Multimodal fusion image classification method based on rls-elm

A technology that integrates images and classification methods. It is used in character and pattern recognition, instruments, computer parts, etc. It can solve the problems of long training time and large storage space, and achieve the effect of improving classification accuracy and classification speed.

Active Publication Date: 2016-06-01
CENT SOUTH UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Secondly, the SVM algorithm that fMRI researchers are keen on has problems such as long training time and large storage space.

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
  • Multimodal fusion image classification method based on rls-elm
  • Multimodal fusion image classification method based on rls-elm
  • Multimodal fusion image classification method based on rls-elm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] The present invention will be further described below in conjunction with the drawings and examples.

[0023] See figure 1 , figure 2 , image 3 , The RLS-ELM-based multimodal fusion image classification method contains the following steps:

[0024] Step 1: Obtain rs-fMRI, sMRI and DTI data of multiple subjects, and perform preprocessing to eliminate the data that does not meet the requirements.

[0025] Step 2: Calculate the ReHo (Regional Homogeneity) value of each voxel in the rs-fMRI data.

[0026] Step 3: Calculate the gray matter density (GrayMatterDensity) value of each voxel in the sMRI data.

[0027] Step 4: Calculate the FA (Fractional Anisotropy) value of each voxel in the DTI data.

[0028] Step 5: Connect the ReHo, gray matter density and FA value of each voxel to form a new feature matrix A.

[0029] Step 6: Perform PCA dimensionality reduction processing on the new feature matrix A.

[0030] Step 7: Train the RLS-ELM classifier to obtain the trained RLS-ELM classific...

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

The invention discloses a multi-modal fusion image sorting method based on RLS-ELM. The method is characterized by comprising the steps that 1. rs-fMRI, sMRI and DTI data of a plurality of tested objects are obtained, preprocessing is carried out, and tested data which do not accord with a provision are removed; 2. ReHo values of voxels in the rs-fMRI data are computed; 3. grey matter density values of voxels in the sMRI data are computed; 4. FA values of voxels in DTI data are computed; 5. the ReHo values, the grey matter density values and the FA values of the voxels are connected to form a new characteristic matrix A; 6. the new characteristic matrix A is subjected to PCA dimension reduction processing; and 7. an RLS-ELM sorter is subjected to training, and a trained RLS-ELM sorter is obtained. According to the multi-modal fusion image sorting method based on RLS-ELM, sorting accuracy and sorting speed are obviously improved, disease early discovering and early treating are achieved, and great significance is achieved in a clinical medicine study process for revealing disease progression is achieved.

Description

Technical field [0001] The invention relates to a multi-modal fusion image classification method, in particular to a multi-modal fusion image classification method aimed at diseases. Background technique [0002] In the past two decades, with the advancement of brain imaging technology, the research of brain science has entered a period of rapid development. Functional magnetic resonance imaging (functional magnetic resonance imaging, fMRI), as a non-invasive brain function detection technology, has become the most widely used brain imaging technology in brain science research due to its good comprehensive performance of time and space resolution. [0003] Resting-state rs-fMRI (resting-statefMRI, rs-fMRI) is the spontaneous regulation of brain BOLD signal generation, reflecting the spontaneous activity of the brain in the resting state, reflecting the association between the brain areas and the brain's operating mechanism, so Clinically, rs-fMRI is more suitable for the research ...

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 Patents(China)
IPC IPC(8): G06K9/62G06K9/46
Inventor 龙军阳洁张祖平张昊
Owner CENT SOUTH UNIV
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