A system and device for determining the probability of autism pattern matching based on EEG features

Abstract

This invention discloses a system and apparatus for determining the probability of autism pattern matching based on electroencephalogram (EEG) features. The system includes: an EEG acquisition module, a data segmentation module, a data determination module, and a probability determination module. The EEG acquisition module acquires n EEG datasets corresponding to n brain regions of the target child within a preset time period. The data segmentation module divides the preset time period into k partitioned time periods and determines the EEG data of each of the n EEG datasets within the k partitioned time periods, resulting in n×k partitioned EEG datasets. The data determination module determines the rhythmic synchronization value and the high-low frequency modulation correlation value of the target child's brain regions corresponding to the same frequency. The probability determination module determines the matching probability between the target child's EEG features and autism patterns, obtaining the target matching probability. Using the implementation method of this application can improve the efficiency of autism assessment.

Description

Technical Field

[0001] This invention relates to the field of electroencephalogram (EEG) data analysis technology, and in particular to a system and device for determining the probability of autism pattern matching based on EEG features. Background Technology

[0002] Autism is a neurodevelopmental disorder that begins in infancy or early childhood. Its core symptoms include impaired social communication, restricted interests, and repetitive, stereotyped behaviors. Current traditional autism assessments largely rely on behavioral observation scales, where professionals subjectively rate children's behavior. These systems are heavily influenced by the assessor's experience and subjective judgment, and the assessment process is time-consuming, making it difficult to meet the needs of large-scale early screening. Therefore, improving the efficiency of autism assessment is a pressing issue. Summary of the Invention

[0003] This application provides a system and device for determining the probability of autism pattern matching based on electroencephalogram (EEG) features, which can improve the efficiency of autism assessment.

[0004] In a first aspect, embodiments of this application provide a probability determination system for autism pattern matching based on electroencephalogram (EEG) features, the system comprising: an EEG acquisition module, a data segmentation module, a data determination module, and a probability determination module, wherein:

[0005] The EEG acquisition module is used to acquire n EEG datasets corresponding to n brain regions of the target child's brain within a preset time period; each EEG dataset corresponds to one brain region, and n is a positive integer;

[0006] The data partitioning module is used to divide the preset time period into k partition time periods; k is a positive integer; the k partition time periods are k consecutive and independent partition time periods;

[0007] Determine the EEG data of each of the n EEG datasets within the k partition time periods to obtain n×k partition EEG data;

[0008] The data determination module is used to determine, based on the n×k partitioned EEG data, the brain region rhythm synchronization value data and the high-low frequency regulation correlation value data of the target child's brain corresponding to the same frequency; the brain region rhythm synchronization value data is used to measure the activity synchronicity of different brain regions on EEG rhythms at the same frequency, and the high-low frequency regulation correlation value data of the same brain region is used to measure the strength of the regulatory relationship between EEG rhythms of different frequencies within the same brain region;

[0009] The probability determination module is used to determine the matching probability between the target child's EEG characteristics and the autism pattern based on the brain region rhythm synchronization value data and the brain region high and low frequency regulation correlation value data, thereby obtaining the target matching probability; the target matching probability is used to assess the risk of the target child having autism.

[0010] In a second aspect, embodiments of this application provide an autism pattern matching probability determination device based on electroencephalogram (EEG) features, the device comprising the system described in the first aspect.

[0011] Implementing this application will have the following beneficial effects:

[0012] As can be seen, the autism assessment system based on a large EEG model described in this application first obtains n EEG datasets corresponding to n brain regions of the target child within a preset time period. Then, the preset time period is divided into k partition time periods, and the EEG data of each of the n EEG datasets within the k partition time periods is determined, resulting in n×k partition EEG data. Then, based on the n×k partition EEG data, the system determines the rhythm synchronization value data of the brain regions with the same frequency and the correlation value data of high and low frequency regulation of the brain regions corresponding to the target child's brain. Finally, based on the rhythm synchronization value data of the brain regions with the same frequency and the correlation value data of high and low frequency regulation of the brain regions, the system determines the matching probability between the target child's EEG characteristics and the autism pattern, thus obtaining the target matching probability used to assess the risk of the target child having autism, effectively improving the efficiency of autism assessment. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the background art, the accompanying drawings used in the embodiments of this application or the background art will be described below.

[0014] Figure 1 This is a schematic diagram of the structure of an autism pattern matching probability determination system based on EEG features provided in an embodiment of this application;

[0015] Figure 2 This is a flowchart of a system for determining the target matching probability through an autism pattern matching probability determination system based on EEG features, provided in an embodiment of this application.

[0016] Figure 3 This is a flowchart provided in an embodiment of the present application, showing how a data determination module determines the degree of rhythm synchronization of brain regions at the same frequency and the degree of correlation between high and low frequency regulation of brain regions.

[0017] Figure 4This is a flowchart provided in an embodiment of the present application for determining the degree of rhythm synchronization of brain regions at the same frequency through a data determination module;

[0018] Figure 5 This is a flowchart provided in an embodiment of the present application for determining the correlation value between high and low frequency regulation in the same brain region through a data determination module;

[0019] Figure 6 This is a flowchart illustrating how a probability determination module determines the target matching probability, as provided in an embodiment of this application.

[0020] Figure 7 This is a schematic diagram of the structure of an auxiliary evaluation model provided in an embodiment of this application;

[0021] Figure 8 This is a schematic diagram of a device for determining the probability of autism pattern matching based on electroencephalogram (EEG) features, provided in an embodiment of this application. Detailed Implementation

[0022] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present application.

[0023] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0024] It should be understood that the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document indicates that the preceding and following related objects are in an "or" relationship. In the embodiments of this application, "multiple" refers to two or more.

[0025] In the embodiments of this application, "at least one item" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. "One or more" refers to one or more items, while "multiple" refers to two or more items. For example, "at least one item" of a, b, or c can represent the following seven cases: a, b, c; a and b; a and c; b and c; a, b, and c. Each of a, b, and c can be an element or a set containing one or more elements.

[0026] The term "connection" in the embodiments of this application refers to various connection methods, such as direct connection or indirect connection, to achieve communication between devices. The embodiments of this application do not impose any limitations on this.

[0027] In this document, the term "implementation" means that a specific feature, structure, or characteristic described in connection with an implementation may be included in at least one implementation of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same implementation, nor is it a separate or alternative implementation mutually exclusive with other implementations. It will be explicitly and implicitly understood by those skilled in the art that the implementations described herein can be combined with other implementations.

[0028] The following describes the relevant content, concepts, meanings, technical problems, technical solutions, and beneficial effects involved in the implementation methods of this application.

[0029] First, let me explain some of the technical terms or phrases used in this application:

[0030] Electroencephalogram (EEG) characteristics: quantitative indicators that reflect the state of brain neural activity, extracted from EEG data of different brain regions of the target child through specific data processing and analysis methods.

[0031] Autism Pattern: Based on extensive research on EEG signals and neurodevelopment of children diagnosed with autism, this set of common and regular abnormal EEG activity features is a standard reference template for determining whether a child's EEG has autism-related abnormalities.

[0032] Please see Figure 1 , Figure 1 This is a schematic diagram of a probability determination system for autism pattern matching based on electroencephalogram (EEG) features, provided in an embodiment of this application. The probability determination system 100 for autism pattern matching based on EEG features includes an EEG acquisition module 101, a data segmentation module 102, a data determination module 103, and a probability determination module 104.

[0033] In this embodiment, the EEG acquisition module 101 is used to acquire n EEG datasets corresponding to n brain regions of the target child's brain within a preset time period, wherein each EEG dataset corresponds to one brain region, and n is a positive integer; the data partitioning module 102 is used to divide the preset time period into k partition time periods, and is also used to determine the EEG data of each of the n EEG datasets within the k partition time periods, to obtain n×k partition EEG data, wherein k is a positive integer, and the k partition time periods are k consecutive and independent partition time periods; the data determination module 103 is used to determine the corresponding frequency of the target child's brain based on the n×k partition EEG data. The system includes brain region rhythm synchronization data and high- and low-frequency modulation correlation data within the same brain region. The high- and low-frequency modulation correlation data measures the synchronicity of activity in different brain regions at the same frequency, while the high- and low-frequency modulation correlation data measures the strength of the regulatory relationship between brain rhythms of different frequencies within the same brain region. A probability determination module 104 determines the matching probability between the target child's EEG characteristics and an autism pattern based on the high- and low-frequency modulation correlation data, thus obtaining a target matching probability. This target matching probability is used to assess the risk of the target child having autism.

[0034] As can be seen, by first collecting complete EEG datasets from n brain regions and dividing them into k consecutive independent time segments, n×k EEG data segments are obtained. This ensures the comprehensiveness of the EEG data and refines the capture of features in the time dimension. Then, based on these data, the rhythm synchronization value of brain regions with the same frequency and the correlation value of high and low frequency regulation in the same brain region are extracted. At the same time, it covers two key EEG feature dimensions: cross-brain region synchronization and intra-brain region regulation, avoiding the one-sidedness of single feature analysis. Finally, the probability determination module integrates the two types of feature data to calculate the target matching probability. This ensures the standardization of data processing and the targeting of feature extraction, and can greatly improve the scientificity, reliability and accuracy of risk assessment through multi-dimensional feature fusion.

[0035] Please see Figure 2 , Figure 2 This is a flowchart illustrating a system for determining the probability of a target match based on an autism pattern matching probability determination system using EEG features, as provided in this application, including but not limited to the following steps:

[0036] S201: Obtain n EEG datasets corresponding to n brain regions of the target child within a preset time period through the EEG acquisition module.

[0037] In this embodiment, each EEG dataset corresponds to a brain region, where n is a positive integer. The EEG acquisition module is the core component of the autism pattern matching probability determination system based on EEG features, responsible for acquiring raw EEG data. The EEG acquisition module includes hardware components such as EEG signal sensors, electrodes, data transmission lines, signal amplification and filtering equipment, and also integrates basic data acquisition control logic. The core purpose of the EEG acquisition module is to accurately acquire EEG data for target children. Specifically, it selects n different brain regions in the target child's brain as EEG data acquisition points, continuously capturing the electrophysiological activity signals generated by these brain regions within a pre-set continuous time period. Each brain region corresponds to a complete and continuous EEG dataset. Through this multi-brain-region, multi-time-period acquisition method, the electrical activity of multiple brain regions of the target child can be comprehensively and realistically obtained.

[0038] In this embodiment, the EEG acquisition module can select n different specific brain regions on the target child's brain as the target areas for data acquisition, and set a pre-planned preset time period as the acquisition cycle. The pre-planned continuous time period can be set by the operator or automatically selected by the autism pattern matching probability determination system based on EEG characteristics. The specific method for determining the preset time period is not limited here. Within this preset acquisition cycle, the EEG acquisition module will continuously capture the EEG activity signals generated by each selected brain region, and independently record and store the signals of each brain region, ultimately forming n complete EEG datasets that correspond one-to-one with the n brain regions. Each dataset fully reflects the electrophysiological activity of the corresponding brain region throughout the preset time period.

[0039] S202: Divide the preset time period into k partition time periods using the data partitioning module.

[0040] In this embodiment, k is a positive integer, and the k partitioned time periods are k consecutive and independent partitioned time periods. The data partitioning module is a key component responsible for time dimension data processing in the autism pattern matching probability determination system based on EEG features. It mainly includes a time segmentation rule setting unit, a time period splitting execution unit, and a partitioned time identification unit. Its core purpose is to standardize and split the preset time period based on the EEG acquisition module. It can split the complete and continuous preset time period into k independent and sequential partitioned time periods according to the system's preset time division logic. At the same time, it assigns a clear time range identifier to each partitioned time period. Then, from the previously collected n EEG datasets, it extracts the corresponding EEG signal part of each dataset within these k partitioned time periods, finally forming n×k partitioned EEG data, allowing the originally complete EEG data to be refined and split according to the time dimension.

[0041] The data segmentation module will divide this continuous and complete time period into k independent time segments according to the preset time period set by the EEG acquisition module. These time segments are connected to each other and do not overlap in time. Each time segment has a clear start and end time. Through this division method, the original complete preset time period is broken down into multiple continuous and independent time segments.

[0042] S203: The data partitioning module determines the EEG data of each of the n EEG datasets within the k partition time periods, resulting in n×k partition EEG data.

[0043] In this embodiment, the data partitioning module first identifies the n EEG datasets acquired by the EEG acquisition module and the k partitioned time periods it has already divided. It then processes each of the n EEG datasets one by one, accurately extracting the corresponding EEG signal segments for each of the n EEG datasets within the k partitioned time periods. In other words, each EEG dataset is divided into k EEG data parts corresponding to different partitioned time periods. Finally, through this partitioning method, the original n complete EEG datasets are divided into n×k partitioned EEG data, which correspond to both different brain regions and different time segments.

[0044] As can be seen, by dividing the preset time period into k consecutive and independent partition time periods through the data segmentation module, and then extracting the EEG data of each EEG dataset within each partition time period to form n×k partition EEG data, it is possible to achieve a detailed segmentation of EEG data in the time dimension, allowing the original complete EEG signal to be accurately divided into time segments, while also preserving the correlation between each brain region and the corresponding time segment. This avoids the blurring of time dimension features in complete long-term EEG data and provides a data foundation for subsequent calculation of the synchronization degree value of brain regions with the same frequency and the correlation degree value of high and low frequency regulation of brain regions by partition time period, thereby improving the pertinence and accuracy of subsequent feature extraction and probability calculation.

[0045] S204: The data determination module determines the brain rhythm synchronization value and high-low frequency regulation correlation value of the target child's brain corresponding to the same frequency brain region based on the n×k partition EEG data.

[0046] In this embodiment, the brain region rhythm synchronization value data of the same frequency is used to measure the activity synchronicity of different brain regions on EEG rhythms at the same frequency, and the brain region high and low frequency regulation correlation value data is used to measure the strength of the regulatory relationship between EEG rhythms of different frequencies within the same brain region.

[0047] The data determination module is the core component of the autism pattern matching probability determination system based on EEG features, responsible for EEG feature extraction and analysis. It mainly includes an EEG data preprocessing unit, a brain region synchronicity calculation unit, a brain region frequency modulation correlation calculation unit, and a feature data integration unit. Its core purpose is to perform in-depth processing and feature mining on the n×k partitioned EEG data output by the data segmentation module. It can first perform preprocessing operations such as noise reduction and standardization on these subdivided EEG data, and then use preset algorithms to calculate the synchronicity of EEG rhythmic activities in different brain regions at the same frequency, as well as the intensity of the modulation relationship between EEG rhythms of different frequencies within the same brain region, thereby generating corresponding brain region rhythm synchronicity value data and brain region high and low frequency modulation correlation value data.

[0048] S205: The probability determination module determines the matching probability between the target child's EEG characteristics and the autism pattern based on the synchronous frequency value data of the brain region with the same frequency and the correlation value data of high and low frequency regulation of the same brain region, and obtains the target matching probability.

[0049] In this embodiment, the target matching probability is used to assess the risk of the target child having autism. The probability determination module is the core component responsible for risk quantification assessment in the autism pattern matching probability determination system based on EEG features. The probability determination module mainly includes an autism pattern feature library, a feature matching algorithm unit, a probability calculation unit, and a result output unit. Its core purpose is to combine two types of key feature data generated by the data determination module to conduct autism risk assessment. Specifically, it first calls the system's built-in autism pattern feature library, which contains standard EEG feature data related to known autism. Then, through a preset matching algorithm, it compares and analyzes the target child's brain region rhythm synchronization value data and brain region high-low frequency modulation correlation value data with the standard data in the autism pattern feature library in multiple dimensions. Subsequently, the probability calculation unit converts the comparison results into specific numerical probabilities, i.e., the target matching probability. Finally, the result output unit presents the target matching probability, providing a basis for assessing the risk of the target child having autism.

[0050] As can be seen, the probability determination module calls the built-in autism pattern feature library to compare the target child's two key EEG feature data—the synchronization degree of brain regions with the same frequency and the correlation degree of high and low frequency regulation in the same brain regions—with autism-related standard data in the library in multiple dimensions. The comparison results are then converted into a quantitative target matching probability. This not only ensures the consistency of the comparison by relying on the standardized feature library, but also avoids the one-sidedness of risk assessment based on a single feature through the comprehensive analysis of the two core features: the synchronization degree of brain regions with the same frequency and the correlation degree of high and low frequency regulation in the same brain regions.

[0051] It should be explained that when assessing the risk of a child with autism using the target matching probability, a preset risk threshold can be used in conjunction with the target matching probability to assess the risk. The target matching probability is a quantitative result derived from the degree of fit between two core features—EEG synchronicity and high- and low-frequency modulation correlation—and the autism pattern. The higher the value, the stronger the match between the target child's EEG characteristics and the autism pattern. A risk threshold validated by a large amount of clinical data can be set. If the target matching probability is higher than the threshold, it indicates that the child has a higher risk of autism, and further professional clinical evaluation should be recommended. If the probability is lower than the threshold, it indicates that the risk is relatively low, and regular observation and follow-up can be carried out in conjunction with the child's daily behavior. If the probability is close to the threshold, a careful judgment should be made based on a comprehensive consideration of EEG feature details, the child's development, and other information. If necessary, repeated testing or the professional intervention evaluation process should be initiated. The entire process should be based on a professional clinical diagnosis as the final basis, and the target matching probability is only used as a reference for auxiliary risk screening.

[0052] As can be seen, the autism pattern matching probability determination system based on EEG features, through the coordinated operation of several modules—EEG acquisition, data segmentation, data determination, and probability determination—can comprehensively collect EEG data from multiple brain regions and time periods. After refined segmentation and targeted feature extraction, the matching probability is finally quantitatively calculated. This system is based on objective EEG physiological data, avoiding potential biases from subjective assessments. Furthermore, by capturing two key features—synchronicity across different brain regions and frequency modulation correlation within the same brain region—it can accurately identify EEG patterns related to autism, making risk assessment more scientific and targeted. The multi-brain-region, multi-time-period approach comprehensively reflects the true state of children's EEG activity, reducing the bias caused by single data or time periods. The final determined matching probability provides an intuitive and reliable reference for autism risk assessment in children, helping relevant personnel to identify potential risks earlier and providing strong support for subsequent intervention or diagnosis, thereby improving the efficiency and accuracy of autism assessment.

[0053] Please see Figure 3 , Figure 3 This application provides a flowchart illustrating how a data determination module determines the degree of rhythmic synchronization in brain regions of the same frequency and the degree of correlation between high and low frequency regulation in the same brain regions, including but not limited to the following steps:

[0054] S301: Determine the EEG data under p frequency bands corresponding to the n×k partitioned EEG data to obtain n×k×p partitioned frequency EEG data.

[0055] In this implementation, p is a positive integer, and one frequency band corresponds to one frequency range. The data determination module first performs frequency division processing on the acquired n×k partitioned EEG data to identify the p different frequency bands corresponding to these data. Each frequency band corresponds to a specific frequency range. Through frequency division, the EEG signals corresponding to each frequency band are extracted from each partitioned EEG data, ultimately obtaining n×k×p partitioned frequency-divided EEG data, which provides the basic data for subsequent calculation of the brain region rhythm synchronization degree value at the same frequency, divided by frequency band.

[0056] S302: Determine the rhythm synchronization value of p brain regions with the same frequency based on the n×k×p segmented frequency EEG data.

[0057] In this embodiment, each frequency band corresponds to a brain region rhythm synchronization value at the same frequency. The data determination module uses n×k×p segmented frequency EEG data as the basis for calculation, and analyzes each of the p frequency bands separately. It can calculate the activity synchronicity of different brain regions at the corresponding frequency EEG rhythms within each frequency band using a preset algorithm. Each frequency band corresponds to a specific value reflecting the degree of synchronization between brain regions within that frequency range, ultimately obtaining p brain region rhythm synchronization values ​​at the same frequency, thereby quantifying the activity synchronicity of different brain regions at the same frequency EEG rhythms.

[0058] S303: Determine the rhythm synchronization degree data of the same frequency brain regions based on the p same frequency brain region rhythm synchronization degree values.

[0059] In this embodiment, the data determination module integrates the obtained p brain region rhythm synchronization values ​​of the same frequency to determine the brain region rhythm synchronization value data of the same frequency.

[0060] S304: Determine the EEG data of q high- and low-frequency pairs corresponding to the n×k partition EEG data to obtain n×k×q partition high-frequency EEG data and n×k×q partition low-frequency EEG data.

[0061] In this embodiment, q is a positive integer, and each high-low frequency pair includes a low-frequency band and a high-frequency band. q high-low frequency pairs are pre-defined, each pair containing a low-frequency band and a corresponding high-frequency band. Then, the EEG signals corresponding to the low-frequency band and the high-frequency band of each high-low frequency pair are extracted from the EEG data of each region, ultimately yielding n×k×q region high-frequency EEG data and n×k×q region low-frequency EEG data.

[0062] S305: Determine the correlation values ​​of high and low frequency regulation in the same brain region based on the n×k×q high-frequency EEG data and the n×k×q low-frequency EEG data.

[0063] In this embodiment, each high-low frequency pair corresponds to a correlation value for high-low frequency regulation within the same brain region. The data determination module analyzes n×k×q high-frequency EEG data and n×k×q low-frequency EEG data, calculating for each of the q high-low frequency pairs. A preset algorithm analyzes the strength of the regulatory relationship between the low-frequency and high-frequency EEG rhythms within each pair, generating a specific quantified value for each pair. Ultimately, the correlation values ​​for high-low frequency regulation within the same brain region are determined.

[0064] S306: Determine the high-low frequency regulation correlation value data of the same brain region based on the q high-low frequency regulation correlation values ​​of the same brain region.

[0065] In this embodiment, the data determination module integrates the obtained high- and low-frequency regulation correlation values ​​of the q brain regions to determine the high- and low-frequency regulation correlation value data of the brain regions.

[0066] As can be seen, by first splitting the regional EEG data into p frequency bands to obtain regional frequency EEG data, and then calculating the rhythm synchronization value of the brain region at the same frequency corresponding to each frequency band and integrating it into the corresponding data, and by splitting it into q high-low frequency pairs to obtain regional high-low frequency EEG data, and calculating the correlation value of high-low frequency regulation of the same brain region corresponding to each high-low frequency pair and integrating it into the corresponding data, the EEG data can be finely decomposed and analyzed in a targeted manner from the frequency dimension. The division into p frequency bands allows for accurate quantification of brain region synchronization at different frequencies, avoiding the limitations of single frequency analysis. The setting of q high-low frequency pairs can comprehensively capture the regulatory relationship of different frequency rhythms within the same brain region, making the extraction of EEG features more comprehensive and in-depth. The separate calculation and integration of the two types of data can fully present the core EEG features of the target child from different dimensions, making the subsequent matching with autism patterns more targeted and accurate.

[0067] Please see Figure 4 , Figure 4 This application provides a flowchart of a method for determining the degree of rhythm synchronization of brain regions at the same frequency using a data determination module, including but not limited to the following steps:

[0068] S401: Determine the n×k sub-band frequency EEG data corresponding to the first frequency band.

[0069] In this embodiment, the first frequency band is any one of the p frequency bands. Since the method for determining the brain region rhythm synchronization level value corresponding to each of the p frequency bands is the same as the method for determining the brain region rhythm synchronization level value corresponding to the first frequency band, this embodiment uses the brain region rhythm synchronization level value corresponding to the first frequency band as an example for explanation. All p brain region rhythm synchronization levels corresponding to the p frequency bands can be determined according to the method used to determine the brain region rhythm synchronization level value corresponding to the first frequency band.

[0070] The data determination module can filter out all EEG data corresponding to the first frequency band from the previously obtained n×k×p segmented frequency EEG data, that is, the n×k segmented frequency EEG data corresponding to the first frequency band.

[0071] S402: Based on the n×k segmented frequency EEG data, determine the reference frequency brain region rhythm synchronization degree value between each pair of brain regions, and obtain a reference frequency brain region rhythm synchronization degree values.

[0072] In this embodiment, 'a' is a positive integer. The data determination module uses the n×k segmented frequency EEG data corresponding to the first frequency band as a basis, and pairs all the brain regions involved to obtain 'a' pairings. Then, it determines the rhythm synchronization degree value of the reference brain region with the same frequency corresponding to each of these 'a' pairings, thus obtaining 'a' reference brain region rhythm synchronization degree values ​​with the same frequency.

[0073] It should be explained that the determination of the rhythmic synchronization values ​​of the a reference brain regions at the same frequency can be illustrated by taking any one of the a paired combinations as an example. Specifically, taking the paired combination of the first brain region and the second brain region as an example, firstly, k segmented frequency EEG data corresponding to the first brain region and k segmented frequency EEG data corresponding to the second brain region are determined. The first brain region and the second brain region are any two brain regions among the n brain regions, which include a paired combinations. Each paired combination is a pair of brain regions among the n brain regions. The paired combination of the first brain region and the second brain region is any one of the a paired combinations. Then, k first reference brain region rhythmic synchronization values ​​between the first brain region and the second brain region are determined. Finally, the reference brain region rhythmic synchronization value between the first brain region and the second brain region is determined based on the k first reference brain region rhythmic synchronization values.

[0074] It should be explained that when determining the k first reference frequency brain region rhythm synchronization values ​​between the first brain region and the second brain region, the first reference frequency brain region rhythm synchronization value corresponding to any one of the k partition time periods (the first partition time period) can be determined first. Then, the k first reference frequency brain region rhythm synchronization values ​​corresponding to the k partition time periods can be determined in the same way as the first reference frequency brain region rhythm synchronization value corresponding to the first partition time period.

[0075] For example, based on k segmented frequency EEG data corresponding to the first brain region, the segmented frequency EEG data corresponding to the first brain region in the first segmented time period is determined, wherein the first segmented time period is any one of the k segmented time periods. Specifically, the segmented frequency EEG data corresponding to the first brain region in the first segmented time period is selected from the k segmented frequency EEG data corresponding to the first brain region.

[0076] For example, the phase value of the first brain region in the first partition is determined based on the partitioned frequency EEG data corresponding to the first partition time period, and a first phase value is obtained. Specifically, the data determination module uses the partitioned frequency EEG data corresponding to the first brain region in the first partition time period as the basis for analysis, processes the EEG data of these specific time periods through a preset phase calculation method, and extracts a quantitative value that can reflect the phase characteristics of the EEG rhythm of the brain region in the first partition time period. This value is the first phase value.

[0077] For example, the segmented frequency EEG data corresponding to the second brain region in the first segmented time period is determined based on the k segmented frequency EEG data corresponding to the second brain region. Specifically, the segmented frequency EEG data corresponding to the second brain region in the first segmented time period is selected from the k segmented frequency EEG data corresponding to the second brain region.

[0078] For example, the phase value of the second brain region in the first partition is determined based on the partitioned frequency EEG data corresponding to the first partition time period, and a second phase value is obtained. Specifically, the data determination module uses the partitioned frequency EEG data corresponding to the second brain region in the first partition time period as the calculation basis, and processes it using the same phase calculation method as the calculation of the first phase value to obtain a quantitative value that can characterize the phase characteristics of the EEG rhythm of the second brain region in the first partition time period. This value is the second phase value.

[0079] For example, a phase difference is determined based on the first phase value and the second phase value. Specifically, the two phase values ​​are calculated according to a preset difference calculation rule to finally obtain a phase difference that can reflect the phase difference between the two brain regions in the first partition time period.

[0080] For example, the first reference frequency brain region rhythm synchronization value corresponding to the first brain region and the second brain region in the first partition time period is determined based on the phase difference value. Specifically, the phase difference value is inversely proportional to the first reference frequency brain region rhythm synchronization value, and can be a preset mapping relationship between the phase difference value and the reference frequency brain region rhythm synchronization value. Based on the mapping relationship and the phase difference value, the first reference frequency brain region rhythm synchronization value corresponding to the first brain region and the second brain region in the first partition time period can be determined.

[0081] It can be seen that by first extracting segmented frequency EEG data of two brain regions in the same time period, then calculating their respective phase values ​​and obtaining phase difference values, and finally determining the synchronization degree of the two brain regions in that time period based on the phase difference values, the synchronization of the two brain regions at the same frequency can be accurately quantified. The limitation of the same time period allows the analysis to focus on the same time dimension, avoiding interference from time period differences. The phase value calculation and difference comparison can intuitively reflect the synchronization state of the electrical activity of the two brain regions. The final first reference brain region rhythm synchronization degree value can accurately capture the synchronization characteristics of the two brain regions in a specific time period.

[0082] It should be explained that, based on the first reference frequency brain region rhythm synchronization degree value corresponding to the first brain region and the second brain region in the first partition time period, k first reference frequency brain region rhythm synchronization degree values ​​corresponding to the first brain region in the k partition time periods can be determined. Thus, the reference frequency brain region rhythm synchronization degree value between the first brain region and the second brain region can be determined based on the k first reference frequency brain region rhythm synchronization degree values. Furthermore, the reference frequency brain region rhythm synchronization degree value between each pair of brain regions can be determined according to the method for determining the reference frequency brain region rhythm synchronization degree value between the first brain region and the second brain region, resulting in a reference frequency brain region rhythm synchronization degree values.

[0083] It can be seen that by first determining the synchronization level of two brain regions within a single time period, then extending it to obtain the synchronization level of two brain regions across all k time periods, then integrating them to obtain the overall reference synchronization level between the two brain regions, and finally generalizing it to all pairwise brain region combinations to obtain a reference synchronization level values, it is ensured that the synchronization characteristics of each pairwise brain region combination can be completely and accurately quantified. All a reference synchronization level values ​​can comprehensively reflect the synchronization distribution of the target child's brain regions in this frequency band.

[0084] S403: Determine the brain region rhythm synchronization value corresponding to the first frequency band based on the a reference brain region rhythm synchronization values ​​of the same frequency.

[0085] In this embodiment, the rhythm synchronization level of the brain region corresponding to the first frequency band can be determined by calculating the average value of the rhythm synchronization level of the a reference brain regions at the same frequency, or the minimum value among the a reference brain regions at the same frequency can be selected to determine the rhythm synchronization level of the brain region corresponding to the first frequency band. The specific determination method is not limited here.

[0086] As can be seen, by first locking any first frequency band among p frequency bands and extracting its corresponding n×k segmented frequency EEG data, then calculating the synchronization degree between all pairs of brain regions to obtain a reference values, and finally integrating these values ​​to determine the synchronization degree value corresponding to the frequency band, a comprehensive and accurate quantification of brain region synchronization within a single frequency band can be achieved. The pairing calculation of brain regions ensures that no synchronization characteristics of any brain region combination are missed, and the integration of a reference values ​​can condense the overall synchronization level of the frequency band, avoiding the one-sidedness brought by single brain region pairing or local data, so that the EEG synchronization characteristics of each frequency band can be presented in a concentrated and objective manner.

[0087] Please see Figure 5 , Figure 5 This application provides a flowchart of a method for determining the correlation between high and low frequency regulation in the same brain region using a data determination module, including but not limited to the following steps:

[0088] S501: Determine the high-frequency EEG data of the n×k partitions and the low-frequency EEG data of the n×k partitions corresponding to the first high-low frequency pair.

[0089] In this embodiment, the first high-low frequency pair is any one of the q high-low frequency pairs. Since the first high-low frequency pair is any one of the q high-low frequency pairs, the q high-low frequency regulation correlation values ​​of the same brain region corresponding to the first high-low frequency pair can be determined according to the method for determining the correlation value of high-low frequency regulation of the same brain region corresponding to the first high-low frequency pair. Here, the correlation value of high-low frequency regulation of the same brain region corresponding to the first high-low frequency pair is used for explanation.

[0090] Select the n×k high-frequency EEG data and n×k low-frequency EEG data corresponding to the first high-low frequency pair from the n×k×q high-frequency EEG data and the n×k×q low-frequency EEG data.

[0091] S502: Based on the high-frequency EEG data of the n×k regions and the low-frequency EEG data of the n×k regions, determine the correlation values ​​of high and low frequency regulation of the k reference brain regions corresponding to each of the n brain regions, and obtain the correlation values ​​of high and low frequency regulation of the n×k reference brain regions.

[0092] In this embodiment, for example, the high-frequency EEG data of the k partitions corresponding to the third brain region are determined based on the n×k partition high-frequency EEG data. Specifically, the third brain region is any one of the n brain regions. The method for determining the high- and low-frequency modulation correlation values ​​of the k reference brain regions corresponding to the third brain region is the same as the method for determining the high- and low-frequency modulation correlation values ​​of the k reference brain regions corresponding to each of the n brain regions. Therefore, the high- and low-frequency modulation correlation values ​​of the n×k reference brain regions corresponding to the n brain regions can be determined according to the method for determining the high- and low-frequency modulation correlation values ​​of the k reference brain regions corresponding to the third brain region. First, it is necessary to obtain the high-frequency EEG data of the k partitions corresponding to the third brain region from the n×k partition high-frequency EEG data. The high-frequency EEG data of the k partitions corresponding to the third brain region is used to determine the high- and low-frequency modulation correlation values ​​of the k reference brain regions corresponding to the third brain region.

[0093] For example, based on the n×k partition low-frequency EEG data, k partition low-frequency EEG data corresponding to the third brain region are determined. Specifically, it is necessary to obtain the k partition low-frequency EEG data corresponding to the third brain region from the n×k partition low-frequency EEG data. The k partition low-frequency EEG data corresponding to the third brain region are also used to determine the correlation values ​​of high and low frequency regulation of the k reference brain regions corresponding to the third brain region.

[0094] For example, based on the high-frequency EEG data of k partitions corresponding to the third brain region and the low-frequency EEG data of k partitions corresponding to the third brain region, the correlation values ​​of high and low frequency regulation of k reference brain regions corresponding to the third brain region are determined. Specifically, in this embodiment, when determining the correlation values ​​of high and low frequency regulation of k reference brain regions corresponding to the third brain region based on the high-frequency EEG data of k partitions corresponding to the third brain region and the low-frequency EEG data of k partitions corresponding to the third brain region, any partition time period can be arbitrarily selected from the k partition time periods. Then, the correlation value of high and low frequency regulation of the third brain region corresponding to the reference brain regions in that partition time period can be determined. The correlation value of high and low frequency regulation of the third brain region and the k reference brain regions corresponding to the k partition time periods can be determined according to the method for determining the correlation value of high and low frequency regulation of the reference brain regions corresponding to that partition time period.

[0095] For example, based on the high-frequency EEG data of k partitions corresponding to the third brain region and the low-frequency EEG data of k partitions corresponding to the third brain region, the high-frequency EEG data of the second partition time period and the low-frequency EEG data of the second partition time period are determined. Specifically, the second partition time period is any one of the k partition time periods. In this embodiment, since the second partition time period is any one of the k partition time periods, the determination of the high- and low-frequency modulation correlation value of the reference brain region corresponding to the third brain region in the second partition time period is explained using the method of determining the high- and low-frequency modulation correlation value of the reference brain region in the second partition time period as an example. Further, the method of determining the high- and low-frequency modulation correlation value of the k reference brain regions corresponding to the third brain region is the same as the method of determining the high- and low-frequency modulation correlation value of the reference brain region in the second partition time period.

[0096] The high-frequency EEG data and low-frequency EEG data of the second partition time period are obtained from the high-frequency EEG data of the k partitions corresponding to the third brain region and the low-frequency EEG data of the k partitions corresponding to the third brain region, so as to facilitate the subsequent determination of the correlation value of high and low frequency regulation of the third brain region in the second partition time period.

[0097] For example, the modulation index of the third brain region within the second time period is calculated based on the high-frequency EEG data and low-frequency EEG data corresponding to the second time period. Specifically, the high-frequency and low-frequency EEG data of the second time period are preprocessed to remove interference noise and make the signal cleaner. Then, the low-frequency EEG signal is divided into different phase intervals according to its rhythm period. Next, the amplitude change of the high-frequency EEG signal in each phase interval of the low-frequency signal is observed, and the distribution of the high-frequency signal amplitude in different low-frequency phase intervals is statistically analyzed. Then, the concentration of this distribution can be quantified using a preset calculation method. For example, whether the high-frequency signal amplitude is more concentrated in a specific phase interval of the low-frequency signal. The higher the concentration, the stronger the modulation effect of the low-frequency signal on the high-frequency signal, and the larger the modulation index. Thus, the modulation index can be calculated based on the third time period. The modulation index of the third brain region within the second time period is calculated using the high-frequency EEG data corresponding to the second time period and the low-frequency EEG data corresponding to the second time period. Alternatively, the modulation index of the third brain region within the second time period can be determined by using the low-frequency EEG data corresponding to that time period as the object, extracting the instantaneous phase of the low-frequency signal through Hilbert transform, and then uniformly dividing these instantaneous phases into m phase intervals. Simultaneously, a Hilbert transform is performed on the high-frequency EEG data corresponding to that time period to extract the instantaneous amplitude of the high-frequency signal. Then, the average value of the high-frequency instantaneous amplitude within each low-frequency phase interval is calculated to form a distribution of the high-frequency amplitude mean for each phase interval. Based on this distribution, Shannon entropy is calculated to quantify the concentration of the amplitude distribution. Finally, the calculated Shannon entropy is normalized using the maximum possible value of the Shannon entropy. The result is the modulation index of the third brain region within that time period. The specific method for determining the modulation index is not limited here.

[0098] For example, the correlation value of high and low frequency regulation of the third brain region in the second partition time period is determined based on the modulation index. Specifically, the modulation index is proportional to the correlation value of high and low frequency regulation of the reference brain region. It can be a preset mapping relationship between the modulation index and the correlation value of high and low frequency regulation of the same brain region. Based on the mapping relationship and the modulation index, the correlation value of high and low frequency regulation of the reference brain region in the second partition time period can be determined.

[0099] It should be explained that the high-low frequency regulation correlation values ​​of the k reference brain regions corresponding to the third brain region can be determined according to the method for determining the correlation values ​​of high and low frequency regulation of the reference brain regions corresponding to the third brain region in the second partition time period.

[0100] It can be seen that by first locking high-frequency and low-frequency EEG data of the same brain region and the same time period, the analysis is focused on the same dimension to avoid interference. Then, by calculating the modulation index, the modulation effect between low frequency and high frequency is quantified, so that the final reference correlation value can truly reflect the interaction strength of high and low frequency signals in the brain region at that time period.

[0101] S503: Determine the correlation value of high and low frequency regulation of the first high and low frequency pair in the same brain region based on the correlation values ​​of high and low frequency regulation of the n×k reference brain regions.

[0102] In this embodiment, the data determination module integrates the calculated high- and low-frequency regulation correlation values ​​of n×k reference brain regions. These scattered brain regions and time-period values ​​can be summarized and analyzed through preset statistical rules or integration algorithms. Finally, a high- and low-frequency regulation correlation value that can comprehensively reflect the overall level of high- and low-frequency regulation correlation of all brain regions under the first high- and low-frequency pair in all time periods is generated. That is, the high- and low-frequency regulation correlation value of the brain regions corresponding to the first high- and low-frequency pair.

[0103] As can be seen, by first locking onto any high-low frequency pair and extracting the corresponding whole-brain time-segmented high-low frequency EEG data, the analysis is ensured to focus on specific high-low frequency combinations to avoid cross-interference. Then, a reference correlation value is calculated for each time segment of each brain region, covering all brain regions and time segments without omitting key features. Finally, these values ​​are integrated to obtain the overall correlation value of the high-low frequency pair, which improves the reliability of determining the correlation value of high-low frequency regulation in the same brain region.

[0104] Please see Figure 6 , Figure 6 This application provides a flowchart of a probability determination module for determining the target matching probability, including but not limited to the following steps:

[0105] S601: Input the brain region rhythm synchronization value data of the same frequency into the preset synchronization feature processing model to obtain the first matching probability.

[0106] In this embodiment, the preset synchronicity feature processing model is an algorithm model in the probability determination module that is specifically used to process the rhythm synchronization value data of brain regions with the same frequency. It is trained based on a large amount of labeled autism-related EEG synchronicity feature data and has built-in standard parameters of synchronicity features corresponding to autism patterns. The core function of the preset synchronicity feature processing model is to receive the input rhythm synchronization value data of brain regions with the same frequency, and through the feature comparison algorithm built into the model, to perform multi-dimensional matching analysis on the EEG synchronicity features of the target child and the synchronicity features of the autism pattern, quantify the degree of fit between the two, and finally output a probability value that reflects the matching level of synchronicity features, which is the first matching probability.

[0107] S602: Input the high and low frequency regulation correlation data of the same brain region into the preset regulatory feature processing model to obtain the second matching probability.

[0108] It should be explained that, in this embodiment, the preset synchronicity feature processing model and the preset regulatory feature processing model together constitute the auxiliary assessment model. This auxiliary assessment model is a specialized model for accurately matching children's EEG characteristics with autism patterns. Its core consists of two complementary sub-models: the preset synchronicity feature processing model and the preset regulatory feature processing model. Please refer to [link to relevant documentation]. Figure 7 , Figure 7 This is a schematic diagram of the structure of an auxiliary evaluation model provided in the embodiments of this application. The auxiliary evaluation model 700 includes a preset synchronization feature processing model 701 and a preset control feature processing model 702.

[0109] The preset synchronization feature processing model 701 can use mature algorithms such as support vector machines, random forests, logistic regression, shallow fully connected neural networks, or gradient boosting trees. These models have been trained on a large amount of EEG synchronization feature annotation data of autistic and normal children, and can adapt to the high-dimensional characteristics and quantitative matching requirements of brain region rhythm synchronization value data of the same frequency. Among them, support vector machines are good at distinguishing feature boundaries, random forests can reduce the risk of overfitting, logistic regression balances simplicity and interpretability, neural networks can explore complex nonlinear relationships, and gradient boosting trees are highly sensitive to subtle differences in features. The model can be flexibly selected according to the actual data scale, feature dimension, and accuracy requirements. The core function is to accurately compare the EEG synchronization features of the target child with the synchronization features of the autism pattern and output the corresponding first matching probability.

[0110] The preset synchronicity feature processing model 701 focuses specifically on the same frequency rhythm synchronization features between brain regions. It is responsible for receiving the same frequency brain region rhythm synchronization degree value data. The same frequency brain region rhythm synchronization degree value data comes from the synchronization quantification results of all pairs of brain regions in a specific frequency band and the whole region time period. The model outputs the first matching probability reflecting the fit of this dimension by comparing the brain region synchronization features of the target child with the standard synchronization features of the autism pattern. The preset regulatory feature processing model 702 focuses on the high and low frequency regulatory correlation features within a single brain region. It receives the high and low frequency regulatory correlation degree value data of the same brain region. The high and low frequency regulatory correlation degree value data of the same brain region comes from the regulation correlation quantification results of all high and low frequency pairs, the whole brain region and the whole region time period. The model outputs the second matching probability representing the fit of this dimension by comparing the brain region high and low frequency regulatory features of the target child with the standard regulatory features of the autism pattern.

[0111] In this embodiment, the preset regulatory feature processing model 702 is an algorithm model in the auxiliary evaluation model 700 that specifically processes the correlation value data of high and low frequency regulation in the same brain region. It has also been trained with a large amount of autism-related EEG regulatory feature data and stores the standard parameters of high and low frequency regulation correlation features corresponding to the autism pattern. Its core function is to receive the input high and low frequency regulation correlation value data of the same brain region, compare the EEG regulatory features of the target child with the regulatory features of the autism pattern through specific feature matching logic, calculate the degree of consistency between the two, and finally output a probability value that can characterize the matching level of regulatory features, namely the second matching probability.

[0112] In this embodiment, the preset regulatory feature processing model 702 can be a mature algorithm model adapted to EEG regulatory feature analysis, such as logistic regression, random forest, support vector machine, gradient boosting tree, or shallow fully connected neural network. These models are trained with a large amount of labeled data of high and low frequency regulatory association features related to autism, and can accurately adapt to the feature attributes of high and low frequency regulatory association values ​​in the same brain region. Among them, logistic regression has strong interpretability and high computational efficiency, random forest and gradient boosting tree can capture complex interaction relationships between features and resist overfitting, support vector machine is suitable for distinguishing the boundaries of high-dimensional regulatory features, and neural network can explore nonlinear regulatory laws. It can be flexibly selected according to the data scale, feature complexity and accuracy requirements. The core is to output the corresponding second matching probability by comparing the EEG regulatory features of the target child with the regulatory features of the autism pattern.

[0113] S603: Determine the target matching probability based on the first matching probability and the second matching probability.

[0114] In this embodiment, for example, a first weight corresponding to the first matching probability and a second weight corresponding to the second matching probability are determined, wherein the sum of the first weight and the second weight is 1. Specifically, the probability determination module assigns corresponding weights to the first matching probability and the second matching probability according to the importance of the rhythm synchronization features of the same frequency brain regions and the high-low frequency regulation association features of the same brain regions in autism pattern matching, namely the first weight and the second weight. The sum of the values ​​of these two weights is fixed to 1, ensuring the rationality of the weight allocation and allowing the two types of matching probabilities to form an effective complement in the calculation of the final target matching probability.

[0115] For example, the target matching probability is determined based on the first matching probability and the second matching probability. Specifically, the target matching probability is calculated according to the following formula:

[0116] Target matching probability = first matching probability × first weight + second matching probability × second weight;

[0117] The target matching probability can be determined based on the first matching probability and the second matching probability according to the above formula.

[0118] It can be seen that by processing two different types of EEG features in different dimensions through separate models and then determining the target matching probability by weighted fusion, the matching value of synchronous features in the same frequency brain region and high- and low-frequency regulatory correlation features in the same brain region can be fully utilized. At the same time, the importance of different features in autism pattern matching can be highlighted by reasonably allocating weights. The setting of a weight sum of 1 ensures the standardization of weighted calculation and the rationality of results, avoids the one-sidedness brought about by single feature analysis, and allows the matching information of the two types of features to complement each other. The final target matching probability can more comprehensively and accurately reflect the degree of fit between the target child's EEG features and the autism pattern, effectively improving the scientificity and reliability of autism risk assessment.

[0119] In summary, the autism assessment system based on a large EEG model described in this application first acquires n EEG datasets corresponding to n brain regions of the target child within a preset time period. Then, the preset time period is divided into k partitioned time periods, and the EEG data of each of the n EEG datasets within the k partitioned time periods is determined, resulting in n×k partitioned EEG data. Based on these n×k partitioned EEG data, the system determines the rhythm synchronization value of the target child's brain regions at the same frequency and the correlation value of high- and low-frequency regulation within the same brain regions. Finally, based on the rhythm synchronization value and the correlation value of high- and low-frequency regulation within the same brain regions, the system determines the matching probability between the target child's EEG characteristics and autism patterns, thus obtaining a target matching probability for assessing the risk of the target child having autism, effectively improving the efficiency of autism assessment.

[0120] Please see Figure 8 , Figure 8 This is a schematic diagram of a device for determining the probability of autism pattern matching based on EEG features, provided in an embodiment of this application. As can be seen, the device 800 for determining the probability of autism pattern matching based on EEG features includes a system for determining the probability of autism pattern matching based on EEG features. This system for determining the probability of autism pattern matching based on EEG features can be the system described in any of the above embodiments.

[0121] It should be noted that, for the sake of simplicity, the aforementioned methods are described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0122] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0123] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0124] Those skilled in the art will understand that all or part of the processes in the methods described above can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0125] The steps of the methods or algorithms described in this application can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disk, portable hard disk, read-only optical disk (CD-ROM), or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Furthermore, the ASIC can reside in a terminal device or management device. Alternatively, the processor and storage medium can exist as discrete components in the terminal device or management device.

[0126] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in the embodiments of this application can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, in the form of a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated.

[0127] The aforementioned computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media.

[0128] The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0129] The modules / units included in the various devices and products described in the above embodiments can be software modules / units, hardware modules / units, or a combination of both. For example, for devices and products applied to or integrated into a chip, all modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits. For devices and products applied to or integrated into a chip module, all modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The implementation is achieved through a software program that runs on the processor integrated within the chip module. The remaining modules / units (if any) can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into terminal equipment, each of their modules / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components within the terminal equipment. Alternatively, at least some modules / units can be implemented through a software program that runs on the processor integrated within the terminal equipment, while the remaining modules / units (if any) can be implemented using hardware methods such as circuits.

[0130] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this application. It should be understood that the above description is only a specific embodiment of the embodiments of this application and is not intended to limit the protection scope of the embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solution of the embodiments of this application should be included within the protection scope of the embodiments of this application.

Claims

1. An electroencephalogram feature-based autism pattern matching probability determination system, characterized by, The autism pattern matching probability determination system based on the electroencephalogram features comprises an electroencephalogram acquisition module, a data division module, a data determination module and a probability determination module. The electroencephalogram acquisition module is configured to acquire n sets of electroencephalogram data corresponding to n brain regions of a brain of a target child within a preset time period; each set of electroencephalogram data corresponds to a brain region, and n is a positive integer; The data division module is configured to divide the preset time period into k sub-periods; k is a positive integer, and the k sub-periods are k continuous and independent sub-periods; The data determination module is configured to determine electroencephalogram data of each set of electroencephalogram data in the n sets of electroencephalogram data within the k sub-periods to obtain n×k sets of sub-period electroencephalogram data; The data determination module is configured to determine same-frequency brain region rhythm synchronization degree value data and same-brain region high-low frequency regulation correlation degree value data corresponding to the brain of the target child based on the n×k sets of sub-period electroencephalogram data; the same-frequency brain region rhythm synchronization degree value data is used to measure the activity synchronization of different brain regions at the same frequency, and the same-brain region high-low frequency regulation correlation degree value data is used to measure the regulation relationship strength between different frequencies of electroencephalogram rhythms in the same brain region; The probability determination module is configured to determine a matching probability of electroencephalogram features of the target child and an autism pattern based on the same-frequency brain region rhythm synchronization degree value data and the same-brain region high-low frequency regulation correlation degree value data to obtain a target matching probability; the target matching probability is used to evaluate the risk of autism of the target child; In the determination of the same-frequency brain region rhythm synchronization degree value data corresponding to the brain of the target child based on the n×k sets of sub-period electroencephalogram data, the data determination module is further configured to: Determine electroencephalogram data in p frequency bands corresponding to the n×k sets of sub-period electroencephalogram data to obtain n×k×p sets of sub-period frequency division electroencephalogram data; p is a positive integer, and one frequency band corresponds to one frequency range; Determine p same-frequency brain region rhythm synchronization degree values based on the n×k×p sets of sub-period frequency division electroencephalogram data; each frequency band corresponds to one same-frequency brain region rhythm synchronization degree value; Determine the same-frequency brain region rhythm synchronization degree value data based on the p same-frequency brain region rhythm synchronization degree values; In the determination of the p same-frequency brain region rhythm synchronization degree values based on the n×k×p sets of sub-period frequency division electroencephalogram data, the data determination module is further configured to: Determine n×k sets of sub-period frequency division electroencephalogram data corresponding to a first frequency band; the first frequency band is any one of the p frequency bands; Determine a reference same-frequency brain region rhythm synchronization degree value between two brain regions based on the n×k sets of sub-period frequency division electroencephalogram data to obtain a reference same-frequency brain region rhythm synchronization degree value; a is a positive integer; Determine the same-frequency brain region rhythm synchronization degree value corresponding to the first frequency band based on the a reference same-frequency brain region rhythm synchronization degree values; In the determining the reference same-frequency brain region rhythm synchronization degree value between the two brain regions based on the n×k partitioned frequency brain electrical data, the data determining module is specifically configured to: determine the k partitioned frequency brain electrical data corresponding to the first brain region and the k partitioned frequency brain electrical data corresponding to the second brain region; the first brain region and the second brain region are any two brain regions in the n brain regions, the n brain regions include a pair combination, the a pair combination is a different pair combination between the two brain regions in the n brain regions, and the pair combination of the first brain region and the second brain region is any one of the a pair combinations; determine the first reference same-frequency brain region rhythm synchronization degree value corresponding to the first brain region and the second brain region in the first partition time period based on the phase difference value; the phase difference value is inversely proportional to the first reference same-frequency brain region rhythm synchronization degree value, the phase difference value is the difference between the phase value of the first brain region in the first partition time period and the phase value of the second brain region in the first partition time period, and the first partition time period is any one of the k partition time periods; determine the reference same-frequency brain region rhythm synchronization degree value between the first brain region and the second brain region based on the k first reference same-frequency brain region rhythm synchronization degree values.

2. The system of claim 1, wherein, In the determining the same brain region high-low frequency regulation correlation degree value data corresponding to the brain of the target child based on the n×k partitioned brain electrical data, the data determining module is further specifically configured to: determine the brain electrical data under q high-low frequency pairs corresponding to the n×k partitioned brain electrical data, to obtain n×k×q partitioned high-frequency brain electrical data and n×k×q partitioned low-frequency brain electrical data; q is a positive integer, and each high-low frequency pair includes a low frequency band and a high frequency band; determine q same brain region high-low frequency regulation correlation degree values based on the n×k×q partitioned high-frequency brain electrical data and the n×k×q partitioned low-frequency brain electrical data; each high-low frequency pair corresponds to a same brain region high-low frequency regulation correlation degree value; determine the same brain region high-low frequency regulation correlation degree value data based on the q same brain region high-low frequency regulation correlation degree values.

3. The system of claim 1, wherein, The data determining module is further specifically configured to: determine the k first reference same-frequency brain region rhythm synchronization degree values between the first brain region and the second brain region in the following steps S1-S6: S1, determine the partitioned frequency brain electrical data corresponding to the first brain region in the first partition time period based on the k partitioned frequency brain electrical data corresponding to the first brain region; S2, determine the phase value of the first brain region in the first partition time period based on the partitioned frequency brain electrical data corresponding to the first brain region in the first partition time period, to obtain a first phase value; S3, determine the partitioned frequency brain electrical data corresponding to the second brain region in the first partition time period based on the k partitioned frequency brain electrical data corresponding to the second brain region; S4, determine a phase value of the second brain area in the first partition time period based on the partition band frequency electroencephalogram data corresponding to the first partition time period of the second brain area, to obtain a second phase value; S5, determine the phase difference value based on the first phase value and the second phase value; S6, determine the first reference same brain area high-low frequency regulation correlation degree value corresponding to the first partition time period of the first brain area and the second brain area based on the phase difference value.

4. The system of claim 2, wherein, In the aspect of determining the q same brain area high-low frequency regulation correlation degree values based on the n×k×q partition high frequency electroencephalogram data and the n×k×q partition low frequency electroencephalogram data, the data determination module is specifically configured to: determine n×k partition high frequency electroencephalogram data and n×k partition low frequency electroencephalogram data corresponding to a first high-low frequency pair; the first high-low frequency pair is any one of the q high-low frequency pairs; determine k reference same brain area high-low frequency regulation correlation degree values corresponding to each brain area in the n brain areas based on the n×k partition high frequency electroencephalogram data and the n×k partition low frequency electroencephalogram data, to obtain n×k reference same brain area high-low frequency regulation correlation degree values; determine a same brain area high-low frequency regulation correlation degree value corresponding to the first high-low frequency pair based on the n×k reference same brain area high-low frequency regulation correlation degree values.

5. The system of claim 4, wherein, In the aspect of determining the k reference same brain area high-low frequency regulation correlation degree values corresponding to each brain area in the n brain areas based on the n×k partition high frequency electroencephalogram data and the n×k partition low frequency electroencephalogram data, the data determination module is specifically configured to: determine k partition high frequency electroencephalogram data corresponding to a third brain area based on the n×k partition high frequency electroencephalogram data; the third brain area is any one of the n brain areas; determine k partition low frequency electroencephalogram data corresponding to the third brain area based on the n×k partition low frequency electroencephalogram data; determine k reference same brain area high-low frequency regulation correlation degree values corresponding to the third brain area based on the k partition high frequency electroencephalogram data corresponding to the third brain area and the k partition low frequency electroencephalogram data corresponding to the third brain area.

6. The system of claim 5, wherein, In the aspect of determining the k reference same brain area high-low frequency regulation correlation degree values corresponding to the third brain area based on the k partition high frequency electroencephalogram data corresponding to the third brain area and the k partition low frequency electroencephalogram data corresponding to the third brain area, the data determination module is specifically configured to: determine partition high frequency electroencephalogram data corresponding to a second partition time period and partition low frequency electroencephalogram data corresponding to the second partition time period based on the k partition high frequency electroencephalogram data corresponding to the third brain area and the k partition low frequency electroencephalogram data corresponding to the third brain area; the second partition time period is any one of the k partition time periods; calculate a modulation index of the third brain area in the second partition time period according to the partition high frequency electroencephalogram data corresponding to the second partition time period and the partition low frequency electroencephalogram data corresponding to the second partition time period; determine a reference homologous brain high-low frequency regulation correlation degree value corresponding to the second sub-region time period based on the modulation index; the modulation index is proportional to the reference homologous brain high-low frequency regulation correlation degree value.

7. The system of any of claims 2-6, wherein, In the aspect of determining the target child's brain electrical feature matching probability with the autism pattern based on the homologous brain high-low frequency regulation correlation degree value data and the homologous brain high-low frequency regulation correlation degree value data, the probability determination module is specifically configured to: input the homologous brain high-low frequency regulation correlation degree value data into a preset regulation feature processing model to obtain a second matching probability; input the homologous brain high-low frequency regulation correlation degree value data into a preset regulation feature processing model to obtain a second matching probability; determine the target matching probability based on the first matching probability and the second matching probability.

8. The system of claim 7, wherein, In the aspect of determining the target matching probability based on the first matching probability and the second matching probability, the probability determination module is specifically configured to: determine a first weight corresponding to the first matching probability and a second weight corresponding to the second matching probability; the sum of the first weight and the second weight is 1; determine the target matching probability based on the first weight, the second weight, the first matching probability and the second matching probability.

9. A device for determining the probability of autism pattern matching based on electroencephalogram (EEG) features, characterized in that, The system comprises any one of claims 1-8. The system comprises any one of claims 1-8.

Images