threshold. If L≥g, the current waveform segment is a normal waveform segment; otherwise, L<g, the current waveform segment is an interference feature.

[0046] Preferably, the operation of the anomaly alarm module in determining the anomaly risk level based on the anomaly result tracing timestamp is as follows:

[0047] If an abnormal result occurs in an adjacent brain state when the current brain state is in the delta wave state, a level one warning will be generated and alerted by a green light signal.

[0048] If the current brain state is set to delta wave, and an abnormal result occurs in the interval brain state, a level two warning will be generated and alerted by a yellow light signal.

[0049] If an abnormal result occurs when the current brain state is in the delta wave position, and there is an interval of two brain states between the current brain state and the current brain state, a three-level warning will be generated and a red light signal will be used to alert the user.

[0050] Furthermore, the severity of situations increases from Level 1 to Level 3 warnings.

[0051] This invention provides an intelligent brainwave acquisition and analysis system. Compared with existing technologies, it has the following advantages:

[0052] Beneficial effects:

[0053] 1. This intelligent EEG acquisition and analysis system utilizes a data preprocessing unit to preprocess, classify, and transform the acquired data. A feature extraction and analysis unit extracts digital signals and constructs analysis curves. Timestamps are used to segment and amplify the analysis curves, and frequency fluctuations are combined to determine the user's brain activity state. These fluctuations are then compared with thresholds set for different states to identify distinguishing features. A result judgment unit removes interfering features and marks abnormal results. This effectively improves the efficiency of acquisition and analysis, enhances the accuracy and reliability of EEG signal analysis, and provides timely and effective feedback on the real-time state of brain signals, offering users more precise health management and decision support.

[0054] 2. This intelligent EEG acquisition and analysis system segments the analysis curve into multiple waveform segments using a time period window. A band magnification window is used to magnify the extracted waveform segments. Then, the source-acquired data is interpolated at each adjacent time point to optimize the waveform segments, resulting in smoother lines. This allows for optimization of rapidly changing waveform segments, making the obtained waveform segments more accurate and facilitating subsequent analysis. Simultaneously, it enables deeper mining of EEG signal characteristics and information, ensuring the accuracy of data analysis.

[0055] 3. This intelligent EEG acquisition and analysis system, through the result judgment unit, removes interference features and marks abnormal results. It can judge interference features and perform removal operations based on the amplitude of the curve and the changes in the spacing between amplitudes. At the same time, it can perform abnormal warning operations after judging whether there are abnormal results in the brain state. This not only enables faster monitoring and management operations, but also enables better decision-making for users, ensuring efficient health monitoring and management. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of the data acquisition and analysis system of the present invention. Detailed Implementation

[0057] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Please see Figure 1 This invention provides two technical solutions:

[0059] Example 1: A brainwave intelligent acquisition and analysis system, comprising:

[0060] The data acquisition module generates category sequences to add user and device information, and uses an electrode array to attach to the user's head to collect brainwave signals. The data is then transmitted and stored via wireless communication technology.

[0061] The data analysis module uses the data preprocessing unit to preprocess, classify, and transform the collected data. The feature extraction and analysis unit extracts the digital signals and constructs analysis curves. The analysis curves are segmented and amplified using timestamps. The activity state of the user's brain is determined by segmenting and comparing it with the thresholds set in different states to determine distinguishing features. The result judgment unit removes interference features and marks abnormal results.

[0062] The anomaly alarm module determines the anomaly risk level based on the timestamp of the anomaly result and performs alarm notification operations based on the anomaly risk level.

[0063] The data update module updates the collected and stored data in real time for any modifications, additions, or deletions of information.

[0064] The data display interface presents the results to the user in the form of charts, curves, and text descriptions.

[0065] Furthermore, the data display interface makes it easy for users to view and understand the analysis results, while also providing data storage and query functions to facilitate long-term health monitoring and management.

[0066] By utilizing a data preprocessing unit to preprocess, classify, and transform the collected data, and a feature extraction and analysis unit to extract digital signals and construct analysis curves, the analysis curves are segmented and amplified using timestamps. Furthermore, by combining frequency fluctuations to segment and determine the user's brain activity state, and comparing this segment with thresholds set for different states to identify distinguishing features, and a result judgment unit to remove interfering features and mark abnormal results, the efficiency of data acquisition and analysis is effectively improved. This also enhances the accuracy and reliability of EEG signal analysis, providing timely and effective feedback on the real-time state of brain signals, and offering users more precise health management and decision support.

[0067] In this embodiment of the invention, the data preprocessing unit performs preprocessing, classification, and signal conversion on the collected data as follows:

[0068] First, determine the user information that needs to be tested, then associate and add the corresponding device information based on the user information, and finally extract the EEG signal data collected by the device to realize the association between user information, device information and EEG signal data.

[0069] After acquiring brainwave signals, the analog signals need to be converted into digital signals through pre-amplification, filtering, signal amplification, and analog-to-digital conversion.

[0070] In this embodiment of the invention, the operation of the feature extraction and analysis unit in extracting digital signals and constructing analysis curves is as follows:

[0071] Extract the digital signal and the corresponding timestamp at the time of acquisition;

[0072] Using the timestamp of the collected EEG signal as the horizontal axis and the frequency of the EEG signal as the vertical axis, with the starting points of the horizontal and vertical axes intersecting, the frequency values ​​of the corresponding EEG signal under the timestamp are extracted to form an analysis curve.

[0073] In this embodiment of the invention, the operation of segmenting and amplifying the analysis curve using timestamps by the feature extraction and analysis unit is as follows:

[0074] Based on the recorded timestamps, the same time period window is set to segment the analysis curve into multiple waveform segments. The bottom length of the time period window is J, the corresponding timestamp interval is △T, and the bottom length of the time period window is equal to the length of the time period on the horizontal axis, while the side length of the time period window is an integer value higher than the entire analysis curve.

[0075] Extract waveform segments where the rise and fall of the curve cannot be known, set a band magnification window to magnify the extracted waveform segments, and the bottom length of the band magnification window is equal to the length of the time period on the horizontal axis, while the side length of the band magnification window is the vertical distance from the horizontal axis to the peak of the current waveform segment.

[0076] Based on the timestamps in the waveform segment, multiple time nodes are marked. The overall waveform segment is magnified n times using a band magnification window. Then, the source data is collected and interpolation is performed on each adjacent time node to optimize the waveform segment, resulting in a waveform segment with smoother lines.

[0077] The analysis curve is segmented into multiple waveform segments by using a time period window. A band magnification window is used to magnify the extracted waveform segments. Then, the source data is collected and interpolation is performed on each adjacent time node to optimize the waveform segments, resulting in smoother waveform segments. This allows for optimization of rapidly changing waveform segments, making the obtained waveform segments more accurate and facilitating subsequent analysis. It also enables a deeper exploration of the characteristics and information of EEG signals, ensuring the accuracy of data analysis.

[0078] In this embodiment of the invention, the steps for interpolating adjacent time nodes are as follows:

[0079] Tracing the timestamps corresponding to the current adjacent time nodes, and extracting data related to the time and EEG digital signals between the timestamps;

[0080] Set the number of data points to be inserted, mark the data between timestamps at equal intervals according to the number of data points, determine the timestamps according to the time division order, insert the corresponding timestamp values, and smoothly connect the data points according to the timestamp order to form a smooth waveform segment.

[0081] In this embodiment of the invention, the operation of the feature extraction and analysis unit in determining the user's brain activity state by segmenting based on frequency fluctuations and comparing it with thresholds set for different states to determine distinguishing features is as follows:

[0082] The analysis of each waveform segment is performed step by step. The feature extraction window is set in the time period window. The side length of the feature extraction window is the same as that of the time period window, while the bottom length of the feature extraction window is set to the same value as the data point that is reduced by n times.

[0083] The feature extraction window moves horizontally along the horizontal axis, and the moving distance is the bottom length of the feature extraction window. Feature recognition is performed on the area that the feature extraction window is fixed in each time. The recognized feature content has features with both rising and falling curves. Specifically, the results include the following:

[0084] Result 1: If the amplitude increases first and then decreases, there is a high amplitude point, and the value of the high amplitude point P1 and the corresponding timestamp t1 are recorded.

[0085] Result 2: If the amplitude decreases first and then increases, there is a low amplitude point, and the value of the low amplitude point P2 and the corresponding timestamp t2 are recorded.

[0086] Result 3: If multiple high-amplitude points, multiple low-amplitude points, or both high-amplitude and low-amplitude points are present, then the waveform segments within the current time period exhibit distinct characteristics. Record the values ​​of each low-amplitude or high-amplitude point as Q. m Or R m and the corresponding timestamp T k ;

[0087] The range of the current waveform segment is determined based on the values ​​of the high and low amplitude points in Result 1 and Result 2. Normal EEG signal images under different states are extracted and matched with the range of the waveform segment to obtain the current brain activity state. The difference between the timestamps between amplitude points is determined to be interference features or abnormal features based on the distance between the timestamps on the horizontal axis.

[0088] In this embodiment of the invention, the operation of extracting normal EEG signal images and matching the range of waveform segments under different states to determine the current brain activity state is as follows:

[0089] The analysis curves of normal EEG signals were extracted as comparison data, and the comparison data were divided into four states: delta wave [a, b], theta wave [b, c], alpha wave [c, d], and beta wave [d, e]. The brain signal states were gradually excited from delta wave to beta wave, and 0 < a < b < c < d < e.

[0090] The values ​​of the maximum high amplitude point and the minimum low amplitude point of the current waveform segment are recorded as [P1, P2]. The values ​​of [P1, P2] are matched with the comparison data, and the interval in which the value is located corresponds to the current brain activity state.

[0091] Furthermore, the normal state of the aforementioned electroencephalogram (EEG) signals is shown in Table 1:

[0092] Table 1. Normal Electroencephalogram (EEG) Signals

[0093]

[0094] In this embodiment of the invention, the removal of interference features and the marking of abnormal results by the result judgment unit are as follows:

[0095] If there are multiple high amplitude points, multiple low amplitude points, or both high amplitude points and low amplitude points in result three, then the current wave segment has distinctive characteristics.

[0096] The interference feature removal operation is as follows: if the current waveform segment is in the θ wave, α wave, or β wave state, then Q... m The waveform segments ∈ [a, b] are interference features. If the current waveform segment is in the delta wave state, and Q m The waveform segment ∈ [a, b] is determined to be a normal waveform segment by the distance between the timestamps of the amplitude points on the horizontal axis;

[0097] Similarly, if the current waveform segment is in the delta, theta, alpha, or beta state, then R m The waveform segment >e is an interference characteristic;

[0098] Furthermore, once identified as a interference feature, the analysis curve of the interference feature is removed;

[0099] The abnormal result marking operation is as follows: after determining the brain activity state corresponding to the current wave segment, determine the low amplitude point Q. m Or high point R m If a value exists that differs from the range of brain activity states, then the timestamp of the abnormal result is marked.

[0100] In this embodiment of the invention, the operation of determining whether a waveform segment is a normal waveform segment by the distance between the timestamps of amplitude points on the horizontal axis is as follows:

[0101] The timestamps of adjacent low-amplitude points are extracted sequentially, and the corresponding distances on the horizontal axis are calculated based on the timestamps. The calculation formula is as follows:

[0102] L=(J / △T)×(T k -T k-1 );

[0103] L is the calculated distance on the horizontal axis from the current adjacent low-amplitude point, and T is the distance from the horizontal axis. k-1 The timestamp is located at the adjacent, lower amplitude point on the front side.

[0104] Set the spacing threshold of adjacent low-amplitude points in a normal waveform segment that satisfies the delta wave as g. Compare the calculated distance with the spacing threshold. If L≥g, the current waveform segment is a normal waveform segment; otherwise, L<g, the current waveform segment is an interference feature.

[0105] In this embodiment of the invention, the operation of the anomaly alarm module to determine the anomaly risk level based on the anomaly result tracing timestamp is as follows:

[0106] If an abnormal result occurs in an adjacent brain state when the current brain state is in the delta wave state, a level one warning will be generated and alerted by a green light signal.

[0107] If the current brain state is set to delta wave, and an abnormal result occurs in the interval brain state, a level two warning will be generated and alerted by a yellow light signal.

[0108] If an abnormal result occurs when the current brain state is in the delta wave position, and there is an interval of two brain states between the current brain state and the current brain state, a three-level warning will be generated and a red light signal will be used to alert the user.

[0109] Furthermore, the severity of situations increases from Level 1 to Level 3 warnings.

[0110] By removing interference features and marking abnormal results through the result judgment unit, interference features can be judged and removed based on the amplitude of the curve and the changes in the spacing between amplitudes. At the same time, after judging whether there are abnormal results in the brain state, an abnormal warning operation is performed accordingly. This not only enables faster monitoring and management operations, but also enables better decision-making for users, ensuring efficient health monitoring and management.

[0111] Example 2 differs from Example 1 in that: multiple users are set up to complete real-time data acquisition operations according to the specified requirements of the experiment. This allows for the identification of data anomalies and interference. By implementing the existing intelligent EEG acquisition and analysis system and the intelligent EEG acquisition and analysis system of this invention, the real-time data acquisition from users is practically applied. The time required to complete the analysis and the accuracy of the final judgment are recorded, as shown in Table 2.

[0112] Table 2 Results Record Table

[0113]

[0114] In summary, by using the intelligent EEG acquisition and analysis system of this invention to perform data analysis, the total time required to complete the analysis is shorter, and the accuracy of comparing the obtained results with known results is higher, thus enabling better application in practical operations.

[0115] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0116] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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 process, method, article, or apparatus.

[0117] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A brainwave intelligent acquisition and analysis system, characterized in that: include: The data acquisition module generates category sequences to add user and device information, and uses an electrode array to attach to the user's head to collect brainwave signals. The data is then transmitted and stored via wireless communication technology. The data analysis module uses the data preprocessing unit to preprocess, classify, and transform the collected data. The feature extraction and analysis unit extracts the digital signals and constructs analysis curves. The analysis curves are segmented and amplified using timestamps. The activity state of the user's brain is determined by segmenting and comparing it with the thresholds set in different states to determine distinguishing features. The result judgment unit removes interference features and marks abnormal results. The anomaly alarm module determines the anomaly risk level based on the timestamp of the anomaly result and performs alarm notification operations based on the anomaly risk level. The data update module updates the collected and stored data in real time for any modifications, additions, or deletions of information. The data display interface presents the results to the user in the form of charts, curves, and text descriptions; The feature extraction and analysis unit extracts digital signals and constructs analysis curves as follows: Extract the digital signal and the corresponding timestamp at the time of acquisition; Using the timestamp of the collected EEG signal as the horizontal axis and the frequency of the EEG signal as the vertical axis, with the starting points of the horizontal and vertical axes intersecting, the frequency values ​​of the corresponding EEG signal under the timestamp are extracted to form an analysis curve. The feature extraction and analysis unit performs segmented amplification of the analysis curve using timestamps as follows: Based on the recorded timestamps, the analysis curve is segmented into multiple waveform segments by setting the same time period window. The bottom length of the time period window is J, the corresponding timestamp interval is △T, and the bottom length of the time period window is equal to the length of the time period on the horizontal axis, while the side length of the time period window is an integer value higher than the entire analysis curve. Extract waveform segments where the rise and fall of the curve cannot be known, set a band magnification window to magnify the extracted waveform segments, and the bottom length of the band magnification window is equal to the length of the time period on the horizontal axis, while the side length of the band magnification window is the vertical distance from the horizontal axis to the peak of the current waveform segment. Based on the timestamps in the waveform segment, multiple time nodes are marked. The overall waveform segment is magnified n times using a band magnification window. Then, the source data is collected and interpolation is performed on each adjacent time node to optimize the waveform segment, resulting in a waveform segment with smoother lines. The feature extraction and analysis unit combines frequency fluctuations to segment and determine the user's brain activity state, compares this state with thresholds set for different states, and determines distinguishing features by: The analysis of each waveform segment is performed step by step. The feature extraction window is set in the time period window. The side length of the feature extraction window is the same as that of the time period window, while the bottom length of the feature extraction window is set to the same value as the data point that is reduced by n times. The feature extraction window moves horizontally along the horizontal axis, and the moving distance is the bottom length of the feature extraction window. Feature recognition is performed on the area that the feature extraction window is fixed in each time. The recognized feature content has features with both rising and falling curves. Specifically, the results include the following: Result 1: If the amplitude increases first and then decreases, there is a high amplitude point, and the value of the high amplitude point P1 and the corresponding timestamp t1 are recorded. Result 2: If the amplitude decreases first and then increases, there is a low amplitude point, and the value of the low amplitude point P2 and the corresponding timestamp t2 are recorded. Result 3: If multiple high-amplitude points, multiple low-amplitude points, or both high-amplitude and low-amplitude points are present, then the waveform segments within the current time period exhibit distinct characteristics. Record the values ​​of each low-amplitude or high-amplitude point as Q. m or R m and the corresponding timestamp T k ; The range of the current waveform segment is determined based on the values ​​of the high and low amplitude points in Result 1 and Result 2. Normal EEG signal images under different states are extracted and matched with the range of the waveform segment to obtain the current brain activity state. The difference between the timestamps between amplitude points is determined to be interference features or abnormal features based on the distance between the timestamps on the horizontal axis.

2. The intelligent brainwave acquisition and analysis system according to claim 1, characterized in that: The data preprocessing unit performs the following operations to preprocess, classify, and transform the collected data: First, determine the user information that needs to be tested, then associate and add the corresponding device information based on the user information, and finally extract the EEG signal data collected by the device to realize the association between user information, device information and EEG signal data. After acquiring brainwave signals, the analog signals need to be converted into digital signals through pre-amplification, filtering, signal amplification, and analog-to-digital conversion.

3. The intelligent brainwave acquisition and analysis system according to claim 1, characterized in that: The steps for interpolating adjacent time nodes are as follows: Tracing the timestamps corresponding to the current adjacent time nodes, and extracting data related to the time and EEG digital signals between the timestamps; Set the number of data points to be inserted, mark the data between timestamps at equal intervals according to the number of data points, determine the timestamps according to the time division order, insert the corresponding timestamp values, and smoothly connect the data points according to the timestamp order to form a smooth waveform segment.

4. The intelligent brainwave acquisition and analysis system according to claim 1, characterized in that: The operation of extracting normal EEG signal images and matching the range of waveform segments under different states to determine the current brain activity state is as follows: The analysis curves of normal EEG signals were extracted as comparison data, and the comparison data were divided into four states: delta wave [a, b], theta wave [b, c], alpha wave [c, d], and beta wave [d, e]. The brain signal states were gradually excited from delta wave to beta wave, and 0 < a < b < c < d < e. The values ​​of the maximum high amplitude point and the minimum low amplitude point of the current waveform segment are recorded as [P1, P2]. The values ​​of [P1, P2] are matched with the comparison data, and the interval in which the value is located corresponds to the current brain activity state.

5. The intelligent brainwave acquisition and analysis system according to claim 1, characterized in that: The operation of removing interference features and marking abnormal results by the result judgment unit is as follows: If there are multiple high amplitude points, multiple low amplitude points, or both high amplitude points and low amplitude points in result three, then the current wave segment has distinctive characteristics. The interference feature removal operation is as follows: if the current waveform segment is in the θ wave, α wave, or β wave state, then Q... m The waveform segments ∈ [a, b] are interference features. If the current waveform segment is in the delta wave state, and Q m The waveform segment ∈ [a, b] is determined to be a normal waveform segment by the distance between the timestamps of the amplitude points on the horizontal axis; Similarly, if the current waveform segment is in the delta, theta, alpha, or beta state, then R m The waveform segment >e is an interference characteristic; Furthermore, once identified as a interference feature, the analysis curve of the interference feature is removed. The abnormal result marking operation is as follows: after determining the brain activity state corresponding to the current wave segment, determine the low amplitude point Q. m Or high point R m If a value exists that differs from the range of brain activity states, then the timestamp of the abnormal result is marked.

6. The intelligent brainwave acquisition and analysis system according to claim 5, characterized in that: The operation of determining whether a waveform segment is normal based on the distance between the timestamps on the horizontal axis is as follows: The timestamps of adjacent low-amplitude points are extracted sequentially, and the corresponding distances on the horizontal axis are calculated based on the timestamps. The calculation formula is as follows: L=(J / △T)×(T k -T k-1 ); L is the calculated distance on the horizontal axis from the current adjacent low-amplitude point, and T is the distance from the horizontal axis. k-1 The timestamp is located at the adjacent, lower amplitude point on the front side. Set the spacing threshold of adjacent low-amplitude points in a normal waveform segment that satisfies the delta wave as g. Compare the calculated distance with the spacing threshold. If L≥g, the current waveform segment is a normal waveform segment; otherwise, L<g, the current waveform segment is an interference feature.

7. The intelligent brainwave acquisition and analysis system according to claim 4, characterized in that: The operation of the anomaly alarm module to determine the anomaly risk level based on the anomaly result tracing timestamp is as follows: If an abnormal result occurs in an adjacent brain state when the current brain state is in the delta wave state, a level one warning will be generated and alerted by a green light signal. If the current brain state is set to delta wave, and an abnormal result occurs in the interval brain state, a level two warning will be generated and alerted by a yellow light signal. If an abnormal result occurs when the current brain state is in the delta wave position, and there is an interval of two brain states between the current brain state and the current brain state, a three-level warning will be generated and a red light signal will be used to alert the user. Furthermore, the severity of situations has increased from Level 1 to Level 3 warnings.

Images