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

Medical MRI image up-sampling method based on adaptive local steering kernel

An adaptive, image-based technology, applied in the field of image processing, can solve the problems of limited generalization ability and large training data sets, and achieve the effect of good generalization ability

Inactive Publication Date: 2020-12-04
CHENGDU UNIV OF INFORMATION TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although their performance is state-of-the-art, these deep learning based methods also have some disadvantages
First, deep learning-based methods often require large training datasets
Second, the generalization ability of deep learning-based methods is also very limited, requiring a tight distribution of training data and test data

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
  • Medical MRI image up-sampling method based on adaptive local steering kernel
  • Medical MRI image up-sampling method based on adaptive local steering kernel
  • Medical MRI image up-sampling method based on adaptive local steering kernel

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach

[0039] Figure 1 Is a flowchart of the up-sampling method of the present invention. The specific implementation method is as follows:

[0040] Step 1: Input data and prepare MRI images to be processed that need up-sampling and denoising.

[0041] Step 2: Calculate the LSK of the local steering nucleus to obtain the local structural information of the MRI image to be processed. The local turning kernel is calculated pixel by pixel, and all the pixels in a domain are used when calculating the LSK of the central pixel in a domain.

[0042] Figure 2 It is a schematic diagram of LSK shapes of adjacent sampled voxels of three image structures. Figure 2 (a) It is a noiseless image, Figure 2 (b) is that the noise intensity is 5% of the maximum intensity, Figure 2 (c) is that the noise intensity is 9% of the maximum intensity. such as Figure 2 As shown, white dots represent the target voxels, and the sampled voxels are contained in a small box centered on the target voxels. Their correspo...

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 an MRI image up-sampling method based on an adaptive local steering kernel, and the method comprises the steps: inputting a to-be-processed MRI image which is prepared for up-sampling and denoising, and firstly adjusting an interpolation weight through a local steering kernel LSK obtained through calculation, so as to adapt to the geometric structure of the image; enhancing the trend of LSK distribution on the noise image through adaptive sharpening, so that noise removal and edge detail enhancement are facilitated; and finally, using Rian deviation correction for correcting deviation generated by asymmetry of Rian noise distribution when a weighted average framework is directly applied to an MRI data set, so that an up-sampling result image obtained through the reconstruction process can keep more context details while denoising is conducted.

Description

Technical field [0001] The invention relates to the field of image processing, in particular to a medical MRI image upsampling method based on adaptive local steering kernel. technical background [0002] One of the main goals of medical imaging is to automatically extract and model the interesting three-dimensional anatomical region (ROIs) in human body. To achieve this goal, people have developed various imaging methods, including computed tomography and magnetic resonance imaging (MRI). MRI can obtain good contrast of soft tissues, so it can effectively observe the difference between normal tissues and diseased tissues. High-resolution images provide a more comprehensive understanding of anatomy at the cost of reducing signal-to-noise ratio (SNR) and increasing imaging time. However, clinically, MRI scanning is usually fast, because longer scanning time will increase the cost, lead to patient discomfort and induce motion artifacts in the images. Therefore, the resolution of cl...

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): G06T5/00G06T5/20
CPCG06T5/20G06T2207/10088G06T2207/20024G06T2207/20192G06T5/73G06T5/70
Inventor 胡靖李欣妍王小龙李琳珂吴锡
Owner CHENGDU UNIV OF INFORMATION 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