Magnetic resonance image classification method and system based on causal relationship

A magnetic resonance image and causal relationship technology, applied in image analysis, image enhancement, image data processing, etc., can solve problems such as inaccurate classification results and failure to consider differences in brain working mechanisms, and achieve high accuracy

Pending Publication Date: 2021-11-30
GUANGDONG UNIV OF TECH
View PDF1 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the defects of inaccurate classification results due to the lack of consideration of differences in brain working mechanisms when classifying brain magnetic resonance images in the prior art, the present invention provides a method and system for classifying magnetic resonance images based on causality, which can be more close Combined with the actual working mechanism of the brain, reflecting the differences in the brain working mechanism of different subjects, the accuracy of the classification results is higher

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
  • Magnetic resonance image classification method and system based on causal relationship
  • Magnetic resonance image classification method and system based on causal relationship
  • Magnetic resonance image classification method and system based on causal relationship

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0079] This embodiment provides a method for classifying magnetic resonance images based on causality, such as figure 1 As shown, the method includes:

[0080] S1: Acquire a magnetic resonance image with label information;

[0081] S2: Process the magnetic resonance image to obtain the BOLD signal of different brain regions;

[0082] S3: Input the BOLD signal of different brain regions as a variable into the causal algorithm to construct a causal connection map of different brain regions;

[0083] S4: Construct features using causal connection graphs;

[0084] S5: Screen the constructed features to obtain the optimal features;

[0085] S6: Classify the magnetic resonance images to be classified based on the optimal features, and obtain a classification result.

[0086] Magnetic resonance imaging technology is used to obtain magnetic resonance images with label information. According to the health status, the label information is divided into disease label and normal label....

Embodiment 2

[0125] This embodiment provides a causality-based magnetic resonance image classification system, which is used to implement a causality-based magnetic resonance image classification method as described in Embodiment 1, such as figure 2 As shown, the system includes:

[0126] A data acquisition module, configured to acquire magnetic resonance images with label information;

[0127] The preprocessing module is used to process the magnetic resonance images to obtain BOLD signals of different brain regions;

[0128] The causal map building module is used to input the BOLD signal of different brain regions as a variable into the causal algorithm to construct a causal connection map of different brain regions;

[0129] Feature construction module, using causal connection graph to construct features;

[0130] The feature screening module is used to screen the constructed features to obtain the optimal features;

[0131] The diagnosis module classifies the magnetic resonance imag...

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 magnetic resonance image classification method and system based on a causal relationship, and relates to the technical field of magnetic resonance image classification. The method comprises the steps: obtaining a magnetic resonance image with label information; processing the magnetic resonance image to obtain BOLD signals of different brain regions; inputting the BOLD signals of the different brain regions as variables into a causal algorithm, and constructing a causal connection diagram of the different brain regions; constructing features by using a causal connection graph, and screening the constructed features to obtain optimal features; and based on the optimal feature, classifying the magnetic resonance image to be classified to obtain a classification result. The actual working mechanism of the brain can be better fitted, the difference of the working mechanisms of the brain of different subjects is reflected, the accuracy of the classification result is higher, and powerful suggestions are provided for subsequent diagnosis and treatment.

Description

technical field [0001] The present invention relates to the technical field of magnetic resonance image classification, and more specifically, to a method and system for magnetic resonance image classification based on causality. Background technique [0002] Existing brain disease diagnosis methods include correlation analysis methods, kernel methods, graph embedding methods, and methods based on deep learning. These methods have achieved certain results in the field of brain disease diagnosis, but there are still some defects. These methods are based on Correlation is used to classify the population, but the correlation analysis is quite different from the real working mechanism of the brain, and cannot correctly reflect the causal connection mode between brain regions; moreover, since the correlation analysis cannot reflect the correct causal relationship, there is no guarantee The accuracy of the predicted diagnosis. [0003] The Chinese patent application CN111916204A ...

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
IPC IPC(8): G06K9/00G06K9/62G06T7/00
CPCG06T7/0012G06T2207/10088G06F2218/08G06F18/241G06F18/214
Inventor 黄晓楷杨泽勤
Owner GUANGDONG UNIV 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