Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A fmri feature extraction and recognition method based on tensor decomposition

A technology of feature extraction and tensor decomposition, applied in character and pattern recognition, instrumentation, calculation, etc., can solve the problems of dimensional disaster, small number of subjects, difficult to describe the dynamic information of adjacent functional connection matrix, etc. The effect of counting disasters and improving robustness

Active Publication Date: 2018-11-13
NANJING UNIV OF TECH
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This type of method will obtain a (90*89) / 2-dimensional vector after vectorizing the functional connectivity matrix of 90 brain regions. There is a "curse of dimensionality" problem, and the feature vectors extracted by algorithms such as PCA and ICA are also difficult to describe the dynamic information between adjacent functional connectivity matrices

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 fmri feature extraction and recognition method based on tensor decomposition
  • A fmri feature extraction and recognition method based on tensor decomposition
  • A fmri feature extraction and recognition method based on tensor decomposition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] Assume that there are currently 30 healthy adults and 30 healthy children under 10 years old. All subjects are right-handed, and the brain function image data is scanned by the same instrument, and the number of scanning sequences is 224. In order to test the effectiveness of the algorithm, half of the adult and child test data were taken out as the training set data, and the remaining half was used as the test set. The specific algorithm is as follows, the process is as follows figure 1 shown.

[0028] The first step is to preprocess all the data, remove the first ten unsteady-state scan sequences, and use SPM8 software to perform time layer correction, head motion correction, and smoothing on the original fMRI data, and perform 00.1 on the image after removing the influence of low-frequency drift. ~0.1Hz filtering; Finally, the influence of whole brain signal, cerebrospinal fluid signal, and head movement covariate parameters is eliminated by DPARSF software;

[002...

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 an fMRI feature extraction and recognition method based on tensor decomposition. The present invention constructs the fMRI dynamic functional connectivity matrix, and utilizes the multiple linear principal component analysis method to decompose the tensor, extracts the feature tensor of the dynamic functional connectivity matrix, and finally vectorizes the feature tensor as the input of the support vector machine to identify different The steps of fMRI data, preprocessing of the entire image sequence and matrix construction are all realized by Matlab program. The invention can not only overcome the "curse of dimensionality" problem, but also dig out the dynamic information of the brain function connection network.

Description

technical field [0001] The present invention relates to a processing and recognition method of brain functional MRI image (fMRI), which is a fMRI feature extraction and recognition method based on tensor decomposition; in particular, it relates to digital image processing, dynamic functional network construction, tensor decomposition and pattern Identification and other knowledge fields and processing methods. The results of fMRI processing by this method can be used to classify different types of fMRI data, such as adult and child data, drunk and sober tester data, tester viewing data of different emotional photos, etc. It belongs to machine learning and pattern recognition methods for high-dimensional image sequence data. Background technique [0002] Functional Magnetic Resonance Imaging (fMRI) technology has been widely concerned by researchers in recent years. The technology achieves the purpose of non-invasively observing changes in brain activity by detecting changes...

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): G06K9/00G06K9/46
Inventor 梅雪黄嘉爽李微微马士林
Owner NANJING UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com