A Method for Extracting Tissue Fiber Bundle Structure Information Based on Adaptive Diffusion Basis Function Decomposition

A technology of function decomposition and structural information, applied in the analysis of materials, material analysis through resonance, magnetic resonance measurement, etc., can solve the problems of long imaging time, unable to obtain stable solutions, unable to effectively fit diffusion weighted signals, etc. high precision effect

Inactive Publication Date: 2011-12-28
HARBIN INST OF TECH
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Although the GMM model can well explain the diffusion phenomenon in the case of two or two previous fibers, it cannot effectively fit the diffusion-weighted signal
There are three main problems: 1. This method requires more diffusion-weighted images with different diffusion-weighted gradient directions (usually the number of directions is greater than 60), which will result in longer imaging time
2. There is a model selection problem in this method, that is, you must choose to fit a single Gaussian model or a Gaussian mixture model, or fit both at the same time and then choose one that can better explain the diffusion-weighted signal
3. In the case of more than two fiber bundles (ie N>2), no stable solution can be obtained
However, in this method, the construction of the diffusion basis function is to use a set of discrete tensors with fixed eigenvalues. Due to the differences in the diffusion characteristics of water molecules in different regions of the tissue, the use of tensors with fixed eigenvalues ​​to construct the diffusion basis function will reduce the accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Method for Extracting Tissue Fiber Bundle Structure Information Based on Adaptive Diffusion Basis Function Decomposition
  • A Method for Extracting Tissue Fiber Bundle Structure Information Based on Adaptive Diffusion Basis Function Decomposition
  • A Method for Extracting Tissue Fiber Bundle Structure Information Based on Adaptive Diffusion Basis Function Decomposition

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0035] Specific Embodiment 1: In this embodiment, a method for extracting tissue fiber bundle structure information based on adaptive diffusion basis function decomposition, the steps to realize the method are as follows:

[0036] 1. Construct a set of tensor basis functions of adaptive eigenvalues; the construction method is:

[0037] a. Evenly select M points on the unit hemisphere, and the coordinates of these points constitute M unit direction column vectors {g j ,j=1,2,...,M};

[0038] In the present embodiment, 260 points approximately uniformly distributed are selected on the unit hemisphere, that is, M=260; the specific selection method is to divide the seven-fold checkerboard ( figure 1 ) mosaic into the icosahedron ( figure 2), take 260 grid points on the upper hemisphere of all seven-fold checkerboards, and normalize them to form 260 unit direction vectors;

[0039] b. Let the tensor eigenvalue be λ=[λ 1 ,λ 2 ,λ 3 ], constructing a set of tensors in, ...

specific Embodiment approach 2

[0056] Specific implementation mode 2: In this implementation mode, a method for extracting tissue fiber bundle structure information based on adaptive diffusion basis function decomposition is described in detail by taking simulation data as an example:

[0057] The generation of the simulation data in this embodiment is to use the Gaussian mixture model method to simulate the situation that there are two bundles of fibers intersecting in the voxel and the angle between the two bundles of fibers is 90 degrees. The three eigenvalues ​​of the diffusion tensor [λ s1 ,λ s2 ,λ s3 ] is set according to the following rules: λ s1 Take 0.5×10 -3 to 1.8×10 -3 mm 2 A random value between / s, another λ s2 =λ s3 And the range is 0.2~0.4×λ s1 The random value of ; the number of diffusion gradient wave vectors in this simulation data is set to 30, that is, q k take q 1 ,q 2 ,...,q 30 , the effective diffusion time is set to τ, and the magnetic resonance simulation signal value S ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A method for extracting tissue fiber bundle structure information based on adaptive diffusion basis function decomposition, which belongs to the field of magnetic resonance diffusion imaging data processing. The purpose of the present invention is to more accurately extract tissue fiber structure information from acquired magnetic resonance diffusion weighted data . Method: 1. Construct a set of tensor basis functions of adaptive eigenvalues; 2. Express the diffusion weighted signal by the linear combination of the constructed tensor basis functions, and then list the diffusion weighted signal represented by the linear combination of tensor basis functions and the actual measured The minimum objective function of the error between diffusion weighted signals; 3. Use an iterative method to solve the weighted coefficients and eigenvalues ​​that minimize the value of the minimized objective function in step 2; 4. Perform the weighted coefficients and eigenvalues ​​obtained in step 3 After post-processing, the running direction of the fiber bundle in the voxel is obtained, which is completed. The advantage of the present invention is that the accuracy of extracting tissue fiber structure information from acquired magnetic resonance diffusion weighted data is high.

Description

technical field [0001] The invention relates to a method for extracting tissue fiber bundle structure information, and belongs to the field of magnetic resonance diffusion imaging data processing. Background technique [0002] Magnetic resonance diffusion imaging is currently the only method that can measure the diffusion motion and imaging of water molecules in tissues in vivo. It detects the microstructure of tissues by measuring and quantifying the diffusion information of water molecules in tissues. The diffusion information of water molecules along different directions is contained in a set of diffusion weighted images (DWI) with different diffusion weighted gradient directions, which can be reconstructed by modeling the diffusion function. The structural information of the fiber bundle mainly refers to the running direction of the fiber bundle in the voxel, and the main diffusion direction expressed by the diffusion function indicates the running direction of the tissu...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G01R33/20
Inventor 刘宛予楚春雨张延丽黄建平
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap