CN109035265A - A kind of novel structure brain connection map construction method - Google Patents
A kind of novel structure brain connection map construction method Download PDFInfo
- Publication number
- CN109035265A CN109035265A CN201811125843.9A CN201811125843A CN109035265A CN 109035265 A CN109035265 A CN 109035265A CN 201811125843 A CN201811125843 A CN 201811125843A CN 109035265 A CN109035265 A CN 109035265A
- Authority
- CN
- China
- Prior art keywords
- brain
- map
- connection map
- area
- construction method
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30016—Brain
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Image Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A kind of novel structure brain connection map construction method, brain connection map research are the guiding key links of brain function and cerebral disease research, how to construct and verify hot spot and forward position that structure brain connection map has become current brain science research.Recent studies have shown that, the construction method of existing structure brain connection map still has several drawbacks: the robustness of the hypothesiss driving method based on model is high, independent of model the correctness based on mind map spectral method be difficult to verify, that the full brain mesostructure brain based on data-driven method building connects map is not fine etc. enough.In order to overcome the shortcomings of existing methods, the invention proposes methods that is a kind of novel and being easy to verify --- and it using Cooperative Clustering algorithm while identifying the structure sub-district of some specific brain regions structure and its connecting brain area, realize the structure brain connection map construction of fining;Two kinds of approach of different angle are taken to demonstrate the correctness of structure brain connection map simultaneously.
Description
Technical field
The invention belongs to fields of biomedicine, are related to a kind of brain connection map construction method based on Cooperative Clustering, especially
It is to be related to a kind of dispersion tensor image brain connection map construction method based on Cooperative Clustering.
Background technique
From in the 1990s, the connection map construction research of dispersion tensor image brain, which becomes, explores fiber knot between brain area
The important channel of structure connection relationship.It can be seen that the connection map construction research of structure brain is that brain physics dissection connection relationship is visited
One basis of rope, currently has become hot spot and the forward position of brain science research.Summary and induction research both domestic and external, the connection of structure brain
Map construction the method that is, based on seed region, the method based on brain map and is based on data-driven there are mainly three types of method
Method.
Method based on seed region is a kind of hypothesis driving method dependent on model.It is false according to the priori to study a question
If selecting brain area as seed region, then track and identify that there are the brain areas that fiber is connect with seed region, to complete seed
The building of region brain area structure connection map.In the connection map research of structure brain, the advantages of method based on seed region, is
It is easy to use, be easy to explain and be widely used.But there is also deficiency, the robusts of the structure brain connection map of such as building for this method
Property is higher to the spatial position of seed region and the dependence of size.
Method based on brain map is a kind of method of building structure brain connection map independent of model.This method will
Brain area in brain map is as node, and the weight on side is based on structure connection definition between every two node, is then examined using module
By all node divisions into each network module, node and its interconnection in each network module are just constituted method of determining and calculating
One modular structure brain network, i.e. structure brain connect map.In addition, realizing the hinge section of global communication between network module
Point and its connection are similarly an organic component of structure brain connection map.Although the method based on brain map is domestic and international
Researcher relatively broadly for brain connection group in the brain science research of means, but the used brain map of this method
Source and brain area quantity directly affect the space topological feature of constructed structure brain connection map (or structure brain network)
It is explained with Neurobiology.
Method based on data-driven is another method independent of the building structure brain connection map of model, main
It include the method based on independent component analysis, the method based on cluster.Full brain is divided into multiple sub-networks by this method, each
Brain area in a sub-network and its a part for being interconnected to constitute brain structure connection map.Although above-mentioned be based on data-driven
Method constructed between two seed methods or between different data collection structure brain connection map all have higher robust
Property and interpretation, and promote the brain science research with brain connection group for means, but it is existing based on data-driven
Method can't obtain the structure connection map of some specific brain regions fine structure.
There is also clearly disadvantageous: the method pair based on seed region for the structure brain connection map of existing various method buildings
The selection of seed region is more sensitive;Method based on brain map depends on selected brain map;Side based on data-driven
For method although the defect of first two method is not present, the existing method based on data-driven is only limitted to the knot on full brain scale
Structure brain connects map construction.
Summary of the invention
For the deficiency of existing method, it is an object of the invention to propose a kind of to be different from traditional novelty and be easy to verify
The method based on data-driven, come solve some specific brain regions fine structure structure connect map Construct question.
To achieve the goals above, the present invention identifies the structon of some specific brain regions structure using Cooperative Clustering algorithm simultaneously
Area and its connection brain area, i.e., the structure that brain fine structure is investigated in building connect map.The technical solution of the invention such as Fig. 1 institute
Show, is described in detail below.
In the structure brain connection map Cooperative Clustering construction method proposed by the present invention based on dispersion tensor image, preferentially
Selecting research obtains more mature thalamus, cerebellum, Basal ganglia etc. as investigation brain structure.Map construction is connected with the structure of thalamus
For, the technology path that the present invention takes such as Fig. 1.Firstly, brain structure thalamus (Fig. 1 (a)) to be investigated is extracted, it can be by automatic
Partitioning algorithm extracts.Then, each voxel of thalamus and its function major loop (brain skin are calculated using fibre-tracking algorithm
Layer) probability connection between each voxel, the structure connection matrix between thalamus and cerebral cortex major loop is rearranged, referred to as
The main ring line structure connection matrix (Fig. 1 (b)) of thalamus voxel.Finally, based on Cooperative Clustering to the main ring line structure of thalamus voxel
Connection matrix is clustered simultaneously along row and column direction, and the main ring line structure connection of thalamus voxel is rearranged according to cluster classification
Matrix (Fig. 1 (c)), Fig. 1 (d) are that the thalamic structures constructed based on Cooperative Clustering connect map schematic diagram, the i.e. mound of clustering recognition
Brain structure sub-district and its connection brain area.
It, can be on the basis of carrying out structure brain connection map to the brains structure such as thalamus, cerebellum, Basal ganglia and being built into function
It is gradually generalized to the structure connection map construction research of other brain structures, and then is constantly improve proposed by the present invention poly- based on collaboration
The structure brain connection map construction method of class simultaneously makes it have universality.
In addition, being verified an always problem to the correctness of the structure brain connection map of building.The present invention takes
Proof scheme be described as follows.The present invention takes the verifying approach of two kinds of different angles.Firstly, selection brain subregion and sub-district connection
The brain structure (such as thalamus, cerebellum, Basal ganglia) that brain area is studied more mature is verified.Secondly as structure connection is
There is reasons, the present invention such as homology will be based on tracer technique institute for the approximation of dissection connection and the brain of estimation, human and animal
The animal brain dissection connection map of acquisition connects map as goldstandard, to verify the structure of mankind's brain of building.
Detailed description of the invention
Fig. 1 is the technical side of the dispersion tensor image brain connection map construction method proposed by the present invention based on Cooperative Clustering
Case.
Fig. 2 is the 7 thalamuses segmentation sub-district obtained using proposition method of the present invention.
Specific embodiment
The invention proposes a kind of, and the dispersion tensor image brain based on Cooperative Clustering connects map construction method, this method
Technical solution is as shown in Figure 1.It is configured to example with the structure connection map of thalamus, specific embodiment is described as follows.
(1) acquisition obtains dispersion tensor image data.The dispersion tensor image of 1 subject, space point are collected
Resolution and disperse sensitising gradient direction number are respectively 1.25 × 1.25 × 1.25 cubic millimeters and 270.
(2) brain structure thalamus and its function major loop (cerebral cortex in addition to Reil's island) to be investigated are extracted.It uses
FreeSurfer automatically extracts the thalamus and cerebral cortex of this subject.
(3) structure connection matrix is calculated.The each voxel of thalamus and its function main ring are calculated using probtrackx2 tool
Probability connection between road (cerebral cortex) each voxel, using the connection probability as main ring line structure connection matrix element
Value.
(4) building structure brain connects map.Using Cooperative Clustering algorithm to ipsilateral thalamus and its cerebral cortex function master
The structure connection matrix of loop is clustered.Pass through the structure sub-district and its connection brain area of clustering recognition thalamus, i.e. realization thalamus
Structure connect map construction.
(5) building obtains the structure connection map of thalamus.Thalamus is divided into 7 structure sub-districts, including thalamus temporo antinion
Area, thalamus abdomen medial prefrontal cortical area, cortex of frontal lobe area, the area thalamus E Ding, thalamus motor area, thalamus default net on the outside of thalamus abdomen
Network area and thalamus visual area.
(6) thalamic structures of verifying building connect map.On the outside of temporo antinion network, abdomen medial prefrontal cortical networks, abdomen
Cortex of frontal lobe network, volume top network, movement network, default network, the corresponding thalamic structures sub-district being connected of visual web, are divided
It is not named as thalamus temporo antinion area, thalamus abdomen medial prefrontal cortical area, cortex of frontal lobe area, thalamus E Dingqu, mound on the outside of thalamus abdomen
Brain motor area, thalamus default network area and thalamus visual area (Fig. 2).
Claims (6)
1. a kind of novel structure brain connects map construction method, which is characterized in that the method is a kind of poly- based on collaboration
The dispersion tensor image brain of class connects map construction method.
2. a kind of novel structure brain according to claim 1 connects map construction method, which is characterized in that the side
Method can construct more robust structure brain connection map in group level.
3. a kind of novel structure brain according to claim 1 connects map construction method, which is characterized in that the side
Method can not only construct the structure connection map of full brain, but also can carry out structure to some specific brain regions structure and connect map structure
It builds.
4. according to claim 3 carry out structure connection map Construct question to some specific brain regions structure, which is characterized in that
The method can cut the structon differentiation that some specific brain regions structure is refined, and identify the knot for dividing sub-district simultaneously
Structure connects brain area.
5. the structon area segmentation problem according to claim 4 refined to some specific brain regions structure, feature
Be, for study comparative maturity brain structure, the function of such brain segmentation of structures sub-district can connect according to its structure
Brain area is connect further to be verified.
6. the structon area segmentation problem according to claim 4 refined to some specific brain regions structure, feature
It is, for unknown brain structure or studies to obtain also jejune brain structure, the function of such brain segmentation of structures sub-district can
Speculated with connecting brain area according to its structure.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811125843.9A CN109035265B (en) | 2018-09-26 | 2018-09-26 | Structural brain connection map construction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811125843.9A CN109035265B (en) | 2018-09-26 | 2018-09-26 | Structural brain connection map construction method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109035265A true CN109035265A (en) | 2018-12-18 |
CN109035265B CN109035265B (en) | 2020-10-27 |
Family
ID=64618256
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811125843.9A Active CN109035265B (en) | 2018-09-26 | 2018-09-26 | Structural brain connection map construction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109035265B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101650827A (en) * | 2009-09-09 | 2010-02-17 | 西北工业大学 | Mixed brain white matter nerve fiber automatic cluster and marking method |
CN106023194A (en) * | 2016-05-18 | 2016-10-12 | 西安交通大学 | Amygdaloid nucleus spectral clustering segmentation method based on resting state function connection |
CN106650818A (en) * | 2016-12-30 | 2017-05-10 | 太原理工大学 | Resting state function magnetic resonance image data classification method based on high-order super network |
CN107146228A (en) * | 2017-03-22 | 2017-09-08 | 东南大学 | A kind of super voxel generation method of brain magnetic resonance image based on priori |
CN107330267A (en) * | 2017-06-28 | 2017-11-07 | 首都医科大学宣武医院 | White matter fiber brain atlas construction method using diffusion tensor imaging medical image |
CN107658018A (en) * | 2017-10-12 | 2018-02-02 | 太原理工大学 | A kind of fusion brain network establishing method based on structure connection and function connects |
CN108182473A (en) * | 2017-12-12 | 2018-06-19 | 中国科学院自动化研究所 | Full-dimension distributed full brain modeling system based on class brain impulsive neural networks |
-
2018
- 2018-09-26 CN CN201811125843.9A patent/CN109035265B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101650827A (en) * | 2009-09-09 | 2010-02-17 | 西北工业大学 | Mixed brain white matter nerve fiber automatic cluster and marking method |
CN106023194A (en) * | 2016-05-18 | 2016-10-12 | 西安交通大学 | Amygdaloid nucleus spectral clustering segmentation method based on resting state function connection |
CN106650818A (en) * | 2016-12-30 | 2017-05-10 | 太原理工大学 | Resting state function magnetic resonance image data classification method based on high-order super network |
CN107146228A (en) * | 2017-03-22 | 2017-09-08 | 东南大学 | A kind of super voxel generation method of brain magnetic resonance image based on priori |
CN107330267A (en) * | 2017-06-28 | 2017-11-07 | 首都医科大学宣武医院 | White matter fiber brain atlas construction method using diffusion tensor imaging medical image |
CN107658018A (en) * | 2017-10-12 | 2018-02-02 | 太原理工大学 | A kind of fusion brain network establishing method based on structure connection and function connects |
CN108182473A (en) * | 2017-12-12 | 2018-06-19 | 中国科学院自动化研究所 | Full-dimension distributed full brain modeling system based on class brain impulsive neural networks |
Non-Patent Citations (3)
Title |
---|
SINGHAI JYOTI,ET AL: "《Collaborative rough-fuzzy clustering: An application to intensity nonuniformity correction in brain MR images》", 《IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS》 * |
姚旭峰,等: "《磁共振弥散张量成像纤维素束踪算法的研究进展》", 《综述与进展》 * |
许东峰,等: "《基于DTI丘脑分割的新方法》", 《中国CT和MIR杂志》 * |
Also Published As
Publication number | Publication date |
---|---|
CN109035265B (en) | 2020-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103886328B (en) | Based on the functional magnetic resonance imaging data classification method of brain mixed-media network modules mixed-media architectural feature | |
Szathmary et al. | Are the Biological Differences Between North American Indians and Eskimos Truly Profound?[and Comments and Reply] | |
CN109102492A (en) | A kind of functional magnetic resonance imaging brain connection map construction method based on Cooperative Clustering | |
Wig et al. | Concepts and principles in the analysis of brain networks | |
CN107392907A (en) | Parahippocampal gyrus function division method based on tranquillization state FMRI | |
WO2020103683A1 (en) | Method and system for individualized prediction of mental illness on basis of brain function map monkey-human cross-species migration | |
CN103345749B (en) | A kind of brain network function connectivity lateralization detection method based on modality fusion | |
CN108229066A (en) | A kind of Parkinson's automatic identifying method based on multi-modal hyper linking brain network modelling | |
CN107944472B (en) | A kind of airspace operation situation calculation method based on transfer learning | |
CN106407277A (en) | Internet of vehicles data-based attribute analysis method for vehicle owner parking point after being clustered | |
CN102509123A (en) | Brain functional magnetic resonance image classification method based on complex network | |
CN113592836A (en) | Deep multi-modal graph convolution brain graph classification method | |
Qi et al. | Parallel group ICA+ ICA: Joint estimation of linked functional network variability and structural covariation with application to schizophrenia | |
CN105117731A (en) | Community partition method of brain functional network | |
CN102496159A (en) | Encephalic region causal connection detection method combining functional magnetic resonance imaging (FMRI) and magnetoencephalography (MEG) | |
CN106650818A (en) | Resting state function magnetic resonance image data classification method based on high-order super network | |
CN109582754A (en) | The method for carrying out urban subject functional areas central detector using POI data | |
CN112418337A (en) | Multi-feature fusion data classification method based on brain function hyper-network model | |
CN110223275A (en) | A kind of cerebral white matter fiber depth clustering method of task-fMRI guidance | |
CN107301409A (en) | Learn the system and method for processing electrocardiogram based on Wrapper feature selectings Bagging | |
CN107590766A (en) | A kind of method of discrimination of the land used combination form related to road traffic accident risk | |
CN109671055A (en) | Pulmonary nodule detection method and device | |
Kaneko et al. | Spatial organization of occipital white matter tracts in the common marmoset | |
D’Souza et al. | A matrix autoencoder framework to align the functional and structural connectivity manifolds as guided by behavioral phenotypes | |
CN110287995A (en) | The multiple features learning network model method of whole day overhead transport jam situation classification |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |