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

Brain nuclear magnetic resonance abnormal image visualization method based on 2D CAM

A kind of nuclear magnetic resonance, abnormal image technology, applied in image enhancement, image analysis, image data processing and other directions, can solve the problem of invisible model learning and so on

Active Publication Date: 2020-04-10
HUBEI UNIV OF TECH
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although you can see the outline of a certain image category on these backpropagated images, you can hardly see what the model has learned

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
  • Brain nuclear magnetic resonance abnormal image visualization method based on 2D CAM
  • Brain nuclear magnetic resonance abnormal image visualization method based on 2D CAM
  • Brain nuclear magnetic resonance abnormal image visualization method based on 2D CAM

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0056] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0057] A method for visualization of brain MRI abnormal images based on 2D CAM, the specific method is as follows:

[0058] 1) Collect abnormal brain MRI images of patients as training samples;

[0059] 2) Use the training samples to train the 2D CAM, and determine the network parameters after training, namely the coefficient matrix W and the bias vector b value;

[0060] 21) Construct a 2D CAM model and initialize the network parameters randomly

[0061] Build 2D CAM models such as figure 1 As shown, the 2D CAM model includes an input layer, a convolution layer, a pooling layer, a global average pooling layer, a fully connected layer And the output layer (output layer), and initialize the 2DCAM model, that is, initialize the coefficient matrix W and the bias vector b value corresponding to all hidden layers and the output layer, s...

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 brain nuclear magnetic resonance abnormal image visualization method based on 2D CAM. Collecting a brain nuclear magnetic resonance abnormal image of the patient as a training sample; training the two-dimensional class activation mapping 2D CAM by using the training sample; determining trained network parameters, i.e., a coefficient matrix W and a bias vector b value; andcreating a visual heat map according to different magnetic resonance images. A brain nuclear magnetic resonance abnormal image of a patient is processed on the basis of a traditional CAM model, automatic recognition and detection are achieved, the visualization effect is good, and a medical researcher can be assisted in quantitative analysis and research conveniently.

Description

technical field [0001] The invention belongs to the technical field of nuclear magnetic resonance image disease visualization, and in particular relates to a 2D CAM-based visualization method for abnormal brain magnetic resonance images. Background technique [0002] As an emerging field of machine learning, deep learning has gradually made great achievements in various fields such as computer vision, audio and video processing, natural language processing, and precise navigation in recent years. Its main starting point is to roughly simulate the human neural network system. Layer-by-layer feature extraction is used to complete the corresponding abstract information induction and summary. Compared with traditional support vector machines and maximum entropy, these methods can only be called shallow learning. Shallow learning usually requires manual design of abstract features by means of mathematical derivation to complete the corresponding recognition, etc. application. ...

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 Applications(China)
IPC IPC(8): G06T7/00G06K9/62G06N3/04G06N3/08
CPCG06T7/0012G06N3/082G06T2207/10088G06T2207/20081G06T2207/20084G06T2207/30016G06N3/045G06F18/241Y02A90/30
Inventor 柯丰恺刘欢平赵大兴孙国栋冯维
Owner HUBEI 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