Electroencephalogram data compression method based on data arrangement optimization

Abstract

The application provides an electroencephalogram data compression method based on data arrangement optimization, and belongs to the field of electroencephalogram data compression, and is used for solving the problem that the traditional compression method in the related art does not fully utilize the rhythm characteristics and phase synchronism of electroencephalogram data, resulting in limited compression efficiency and easy loss of key physiological characteristics. The method extracts rhythm and analyzes phase synchronism of the collected electroencephalogram data, optimizes data arrangement according to the analysis result, makes the data with strong correlation adjacent, dynamically adjusts the compression parameters to perform compression, and recovers and corrects the phase through index information during decompression. It can improve the compression rate while retaining key characteristics such as neural synchronism, and is suitable for efficient transmission and storage in multiple scenes.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of electroencephalogram data compression, and in particular to an electroencephalogram data compression method based on data arrangement optimization. BACKGROUND

[0002] Electroencephalogram data has the characteristics of high sampling rate and multi-channel, and the data volume is huge, which brings great pressure to transmission and storage. Traditional compression methods mostly directly apply general compression algorithms, without fully utilizing the inherent characteristics of electroencephalogram data such as rhythm characteristics and phase synchronism, resulting in insufficient mining of data correlation, limited compression efficiency, and difficulty in reducing transmission bandwidth and storage demand while ensuring data integrity. SUMMARY

[0003] The present application provides an electroencephalogram data compression method based on data arrangement optimization, which can optimize data arrangement by mining the phase synchronism characteristics of electroencephalogram data and improve compression efficiency.

[0004] In a first aspect, the present application provides an electroencephalogram data compression method based on data arrangement optimization. The collected electroencephalogram data is subjected to correlation analysis, the data is arranged according to the analysis results, the data with strong correlation is distributed adjacent to each other, and then the arranged data is subjected to compression processing.

[0005] By adopting the above technical solution, the data is optimally arranged through correlation analysis of electroencephalogram data, the data with strong correlation is concentrated, the concentration of data redundancy is enhanced, and the efficiency of subsequent compression processing is improved, while reducing the data volume and retaining key information.

[0006] Further, the correlation analysis includes rhythm characteristic extraction of electroencephalogram data, separation of different rhythm components, and analysis of phase synchronism of each rhythm component.

[0007] By adopting the above technical solution, the correlation analysis is more in line with the physiological characteristics of electroencephalogram data by extracting rhythm characteristics and analyzing phase synchronism, and the accuracy of correlation judgment is improved, providing a more reliable basis for data arrangement.

[0008] Further, the analysis of the phase synchronism of each rhythm component is achieved by real-time calculation of the phase difference stability of different leads under the same rhythm through a sliding window, to obtain a phase synchronism index.

[0009] By adopting the above technical solution, the real-time calculation of the sliding window can adapt to the non-stationarity of electroencephalogram data, dynamically capture the phase synchronism changes, and make the phase synchronism index more in line with the real-time characteristics of the data, improving the timeliness of the analysis.

[0010] Further, the length of the sliding window is determined based on the period of the corresponding rhythm component, so that at least two complete rhythm periods are contained in the window.

[0011] By adopting the above technical solution, the window length is determined based on the rhythm period, ensuring that sufficient phase feature samples are contained in the window, improving the accuracy of phase difference stability calculation, and providing reliable data basis for phase synchronization analysis.

[0012] Further, the data is arranged according to the analysis result, that is, the electroencephalogram data is blocked according to the rhythm component, and then the data arrangement order is determined in each block according to the phase synchronization index.

[0013] By adopting the above technical solution, the block is blocked according to the rhythm, avoiding the interference of different rhythm components, and the data redundancy is further concentrated based on the phase synchronization arrangement in the block, so that the compression algorithm is more likely to capture redundant information.

[0014] Further, the data arrangement order in each block is determined according to the phase synchronization index, including calculating the global phase coordination degree of the data with other data and the local time sequence phase continuity degree, and comprehensively obtaining the arrangement priority according to the two degrees, and arranging according to the priority.

[0015] By adopting the above technical solution, the arrangement order is determined by comprehensively considering the global coordination degree and the local continuity degree, which not only ensures the overall correlation of the data, but also avoids the time sequence breakage, and balances the redundancy concentration and data integrity.

[0016] Further, the compressed data is arranged according to the rhythm characteristics of the arranged data, and the parameters of the compression algorithm are dynamically adjusted, and then the adjusted compression algorithm is applied for processing.

[0017] By adopting the above technical solution, the compression algorithm parameters are adapted to the data rhythm characteristics, improving the capture ability of the compression algorithm for the redundant information in the rhythm, and further improving the compression efficiency.

[0018] Further, the parameters of the compression algorithm include the size of the sliding window, and the size and the period of the corresponding rhythm component have a preset proportional relationship.

[0019] By adopting the above technical solution, the sliding window of the compression algorithm is adapted to the rhythm period, which can more accurately capture the rhythm redundancy information, and enhance the pertinence and effectiveness of the compression algorithm.

[0020] Further, after the arranged data is compressed, index information recording the data arrangement rule is generated, and when decompressing, the original arrangement of the data is restored according to the index information, and the phase is corrected.

[0021] By adopting the technical scheme, the integrity of the original structure and phase characteristics of the decompressed data is ensured, the usability of the compressed data is ensured, and the subsequent analysis and application requirements are met.

[0022] Further, the collected electroencephalogram data is from a multi-channel electrode array, the electrodes are distributed according to a standardized spatial layout, and the collection process is in a controllable electromagnetic interference environment.

[0023] By adopting the technical scheme, the quality and spatial distribution of the original electroencephalogram data are ensured, a reliable data foundation is provided for subsequent correlation analysis and arrangement optimization, and the influence of noise on the compression effect is reduced.

[0024] To sum up, the present application at least contains the following beneficial effects:

[0025] 1. A phase synchronization-based electroencephalogram data compression method is provided, which improves the compression efficiency and ensures data integrity;

[0026] 2. The rhythm block and phase synchronization analysis enhance the accuracy of data correlation mining;

[0027] 3. By dynamically adjusting the compression parameters and phase correction, the characteristics of electroencephalogram data are adapted and the quality of decompressed data is guaranteed.

[0028] It should be understood that the content described in the summary section is not intended to limit the key or important features of the embodiments of the present application, nor to limit the scope of the present application. Other features of the present application will become apparent through the following description. BRIEF DESCRIPTION OF DRAWINGS

[0029] The above and other features, advantages, and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description when taken in conjunction with the accompanying drawings, in which like reference characters designate like elements in which:

[0030] Figure 1 An exemplary operating environment in which embodiments of the present application can be implemented is shown.

[0031] Figure 2 A flowchart of an electroencephalogram data compression method based on data arrangement optimization in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0032] To make the purposes, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0033] In addition, the term "and / or" in this document merely describes an association relationship of associated objects, and indicates that there can be three relationships, for example, A and / or B can represent three cases of A existing alone, A and B existing simultaneously, and B existing alone. In addition, the character " / " in this document generally represents an "or" relationship between the front and rear associated objects.

[0034] The present application provides an electroencephalogram data compression method based on data arrangement optimization, which optimizes data arrangement by mining electroencephalogram rhythm and phase synchronization, improves compression efficiency, fits physiological characteristics of electroencephalogram, guarantees data integrity, and facilitates transmission and storage.

[0035] The present application discloses an electroencephalogram data compression method based on data arrangement optimization.

[0036] Figure 1 An exemplary operating environment schematic diagram in which embodiments of the present application can be implemented is shown.

[0037] With reference to Figure 1 The operating environment includes an electroencephalogram acquisition module, which adopts a multi-channel electrode array conforming to the standardized spatial layout of the international 10-20 system, is used for synchronously acquiring original electroencephalogram signals of multiple leads, and provides multi-dimensional data input for subsequent phase synchronization analysis. The operating environment needs to be in an electromagnetic shielding environment, is equipped with anti-interference cables and filtering devices, so as to reduce power frequency interference and equipment electromagnetic radiation, guarantee the signal-to-noise ratio of the original electroencephalogram signals, and avoid interference of noise on phase feature extraction. The operating environment also includes a computing processing unit, which needs to have sufficient computing power, can support real-time operations such as rhythm feature extraction, sliding window phase difference calculation, arrangement priority determination, and compression algorithm parameter adjustment, and ensure the timeliness of data processing. In addition, the operating environment needs to be configured with calibration and quality control devices, such as an electrode impedance calibrator, which is used for calibrating the contact state of the electrodes and the scalp, ensuring stable signal transmission, and providing a reliable original data benchmark for phase synchronization analysis. The devices are connected through special data interfaces, the signals acquired by the electroencephalogram acquisition module are transmitted to the computing processing unit through anti-interference cables, the calibration device is connected with the electroencephalogram acquisition module to realize real-time quality control, and the overall environment supports the whole process implementation of the "electroencephalogram data compression method based on data arrangement optimization" through collaborative work.

[0038] Figure 2 A flowchart of an EEG data compression method based on data arrangement optimization is shown in an embodiment of this application.

[0039] Reference Figure 2 The method specifically includes the following steps:

[0040] S1: Perform correlation analysis on the collected EEG data.

[0041] In this step, the correlation analysis includes extracting rhythmic features from the EEG data, separating different rhythmic components, and then analyzing the phase synchronicity of each rhythmic component. The analysis of the phase synchronicity of each rhythmic component is achieved by calculating the phase difference stability of different leads under the same rhythm in real time through a sliding window to obtain a phase synchronicity index. The length of the sliding window is determined based on the period of the corresponding rhythmic component, so that the window contains at least two complete rhythmic cycles. The acquired EEG data comes from a multi-channel electrode array, the electrode distribution conforms to a standardized spatial layout, and the acquisition process is conducted in an environment with controllable electromagnetic interference.

[0042] Specifically, the collected EEG data are multi-channel time-series signals, assuming the first... One lead (The total number of leads, determined by the number of channels in the electrode array) at time... The total number of sampling points is determined by the sampling frequency. The raw data (determined by the collection time) is sampling frequency To ensure the integrity of the high-frequency rhythm, rhythm feature extraction is achieved using wavelet transform. Wavelet coefficients are obtained by performing continuous wavelet transform. ,in This is a scale parameter (inversely proportional to frequency). The time shift parameter is determined by preset different rhythms. (like Wave 8- Wave The frequency range (etc.) is used to select coefficients of the corresponding scale and then subjected to inverse wavelet transform. Obtain rhythmic components ,Right now ,in For rhythm The corresponding wavelet coefficients.

[0043] Phase synchronization analysis requires first extracting the instantaneous phase using the Hilbert transform, and then analyzing the rhythm components. Perform Hilbert transform to obtain analytic signal ,in For Hilbert transform operators, Imaginary unit, instantaneous phase The range of values ​​is The length of the sliding window By rhythm cycle The frequency (center frequency of the rhythm) is determined by the following formula: ,in For the floor function, ensure that the window contains at least two complete cycles.

[0044] Within each sliding window, the phase lock value (PLV) for different leads is calculated as an indicator of phase synchronization. and The formula for calculating PLV is: ,in For a moment The phase difference is determined by adding or subtracting... Mapped to To avoid overflow, the PLV value range is: The closer the value is to 1, the stronger the phase synchronization. During the acquisition process, the electrode array must conform to the international 10-20 system layout to ensure the standardization of spatial coordinates, and environmental electromagnetic interference must be controlled within a certain signal-to-noise ratio. Within the range, temperature To stabilize electrode impedance and ensure the reliability of raw data.

[0045] S2: Arrange the data according to the analysis results, so that data with strong correlations are distributed adjacently.

[0046] In this step, the method of arranging the data according to the analysis results involves first dividing the EEG data into blocks according to rhythm components, and then determining the data arrangement order within each block based on the phase synchronization index. The method of determining the data arrangement order within each block based on the phase synchronization index includes calculating the global phase coherence degree of the data with other data and its own temporal phase continuity, combining the two to obtain the arrangement priority, and sorting according to the priority.

[0047] When dividing into blocks according to rhythmic components, each rhythm All corresponding lead data were divided into independent data blocks. The block contains all leads for that rhythm. In the full time series EEG data For data containing multiple rhythmic components, the energy proportion of each rhythm can be considered. ( Identify the dominant rhythm block, prioritizing those with an energy percentage higher than a preset threshold (e.g., The dominant rhythm blocks are arranged in a refined manner.

[0048] In each block Within this, the calculation of global phase coherence is based on the phase locking value obtained in step S1. For a certain lead In the window The data, whose global phase coherence is defined as the correlation with all other leads within the block. Average PLV: ,in The total number of leads within the block, with a value range of [value missing]. A higher value indicates stronger overall synergy between the data and other data within the block.

[0049] Self-series phase continuity is used to measure the phase stability of data in the time dimension, for leads. At any moment phase Its difference from the previous moment The phase difference is After mapping (ensuring that in) (Internal), the formula for calculating the temporal phase continuity is: The range of values ​​is A higher value indicates a smoother change in the timing phase.

[0050] The overall priority order is obtained by weighted summation: ,in Weighting coefficients (range of values) Prioritize ensuring global collaboration. Data within a block is processed according to priority. The descending order arrangement ensures that high-priority data are distributed adjacently, forming a compact, redundant, centralized sequence.

[0051] S3: Perform compression processing on the sorted data.

[0052] In this step, the compression processing of the arranged data is performed by dynamically adjusting the parameters of the compression algorithm according to the rhythmic characteristics of the arranged data, and then applying the adjusted compression algorithm for processing; the parameters of the compression algorithm include the sliding window size, which is proportional to the period of the corresponding rhythmic component.

[0053] The sorted data is arranged in rhythm blocks. By splicing them together sequentially, a continuous data stream is formed. ( (Total number of rhythms), where the data within each rhythm block has been prioritized. The data is sorted and centrally distributed with high coherence. A general compression algorithm based on a sliding window (such as LZ77) is chosen, with the sliding window size as its core parameter. This parameter is related to the rhythm cycle The preset proportion relationship is calculated by the following formula , wherein is a preset proportion coefficient (usually 2, consistent with the window setting in step S1), is a sampling frequency (unit: Hz), and is ensured to be an integer. The unit of is the number of sampling points, and

[0054] complete rhythm cycles are contained in the window. Taking the LZ77 algorithm as an example, the adjusted sliding window can more accurately capture the repeated sequences formed by the periodicity of the rhythm in the arranged data: in the window, the algorithm compares the similarity of the current data and the historical data, and replaces the original data with a “distance-length” pointer for the repeated sequence. Since the arranged data has concentrated the data with high phase synchronism, the repeated patterns (such as the periodic fluctuations of the wave) in the same rhythm cycle are more easily identified in the window, and the coding efficiency is significantly improved. For multi-rhythm data, different rhythm blocks can be dynamically switched according to the wave period (for example, the wave period is shorter), and adaptive compression is realized by inserting parameter switching markers in the compressed data stream.

[0055] During compression, the pointers and non-repeated data output by the LZ77 algorithm can also be secondarily encoded in combination with entropy encoding (such as Huffman encoding), and the construction of the encoding table is based on the probability distribution of the arranged data (the numerical distribution of high-priority data is more concentrated, and the entropy value is lower). Through the above parameter adjustment and algorithm adaptation, the compression process can fully utilize the redundancy structure of the arranged data to reduce the reconstruction error under the same compression rate, or to improve the compression efficiency under the same error constraint.

[0056] S4: After performing compression processing on the arranged data, index information recording the data arrangement rule is generated, and in decompression, the original arrangement of the data is restored according to the index information, and the phase is corrected.

[0057] The method of this step specifically includes: generating index information recording the data arrangement rule, and in decompression, the original arrangement of the data is restored according to the index information, and the phase is corrected.

[0058] The generated index information contains multi-dimensional arrangement rule records, specifically including: the identification of each rhythm block (such as the rhythm type , energy proportion ), the arrangement order of the leads in the block (recorded from the original lead number mapping relationship of the post-arrangement serial number, partition parameters of the time window (such as window length , starting time ), and calculation parameters of the arrangement priority (such as weight coefficient ). The index information is stored in a structured table form, and each entry corresponds to the arrangement rule of a data block, ensuring accurate positioning of the original space-time coordinates of each data block during decompression.

[0059] In the decompression stage, first, according to the rhythm block identifier and lead mapping relationship in the index information, the compressed data is rearranged according to the original rhythm block and lead order, and restored to the multi-channel time sequence structure before block division. At this time, due to the division of the time window and the adjustment of the lead order in the arrangement process, the phase of the data may have a space-time offset, and phase correction is needed: for the time dimension, the offset between the post-arrangement time and the original time is calculated, and the phase is adjusted according to the rhythm period , and the correction formula is , to ensure the continuity of the phase with time; for the spatial dimension, for the phase difference offset caused by lead arrangement, according to the historical record of the original phase lock value , the correction coefficient is calculated (wherein is the original phase difference, and is the post-arrangement phase difference), and the phase relationship between leads is corrected through to restore the original spatial synchronicity.

[0060] After correction, the consistency of the phase characteristics of the restored data with the original data needs to be verified, which can be done by calculating the deviation (such as mean square error MSE ) between the corrected phase lock value and the original , and ensuring that the deviation is lower than a preset threshold (such as 0.05), to ensure the integrity of the physiological characteristics of the decompressed data.

[0061] Through multi-channel standardized acquisition and electromagnetic shielding environment control, the interference of noise on the electroencephalogram data can be reduced, the signal-to-noise ratio and spatial layout specification of the original signal are ensured, and reliable data basis is provided for subsequent analysis; rhythm characteristics are extracted and phase synchronization (PLV) is calculated based on a sliding window (adapted to the rhythm period), which is more in line with the essence of phase coordination of neural activity than traditional amplitude analysis, can accurately capture the internal correlation of different leads and time points, and the sliding window can dynamically adapt to the non-stationarity of electroencephalogram, and real-time tracking of synchronization changes; sorting according to rhythm blocks combined with global phase coordination and local time continuity can avoid the redundant interference of different rhythms, arrange high synchronization data closely, and strengthen the redundancy concentration; dynamically adjusting the compression algorithm parameters (such as the sliding window and the rhythm period are proportional), can make the compression algorithm more accurately identify the rhythmic repetitive pattern, and improve the redundancy capture efficiency; generating index information and decompression phase correction can restore the original spatio-temporal structure of the data and correct the phase offset caused by the arrangement, and ensure the data integrity.

[0062] In summary, from the reliability guarantee of data acquisition, to the physiological fitting of correlation analysis, to the redundancy concentration of arrangement optimization, finally through compression adaptation and decompression correction to balance efficiency and integrity, the overall technical means can be derived as follows: the method can improve the compression rate of electroencephalogram data while retaining key physiological characteristics (such as neural synchronization and rhythm periodicity), and has stronger adaptability to non-stationary electroencephalogram signals, and is suitable for efficient transmission and storage in multiple scenarios such as clinical monitoring and wearable devices.

[0063] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the action order described, because according to the embodiments of the present application, certain steps can be performed in other order or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0064] In summary, the present application at least contains the following beneficial effects:

[0065] 1. By mining the phase synchronization and rhythm characteristics of electroencephalogram data to optimize data arrangement, the capture efficiency of the compression algorithm for neural activity coordination redundancy is improved, key physiological characteristics are retained while the compression rate is improved;

[0066] 2. Dynamic sliding window and block arrangement strategy are used to adapt to the non-stationarity of electroencephalogram signals (such as the switching of resting state and task state), which enhances the adaptability of the method to electroencephalogram data in different scenarios;

[0067] 3. The combination of index information and phase correction mechanism can accurately restore the original space-time structure and phase relationship of the data in the decompression process, guarantee the integrity and availability of the compressed data, and meet the needs of clinical analysis and scientific research.

[0068] The above description is merely the preferred embodiments of the present application and the principle of the applied technology. It should be understood by those skilled in the art that the disclosed range in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combinations of the above technical features or equivalent features without departing from the disclosed concept. For example, the above features can be replaced with the technical features disclosed in the present application (but not limited to) having similar functions to form technical solutions.

Claims

1. An electroencephalogram data compression method based on data arrangement optimization, characterized in that, The collected electroencephalogram data is subjected to correlation analysis, and the data is arranged according to the analysis result, so that the data with strong correlation is distributed adjacently, and then the arranged data is subjected to compression processing; The correlation analysis includes rhythm feature extraction of the electroencephalogram data, separation of different rhythm components, and analysis of phase synchronization of each rhythm component; The data is arranged according to the analysis result, which is to block the electroencephalogram data according to rhythm components, and then determine the data arrangement order in each block according to the phase synchronization index; The data arrangement order in each block is determined according to the phase synchronization index, which includes calculating the global phase coordination degree of the data with other data and the self time sequence phase continuity degree, comprehensively obtaining the arrangement priority according to the two, and sorting according to the priority; When the rhythm components are divided into blocks, each rhythm Corresponding to all lead data is divided into independent data blocks , which contains all leads under the rhythm In the full time sequence Electroencephalogram data ; for data with multiple rhythm components, the dominant rhythm block can be determined according to the energy proportion of each rhythm ( ) The dominant rhythm block with energy proportion higher than the preset threshold is preferentially arranged in detail; In each block , the calculation of global phase coherence is based on the pre-acquired phase-locking values , for a certain lead , in the window , the global phase coherence of the data is defined as the mean of the PLV of the data with all other leads in the block : , where is the total number of leads in the block, and the value range is , the higher the value, the stronger the overall coherence of the data with other data in the block. The self-timing phase continuity is used to measure the phase stability of data in time dimension, and is used for lead At time point The phase of the current time point is different from the phase of the previous time point by After mapping processing (ensuring within ), the calculation formula of the timing phase continuity is , the value range is , and the higher the value is, the more gentle the timing phase change is. The comprehensive priority is obtained by weighted summation: wherein is a weight coefficient, and the global cooperativity is preferentially guaranteed; the data in the block are arranged in descending order of the priority , so that the high-priority data are distributed adjacently to form a compact redundant concentrated sequence.

2. The method for electroencephalogram data compression based on data permutation optimization according to claim 1, characterized in that, The analysis of the phase synchronization of each rhythm component is to calculate the phase difference stability of different leads under the same rhythm through a sliding window in real time, and obtain the phase synchronization index.

3. The method for electroencephalogram data compression based on data permutation optimization according to claim 2, characterized in that, The length of the sliding window is determined based on the period of the corresponding rhythm component, so that at least two complete rhythm periods are contained in the window.

4. The method for electroencephalogram data compression based on data permutation optimization according to claim 1, characterized in that, The arranged data is subjected to compression processing, which is to dynamically adjust the parameters of the compression algorithm according to the rhythm characteristics of the arranged data, and then apply the adjusted compression algorithm for processing.

5. The method for electroencephalogram data compression based on data permutation optimization according to claim 4, characterized in that, The parameters of the compression algorithm include the size of the sliding window, which has a preset proportional relationship with the period of the corresponding rhythm component.

6. The method for electroencephalography data compression based on data permutation optimization according to claim 1, characterized in that, After the arranged data is subjected to compression processing, index information recording the data arrangement rule is generated, and when decompression is performed, the original arrangement of the data is restored according to the index information, and the phase is corrected.

7. The method for electroencephalography data compression based on data permutation optimization according to claim 1, characterized in that, The collected electroencephalogram data comes from a multi-channel electrode array, the electrodes are distributed according to a standardized spatial layout, and the collection process is in a controllable electromagnetic interference environment.

Images