Methods and devices for locating object in CT imaging

Patent metadata

Field	Value
Publication number	US-9058658-B2
Application number	US-201314037236-A
Country	US
Kind code	B2
Filing date	Sep 25, 2013
Priority date	Sep 29, 2012
Publication date	Jun 16, 2015
Grant date	Jun 16, 2015

Abstract

The present disclosure provides methods and devices for locating a plurality of interested objects in CT imaging. Location of the interested objects in the three-dimensional space can be determined by using three projection images that are substantially perpendicular to each other. The method can rapidly locate interested objects in a CT image without pre-reconstruction of the CT image even if there are a plurality of interested objects in the field of view. The algorithm does not involve interactive steps. The method is rapid and effective, and thus applicable to industrial applications.

First claim

What is claimed is:

1. A method for locating a plurality of interested objects in Computed Tomography (CT) imaging, comprising:
computing from projection data a first projection image at an observation direction perpendicular to a CT tomographic plane by means of a projection synthesis algorithm;
selecting two projection images, referred as a second projection image and a third projection image, from a cone beam projection perpendicular to the observation direction, the second projection image being substantially orthogonal to the third projection image;
determining locations of the plurality of interested objects in the first, second and third projection images; and
computing locations of each of interested objects in the three-dimensional space based on the locations of the plurality of interested objects in the first, second and third projection images.

2. The method according to claim 1 , wherein calculating the first projection image comprises:
calculating a partial derivative of Radon data corresponding to the first projection image based on the projection data by using the relationship between the projection data and Radon transform values; and
calculating the first projection image based on a filter back projection algorithm and the partial derivative of Radon data.

3. The method according to claim 1 , wherein selecting two projection images comprises:
selecting the second projection image and the third projection image based on the first projection image, so that the overlapping region of the plurality of interested objects in the second and third projection images are minimal.

4. The method according to claim 3 , wherein selecting the second projection image and the third projection image based on the first projection image comprises:
segmenting the first projection image to obtain a binary image containing only information on an interested region;
performing a forward projection of fan beams on the binary image to obtain a sonogram of fan beam projection, where a fan angle of the applied fan beam is equal to an open angle of a corresponding light source target point of a central slice of a cone beam system;
counting peaks by a peak founding algorithm for each column of the sinogram; and
selecting two projection angles that are 90 degrees relative to each other from projection angles at which the count of peaks is equal to the number of interested objects in the first projection image, thereby determining the second projection image and the third projection image.

5. The method according to claim 1 , wherein determining locations of the is plurality of interested objects in the first, second and third projection images comprises:
partitioning the first projection image with respect to the respective interested objects and determining center of gravity of each of the interested objects in the first projection image; and
partitioning the second and third projection images with respect to the respective interested objects and determining center of gravity of each of the interested objects in the second and third projection images.

6. A device for locating a plurality of interested objects in Computed Tomography (CT) imaging, comprising:
a computer configured to compute a first projection image at an observation direction perpendicular to a CT tomographic plane by means of a projection synthesis algorithm;
a selector configured to select two projection images, referred as a second projection image and a third projection image, from a cone beam projection perpendicular to the observation direction, the second projection image being substantially orthogonal to the third projection image;
a location determining component configured to determine locations of the plurality of interested objects in the first, second and third projection images; and
a computer configured to compute locations of each of interested objects in the three-dimensional space based on the locations of the plurality of interested objects in the first, second and third projection images.

7. The device according to claim 6 , wherein means for calculating the first projection image comprises:
a calculator configured to calculate a partial derivative of Radon data corresponding to the first projection image based on the projection data by using the relationship between the projection data and Radon transform values; and
a calculator configured to calculate the first projection image based on a filter back projection algorithm and the partial derivative of Radon data.

8. The device according to claim 6 , wherein means for selecting two projection images comprises:
a selector configured to select the second projection image and the third projection image based on the first projection image, so that the overlapping region of the plurality of interested objects in the second and third projection images are minimal.

9. The device according to claim 8 , wherein means for selecting the second projection image and the third projection image based on the first projection image comprises:
a segmenter configured to segment the first projection image to obtain a binary image containing only information on an interested region;
a projector configured to perform a forward projection of fan beams on the binary image to obtain a sonogram of fan beam projection, where a fan angle of the applied fan beam is equal to an open angle of a corresponding light source target point of a central slice of a cone beam system;
a counter configured to count peaks by a peak founding algorithm for each column of the sinogram; and
a selector configured to select two projection angles that are 90 degrees relative to each other from projection angles at which the count of peaks is equal to the number of interested objects in the first projection image, thereby determining the second projection image and the third projection image.

10. The device according to claim 6 , wherein the location determining component configured to determine locations of the plurality of interested objects in the first, second and third projection images comprises:
a first partitioner configured to partition the first projection image with respect to the respective interested objects and determining center of gravity of each of the interested objects in the first projection image; and
a second partitioner configured to partition the second and third projection images with respect to the respective interested objects and determining center of gravity of each of the interested objects in the second and third projection images.

Assignees

• Univ Tsinghua
• Nuctech Co Ltd

Inventors

• Li Liang
• Zhang Li
• Chen Zhiqiang
• Zhao Ziran
• Xing Yuxiang
• Xiao Yongshun
• Wang Qingli

Classifications

• G06T12/20 Primary

  Inverse problem, i.e. transformations from projection space into object space · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• A61B6/5258

  involving detection or reduction of artifacts or noise · CPC title

• G16H40/63

  for local operation · CPC title

• A61B6/4085

  Cone-beams · CPC title