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

Cross-modal nuclear magnetic resonance super-resolution network and image super-resolution method

A nuclear magnetic resonance and super-resolution technology, applied in image data processing, graphic image conversion, biological neural network model, etc. The effect of improving generalization

Active Publication Date: 2022-04-12
ZHEJIANG LAB
View PDF13 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is mainly to solve the problem of cross-modal nuclear magnetic resonance image super-resolution in the current nuclear magnetic resonance image super-resolution task based on deep learning

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 nuclear magnetic resonance super-resolution network and image super-resolution method
  • Cross-modal nuclear magnetic resonance super-resolution network and image super-resolution method
  • Cross-modal nuclear magnetic resonance super-resolution network and image super-resolution method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0055] The learning objectives of the present invention are:

[0056] ; ; ; ;

[0057] X is the input low-resolution image, Y is the high-resolution image output, S is the network upsampling function, , , Respectively, the number of channels, length, and width of the input feature, As the factor used above, , represent the output space and the input space, respectively.

[0058] In this regard, the present invention starts from the super-resolution of cross-modal NMR images, designs a cross-modal high-frequency deformable network that can take into account both global and local perception characteristics, and uses its internal spatial feature adaptive module to eliminate the gap between different modal features. The difference between different modes is further strengthened, and the high-frequency structure prior and inter-modal context prior are designed. The high-frequency gradient image is used as the network learning target, and the different frequenc...

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

According to the cross-modal nuclear magnetic resonance super-resolution network and the image super-resolution method provided by the invention, a high-resolution T2WI image is reconstructed by introducing a T1WI nuclear magnetic resonance image as auxiliary information, so that information exchange and complementation among different modals are realized, the information fusion capability is remarkably improved, and feature expression is enhanced. The low-frequency and high-frequency information reconstruction task of the T2WI modal image is divided and treated, the perception ability of the network to the characteristics is enhanced by using the local perception characteristic of the convolution and the global perception characteristic of the deformable network, and the reconstruction effect of the high-frequency information is effectively improved. And a designed inter-modal multi-head attention module effectively fuses feature non-local information by utilizing spatial feature self-similarity, so that the generalization of the model is effectively improved.

Description

technical field [0001] The invention relates to the field of nuclear magnetic resonance image super-resolution, in particular to a cross-modal nuclear magnetic resonance super-resolution network and an image super-resolution method. Background technique [0002] With its applicability in capturing the pathological details of human soft tissues, MRI images are considered to be one of the most widely used data in computer-aided diagnosis and brain function exploration, which can effectively help medical researchers in disease diagnosis and brain function analyze. High-resolution MRI images can provide effective auxiliary information for pathological analysis. On the contrary, low-resolution MRI images may cause difficulties in the diagnosis of some diseases (such as tumors with small lesions) in clinical diagnosis, which is not conducive to the disease. The timely diagnosis of the disease brings health risks, and it also brings challenges for researchers to explore the intern...

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): G06T3/40G06N3/04G06V10/80G06V10/82
Inventor 程乐超王良方超伟张鼎文
Owner ZHEJIANG LAB
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