A method for detecting brain function localization state based on task-state EEG source functional connectivity map

By collecting EEG signals from different task states, tracing and connecting maps are constructed, and the overlap of brain function connection maps is compared, the problem of difficulty in comprehensively detecting brain function positioning status in the existing technology is solved, and the brain function positioning status of a scientific system is realized Test.

CN115089195BInactive Publication Date: 2024-11-29SOUTH CHINA UNIV OF TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202210733090.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-06-27
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing technology is difficult to scientifically and comprehensively detect brain function localization status, resulting in a lack of scientific basic theoretical guidance. Most of the research is limited to local areas, so it is impossible to fully determine brain function localization status.

Method used

By collecting EEG signals from different task states, tracing the EEG signals, building a brain functional connection map, comparing the spatial distribution overlap of brain functional connection maps from different task states, and determining the brain functional positioning status.

Benefits of technology

This method is based on the basic theory of bioelectricity and reflective arc, has the advantage of scientific microscopic theoretical basis, and can systematically and comprehensively detect the brain function positioning state, overcoming the limitation that traditional methods can only detect in a local range.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115089195B_ABST
    Figure CN115089195B_ABST
Patent Text Reader

Abstract

The present invention relates to a method for detecting brain function localization state based on task state EEG source function connection diagram, and the method comprises the steps of: S1, collecting EEG signals of various task states, including action, language, vision, and hearing tasks; S2, using sLORETA algorithm to trace the source of EEG signals, performing source function connection analysis on EEG signals of various task states, and calculating the PLV values ​​of each frequency band between each brain region signal to obtain an adjacency matrix, and constructing a brain function connection diagram; S3, performing intersection and symmetric difference operations on the adjacency matrices obtained in each task state, comparing the spatial distribution overlap of the brain function connection diagrams of each task state, and determining the brain function localization state. The method of the present invention can detect the brain function localization state scientifically, systematically, and comprehensively, and can be widely used in basic research of neuroscience such as brain function localization based on electroencephalogram analysis.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of neuroscience research based on electroencephalogram analysis, and in particular to a method for detecting the localization state of brain function based on a task-state EEG source functional connection diagram. Background Art

[0002] Whether and how brain function is localized has always been an unresolved basic theoretical problem in the field of neuroscience. Since the study of brain function, many studies have used traditional methods such as dissection, electrical stimulation, and chemical stimulation to detect the localization status of brain function, and proposed a series of theories of brain function localization status, such as the partition theory, the holistic theory, the functional system theory, the functional module theory, and the neural network theory. The localization theory believes that a certain area of ​​the brain is responsible for a certain function; the holistic theory believes that brain function is the brain as a whole, and the strength of the function is related to the size of the brain but not to the specific part of the brain; the functional system theory, the functional module theory, and the neural network theory all emphasize the partition of brain function and take into account the connection between brain functional areas. These theories reflect different brain function localization states, and the theories contradict each other. It is impossible to come up with a recognized brain function localization state, which makes the application of brain function localization in many fields of neuroscience lack scientific basic theoretical guidance. Therefore, there are many defects in practical applications. Moreover, most of the current research on brain function localization is limited to the study of local areas of the cerebral cortex, and cannot scientifically and systematically determine the brain function localization state. Therefore, there is an urgent need to propose a scientific, systematic and comprehensive detection method to determine the localization status of brain function. Summary of the invention

[0003] In order to solve the technical problems existing in the prior art, the present invention provides a method for detecting the localization state of brain function based on the task-state EEG source functional connection diagram. By collecting EEG signals in different task states, tracing the EEG signals, constructing a brain function connection diagram, and comparing the spatial distribution overlap of the brain function connection diagrams in different task states, the localization state of brain function is determined. This method detects the localization state of brain function based on the basic theories of bioelectricity and reflex arcs, and has the superiority of scientific microscopic theoretical basis; and the collected EEG signals cover a wide range, and can systematically and comprehensively detect the localization state of brain function.

[0004] The method of the present invention is implemented by the following technical scheme: a method for detecting the localization state of brain function based on the task-state EEG source functional connection diagram, comprising the following steps:

[0005] S1. Collect EEG signals in various tasks, including movement, language, vision, and hearing tasks;

[0006] S2. Use the sLORETA algorithm to trace the source of EEG signals, perform source functional connectivity analysis on EEG signals in various task states, calculate the PLV values ​​of each frequency band between signals in each brain region to obtain the adjacency matrix, and construct a brain functional connectivity map;

[0007] S3. Perform intersection and symmetric difference operations on the adjacency matrices obtained for each task state, compare the spatial distribution overlap of the brain function connection diagrams of each task state, and determine the brain function localization state.

[0008] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0009] 1. The present invention collects EEG signals of different task states, traces the EEG signals, constructs a brain function connection map, compares the spatial distribution overlap of the brain function connection map of different task states, and determines the brain function localization state. This method detects the brain function localization state based on the basic theory of bioelectricity and reflex arc, has the superiority of scientific microscopic theoretical basis, and the acquisition method of EEG signals is safe and convenient, thereby ensuring that the data source of EEG signals is sufficient.

[0010] 2. The present invention uses a brain function connection method to detect brain function status, which can systematically and comprehensively detect brain function status; this method can use high-density EEG electrodes to collect EEG signals across the entire brain on the scalp surface to study the response states of various brain regions when performing different tasks, overcoming the problem that traditional research methods such as electrical stimulation and ECoG can only detect the localization state of brain function in a local area. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 is a flow chart of the method of the present invention;

[0012] Figure 2 It is a schematic diagram of constructing an EEG functional connection map according to the present invention;

[0013] Figure 3 This is a comparison experiment result diagram of the EEG source functional connection diagram of the present invention. DETAILED DESCRIPTION

[0014] The present invention is further described in detail below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.

[0015] Example

[0016] like Figure 1 As shown, the method for detecting the brain function localization state based on the task-state EEG source functional connection diagram in this embodiment includes the following steps:

[0017] S1. Collect EEG signals in various tasks, including movement, language, vision, hearing and other tasks;

[0018] S2. Use the sLORETA algorithm to trace the source of EEG signals, perform source functional connectivity analysis on EEG signals in various task states, calculate the PLV values ​​of different frequency bands between signals in each brain region to obtain the adjacency matrix, and construct a brain functional connectivity map;

[0019] S3. Perform intersection and symmetric difference operations on the adjacency matrices obtained for each task state, compare the spatial distribution overlap of the brain function connection diagrams of each task state, and determine the brain function localization state.

[0020] In this embodiment, the process of EEG signal collection in step S1 is as follows:

[0021] When the human body performs functional activities, function-related EEG signals are generated in the cerebral cortex. The EEG signals are based on the microscopic theories of bioelectricity and reflex arcs. The spatial distribution of EEG signals is analyzed to detect the localization state of brain function. By collecting EEG signals of the subjects when they are performing or accepting specific tasks, including resting, manual, language, auditory and visual, the subjects are asked to perform eye-opening, eye-closing, and manual opening and closing tasks according to prompts. In the task state, the BCI2000 system is used to record 64-channel EEG.

[0022] like Figure 2 As shown, in this embodiment, the specific process of step S2 is as follows:

[0023] S21, perform EEG signal preprocessing, intercept the EEG signals corresponding to the eyes open, eyes closed, and opening and closing manual tasks in each lead to generate EEG samples; use a common average reference to eliminate the error caused by the change of the original reference electrode; perform 1Hz high-pass filtering and 49-51Hz notch filtering to remove power frequency interference; perform baseline drift removal to eliminate the deviation of EEG relative to the baseline, and use independent component analysis to remove obvious artifacts such as electrooculography to obtain a relatively clean EEG signal;

[0024] S22. EEG source tracing includes solving two parts: the EEG positive problem and the EEG inverse problem. The EEG positive problem describes how the EEG signals generated by the cerebral cortex are transmitted to the scalp, while the EEG inverse problem traces the signals on the cerebral cortex through the observed scalp potentials. Among them, the solution process of the EEG positive problem includes three models: the source model, the head model and the head surface electrode model. The source model adopts a distributed source model. In the source model, the signal sources are evenly distributed in the cortex, and their direction is the direction of the normal vector. The number of sources is about 15,000. The head model uses the cranial MRI data corresponding to the ICBM152 template brain to establish a real head model. The model includes three layers: cerebral cortex, skull and scalp. The conductivity of the skull is about one eightieth of the conductivity of the cerebral cortex and scalp. The head surface electrode model uses the spatial coordinates of the EEG electrode, that is, the 10-10 or 10-20 standard electrode lead, and aligns with the MRI spatial coordinate correction to obtain the position of each electrode on the head surface. The OpenMEEG algorithm is used to calculate the guidance field matrix by combining the three models. The inverse problem solving process is to combine the preprocessed EEG signal and the guidance field matrix, and use the sLORETA inverse solution algorithm to trace the source to obtain the EEG signal after tracing back to the cerebral cortex. The tracing process is applicable to EEG samples of different tasks.

[0025] S23. Construct EEG functional connectivity diagrams in multiple task states. Extract source signals from various brain regions based on the Desikan-Killiany brain template to obtain signals from 68 brain regions; extract signals from the α, β and γ frequency bands of each brain region respectively; calculate the PLV values ​​of functional connectivity indicators between different brain regions in each frequency band to obtain an adjacency matrix, and perform thresholding, wherein the threshold selection criterion is to first calculate the value of the adjacency matrix of each task while retaining the top 5% intensity, and then average the values ​​of each task to obtain the threshold; use each brain region in the Desikan-Killiany template as a node, and the values ​​in the thresholded adjacency matrix as an edge to establish a brain functional connectivity diagram.

[0026] Specifically, taking the α-band signal at rest with eyes closed as an example, the PLV value of the functional connectivity index of each brain region in the α-band was calculated to obtain the adjacency matrix; the threshold was set to retain the value of about the first 5% of the connections in each task state adjacency matrix, and the combined threshold of the two tasks was set to 0.6316, and the adjacency matrix was thresholded to obtain a binary matrix; each brain region was used as a node, and the values ​​corresponding to each brain region in the binary matrix were used as edges to construct a brain functional connection map.

[0027] In this embodiment, the specific process of determining the brain function localization state in step S3 is as follows:

[0028] If the spatial distribution of connection strength of the functional connection diagrams corresponding to each task state does not overlap, then the state of brain function localization is the partition theory; if the spatial distribution of connection strength of the functional connection diagrams corresponding to each task state basically overlaps, then the state of brain function localization is the holistic theory; if the spatial distribution of connection strength of the functional connection diagrams corresponding to each task state partially overlaps, then the state of brain function localization is the mixed theory.

[0029] In this embodiment, the brain functional connection diagrams in the α frequency band during the eyes closed resting state and the manual task obtained in step S2 are used to perform intersection and symmetric difference operations on the adjacency matrices of the two tasks, and compare the brain functional connection diagrams in the two different task states, such as Figure 3 As shown, the brain functional connection diagrams of the two task states are partially overlapping, indicating that the brain functional localization state is a mixed theory. Multiple brain regions are required to participate in the execution of a certain functional task, and the same brain region is also involved in the execution of different tasks.

[0030] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for detecting brain function localization state based on task-state EEG source functional connection diagram, characterized in that: The following steps are involved: S1. Collect EEG signals in various tasks, including movement, language, vision, and hearing tasks; S2. Use the sLORETA algorithm to trace the source of EEG signals, perform source functional connectivity analysis on EEG signals in various task states, calculate the PLV values ​​of each frequency band between signals in each brain region to obtain the adjacency matrix, and construct a brain functional connectivity map; S3, performing intersection and symmetric difference operations on the adjacency matrices obtained for each task state, comparing the spatial distribution overlap of the brain function connection diagrams of each task state, and determining the brain function localization state; The specific process of step S2 is as follows: S21, performing EEG signal preprocessing, intercepting the EEG signals corresponding to the eyes opening, eyes closing, and opening, closing manual tasks in each lead to generate EEG samples; A common average reference is used for EEG samples to eliminate the error caused by the change of the original reference electrode; Perform high-pass filtering and notch filtering to remove power frequency interference; Baseline drift was removed to eliminate the deviation of EEG relative to the baseline, and independent component analysis was used to remove eye artifacts to obtain clean EEG signals; S22. EEG source tracing includes solving two parts: the EEG positive problem and the EEG inverse problem. The EEG positive problem describes how the EEG signals generated by the cerebral cortex are transmitted to the scalp, while the EEG inverse problem traces the signals on the cerebral cortex through the observed scalp potentials. Among them, the solution process of the EEG positive problem includes three models: the source model, the head model and the head surface electrode model. The source model adopts a distributed source model. In the source model, the signal sources are evenly distributed in the cortex, and their direction is the direction of the normal vector. The number of sources is about 15,000. The head model uses the cranial MRI data corresponding to the ICBM152 template brain to establish the true The real head model consists of three layers: the cerebral cortex, the skull and the scalp. The conductivity of the skull is about one eightieth of the conductivity of the cerebral cortex and the scalp. The head surface electrode model uses the spatial coordinates of the EEG electrodes, i.e., the 10-10 or 10-20 standard electrode leads, and aligns with the MRI spatial coordinates to obtain the position of each electrode on the head surface. The OpenMEEG algorithm is used to calculate the guidance field matrix by combining the three models. The inverse problem solving process is to combine the preprocessed EEG signal and the guidance field matrix, and use the sLORETA inverse solution algorithm to trace the source to obtain the EEG signal after tracing back to the cerebral cortex. S23. Construct EEG functional connectivity diagrams in multiple task states, extract source signals of each brain region based on the Desikan-Killiany brain template to obtain signals of each brain region; extract signals of the α, β and γ frequency bands of each brain region respectively; calculate the PLV values ​​of the functional connectivity index between each brain region in each frequency band to obtain the adjacency matrix, and perform thresholding, using each brain region in the Desikan-Killiany template as a node and the values ​​in the adjacency matrix after thresholding as edges to establish a brain functional connectivity diagram.

2. The method for detecting brain function localization state based on task-state EEG source functional connection diagram according to claim 1, characterized in that: The process of EEG signal acquisition in step S1 is as follows: By collecting EEG signals of the subjects when they are performing or accepting specific tasks, including resting, manual, language, auditory and visual signals, the subjects are asked to perform eye opening, eye closing, and open and close manual tasks according to prompts; in the task state, the BCI2000 system is used to record 64-channel EEG.

3. The method for detecting brain function localization state based on task-state EEG source functional connection diagram according to claim 1, characterized in that: The threshold in step S23 is selected by first calculating the value of the adjacency matrix of each task when the top 5% strength is retained, and then averaging the values ​​of each task to obtain the threshold.

4. The method for detecting brain function localization state based on task-state EEG source functional connection diagram according to claim 1, characterized in that: The specific process of determining the brain function localization state in step S3 is as follows: If the spatial distribution of connection strength of the functional connection diagrams corresponding to each task state does not overlap, then the state of brain function localization is the partition theory; if the spatial distribution of connection strength of the functional connection diagrams corresponding to each task state overlaps, then the state of brain function localization is the holistic theory; if the spatial distribution of connection strength of the functional connection diagrams corresponding to each task state partially overlaps, then the state of brain function localization is the mixed theory.

Citation Information

Patent Citations

  • Automatic analysis method and system based on resting-state EEG frequency domain characteristics and brain network

    CN113576491A

  • Brain function area positioning method and system combining spectral clustering and event correlation detection

    CN114081507A