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

Cross-modal medical image synthesis method based on parallel generative network

A synthesis method and technology of medical images, applied in image enhancement, image analysis, image data processing and other directions, can solve problems such as difficult to achieve, lack of training data and accurate understanding of related experimental phenomena.

Active Publication Date: 2020-05-19
ZHEJIANG UNIV OF TECH
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The existing technical defects are: the existing data-driven synthesis method lacks an accurate understanding of the potential basic mechanism of training data and related experimental phenomena; Usually the premise is that the two images have been registered, but this is difficult to achieve on deformable organs and human tissues

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
  • Cross-modal medical image synthesis method based on parallel generative network
  • Cross-modal medical image synthesis method based on parallel generative network
  • Cross-modal medical image synthesis method based on parallel generative network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] The present invention will be further described below in conjunction with the accompanying drawings.

[0045] refer to figure 1 and figure 2 , a method for cross-modal medical image synthesis based on parallel generative networks, comprising the following steps:

[0046] 1) Establish the basis for grouping based on expert experience, and use the small sample training set of the corresponding modality to classify the CNN-based grouper C A 、C B For training, the process is as follows:

[0047] 1.1) According to expert experience, specify groups 1, 2, 3...n of the MR data set, a total of n groups of representative images (several pictures in each group), where the feature of each group is the space represented by the image Anatomical structure, CT reference map obtained in the same way as above;

[0048] 1.2) Use the training set of MR to group the C A For training, use the training set of CT to group the C B Carrying out training, the grouper performs group proba...

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 cross-modal medical image synthesis method based on a parallel generative network, and the method comprises the steps: obtaining a common group vector feature space of cross-modal medical images through preprocessing, and obtaining a plurality of training sets for training a synthesizer; the training synthesizer is used for carrying out cross-modal synthesis training on the pair of training sets on each group of training sets; if the feature difference value of the sample pair is within an error allowable range, outputting a composite graph; and if the characteristicdifference value is greater than the allowable error range, performing weighted combination on the two regular terms and the original loss function of the synthesizer to guide the synthesizer to continue training. According to the invention, the quality of medical image whole-image synthesis of the easily-deformed part is improved.

Description

technical field [0001] The invention relates to a method for synthesizing cross-modal medical images. Background technique [0002] In recent years, various image synthesis techniques have been proposed and widely used in the medical field, and with the improvement of hardware equipment and the increase of medical image data, they will play a greater role. According to the characteristics of the driving mechanism, they can be divided into two categories: hypothesis-driven mechanism models and data-driven phenomenological models. Among them, deep neural networks, especially generative adversarial networks, as important representatives of phenomenological models, have strong Data-driven capabilities to synthesize the images we need to solve medical tasks with little knowledge of the underlying mechanisms. [0003] The existing technical defects are: the existing data-driven synthesis method lacks an accurate understanding of the potential basic mechanism of training data and ...

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): G06T5/50G06K9/62
CPCG06T5/50G06T2207/10088G06T2207/10081G06T2207/20081G06T2207/20084G06F18/24
Inventor 管秋陈奕州张跃耀胡海根徐新黎楼海燕陈峰徐涵杰陆正威
Owner ZHEJIANG 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