Multidimensional Feature Classification Method Based on Brain Magnetic Resonance Image

A technology of nuclear magnetic resonance images and multi-dimensional features, applied in the field of image processing, can solve problems such as the inability to analyze functional connections of specific parts, and achieve the effect of high accuracy and strong applicability

Active Publication Date: 2019-05-28
SUZHOU INST OF BIOMEDICAL ENG & TECH CHINESE ACADEMY OF SCI
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing features based on brain magnetic resonance images are mainly based on voxel-based morphological features. The classification algorithm established by using this feature can only find out the structural changes of specific parts, but cannot analyze the changes of functional connections of specific parts. At present, there is no good classification algorithm for clinical use that can simultaneously analyze the changes in human body structure and the structure-functional connection on the basis of brain MRI images, which requires extracting higher-dimensional information from brain MRI images. feature and establish a classifier to realize the classification algorithm of related diseases

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
  • Multidimensional Feature Classification Method Based on Brain Magnetic Resonance Image
  • Multidimensional Feature Classification Method Based on Brain Magnetic Resonance Image
  • Multidimensional Feature Classification Method Based on Brain Magnetic Resonance Image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] On the basis of the multidimensional feature classification method for brain MRI images provided in the above embodiments, this embodiment provides an example of a multidimensional feature classification method based on brain MR images.

[0068] First, a brain MR image is divided into regions to extract several ROI features. A ROI feature includes several types of labeling features for marking a brain MR image, such as gray matter volume, white matter volume, cerebrospinal fluid volume, cerebral cortex thickness, and brain volume. Marker features of classes such as cortical surface area. In order to eliminate individual differences, the above-mentioned types of marker features were normalized. Correspondingly, the normalization of gray matter volume, white matter volume, and cerebrospinal fluid volume was performed by dividing the volume of each ROI by the total intracranial volume to make individual differences Minimized; cortical thickness was normalized by dividing t...

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-dimensional feature classification method based on a brain MR image. The brain MR image is divided into regions to extract several ROI features; ; Several ROI features and several related features are extracted from several brain MR images to form a ROI feature set and a related feature set; the optimal ROI feature subset and the optimal correlation feature set are selected respectively for the ROI feature set and the related feature set by the hybrid feature algorithm. Feature subset; set the weight factor of the function ratio of ROI features in the classifier, and integrate the optimal ROI feature subset and the optimal related feature subset through the weight factor and the multi-core SVM model to form a multi-core classifier. The invention obtains high-dimensional features, can simultaneously analyze local changes and functional connection changes caused by related diseases, has high classification accuracy, and can assist in diagnosing different diseases.

Description

technical field [0001] The present invention relates to the technical field of image processing, and more specifically, the present invention relates to a multidimensional feature classification method based on brain magnetic resonance images. Background technique [0002] Medical imaging refers to the technology and processing process of obtaining internal tissue images of the human body or a certain part of the human body in a non-invasive manner for medical treatment or medical research. It includes the following two relatively independent research directions: medical imaging system and medical image processing. The former refers to the process of image formation, including the research on imaging mechanism, imaging equipment, imaging system analysis and other issues; the latter refers to the further processing of the obtained image, the purpose of which is to make the original image that is not clear enough Restoration, or to highlight certain feature information in the...

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): G06T7/00G06K9/32G06K9/46G06K9/62
CPCG06T7/0012G06T2207/30016G06T2207/10088G06T2207/20081G06V10/25G06V10/44G06F18/2411
Inventor 彭博戴亚康史文博周志勇佟宝同
Owner SUZHOU INST OF BIOMEDICAL ENG & TECH CHINESE ACADEMY OF SCI
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