Dynamic brain function network analysis method and system based on time-space joint state

Abstract

The application belongs to the technical field of medical image analysis, and provides a dynamic brain function network analysis method and system based on a time-space joint state. The method comprises the following steps: in the time dimension, according to a preset sliding window, a four-dimensional resting-state functional magnetic resonance image data segment after preprocessing is intercepted, and the Pearson correlation coefficient of the time sequence signals between any two brain regions in the window segment is calculated to obtain a dynamic brain function connection matrix; the independent component analysis method is used to extract the dynamic brain function connection matrix to obtain the individual corresponding independent component and the time sequence corresponding to the independent component; based on the individual corresponding independent component and the time sequence corresponding to the independent component, an efficient forward search strategy and a classifier are used to construct a time-space fusion discrimination model, specific brain dynamic networks related to specific neuropsychiatric diseases are identified, and a decision value is output, so that quantitative measurement of the dynamic brain function network at the individual level is realized.

Description

Technical Field

[0001] This invention belongs to the field of medical image analysis technology, and in particular relates to a dynamic brain functional network analysis method and system based on the joint state of time and space. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Currently, functional magnetic resonance imaging (fMRI) is widely used in basic research and clinical applications of brain function in neuropsychiatric diseases. This technology can record brain functional activity in four-dimensional space, that is, it can characterize the one-dimensional temporal functional activity of voxels in three-dimensional space. Traditional brain functional network analysis methods assume that the brain network is in a steady state, that is, the functional network pattern of the brain does not change over time. It extracts the steady-state brain functional network by measuring the interaction between brain regions based on brain signals during the acquisition period. In the process of analyzing the steady-state brain network, the focus is mainly on the spatial specific patterns of brain function and model construction. However, the steady-state brain functional network is insufficient to reflect the complex time-varying characteristics of the brain system. Dynamic brain network analysis is considered to be beneficial for revealing the imaging features of diseases and the corresponding brain change mechanisms.

[0004] The sliding time window algorithm is the most commonly used method for estimating dynamic brain networks based on resting-state functional magnetic resonance imaging (fMRI). Numerous studies have used the dynamic functional connectivity matrix obtained by this algorithm to describe and compare the network characteristics of patients with neuropsychiatric disorders, including clustering of spatial connectivity states in the brain's dynamic network and dynamic causal analysis. Although many studies have analyzed the relationship between time-varying brain activity and functional connectivity, they generally employ between-group statistical methods or discriminant analysis based on network spatial attributes, lacking effective and comprehensive quantitative indicators and biomarkers to characterize dynamic networks, making it difficult to achieve individual-level analysis and personalized diagnosis and treatment. Furthermore, the black-box decision-making nature of deep neural networks and the requirement for large samples limit their application in decision-making tasks related to neuropsychiatric disorders. Summary of the Invention

[0005] To address the technical problems mentioned above, this invention provides a method and system for dynamic brain functional network analysis based on a combined temporal and spatial state. This method can identify brain network biomarkers related to neuropsychiatric diseases, enabling quantitative analysis of dynamic brain functional networks in populations and opening up new avenues for individualized diagnosis and treatment of neuropsychiatric diseases.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] The first aspect of the present invention provides a method for dynamic brain functional network analysis based on the joint state of time and space.

[0008] Dynamic brain functional network analysis methods based on joint temporal and spatial states include:

[0009] Acquire four-dimensional resting-state functional magnetic resonance imaging data of patients with specific neuropsychiatric diseases and normal controls, and perform preprocessing;

[0010] In the time dimension, preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments are extracted according to a preset sliding window, and the Pearson correlation coefficient of temporal signals between any two brain regions within the window segment is calculated to obtain the dynamic brain functional connectivity matrix.

[0011] Independent component analysis was used to extract the dynamic brain functional connectivity matrix to obtain the individual independent components and the time series corresponding to the independent components.

[0012] Based on the individual independent components and their corresponding time series, a time-space fusion discrimination model is constructed using an efficient forward search strategy and classifier to identify specific dynamic brain networks associated with specific neuropsychiatric diseases and output decision values, thereby achieving quantitative measurement of dynamic brain functional networks at the individual level.

[0013] A second aspect of the present invention provides a dynamic brain functional network analysis system based on the joint state of time and space.

[0014] A dynamic brain functional network analysis system based on joint temporal and spatial states includes:

[0015] The data acquisition and preprocessing module is configured to acquire four-dimensional resting-state functional magnetic resonance imaging data of patients with specific neuropsychiatric diseases and normal control groups, and perform preprocessing.

[0016] The dynamic brain functional connectivity matrix construction module is configured to: in the time dimension, extract preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments according to a preset sliding window, and calculate the Pearson correlation coefficient of time-series signals between any two brain regions within the window segment to obtain the dynamic brain functional connectivity matrix.

[0017] The time-space joint state module is configured to extract the dynamic brain functional connectivity matrix using independent component analysis to obtain the individual's corresponding independent components and the time series corresponding to the independent components.

[0018] The analysis module is configured to: based on the individual's corresponding independent components and the time series corresponding to the independent components, use an efficient forward search strategy and classifier to construct a time-space fusion discrimination model, identify specific brain dynamic networks related to specific neuropsychiatric diseases, and output decision values ​​to achieve quantitative measurement of dynamic brain functional networks at the individual level.

[0019] A third aspect of the present invention provides a computer-readable storage medium.

[0020] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the dynamic brain functional network analysis method based on time-space joint state as described in the first aspect above.

[0021] A fourth aspect of the present invention provides a computer device.

[0022] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the dynamic brain functional network analysis method based on time-space joint state as described in the first aspect above.

[0023] Compared with the prior art, the beneficial effects of the present invention are:

[0024] This invention first collects resting-state functional magnetic resonance imaging (fMRI) data from participants in different groups (patients with a certain neuropsychiatric disease and normal controls), and preprocesses the collected image data. Then, 60-second segments are extracted using overlapping sliding windows in the time dimension, and four-dimensional data for each segment are extracted. Pearson correlation calculations are performed on these segments using a brain mapping template to obtain a functional connectivity matrix. A dynamic brain functional connectivity matrix is ​​constructed based on the functional connectivity matrix corresponding to the retained real time (the time corresponding to the segment window header). Independent component analysis is used to extract individual-specific and spatially corresponding functional connectivity states (spatial information) and their corresponding temporal fluctuation sequences (temporal information) from the dynamic brain functional connectivity matrix. Finally, a multivariate pattern discrimination model is used to construct a temporal-spatial fusion classifier in Riemannian manifold space to identify specific dynamic brain networks associated with specific neuropsychiatric diseases, and outputs decision values ​​to achieve quantitative measurement of dynamic brain functional networks at the individual level. This invention can identify brain network biomarkers associated with neuropsychiatric diseases, realize quantitative analysis of dynamic brain functional networks in populations, and open up new avenues for personalized diagnosis and treatment of neuropsychiatric diseases.

[0025] This invention utilizes quantitative analysis of the temporal and spatial information of dynamic brain functional networks to comprehensively and effectively characterize their changes. Furthermore, it performs multivariate discriminant analysis on brain networks in combined states within a Riemannian manifold space, identifying the most discriminative and specific dynamic brain networks. This quantitatively characterizes the individual-level dynamic characteristics of brain functional networks, enabling spatiotemporal observation of brain functional networks. While studying the physiological mechanisms of neuropsychiatric diseases, the specific indicators output by the model can be used for auxiliary diagnosis, thereby promoting early diagnosis and treatment of related diseases.

[0026] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0027] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0028] Figure 1 This is a flowchart illustrating the construction of independent components of an individual brain functional network pattern based on existing methods, as shown in Embodiment 1 of the present invention.

[0029] Figure 2 This is a flowchart of the dynamic brain functional network discrimination method based on the joint state of time and space as shown in Embodiment 1 of the present invention;

[0030] Figure 3 This is a flowchart of feature search and analysis shown in Embodiment 1 of the present invention. Detailed Implementation

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0032] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0034] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of this disclosure. It should be noted that each block in a flowchart or block diagram may represent a module, segment, or portion of code, which may include one or more executable instructions for implementing the logical functions specified in the various embodiments. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutively represented blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, may be implemented using a dedicated hardware-based system that performs the specified functions or operations, or using a combination of dedicated hardware and computer instructions.

[0035] Example 1

[0036] like Figure 1 As shown, this embodiment provides a dynamic brain functional network analysis method based on the joint state of the time and space domains. This embodiment uses the application of this method to a server as an example for illustration. It is understood that this method can also be applied to terminals, and can also be applied to systems including terminals, servers, and other components, and can be implemented through interaction between the terminal and the server. The server can be an independent physical server, a server cluster composed of multiple physical servers, or a distributed system. It can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network servers, cloud communication, middleware services, domain name services, CDN security services, and big data and artificial intelligence platforms. The terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. The terminal and server can be directly or indirectly connected via wired or wireless communication, which is not limited herein. In this embodiment, the method includes the following steps:

[0037] Acquire four-dimensional resting-state functional magnetic resonance imaging data of patients with specific neuropsychiatric diseases and normal controls, and perform preprocessing;

[0038] In the time dimension, preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments are extracted according to a preset sliding window, and the Pearson correlation coefficient of temporal signals between any two brain regions within the window segment is calculated to obtain the dynamic brain functional connectivity matrix.

[0039] Independent component analysis was used to extract the dynamic brain functional connectivity matrix to obtain the individual independent components and the time series corresponding to the independent components.

[0040] Based on the individual independent components and their corresponding time series, a time-space fusion discrimination model is constructed using an efficient forward search strategy and classifier to identify specific dynamic brain networks associated with specific neuropsychiatric diseases and output decision values, thereby achieving quantitative measurement of dynamic brain functional networks at the individual level.

[0041] The specific solution in this embodiment can be implemented using the following steps:

[0042] 1. Four-dimensional resting-state functional magnetic resonance imaging is collected from participants in different groups (patients with a certain neuropsychiatric disease and normal control groups), and the acquisition time is usually no less than 8 minutes.

[0043] Figure 1 It is a flowchart for constructing independent components of individual brain functional network patterns based on existing methods.

[0044] 2. Preprocessing of the acquired data. Preprocessing of resting-state functional magnetic resonance imaging (fMRI) data mainly includes time-series correction, head motion correction, registration, spatial normalization, spatial smoothing, and filtering.

[0045] 3. Divide the preprocessed fMRI data into N brain regions according to the brain atlas template, and extract the average time series of voxels contained in each brain region. Divide the aforementioned average time series into segments according to the preset sliding window size W (usually 60 seconds), calculate the Pearson correlation coefficient r between the time series signals of any two brain regions within the window segment, and use Fisher's z transform to convert the r value into a z value.

[0046] Assuming T is the average time series length of the original data, the dynamic brain function connectivity matrix P is obtained after sliding window calculation. P is a three-dimensional matrix of N×N×L (L=T / W) representing the temporal network of brain function, where N×N is the functional connectivity matrix within the window and L is the number of layers of its temporal network.

[0047] 4. Obtain all brain functional network pattern components of the subjects through group independent component analysis (ICA). First, all individual dynamic brain functional connectivity matrices / temporal brain functional networks are concatenated in the time domain to obtain a longer-term three-dimensional dataset. Then, independent component analysis is applied to this concatenated three-dimensional temporal network data to generate group independent components. Finally, a reverse reconstruction step maps the group independent components to individual individuals to obtain the corresponding independent components and their corresponding time series. Each independent component corresponds to the spatial functional connectivity state (spatial information) of the dynamic brain functional network, and the time series corresponds to the temporal fluctuations of the functional connectivity state (temporal information). To explore more refined brain network patterns, the number of independent components can be set to more than 50, or the number of components can be estimated based on the data.

[0048] Figure 2 This is a flowchart of the dynamic brain functional network discrimination method based on the joint state of time and space according to the present invention.

[0049] 5. Utilizing an efficient forward selection search mechanism and a support vector machine (SVM) model, a temporal-spatial fusion discriminant model is constructed to identify specific disease-specific dynamic brain networks. From the perspective of multivariate data representation, the functional connectivity states obtained by ICA can be used as basis functions to span a subspace, which can characterize the dynamic functional network pattern (dFNP).

[0050] Assume independent components IC = {ic i If |i=1,…k}, then dFNP can be defined as:

[0051]

[0052] In the spatial domain, IC represents the brain network connectivity state, which is the individual's spatial state (SS); in the temporal domain, IC represents the corresponding time series, which is the individual's temporal state (TS). Therefore, IC can be defined as a combined temporal and spatial state:

[0053] ic i ={(q si ,q ti )|q si ∈R ss ,q ti ∈R ts}

[0054] Where R ss R ts These represent the spatial and temporal state spaces, respectively.

[0055] Since the time-space domain is a non-Euclidean space, the distance between dFNPs in discriminant analysis is defined using either spatial fusion or state fusion. Spatial fusion measures SS and TS in two separate Riemannian spaces, followed by kernel fusion. State fusion projects SS and TS into a single Riemannian space through mapping and merging for measurement. Therefore, the fused dFNP distance can be defined as:

[0056]

[0057] Where f, g, and h are distance functions in different spaces.

[0058] This invention maps SS to a Grassmann manifold for geodesic distance measurement. TS requires further calculation of the temporal coherence matrix of IC (the partial correlation matrix of any two IC time series), and then maps it to a symmetric positive definite (SPD) manifold for geodesic distance measurement. The distance measurement matrices between all individuals in different manifolds are embedded into commonly used kernel functions, such as the radial basis function (RBF) kernel and the sigmoid kernel, as a custom kernel in the SVM model for linear fusion. The fusion ratio coefficient can be selected through cross-validation.

[0059] Based on the fused kernel function, an SVM classifier can be built between a specific disease and a normal control group. Since different brain networks (independent components) have different pattern representation capabilities, classifiers built based on different combinations of brain networks have different classification performances. Therefore, based on an efficient forward search strategy and its classifier performance, the most discriminative combination of brain networks is selected from all independent components.

[0060] Efficient forward search strategies are beneficial for obtaining results quickly when dealing with a large number of independent components and high computational demands. Figure 3 The feature search and analysis flowchart according to the present invention is as follows:

[0061] 1) Assume the dataset has M subjects, and ICA yields k corresponding independent components. For each independent component ic i (i = 1, ..., k), the set of independent components corresponding to M subjects. A classifier C can be built. i Cross-validation can yield results based on dFNP. i =span(ic i The temporal-spatial joint state of the classifier is obtained, and a Riemann manifold kernel fusion SVM classifier is constructed and its classification performance is estimated (the classifier performance can be evaluated by indicators such as the area under the receiver operating characteristic curve and the classification accuracy).

[0062] 2) Based on the estimated classifier performance, adjust the classifier C according to the numerical values. i Arrange in descending order to obtain the sorted independent components ic i '(i=1,…k) reflects the classification performance of each independent component on the dataset in the time-space domain.

[0063] 3) Sequentially install ic i The first n (n=1,…k) ordered independent components are used as dFNPs to construct a Riemannian manifold kernel fusion SVM classifier C. n ', and estimate classification performance. From C nThe dFNP corresponding to the classifier with the best classification performance is selected as the most discriminative dynamic functional network pattern. In this way, only 2k-1 classifiers are needed to obtain the most discriminative specific brain network combination.

[0064] 6. Using the identified specific brain functional network combinations, a time-space fusion discriminant model is constructed to output classification posterior probability values, thereby quantitatively measuring the dynamic state of brain functional networks. The quantified feature values ​​are then correlated with widely used clinical behavioral scales to discuss whether there is a correlation between intrinsic brain networks and epibehavioral behavior, and to further analyze their significance and impact.

[0065] The above description uses the Grassman and SPD manifolds as examples and the SVM discriminant model as an example to illustrate the method of this invention. However, this invention can also be used to construct other machine learning models on other manifolds for dynamic brain functional network analysis.

[0066] Example 2

[0067] This embodiment provides a dynamic brain functional network analysis system based on the joint state of the time domain and the spatial domain.

[0068] A dynamic brain functional network analysis system based on joint temporal and spatial states includes:

[0069] The data acquisition and preprocessing module is configured to acquire four-dimensional resting-state functional magnetic resonance imaging data of patients with specific neuropsychiatric diseases and normal control groups, and perform preprocessing.

[0070] The dynamic brain functional connectivity matrix construction module is configured to: in the time dimension, extract preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments according to a preset sliding window, and calculate the Pearson correlation coefficient of time-series signals between any two brain regions within the window segment to obtain the dynamic brain functional connectivity matrix.

[0071] The time-space joint state module is configured to extract the dynamic brain functional connectivity matrix using independent component analysis to obtain the individual's corresponding independent components and the time series corresponding to the independent components.

[0072] The analysis module is configured to: based on the individual's corresponding independent components and the time series corresponding to the independent components, use an efficient forward search strategy and classifier to construct a time-space fusion discrimination model, identify specific brain dynamic networks related to specific neuropsychiatric diseases, and output decision values ​​to achieve quantitative measurement of dynamic brain functional networks at the individual level.

[0073] It should be noted that the data acquisition and preprocessing module, the dynamic brain functional connectivity matrix construction module, the time-space joint state module, and the analysis module described above are the same examples and application scenarios implemented in the steps of Embodiment 1, but are not limited to the content disclosed in Embodiment 1. It should also be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0074] Example 3

[0075] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the dynamic brain functional network analysis method based on time-space joint state as described in Embodiment 1 above.

[0076] Example 4

[0077] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the dynamic brain functional network analysis method based on time-space joint state as described in Embodiment 1 above.

[0078] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0079] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0080] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0082] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A dynamic brain functional network analysis method based on joint temporal and spatial states, characterized in that, include: Acquire four-dimensional resting-state functional magnetic resonance imaging data of patients with specific neuropsychiatric diseases and normal controls, and perform preprocessing; In the time dimension, preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments are extracted according to a preset sliding window, and the Pearson correlation coefficient of temporal signals between any two brain regions within the window segment is calculated to obtain the dynamic brain functional connectivity matrix. Independent component analysis was used to extract the dynamic brain functional connectivity matrix to obtain the individual independent components and the time series corresponding to the independent components. Based on individual independent components and their corresponding time series, a time-space fusion discrimination model is constructed using an efficient forward search strategy and classifier to identify specific dynamic brain networks associated with specific neuropsychiatric diseases and output decision values, thereby achieving quantitative measurement of dynamic brain functional networks at the individual level. The process of using a classifier specifically includes: From the perspective of multivariate data representation, the independent components of individuals obtained by independent component analysis and the corresponding time series of these components are constructed as a subspace spanned by the joint time-domain and spatial-domain states as basis functions. This subspace represents the dynamic brain functional network pattern. dFNP ; Assuming independent components ,but dFNP Defined as: IC is defined as a joint state in the time and space domains: in Represent the spatial and temporal state spaces, respectively; Defined in discriminant analysis dFNP The distance between them is determined by either spatial fusion or state fusion. Spatial fusion measures the spatial and temporal states of an individual in two separate Riemannian manifolds, followed by kernel fusion. State fusion projects the spatial and temporal states of an individual into a single Riemannian manifold through mapping and merging for measurement. Thus, the fused distance... dFNP Distance is defined as: in f, g, h These are distance functions in different spaces; The spatial state of an individual is mapped onto a Grassmann manifold for geodesic distance measurement. The temporal state of an individual requires further calculation of the temporal cooperability matrix of IC, which is then mapped onto a symmetric positive definite manifold for geodesic distance measurement. The distance measurement matrices between all individuals in different manifolds are embedded into the kernel function and used as a custom kernel in the classifier for linear fusion. Based on the fused kernel function, a classifier is established between patients with specific neuropsychiatric diseases and normal control groups.

2. The dynamic brain functional network analysis method based on time-space joint state as described in claim 1, characterized in that, The preprocessing includes: temporal correction, head motion correction, registration, spatial normalization, spatial smoothing, and filtering.

3. The dynamic brain functional network analysis method based on time-space joint state as described in claim 1, characterized in that, The step of extracting preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments according to a preset sliding window in the time dimension specifically includes: The preprocessed four-dimensional resting-state functional magnetic resonance imaging data is divided into multiple brain regions according to the brain atlas template, and the average time series of voxels contained in each brain region is extracted. The average time series is then segmented according to a preset sliding window.

4. The dynamic brain functional network analysis method based on time-space joint state as described in claim 1, characterized in that, The extraction of the dynamic brain functional connectivity matrix using independent component analysis to obtain the individual's corresponding independent components and the time series corresponding to the independent components specifically includes: By concatenating all individual dynamic brain function connectivity matrices in the time domain, three-dimensional temporal network data is obtained. Independent component analysis was used to generate groups of independent components from the three-dimensional temporal network data. By reconstructing the group's independent components to a single individual, the individual's corresponding independent components and time series are obtained. The independent components correspond to the spatial functional connectivity state of the dynamic brain functional network, and the time series corresponds to the temporal fluctuations of the spatial functional connectivity state.

5. The dynamic brain functional network analysis method based on time-space joint state as described in claim 1, characterized in that, The efficient forward search strategy specifically includes: 1) Assuming the dataset has M subjects, and the independent component analysis method yields k corresponding independent components, for each independent component... The set of independent components corresponding to M subjects Build a classifier The results were obtained through cross-validation. The temporal-spatial joint state is obtained, and a Riemannian manifold kernel-based fusion classifier is constructed and its classification performance is estimated; among them, ; ; 2) Based on the estimated classifier performance, evaluate the classifier according to the numerical values. Arrange in descending order to obtain the sorted independent components. ;in, ; 3) In sequence The former An ordered, independent component was used as a dynamic brain function pattern to construct a Riemannian manifold fusion SVM classifier. And estimate classification performance; from The dynamic brain function pattern corresponding to the classifier with the best classification performance was selected as the most discriminative dynamic brain function network pattern; among them, .

6. The dynamic brain functional network analysis method based on time-space joint state according to claim 1, characterized in that, The process of outputting decision values ​​to achieve quantitative measurement of dynamic brain functional networks at the individual level specifically includes: constructing a time-space fusion discrimination model using the identified specific brain functional network combinations, outputting classification posterior probability values, and achieving quantitative measurement of the state of dynamic brain functional networks.

7. A dynamic brain functional network analysis system based on time-space joint state, employing the dynamic brain functional network analysis method based on time-space joint state as described in claim 1, characterized in that, include: The data acquisition and preprocessing module is configured to acquire four-dimensional resting-state functional magnetic resonance imaging data of patients with specific neuropsychiatric diseases and normal control groups, and perform preprocessing. The dynamic brain functional connectivity matrix construction module is configured to: in the time dimension, extract preprocessed four-dimensional resting-state functional magnetic resonance imaging data segments according to a preset sliding window, and calculate the Pearson correlation coefficient of time-series signals between any two brain regions within the window segment to obtain the dynamic brain functional connectivity matrix. The time-space joint state module is configured to extract the dynamic brain functional connectivity matrix using independent component analysis to obtain the individual's corresponding independent components and the time series corresponding to the independent components. The analysis module is configured to: based on the individual's corresponding independent components and the time series corresponding to the independent components, use an efficient forward search strategy and classifier to construct a time-space fusion discrimination model, identify specific brain dynamic networks related to specific neuropsychiatric diseases, and output decision values ​​to achieve quantitative measurement of dynamic brain functional networks at the individual level.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the dynamic brain functional network analysis method based on the joint state of time and space as described in any one of claims 1-6.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the dynamic brain functional network analysis method based on the joint state of time and space as described in any one of claims 1-6.

Images