Method, device and equipment for accelerating CT iterative reconstruction

A technique for iterative reconstruction and reconstruction of images, applied in 2D image generation, image data processing, instruments, etc., can solve the time-consuming problems of CT iterative reconstruction methods, achieve increased hardware computing costs, wide application, and reduce calculation of back projection values The effect of the number of times

Pending Publication Date: 2021-08-27
灵图光睿(苏州)智能科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

This patented technology allows for faster CT imagery based upon traditional methods that require calculating all three dimensions separately instead of just one dimension at once. It uses an original reference plate made from this technique to create a mask image called masiré pattern, which represents each pixel within its own area. Only objects outside these areas contribute to the final result. By comparing different images taken under certain conditions or parameters, we get more accurate results than current techniques while also saving computational resources compared to previous approaches. Additionally, combining this approach with graphics processing units' hardwares improves overall performance even better when combined into a single device like GPGs.

Problems solved by technology

This patented technical problem addressed in this patents relates to improving the efficiency and accuracy of x ray ComputedTomography (CT). Current methods for analyzable imagery have slow down their process due to the large number of projections needed during each scan. Therefore, there needs a new way to quickly create high-quality three dimensional ("3-D") models from low resolution two dimensions obtained via filtered backprojection techniques without requiring expensive hardware components such as filters.

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
  • Method, device and equipment for accelerating CT iterative reconstruction
  • Method, device and equipment for accelerating CT iterative reconstruction
  • Method, device and equipment for accelerating CT iterative reconstruction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0062] In a CT scanning system, such as a cone-beam CT scanning system, images of objects scanned in three dimensions can be where i, j, k are integers, and satisfy the conditions 1≤i≤I, 1≤j≤J, 1≤k≤K; I, J, K are the pixels of the three-dimensional image in three dimensions number.

[0063] The iterative algorithm is a method for solving linear system equations. On the basis of a given initial solution, a series of approximate solutions can be generated through iteration to approach the optimal solution. The CT image reconstruction system can be expressed by the following linear system equations:

[0064] Af=b, (1)

[0065] Among them, A is the forward projection matrix; f is the object to be reconstructed, which is an unknown variable; b is the projection data collected by the CT system.

[0066] Since the solution of equation (1) is usually an ill-posed problem and cannot be solved directly, it is usually transformed into the following unconstrained problem based on least sq

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 method, a device and equipment for accelerating CT iterative reconstruction. The method comprises the following steps of constructing a mask image Mask of an object based on a positioning sheet, initializing a reconstructed image f0, calculating residual projection data delta b of the forward projection data b' and the actual measurement projection data b, based on the residual error projection data deltab, only conducting back projection on object image pixels in the mask image Mask, and acquiring a residual error reconstruction image deltaf, updating the reconstructed image f0 by using the residual reconstructed image deltaf, and iterating to obtain a reconstructed image f1, judging whether the reconstruction algorithm model meets a convergence condition or not, and if yes, outputting a reconstructed image f1, or otherwise, continuing iterative computation based on the reconstructed image f1 until the reconstruction algorithm model meets a convergence condition, and outputting a reconstructed image fn, wherein n is the number of iterations. According to the method, on the premise of not additionally introducing hardware equipment and not reducing the image quality, the process of reconstructing the CT image by using an iterative reconstruction algorithm is accelerated by accelerating the process of back projection operation.

Description

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

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
Owner 灵图光睿(苏州)智能科技有限公司
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