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

A metal artifact correction method for a head CT image

A CT image and metal artifact technology, applied in the field of metal artifact correction of head CT images, can solve problems such as a large number of calculations, structural deformation of metal artifact parts, performance limitations, etc., to achieve PSNR improvement, SSIM improvement, good metal Effects of Artifact Correction Capability

Pending Publication Date: 2019-05-28
NANJING UNIV OF POSTS & TELECOMM
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Linear interpolation (LI) is a traditional method to reduce metal artifacts, which uses interpolation technology to replace the part affected by metal in the original projection data, so as to achieve the purpose of removing metal artifacts, but this method is easy to produce new artifacts , resulting in structural deformation of parts with severe metal artifacts
Iterative reconstruction (LR) is a traditional method of clinical CT, which is limited by the performance of experimental equipment due to the large amount of calculation required
[0004] In summary, although there are many existing hardening artifact correction methods, they all have their own limitations in specific applications.

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
  • A metal artifact correction method for a head CT image
  • A metal artifact correction method for a head CT image
  • A metal artifact correction method for a head CT image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0057] Experimental results such as Figure 5 As shown, the input image with artifacts is on the left, and the output image after metal artifact correction is on the right. It can be seen that the artifacts in the CT image have been well corrected.

[0058] Meanwhile, the present embodiment uses beam hardening correction (BHC) method and linear interpolation (LI) method for comparison. Image 6 Four comparison results after using different artifact correction methods are shown (the display window is [-1000 1000]). Image 6 (a) is the reference image obtained after using the deformable image registration method, Image 6 (b) is the original image with artifacts; Image 6 (c) is the corrected image after using the BHC method Image 6 (d) is the rectified image using the LI method; Image 6 (e) is the image rectified using our method.

[0059] Table 1 lists the Image 6 References in (c)-(e) Figure 5 SSIM value. Table 2 lists the PSNR of the rectified image relative to t...

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 metal artifact correction method for a head CT image, and the method comprises the steps of firstly, building a database which contains and does not contain the artifact headCT image, and carrying out the preprocessing of data through employing a deformable image registration method before the training of a convolutional neural network (CNN); secondly, constructing a simple 17-layer CNN architecture in training to learn metal artifacts, and adopting a GPU to accelerate the speed of training data, so that the learning efficiency of the network is improved. Meanwhile,experimental results show that the method has good metal artifact correction capability, and the PSNR and SSIM values are also obviously improved.

Description

technical field [0001] The invention belongs to metal artifact correction technology, in particular to a metal artifact correction method for head CT images. Background technique [0002] Computed tomography (CT) is one of the commonly used equipment in the medical and industrial fields. It uses a specific CT reconstruction algorithm to obtain slice images, which plays an important role in disease diagnosis and defect inspection. In CT scanning, as a high-density object, the attenuation coefficient of metal to X-rays is much higher than that of human tissue. If there are high-attenuation substances similar to metals during the scanning process, X-rays will be sharply attenuated when passing through these objects. As a result, the corresponding projection data is distorted, and artifacts appear in the reconstructed image. When there is only a single metal in the imaged object, the metal artifacts appear as radial artifacts around it; when there are many metal objects and the...

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): G06T11/00G06T7/30
Inventor 陈茜谢世朋
Owner NANJING UNIV OF POSTS & TELECOMM
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