Construction method and application of three-dimensional MRI image denoising model based on Wasserstein generative adversarial network

A technology of network model and construction method, applied in image enhancement, image analysis, image coding, etc., can solve problems such as limited denoising performance

Active Publication Date: 2019-08-06
SICHUAN UNIV
View PDF2 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, so far, there are very few studies on deep learning in MRI image denoising, and only one article using a simple end-to-end CNN network model for MRI image processing (Jiang, D., Dou, W., Vosters, L., Xu, X., Sun, Y., Tan, T., 2018. Denoising of 3D magnetic resonance images with multichannel residual learning of convolutional neural network. Japanese Journal of Radiology, 1-9), due to the relatively simple network structure, its Limited denoising performance

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
  • Construction method and application of three-dimensional MRI image denoising model based on Wasserstein generative adversarial network
  • Construction method and application of three-dimensional MRI image denoising model based on Wasserstein generative adversarial network
  • Construction method and application of three-dimensional MRI image denoising model based on Wasserstein generative adversarial network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] This embodiment is based on the construction method of the 3D MRI image denoising model of the Wasserstein generation confrontation network, such as figure 1 As shown, it includes the following steps:

[0058] (1) Construct a training set, using noisy 3D MRI image data and corresponding noise-free 3D MRI image data to construct a training set.

[0059] The data used in this embodiment comes from the IXI dataset: this dataset is a public dataset and can be downloaded from http: / / brain-development.org / ixi-dataset / . This data set comes from three different hospitals and brain images of different patients, including MRI images T1, T2, and PDW three MRI modal images; for the acquisition of noise images, clinical data (noise-free images) and computer simulation noise can be used to get a noisy image. Its simulation formula is as follows

[0060]

[0061] Among them, noise img Represents the synthesized noise image, level represents the noise level, noise 1 and noise ...

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 construction method and application of a three-dimensional MRI image denoising model based on a Wasserstein generative adversarial network. According to the invention, a Wasserstein generative adversarial network is used as a basic model to process an MRI noise image; large-scale data is used for training, a model can automatically learn potential association between a noise image and a non-noise image from the data, confrontation loss, perception loss and MSE loss are introduced into construction of a training model loss function, and the constructed model has a gooddenoising effect on a three-dimensional MRI image.

Description

technical field [0001] The invention belongs to the technical field of MRI (nuclear magnetic resonance imaging) image processing, and relates to an MRI image denoising method, in particular to a construction method and application of a three-dimensional MRI image denoising model based on a Wassertein generative confrontation network. Background technique [0002] In the process of clinical diagnosis, every detail will affect the diagnosis results of doctors, which requires more researchers to invest time and energy in researching new imaging techniques and image processing methods. [0003] As a routine method of medical diagnosis and one of the most advanced imaging technologies in the medical imaging field, MRI has the advantages of no trauma to the human body, no radiation damage, and direct tomographic imaging in any direction. However, under the conditions of high-speed imaging and high-resolution imaging, problems such as eddy current distortion are prone to occur, whi...

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): G06T5/00G06T9/00
CPCG06T5/002G06T9/001G06T9/002G06T2207/10088G06T2207/20081G06T2207/20084
Inventor 张意冉茂松周激流
Owner SICHUAN UNIV
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap