CN112580468A - Method for detecting time-stable community in brain function network - Google Patents
Method for detecting time-stable community in brain function network Download PDFInfo
- Publication number
- CN112580468A CN112580468A CN202011450671.XA CN202011450671A CN112580468A CN 112580468 A CN112580468 A CN 112580468A CN 202011450671 A CN202011450671 A CN 202011450671A CN 112580468 A CN112580468 A CN 112580468A
- Authority
- CN
- China
- Prior art keywords
- brain
- community
- node
- network
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000003925 brain function Effects 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 23
- 210000004556 brain Anatomy 0.000 claims abstract description 81
- 230000006870 function Effects 0.000 claims abstract description 32
- 238000007711 solidification Methods 0.000 claims abstract description 17
- 230000008023 solidification Effects 0.000 claims abstract description 17
- 230000001537 neural effect Effects 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 36
- 239000013598 vector Substances 0.000 claims description 22
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000003064 k means clustering Methods 0.000 claims description 4
- 238000005192 partition Methods 0.000 claims description 4
- 238000007596 consolidation process Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000002407 reforming Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 230000008521 reorganization Effects 0.000 claims description 2
- 238000011160 research Methods 0.000 abstract description 7
- 238000007621 cluster analysis Methods 0.000 abstract 1
- 230000003920 cognitive function Effects 0.000 description 2
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 description 2
- 230000006641 stabilisation Effects 0.000 description 2
- 238000011105 stabilization Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006998 cognitive state Effects 0.000 description 1
- 230000036992 cognitive tasks Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002610 neuroimaging Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for detecting a time-stable community in a brain function network, which mainly comprises the following steps: obtaining preprocessed neural image or electroencephalogram data, defining nodes of a brain function network and extracting a neural signal time sequence of each node; constructing a brain dynamic function network by adopting a sliding time window method; performing cluster analysis on the brain dynamic function network sequence to obtain functional connections under a plurality of brain states, and thus obtaining a switching relation among different brain states; performing optimal community division on the functional network in each brain state, wherein each node of the brain functional network is distributed into a certain community; calculating the community solidification degree of each group of node pairs of the brain function network according to the community division result, and extracting all the node pairs with the community solidification degree equal to 1 to obtain the time-stable community of the brain function network; the method is low in time complexity, and can be used as a basis for the follow-up research of time-stable communities in the brain function network.
Description
Technical Field
The invention belongs to the technical field of biomedical information processing, and particularly relates to a method for detecting a time-stable community in a brain function network.
Background
Complex network science has been widely used to study the mechanisms of cognitive function of the brain. When the brain processes cognitive tasks, brain areas which are coordinated with each other form close functional connection, and a community structure of a brain function network is formed. Within communities, functional connections between brain regions are tight and strong, while functional connections between communities are relatively sparse and weak.
In order to adapt to different cognitive states, the strength of functional connection between different brain areas can be dynamically changed, so that the community membership of the brain areas is changed, and the reorganization of the brain function network community structure is caused. Most of the existing researches focus on the research on the relationship between the structural change of the brain function network community and the cognitive function, and in the researches of other fields of complex networks, the time-stable community has attracted the attention of researchers. The time-stable community refers to a relatively stable community structure formed in a complex network along with the evolution of time, and the community structure cannot be damaged in the reforming process of the network community structure. To study the time-stable communities in the brain function network, the time-stable communities need to be detected and identified first.
Disclosure of Invention
In view of the above-mentioned needs, the present invention is to solve the following problems:
the method for detecting the time-stable community in the brain function network is provided, the detection and the identification of the time-stable community in the brain function network are realized, and a basis is provided for the further research and analysis of the time-stable community in the brain function network.
In order to achieve the purpose, the invention adopts the following technical scheme:
a method for detecting time-stable communities in a brain function network comprises the following steps:
1) acquiring preprocessed neural images or electroencephalogram data, and defining nodes of a brain function network according to a standardized brain partition template or an electroencephalogram channel position;
2) extracting a neural signal time sequence of each node of the brain functional network;
3) dividing a time sequence of each node of the brain functional network into a plurality of subsections which are mutually overlapped and have the same length by adopting a sliding time window method, and constructing a brain dynamic functional network which changes along with time according to a Pearson correlation or mutual information network construction method;
4) reforming the upper triangular element of each brain dynamic function network matrix, which does not contain diagonal elements, into a column vector, and integrating all obtained column vectors into a new function connection matrix FCv according to the time sequence, wherein the matrix contains function connection information of the brain dynamic function network at all times;
5) adopting a K-means clustering algorithm to perform clustering analysis on the FCv matrix, clustering all column vectors of the FCv matrix into a plurality of classes according to the dunne score of a clustering result, wherein each class corresponds to functional connection in one brain state, the column vectors of the FCv matrix belong to one brain state respectively, and obtaining a switching relation between the brain states according to the arrangement sequence of the column vectors in the FCv matrix;
6) calculating the average value of all FCv matrix column vectors contained in each brain state, and restoring the obtained average value column vectors into a square matrix form, namely obtaining a functional connection network in each brain state;
7) performing optimal community division on the function connection network in each brain state, wherein the community division quality is obtained by a modularity function Q:
wherein m represents the number of connecting edges of the brain function connecting network, AijRepresents the strength of the connecting edge between the network node i and the node j, kiRepresents the degree of the node i, if the node i and the node j belong to the same community, delta(s)i,sj) 1, otherwise δ(s)i,sj)＝0；
8) Evaluating the community division result of the functional connection network under each brain state by using a modularity function Q, and finally adopting a community division scheme with the maximum Q value, wherein each node of the functional connection network under each brain state is distributed to a certain community;
9) calculating all brain states by community solidityIn the event switching, one node is the proportion of switching events belonging to the same community. Degree of consortium consolidation SDijThe expression is as follows:
wherein R represents the number of brain state pairs with switching relationship, and C if the node i and the node j belong to the same community in the brain state pi,pCj,p1, otherwise Ci,pCj,p0; SD if node i and node j belong to the same community in all brain state switching eventsij＝1；
10) The community solidification degree matrix of the brain function connection network is obtained by calculating the community solidification degree of each group of node pairs of the brain function connection network, and each element of the matrix represents the community solidification degree of one node pair; and extracting all node pairs with the community solidification degree equal to 1 to obtain the time-stable community in the brain function network.
The method comprises the steps of firstly constructing a brain dynamic function network sequence by adopting a sliding time window method, then extracting upper triangular elements of each function network and arranging the upper triangular elements according to a time sequence to form a new function connection matrix FCv, then classifying FCv column vectors by adopting a K-means clustering algorithm to obtain a plurality of function connection networks in a brain state and a mutual switching relation of the function connection networks along with time, then carrying out optimal community division on the function connection networks in each brain state, finally calculating community solidification degree of each group of node pairs of the brain function networks according to community division results, and extracting all node pairs with community solidification degree equal to 1 to obtain a time stable community of the brain function network.
The invention has the following beneficial effects:
1) compared with the traditional brain dynamic function network research method, the method provided by the invention has the advantages that the upper triangular elements of the brain dynamic function network matrix are firstly extracted before the clustering algorithm is applied, and the time complexity of the subsequent clustering algorithm operation is favorably improved.
2) The time stabilization community of the brain function network is extracted through the community solidification degree, and a foundation can be made for the follow-up research of the time stabilization community in the brain function network.
Drawings
FIG. 1 is a flow chart of an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples. The examples are provided for the purpose of illustration only and are not intended to limit the scope of the invention.
Example (b): a specific embodiment of a method for detecting a time-stable community in a brain function network comprises the following steps:
1) obtaining preprocessed neural image or electroencephalogram data, and defining nodes of a brain function network according to a standardized brain partition template or an electroencephalogram channel position. In this embodiment, functional nuclear magnetic resonance imaging data is adopted, an AAL brain partition template is selected, and a human brain is divided into 90 brain areas, that is, a brain functional network includes 90 nodes;
2) calculating the time sequence average value of all voxels contained in each brain region in functional nuclear magnetic resonance imaging, and extracting neural signal time sequences of 90 nodes of a brain functional network;
3) the present embodiment includes 110 neuroimaging data to be tested. And dividing the time sequence of each node of all tested brain function networks into a plurality of sub-sections which are mutually overlapped and have the same length by adopting a sliding time window method. In this embodiment, the total length of the time series of each brain region is 152, the length of the selected time window is 30, the interval step length of the window is 1, and each subject obtains 122 time series subsections in total;
4) and calculating the Pearson correlation coefficient of the time sequences of any two brain areas in each tested time sequence subsection to construct a brain dynamic function network which changes along with time. In this embodiment, a total of 122 × 110 — 13420 brain function networks are obtained;
5) the upper triangular elements of each brain dynamic function network matrix not containing diagonal elements are reformed into a column vector, and all the obtained column vectors are integrated into a new function connection matrix FCv in time order. In the embodiment, the size of the matrix FCv is 3960 × 13420, and the matrix contains the functional connection information of all tested brain dynamic function networks at all times;
6) and (3) carrying out clustering analysis on the FCv matrix by adopting a K-means clustering algorithm, clustering all column vectors of the FCv matrix into a plurality of classes according to the dunne score of a clustering result, wherein each class corresponds to functional connection in one brain state, the column vectors of the FCv matrix belong to one brain state respectively, and obtaining the switching relation among the brain states according to the arrangement sequence of the column vectors in the FCv matrix. In the present embodiment, 10 kinds of brain states, 28 kinds of brain state switching relations are obtained;
7) calculating the average value of all FCv matrix column vectors contained in each brain state, and restoring the obtained average value column vectors into a square matrix form, namely obtaining a functional connection network in each brain state;
8) performing optimal community division on the function connection network in each brain state, wherein the community division quality is obtained by a modularity function Q:
wherein m represents the number of connecting edges of the brain function connecting network, AijRepresents the strength of the connecting edge between the network node i and the node j, kiRepresents the degree of the node i, if the node i and the node j belong to the same community, delta(s)i,sj) 1, otherwise δ(s)i,sj)＝0；
9) Evaluating the community division result of the functional connection network under each brain state by using a modularity function Q, and finally adopting a community division scheme with the maximum Q value, wherein each node of the functional connection network under each brain state is distributed to a certain community;
10) and calculating the proportion of one node to switching events belonging to the same community in all brain state switching events by adopting community solidity. Degree of consortium consolidation SDijThe expression is as follows:
in the formula, R represents the number of brain state pairs with switching relationship, and if the node i and the node j belong to the same community in the brain state p, Ci,pCj,p1, otherwise Ci,pCj,pIf node i and node j belong to the same community in all brain state switching events, SD is 0ij＝1；
11) The community solidification degree matrix of the brain function connection network is obtained by calculating the community solidification degree of each group of node pairs of the brain function connection network, and each element of the matrix represents the community solidification degree of one node pair; and extracting all node pairs with the community solidification degree equal to 1 to obtain the time-stable community of the brain function network. In this example, a total of 3 time-stable communities are obtained, including 6, and 13 brain regions, respectively.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are merely illustrative of the principles of the present invention, but that various changes and modifications may be made without departing from the spirit and scope of the invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (2)
1. A method for detecting time-stable communities in a brain function network is characterized in that: the method comprises the following steps:
1) acquiring preprocessed neural images or electroencephalogram data, and defining nodes of a brain function network according to a standardized brain partition template or an electroencephalogram channel position;
2) extracting a neural signal time sequence of each node of the brain functional network;
3) dividing a time sequence of each node of the brain functional network into a plurality of subsections which are mutually overlapped and have the same length by adopting a sliding time window method, and constructing a brain dynamic functional network which changes along with time according to a Pearson correlation or mutual information network construction method;
4) reforming the upper triangular element of each brain dynamic function network matrix, which does not contain diagonal elements, into a column vector, and integrating all obtained column vectors into a new function connection matrix FCv according to the time sequence, wherein the matrix contains function connection information of the brain dynamic function network at all times;
5) adopting a K-means clustering algorithm to perform clustering analysis on the FCv matrix, clustering all column vectors of the FCv matrix into a plurality of classes according to the dunne score of a clustering result, wherein each class corresponds to functional connection in one brain state, the column vectors of the FCv matrix belong to one brain state respectively, and obtaining a switching relation between the brain states according to the arrangement sequence of the column vectors in the FCv matrix;
6) calculating the average value of all FCv matrix column vectors contained in each brain state, and restoring the obtained average value column vectors into a square matrix form, namely obtaining a functional connection network in each brain state;
7) performing optimal community division on the function connection network in each brain state, wherein the community division quality is obtained by a modularity function Q:
wherein m represents the number of connecting edges of the brain function connecting network, AijRepresents the strength of the connecting edge between the network node i and the node j, kiRepresents the degree of the node i, if the node i and the node j belong to the same community, delta(s)i，sj) 1, otherwise δ(s)i，sj)＝0；
8) Evaluating the community division result of the functional connection network under each brain state by using a modularity function Q, and finally adopting a community division scheme with the maximum Q value, wherein each node of the functional connection network under each brain state is distributed to a certain community;
9) and calculating the proportion of one node to switching events belonging to the same community in all brain state switching events by adopting community solidity. Degree of consortium consolidation SDijThe expression is as follows:
wherein R represents the number of brain state pairs with switching relationship, and C if the node i and the node j belong to the same community in the brain state pi，pCj，p1, otherwise Ci，pCj，p0; SD if node i and node j belong to the same community in all brain state switching eventsij＝1；
10) The community solidification degree matrix of the brain function connection network is obtained by calculating the community solidification degree of each group of node pairs of the brain function connection network, and each element of the matrix represents the community solidification degree of one node pair; and extracting all node pairs with the community solidification degree equal to 1 to obtain the time-stable community in the brain function network.
2. The method as claimed in claim 1, wherein the method comprises detecting the time-stable community in the reorganization process of the brain function network community by using community solidity; the community solidity of one node pair is defined as the proportion of one node pair to switching events belonging to the same community in all brain state function connection switching events.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011450671.XA CN112580468A (en) | 2020-12-11 | 2020-12-11 | Method for detecting time-stable community in brain function network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011450671.XA CN112580468A (en) | 2020-12-11 | 2020-12-11 | Method for detecting time-stable community in brain function network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112580468A true CN112580468A (en) | 2021-03-30 |
Family
ID=75131174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011450671.XA Pending CN112580468A (en) | 2020-12-11 | 2020-12-11 | Method for detecting time-stable community in brain function network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112580468A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114521905A (en) * | 2022-01-25 | 2022-05-24 | 中山大学 | Electroencephalogram signal processing method and system based on synchronous connection characteristics |
| CN115644893A (en) * | 2022-09-27 | 2023-01-31 | 燕山大学 | Method for classifying electroencephalogram channel communities by means of regional electroencephalogram modeling and diagonal block model |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117731A (en) * | 2015-07-17 | 2015-12-02 | 常州大学 | Community partition method of brain functional network |
| CN109730678A (en) * | 2019-01-28 | 2019-05-10 | 常州大学 | A kind of multilayer cerebral function network module division methods |
| CN111513717A (en) * | 2020-04-03 | 2020-08-11 | 常州大学 | Method for extracting brain functional state |
-
2020
- 2020-12-11 CN CN202011450671.XA patent/CN112580468A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117731A (en) * | 2015-07-17 | 2015-12-02 | 常州大学 | Community partition method of brain functional network |
| CN109730678A (en) * | 2019-01-28 | 2019-05-10 | 常州大学 | A kind of multilayer cerebral function network module division methods |
| CN111513717A (en) * | 2020-04-03 | 2020-08-11 | 常州大学 | Method for extracting brain functional state |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114521905A (en) * | 2022-01-25 | 2022-05-24 | 中山大学 | Electroencephalogram signal processing method and system based on synchronous connection characteristics |
| CN115644893A (en) * | 2022-09-27 | 2023-01-31 | 燕山大学 | Method for classifying electroencephalogram channel communities by means of regional electroencephalogram modeling and diagonal block model |
| CN115644893B (en) * | 2022-09-27 | 2024-05-17 | 燕山大学 | Regional electroencephalogram modeling and diagonal block model electroencephalogram channel community classification method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111127441B (en) | Multi-modal brain image depression recognition method and system based on graph node embedding | |
| CN108446730B (en) | CT pulmonary nodule detection device based on deep learning | |
| Watanabe et al. | Disease prediction based on functional connectomes using a scalable and spatially-informed support vector machine | |
| Xing et al. | Convolutional neural network with element-wise filters to extract hierarchical topological features for brain networks | |
| CN112418337B (en) | Multi-feature fusion data classification method based on brain function hyper-network model | |
| CN112580468A (en) | Method for detecting time-stable community in brain function network | |
| CN105117731A (en) | Community partition method of brain functional network | |
| CN110598793A (en) | Brain function network feature classification method | |
| CN116503680B (en) | Brain image structured analysis and brain disease classification system based on brain atlas | |
| Xiao et al. | Alternating diffusion map based fusion of multimodal brain connectivity networks for IQ prediction | |
| CN113628201A (en) | Deep learning-based pathological section analysis method, electronic device and readable storage medium | |
| CN108596228B (en) | Brain function magnetic resonance image classification method based on unsupervised fuzzy system | |
| CN111325288B (en) | Clustering idea-based multi-view dynamic brain network characteristic dimension reduction method | |
| CN115272295A (en) | Dynamic brain function network analysis method and system based on time domain-space domain combined state | |
| CN107832656B (en) | Brain function state information processing method based on dynamic function brain network | |
| CN113786185A (en) | Static brain network feature extraction method and system based on convolutional neural network | |
| CN111325268A (en) | Image classification method and device based on multi-level feature representation and integrated learning | |
| CN108846407B (en) | Magnetic resonance image classification method based on independent component high-order uncertain brain network | |
| CN107256408B (en) | Method for searching key path of brain function network | |
| Laeli et al. | Tuberculosis detection based on chest x-rays using ensemble method with cnn feature extraction | |
| CN110335682B (en) | Real part and imaginary part combined complex fMRI data sparse representation method | |
| CN113516186A (en) | Modularized feature selection method for brain disease classification | |
| CN113673451A (en) | Graph volume module for extracting image features of tissue cytology pathology pieces | |
| CN106709921A (en) | Color image segmentation method based on space Dirichlet hybrid model | |
| Ye et al. | Alleviating feature confusion in cross-subject human activity recognition via adversarial domain adaptation strategy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210330 |
|
| RJ01 | Rejection of invention patent application after publication |