CN104408713A - Method and system for extracting image characteristics of diffusion tensor - Google Patents
Method and system for extracting image characteristics of diffusion tensor Download PDFInfo
- Publication number
- CN104408713A CN104408713A CN201410627576.0A CN201410627576A CN104408713A CN 104408713 A CN104408713 A CN 104408713A CN 201410627576 A CN201410627576 A CN 201410627576A CN 104408713 A CN104408713 A CN 104408713A
- Authority
- CN
- China
- Prior art keywords
- tensor
- eigenvalue
- image
- dti
- feature
- 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
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
- G06T2207/10092—Diffusion tensor magnetic resonance imaging [DTI]
-
- 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)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to the technical field of medical image processing, and particularly relates to a method and a system for extracting image characteristics of diffusion tensor. The method for extracting the image characteristics of the diffusion tensor comprises the steps of a, calculating a tensor eigenvalue of a DTI (diffusion tensor imaging) image in the principal direction, and analyzing distribution characteristics of tensor eigenvalues of all images; b, constructing a weight matrix according to the distribution characteristics of the tensor eigenvalues, and establishing and optimizing a weight matrix based target function so as to construct a new characteristic space; and c, solving the target function to acquire a multi-linear projection matrix, projecting a DTI tensor eigenvalue based matrix space to multiple linear sub-spaces, and carrying out characteristic extraction and dimensionality reduction on DTI images of the sub-spaces. The method provided by the invention not only can effectively reflect various sample characteristics, but also can reduce the calculation cost, diffusion characteristics of the diffusion tensor image on three principal directions are fully dug on the basis of keeping the original image structure information, purposes of dimensionality reduction and characteristic extraction are realized, and a dimensionality curse is avoided effectively.
Description
Technical field
The invention belongs to technical field of medical image processing, particularly relate to a kind of method and system of dispersion tensor image characteristics extraction.
Background technology
Diffusion tensor (Diffusion Tensor Imaging, DTI), being a kind of new method describing brain structure, is the special shape of Magnetic resonance imaging (MRI).Being different from Magnetic resonance imaging is follow the trail of the hydrogen atom in hydrone, and diffusion tensor is according to the drawing of hydrone moving direction.Diffusion tensor figure (presentation mode is different from former image) can disclose the brain tumor cell that how to affect the nerves and connect, guide healthcare givers to carry out operation on brain, the trickle abnormality change of the regarding brain such as same apoplexy, multiple sclerosis, schizophrenia, Dyslexia and spinal cord can also be disclosed.
Diffusion tensor data are second-order tensor structure in essence, and its each voxel contains the three-dimensional spatial information of hydrone disperse in white matter nerve fiber bundles.Traditional being caused by the learning algorithm Treatment Analysis dispersion tensor image of vector pattern such as destroys some row problems such as initial data structure, and calculated amount is large, and assess the cost height.
Summary of the invention
The invention provides a kind of method and system of dispersion tensor image characteristics extraction, there is destruction initial data structure in the learning algorithm Treatment Analysis dispersion tensor image being intended to solve existing vector pattern, and calculated amount is large, assess the cost high technical matters.
The present invention is achieved in that a kind of method of dispersion tensor image characteristics extraction, comprising:
Step a: calculate the tensor eigenvalue in DTI image principal direction, and analyze the distribution characteristics of the tensor eigenvalue of all images;
Step b: according to the distribution characteristics of tensor eigenvalue structure weight matrix, set up and the objective function optimized based on weight matrix to construct new feature space;
Step c: solve objective function and obtain polyteny projection matrix, is mapped to multilinear subspace by the space of matrices based on DTI tensor eigenvalue, and carries out feature extraction and dimensionality reduction to the DTI image of each subspace.
The technical scheme that the embodiment of the present invention is taked also comprises: in described step b, and described foundation optimizing is specially based on the objective function of weight matrix: the application multiple power series method of development sets up the objective function based on weight matrix.
The technical scheme that the embodiment of the present invention is taked also comprises: in described step c, described in solve objective function and obtain polyteny projection matrix and be specially: alternating least-squares Optimization Solution objective function, obtains polyteny projection matrix.
The technical scheme that the embodiment of the present invention is taked also comprises: in described step c, describedly the space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace is specially: according to polyteny projection matrix, by tensor product computing, space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace, makes each subspace can capture the amount of variability of most of orthogonal multidimensional in new feature space; The described DTI image to each subspace carries out feature extraction and dimensionality reduction is specially: use the DTI image of polyteny core principle component analysis method to each subspace to carry out feature extraction and dimensionality reduction.
The technical scheme that the embodiment of the present invention is taked also comprises: also comprise after described step c: the tensor eigenvalue calculating the DTI image after projection, and according to the tensor property after the signal to noise ratio (S/N ratio) determination dimensionality reduction of tensor eigenvalue.
Another technical scheme that the embodiment of the present invention provides is: a kind of system of dispersion tensor image characteristics extraction, comprise eigenvalue processing module, weight matrix constructing module, feature space constructing module, Feature Mapping module and characteristic extracting module, described eigenvalue processing module for calculating the tensor eigenvalue in DTI image principal direction, and analyzes the distribution characteristics of the tensor eigenvalue of all images; Described weight matrix constructing module is used for the distribution characteristics structure weight matrix according to tensor eigenvalue; Described feature space constructing module for set up and the objective function optimized based on weight matrix to construct new feature space; Described Feature Mapping module is used for solving objective function and obtains polyteny projection matrix, and the space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace; Described characteristic extracting module is used for carrying out feature extraction and dimensionality reduction to the DTI image of each subspace.
The technical scheme that the embodiment of the present invention is taked also comprises: described eigenvalue processing module comprises eigenvalue computing unit and characteristic analysis unit;
Described eigenvalue computing unit is for calculating the tensor eigenvalue of DTI image in three principal directions;
Described characteristic analysis unit is used for the tensor design feature according to dispersion tensor view data, analyzes the distribution characteristics of the tensor eigenvalue of all images.
The technical scheme that the embodiment of the present invention is taked also comprises: described feature space constructing module comprises objective function and sets up unit and feature space tectonic element;
Described objective function sets up unit for applying the objective function of multiple power series method of development foundation based on weight matrix;
Described feature space tectonic element is used for constructing new feature space by objective function optimization.
The technical scheme that the embodiment of the present invention is taked also comprises: described Feature Mapping module comprises projection matrix computing unit and Feature Mapping unit;
Described projection matrix computing unit is used for alternating least-squares Optimization Solution objective function, obtains polyteny projection matrix;
Described Feature Mapping unit is used for according to polyteny projection matrix, by tensor product computing, space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace, makes each subspace can capture the amount of variability of most of orthogonal multidimensional in new feature space.
The technical scheme that the embodiment of the present invention is taked also comprises: described characteristic extracting module comprises feature extraction unit and feature calculation unit;
Described feature extraction unit carries out feature extraction and dimensionality reduction for using the DTI image of polyteny core principle component analysis method to each subspace;
Described feature calculation unit for calculating the tensor eigenvalue of the DTI image after projection, and according to the tensor property after the signal to noise ratio (S/N ratio) determination dimensionality reduction of tensor eigenvalue.
The method and system of the dispersion tensor image characteristics extraction of the embodiment of the present invention are by calculating and analyzing the tensor eigenvalue of DTI image in three principal directions with structure based on weight matrix feature space, the application multiple power series method of development constructs the objective function of weight matrix, by optimizing this objective function to construct optimum feature space, and by tensor product by all Feature Mapping to respective multilinear subspace, make it can effectively reflect Different categories of samples feature, can reduce again assessing the cost; And use polyteny core principle component analysis method to subspace representation and dimensionality reduction, the correlativity of proper vector and the degree of depth, mutual independence can be ensured, can on the basis keeping original image structure information, the dispersive test of abundant excavation dispersion tensor image in three principal directions, realize the object of image dimensionality reduction and feature extraction, effectively prevent dimension disaster.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the method for the dispersion tensor image characteristics extraction of the embodiment of the present invention;
Fig. 2 is the structural representation of the system of the dispersion tensor image characteristics extraction of the embodiment of the present invention.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
Referring to Fig. 1, is the process flow diagram of the method for the dispersion tensor image characteristics extraction of the embodiment of the present invention.The method of the dispersion tensor image characteristics extraction of the embodiment of the present invention comprises the following steps:
Step 100: calculate the tensor eigenvalue of DTI image in three principal directions;
Step 200: according to the tensor design feature of dispersion tensor view data, analyzes the distribution characteristics of the tensor eigenvalue of all images;
Step 300: according to the distribution characteristics structure weight matrix of tensor eigenvalue;
Step 400: the application multiple power series method of development sets up the objective function based on weight matrix;
Step 500: construct new feature space by objective function optimization;
Step 600: alternating least-squares Optimization Solution objective function, obtains polyteny projection matrix;
In step 600, least square method (also known as least square method) is a kind of mathematical optimization techniques, and it finds the optimal function coupling of data by the quadratic sum of minimum error; Utilize least square method can try to achieve unknown data easily, and between the data that these are tried to achieve and real data, the quadratic sum of error is minimum.Least square method also can be used for curve, and some other optimization problem is also expressed by minimization of energy or maximization entropy least square method; The present invention is according to replacing least square principle, and decompose space originally to a series of space-filling curve subspace, mining analysis DTI image is in the impact of the disperse level of interaction of three principal directions.
Step 700: according to polyteny projection matrix, by tensor product computing, is mapped to multilinear subspace by the space of matrices based on DTI tensor eigenvalue, makes each subspace can capture the amount of variability of most of orthogonal multidimensional in new feature space;
In step 700, tensor product: in mathematics, every in category multiple object obtain an object, and the operation meeting certain binding rule and exchange regulation can be considered as " tensor product ", the cartesian product such as gathered, without handing over also, the product of manifold; The present invention by tensor product by all Feature Mapping to respective multilinear subspace, make it can effectively reflect Different categories of samples feature, can reduce again assessing the cost.
Step 800: use polyteny core principle component analysis method (MKPCA) the DTI image to each subspace to carry out feature extraction and dimensionality reduction;
In step 800, the polyteny core principle component analysis method (MKPCA) that the present invention proposes is for the core of feature extraction: utilize the non-linear popularization of one that the principal component analysis method of target kernel function to classics is carried out; Compared with traditional principal component analysis method, MKPCA have effectively can catch data nonlinear characteristic, distribution situation not requirement to luv space data, there is better popularity, the correlativity of proper vector and the degree of depth, mutual independence can also be ensured in addition, can on the basis keeping original image structure information, the dispersive test of abundant excavation dispersion tensor image in three principal directions, realize the object of image dimensionality reduction and feature extraction, and can calculated amount be reduced, avoid dimension disaster.
Step 900: the tensor eigenvalue calculating the DTI image after projection, and according to the tensor property after the signal to noise ratio (S/N ratio) determination dimensionality reduction of tensor eigenvalue.
Referring to Fig. 2, is the structural representation of the system of the dispersion tensor image characteristics extraction of the embodiment of the present invention.The system of the dispersion tensor image characteristics extraction of the embodiment of the present invention comprises eigenvalue processing module, weight matrix constructing module, feature space constructing module, Feature Mapping module and characteristic extracting module; Particularly:
Eigenvalue processing module comprises eigenvalue computing unit and characteristic analysis unit;
Eigenvalue computing unit is for calculating the tensor eigenvalue of DTI image in three principal directions;
Characteristic analysis unit is used for the tensor design feature according to dispersion tensor view data, analyzes the distribution characteristics of the tensor eigenvalue of all images;
Weight matrix constructing module is used for the distribution characteristics structure weight matrix according to tensor eigenvalue;
Feature space constructing module comprises objective function and sets up unit and feature space tectonic element;
Objective function sets up unit for applying the objective function of multiple power series method of development foundation based on weight matrix;
Feature space tectonic element is used for constructing new feature space by objective function optimization;
Feature Mapping module comprises projection matrix computing unit and Feature Mapping unit;
Projection matrix computing unit is used for alternating least-squares Optimization Solution objective function, obtains polyteny projection matrix; Wherein, least square method is a kind of mathematical optimization techniques, and it finds the optimal function coupling of data by the quadratic sum of minimum error; Utilize least square method can try to achieve unknown data easily, and between the data that these are tried to achieve and real data, the quadratic sum of error is minimum; Least square method also can be used for curve, and some other optimization problem is also expressed by minimization of energy or maximization entropy least square method; The present invention is according to replacing least square principle, and decompose space originally to a series of space-filling curve subspace, mining analysis DTI image is in the impact of the disperse level of interaction of three principal directions.
Feature Mapping unit is used for according to polyteny projection matrix, by tensor product computing, space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace, makes each subspace can capture the amount of variability of most of orthogonal multidimensional in new feature space; Wherein, tensor product: in mathematics, every in category multiple object obtain an object, and the operation meeting certain binding rule and exchange regulation can be considered as " tensor product ", the cartesian product such as gathered, without handing over also, the product of manifold; The present invention by tensor product by all Feature Mapping to respective multilinear subspace, make it can effectively reflect Different categories of samples feature, can reduce again assessing the cost.
Characteristic extracting module comprises feature extraction unit and feature calculation unit;
Feature extraction unit carries out feature extraction and dimensionality reduction for using the DTI image of polyteny core principle component analysis method to each subspace; Wherein, the polyteny core principle component analysis method that the present invention proposes is for the core of feature extraction: utilize the non-linear popularization of one that the principal component analysis method of target kernel function to classics is carried out; Compared with traditional principal component analysis method, MKPCA have effectively can catch data nonlinear characteristic, distribution situation not requirement to luv space data, there is better popularity, the correlativity of proper vector and the degree of depth, mutual independence can also be ensured in addition, can on the basis keeping original image structure information, the dispersive test of abundant excavation dispersion tensor image in three principal directions, realize the object of image dimensionality reduction and feature extraction, and can calculated amount be reduced, avoid dimension disaster.
Feature calculation unit for calculating the tensor eigenvalue of the DTI image after projection, and according to the tensor property after the signal to noise ratio (S/N ratio) determination dimensionality reduction of tensor eigenvalue.
The method and system of the dispersion tensor image characteristics extraction of the embodiment of the present invention are by calculating and analyzing the tensor eigenvalue of DTI image in three principal directions with structure based on weight matrix feature space, the application multiple power series method of development constructs the objective function of weight matrix, by optimizing this objective function to construct optimum feature space, and by tensor product by all Feature Mapping to respective multilinear subspace, make it can effectively reflect Different categories of samples feature, can reduce again assessing the cost; And use polyteny core principle component analysis method to subspace representation and dimensionality reduction, the correlativity of proper vector and the degree of depth, mutual independence can be ensured, can on the basis keeping original image structure information, the dispersive test of abundant excavation dispersion tensor image in three principal directions, realize the object of image dimensionality reduction and feature extraction, effectively prevent dimension disaster.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (10)
1. a method for dispersion tensor image characteristics extraction, comprising:
Step a: calculate the tensor eigenvalue in DTI image principal direction, and analyze the distribution characteristics of the tensor eigenvalue of all images;
Step b: according to the distribution characteristics of tensor eigenvalue structure weight matrix, set up and the objective function optimized based on weight matrix to construct new feature space;
Step c: solve objective function and obtain polyteny projection matrix, is mapped to multilinear subspace by the space of matrices based on DTI tensor eigenvalue, and carries out feature extraction and dimensionality reduction to the DTI image of each subspace.
2. the method for dispersion tensor image characteristics extraction according to claim 1, it is characterized in that, in described step b, described foundation optimizing is specially based on the objective function of weight matrix: the application multiple power series method of development sets up the objective function based on weight matrix.
3. the method for dispersion tensor image characteristics extraction according to claim 1, it is characterized in that, in described step c, described in solve objective function and obtain polyteny projection matrix and be specially: alternating least-squares Optimization Solution objective function, obtains polyteny projection matrix.
4. the method for dispersion tensor image characteristics extraction according to claim 1, it is characterized in that, in described step c, describedly the space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace is specially: according to polyteny projection matrix, by tensor product computing, space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace, makes each subspace in new feature space can capture the amount of variability of most of orthogonal multidimensional; The described DTI image to each subspace carries out feature extraction and dimensionality reduction is specially: use the DTI image of polyteny core principle component analysis method to each subspace to carry out feature extraction and dimensionality reduction.
5. the method for the dispersion tensor image characteristics extraction according to claim 1 or 4, it is characterized in that, also comprise after described step c: the tensor eigenvalue calculating the DTI image after projection, and according to the tensor property after the signal to noise ratio (S/N ratio) determination dimensionality reduction of tensor eigenvalue.
6. the system of a dispersion tensor image characteristics extraction, it is characterized in that: comprise eigenvalue processing module, weight matrix constructing module, feature space constructing module, Feature Mapping module and characteristic extracting module, described eigenvalue processing module for calculating the tensor eigenvalue in DTI image principal direction, and analyzes the distribution characteristics of the tensor eigenvalue of all images; Described weight matrix constructing module is used for the distribution characteristics structure weight matrix according to tensor eigenvalue; Described feature space constructing module for set up and the objective function optimized based on weight matrix to construct new feature space; Described Feature Mapping module is used for solving objective function and obtains polyteny projection matrix, and the space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace; Described characteristic extracting module is used for carrying out feature extraction and dimensionality reduction to the DTI image of each subspace.
7. the system of dispersion tensor image characteristics extraction according to claim 6, is characterized in that, described eigenvalue processing module comprises eigenvalue computing unit and characteristic analysis unit;
Described eigenvalue computing unit is for calculating the tensor eigenvalue of DTI image in three principal directions;
Described characteristic analysis unit is used for the tensor design feature according to dispersion tensor view data, analyzes the distribution characteristics of the tensor eigenvalue of all images.
8. the system of dispersion tensor image characteristics extraction according to claim 6, is characterized in that, described feature space constructing module comprises objective function and sets up unit and feature space tectonic element;
Described objective function sets up unit for applying the objective function of multiple power series method of development foundation based on weight matrix;
Described feature space tectonic element is used for constructing new feature space by objective function optimization.
9. the system of dispersion tensor image characteristics extraction according to claim 6, is characterized in that, described Feature Mapping module comprises projection matrix computing unit and Feature Mapping unit;
Described projection matrix computing unit is used for alternating least-squares Optimization Solution objective function, obtains polyteny projection matrix;
Described Feature Mapping unit is used for according to polyteny projection matrix, by tensor product computing, space of matrices based on DTI tensor eigenvalue is mapped to multilinear subspace, makes each subspace in new feature space can capture the amount of variability of most of orthogonal multidimensional.
10. the system of dispersion tensor image characteristics extraction according to claim 6, is characterized in that, described characteristic extracting module comprises feature extraction unit and feature calculation unit;
Described feature extraction unit carries out feature extraction and dimensionality reduction for using the DTI image of polyteny core principle component analysis method to each subspace;
Described feature calculation unit for calculating the tensor eigenvalue of the DTI image after projection, and according to the tensor property after the signal to noise ratio (S/N ratio) determination dimensionality reduction of tensor eigenvalue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410627576.0A CN104408713B (en) | 2014-11-10 | 2014-11-10 | A kind of method and system of dispersion tensor image characteristics extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410627576.0A CN104408713B (en) | 2014-11-10 | 2014-11-10 | A kind of method and system of dispersion tensor image characteristics extraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104408713A true CN104408713A (en) | 2015-03-11 |
| CN104408713B CN104408713B (en) | 2018-01-05 |
Family
ID=52646342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410627576.0A Active CN104408713B (en) | 2014-11-10 | 2014-11-10 | A kind of method and system of dispersion tensor image characteristics extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104408713B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184794A (en) * | 2015-09-07 | 2015-12-23 | 中国科学院深圳先进技术研究院 | CSM assistant analysis system and method based on tensor image |
| CN107292329A (en) * | 2017-04-19 | 2017-10-24 | 西北工业大学 | Event imagination sorting technique based on CI CSP algorithms |
| CN112183599A (en) * | 2020-09-22 | 2021-01-05 | 东南大学 | Multi-station forging data feature extraction method based on UMPCA |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090190817A1 (en) * | 2007-07-20 | 2009-07-30 | Medical College Of Georgia Research Institute, Inc. | Systems and methods for parametric mapping for correction of noise -based systematic bias of DTI metrics, using a DTI mapping phantom |
| CN101763638A (en) * | 2009-12-14 | 2010-06-30 | 西北工业大学 | Method for classifying cerebral white matter fiber tracts in diffusion tensor nuclear magnetic resonance image |
| CN103049901A (en) * | 2012-08-03 | 2013-04-17 | 上海理工大学 | Magnetic resonance diffusion tensor imaging fiber bundle tracking device |
| CN103440512A (en) * | 2013-09-17 | 2013-12-11 | 西安电子科技大学 | Identifying method of brain cognitive states based on tensor locality preserving projection |
-
2014
- 2014-11-10 CN CN201410627576.0A patent/CN104408713B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090190817A1 (en) * | 2007-07-20 | 2009-07-30 | Medical College Of Georgia Research Institute, Inc. | Systems and methods for parametric mapping for correction of noise -based systematic bias of DTI metrics, using a DTI mapping phantom |
| CN101763638A (en) * | 2009-12-14 | 2010-06-30 | 西北工业大学 | Method for classifying cerebral white matter fiber tracts in diffusion tensor nuclear magnetic resonance image |
| CN103049901A (en) * | 2012-08-03 | 2013-04-17 | 上海理工大学 | Magnetic resonance diffusion tensor imaging fiber bundle tracking device |
| CN103440512A (en) * | 2013-09-17 | 2013-12-11 | 西安电子科技大学 | Identifying method of brain cognitive states based on tensor locality preserving projection |
Non-Patent Citations (2)
| Title |
|---|
| 于红芸 等: "SAR图像的快速核主成分分析识别方法", 《中国图象图形学报》 * |
| 郭金玉 等: "基于多线性核主成分分析的掌纹识别", 《光电子 激光》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184794A (en) * | 2015-09-07 | 2015-12-23 | 中国科学院深圳先进技术研究院 | CSM assistant analysis system and method based on tensor image |
| CN105184794B (en) * | 2015-09-07 | 2018-04-17 | 中国科学院深圳先进技术研究院 | A kind of CSM Computer Aided Analysis Systems and method based on tensor image |
| CN107292329A (en) * | 2017-04-19 | 2017-10-24 | 西北工业大学 | Event imagination sorting technique based on CI CSP algorithms |
| CN107292329B (en) * | 2017-04-19 | 2020-10-16 | 西北工业大学 | Event idea classification method based on CI-CSP algorithm |
| CN112183599A (en) * | 2020-09-22 | 2021-01-05 | 东南大学 | Multi-station forging data feature extraction method based on UMPCA |
| CN112183599B (en) * | 2020-09-22 | 2024-02-27 | 东南大学 | UMPCA-based multi-station forging data feature extraction method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104408713B (en) | 2018-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | BPGAN: Brain PET synthesis from MRI using generative adversarial network for multi-modal Alzheimer’s disease diagnosis | |
| CN104361318A (en) | Disease diagnosis auxiliary system and disease diagnosis auxiliary method both based on diffusion tensor imaging technology | |
| Zhang et al. | Predicting brain structural network using functional connectivity | |
| Zipunnikov et al. | Longitudinal high-dimensional principal components analysis with application to diffusion tensor imaging of multiple sclerosis | |
| Odusami et al. | Pixel-level fusion approach with vision transformer for early detection of Alzheimer’s disease | |
| Shahbazi et al. | Using distance on the Riemannian manifold to compare representations in brain and in models | |
| Majka et al. | Possum—a framework for three-dimensional reconstruction of brain images from serial sections | |
| CN104573742A (en) | Medical image classification method and system | |
| Peterfreund et al. | Local conformal autoencoder for standardized data coordinates | |
| CN104408713A (en) | Method and system for extracting image characteristics of diffusion tensor | |
| US20150324994A1 (en) | Method of Automatically Calculating Linking Strength of Brain Fiber Tracts | |
| CN111938592A (en) | Missing multi-mode representation learning algorithm for Alzheimer's disease diagnosis | |
| Shariff et al. | A generative model of microtubule distributions, and indirect estimation of its parameters from fluorescence microscopy images | |
| CN105184794A (en) | CSM assistant analysis system and method based on tensor image | |
| Ghosh et al. | A joint subspace mapping between structural and functional brain connectomes | |
| CN104411230A (en) | Detection of disease-related retinal nerve fiber layer thinning | |
| Carloni et al. | Exploiting causality signals in medical images: A pilot study with empirical results | |
| Lu et al. | Predicting progressions of cognitive outcomes via high-order multi-modal multi-task feature learning | |
| Wang et al. | Diffusion tensor image registration using hybrid connectivity and tensor features | |
| Holenstein et al. | Simulation and evaluation of 3D traction force microscopy | |
| CN109241996B (en) | FMRI brain image analysis method based on gradient characterization similarity and Searchlight | |
| Li et al. | Biomarker extraction based on subspace learning for the prediction of mild cognitive impairment conversion | |
| Carmo et al. | Automated computed tomography and magnetic resonance imaging segmentation using deep learning: a beginner's guide | |
| Palma et al. | MAMBA—Multi‐pAradigM voxel‐Based Analysis: A computational cookbot | |
| Li et al. | Nonlinear Bayesian optimal experimental design using logarithmic Sobolev inequalities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |