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

Magnetic resonance thin-layer image reconstruction method

An image reconstruction and magnetic resonance technology, applied in image data processing, 3D modeling, instruments, etc., can solve the problems of difficult data acquisition, insufficient structural similarity performance, and inability to achieve clinical diagnosis, and achieve the effect of increasing data capacity.

Inactive Publication Date: 2019-07-23
FUDAN UNIV
View PDF4 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, asymptomatic children typically rarely undergo brain MRI scans, so children's brain MRI images are more difficult to obtain than data for adults, let alone high-quality thin-section images
[0004] Currently commonly used reconstruction algorithms (for example, bilinear interpolation, sparse representation, 3D-SRU-Net, etc.) are all directly interpolated in the data space for areas without data, and reconstructed by direct interpolation. Magnetic resonance thin-section images, especially children's magnetic resonance thin-section images, are not good enough in peak signal-to-noise ratio, structural similarity, regularized mutual information and other imaging indicators, and cannot meet the requirements that can help doctors make clinical diagnoses

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 thin-layer image reconstruction method
  • Magnetic resonance thin-layer image reconstruction method
  • Magnetic resonance thin-layer image reconstruction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] based on the following Figure 1 to Figure 6 , specifically explain the preferred embodiment of the present invention.

[0046] Such as figure 1 As shown, the present invention provides a magnetic resonance thin layer image reconstruction method, comprising the following steps:

[0047] Step S1, using the generative adversarial network to fuse the magnetic resonance thick-slice images of the transverse plane and the sagittal plane, and initially generate the corresponding magnetic resonance thin-slice image data;

[0048] Step S2, using the convolutional neural network to correct the details of the preliminarily generated magnetic resonance thin-slice image data, and reconstruct the magnetic resonance thin-slice image data.

[0049] The generated confrontation network (3D-Y-Net-GAN) includes a generator and a conditional discriminator. The convolutional neural network includes a three-dimensional densely connected U-shaped structure (3D-DenseU-Net) and an enhanced re...

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 relates to a magnetic resonance thin-layer image reconstruction method, which comprises the following steps of: fusing a magnetic resonance thick-layer image with a cross section and a sagittal plane by utilizing a generative adversarial network, initially generating corresponding magnetic resonance thin-layer image data, and performing detail correction on the initially generated magnetic resonance thin-layer image data by utilizing a convolutional neural network to reconstruct the magnetic resonance thin-layer image data. According to the method, a more real magnetic resonancethin-layer image can be obtained, the peak signal-to-noise ratio, the structural similarity and the regularization mutual information are greatly improved, the child thin-layer brain magnetic resonance image data capacity can be effectively increased, and a foundation is laid for later research.

Description

technical field [0001] The invention relates to a magnetic resonance thin layer image reconstruction method. Background technique [0002] The data obtained by magnetic resonance imaging can be roughly divided into thin-slice magnetic resonance images and thick-slice magnetic resonance images according to the spatial distance between adjacent scanning layers. Due to its high spatial resolution, thin-slice MRI images are ideal medical images for the study of brain structure and intraoperative navigation of the brain. However, due to problems such as thin-slice scanning efficiency and machine loss, thick-slice magnetic resonance images are widely used in clinical practice, and the data volume of thin-slice magnetic resonance images is relatively limited. There are relatively few thin-section magnetic resonance images of children's brains, but they are of great significance to the study of human brain development. [0003] Compared with adult brain image data, children's MRI ...

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): G06T17/00G06N3/04
CPCG06T17/00G06N3/045
Inventor 余锦华谷家琪汪源源邓寅晖童宇宸
Owner FUDAN UNIV
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