Method and apparatus for three dimensional image reconstruction from multiple stereotactic or isocentric backprojections

• US 5,588,033 A
• Filed: 06/06/1995
• Issued: 12/24/1996
• Est. Priority Date: 06/06/1995
• Status: Expired due to Fees

• Current Assignee: St Jude Children's Research Hospital Incorporated (Abbott Laboratories)
• Original Assignee: St Jude Children's Research Hospital Incorporated (Abbott Laboratories)
• Inventors: Yeung, Kah T. D.
• Primary Examiner(s): Porta, David P.
• Assistant Examiner(s): Bruce, David Vernon
• Application Number:
  US08/466,829
• Time in Patent Office: 567 Days
• Field of Search: 364/413.14, 364/413.15, 364/413.16, 364/413.21, 250/363.02, 250/363.04, 250/363.09, 378/4, 378/10, 378/901
• US Class Current: 378/4
• CPC Class Codes: G06T 11/006 Inverse problem, transformation from projection-space into object-space, e.g. transform methods, back-projection, algebraic methods
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

A method and apparatus for reconstructing a three dimensional image of an object from a plurality of two dimensional radiographic images includes the steps of performing radiography of the object from a point source to produce a two dimensional image on a film, backprojecting the two dimensional image through a reconstruction volume to the point source to produce a backprojected image, repeating these steps for a plurality of viewing angles and determining the intersection of the backprojected images and the reconstruction volume. This intersection defines a three dimensional, reconstructed image of the object for each view. Upon superimposition of these three dimensional reconstructed images, the voxels that all the images have in common define a final reconstructed image of the object. Additionally, the present invention includes a stereotactic localizer frame that has eight sides for providing at least four views of the object. The eight faces allow radiography of the object to be performed at almost any viewing angle. This method can be used in clinical applications for the reconstruction of three dimensional structures within the patient'"'"'s body through multiple two dimensional radiographic views of the patient.

Claims

1. A method for reconstructing a three dimensional image of an object from a plurality of radiographic images, comprising the steps of:

backprojecting each radiographic image through a reconstruction volume to produce a plurality of corresponding backprojected images; and

determining an intersection of said plurality of backprojected images in said reconstruction volume, said intersection defining a three dimensional, reconstructed image of the object.

2. The method of claim 1, wherein, for each radiographic image, said backprojecting step comprises:

determining a position, relative to said reconstruction volume, of a point source used to produce said radiographic image;

determining a position, relative to said reconstruction volume, of an image plane containing said radiographic image;

digitizing said radiographic image to produce a digitized image, said digitized image being composed of a plurality of pixels; and

backprojecting each pixel of said digitized image through said reconstruction volume to said point source to form raylines.

3. The method of claim 2, wherein said step of determining said intersection comprises:

defining a reconstruction volume in three dimensional space, said reconstruction volume having a plurality of voxels; and

determining, for each backprojected image, the intersection of said raylines with said plurality of voxels to provide a plurality of intersected voxels, wherein each intersected voxel defines a portion of the three dimensional, reconstructed image of the object.

4. The method of claim 3, wherein said step of determining the intersection of said raylines with said plurality of voxels comprises:

(a) determining, for one of said plurality of backprojected images, the intersection of said raylines with said plurality of voxels to provide a first set of intersected voxels;

(b) determining, for another of said backprojected images, the intersection of said raylines with said plurality of voxels to provide a second set of intersected voxels;

(c) superimposing said first set of intersected voxels and said second set of intersected voxels to form a subset of shared intersected voxels;

(d) repeating steps (b) and (c) for each of the remaining backprojected images from said plurality of backprojected images, to produce a final subset of shared intersected voxels, wherein said final subset of shared intersected voxels defines the three dimensional, reconstructed image of the object.

6. The method of claim 3, further comprising the step of:

displaying the three dimensional, reconstructed image of the object, wherein each intersected voxel is displayed using a first intensity and each voxel that is not intersected by a rayline from each of said plurality of backprojected images is displayed using a second intensity.

7. The method of claim 3, further comprising the step of:

displaying the three dimensional, reconstructed image of the object, wherein each intersected voxel is displayed using a first color and each voxel that is not intersected by a rayline from each of said plurality of backprojected images is displayed using a second color.

5. The method of 1, wherein said backprojecting step comprises:

backprojecting at least six radiographic images through a reconstruction volume to produce at least six corresponding backprojected images.

8. A method for three dimensional imaging of an object, comprising the steps of:

performing radiography of the object from a point source to produce a radiographic image on an image plane;

backprojecting said radiographic image through a reconstruction volume to said point source to produce a backprojected image;

repeating steps (a) and (b) for a plurality of imaging angles to produce a plurality of backprojected images; and

determining an intersection of said plurality of backprojected images in said reconstruction volume, said intersection defining a three dimensional, reconstructed image of the object.

9. The method of claim 8, wherein, for each radiographic image, said backprojecting step comprises:

determining a position, relative to said reconstruction volume, of said point source used to produce said radiographic image;

determining a position, relative to said reconstruction volume, of an image plane containing said radiographic image;

digitizing said radiographic image to produce a digitized image, said digitized image being composed of a plurality of pixels; and

backprojecting each pixel of said digitized image through said reconstruction volume to said point source to form raylines.

10. The method of claim 9, wherein said step of determining said intersection comprises:

defining a reconstruction volume in three dimensional space, said reconstruction volume having a plurality of voxels; and

determining, for each backprojected image, the intersection of said raylines with said plurality of voxels to provide a plurality of intersected voxels, wherein each voxel that is intersected by a rayline from each of said plurality of backprojected images defines a portion of the three dimensional, reconstructed image of the object.

11. The method of claim 8, wherein said backprojecting step comprises:

backprojecting at least six radiographic images through a reconstruction volume to produce at least six corresponding backprojected images.

12. An apparatus for reconstructing a three dimensional image of an object from a plurality of radiographic images, comprising:

means for backprojecting each radiographic image through a reconstruction volume to produce a plurality of corresponding backprojected images; and

means for determining an intersection of said plurality of backprojected images in said reconstruction volume, said intersection defining a three dimensional, reconstructed image of the object.

13. The apparatus of claim 12, wherein said backprojecting means comprises:

means for determining a position, relative to said reconstruction volume, of a point source used to produce said radiographic image;

means for determining a position, relative to said reconstruction volume, of an image plane containing said radiographic image;

means for digitizing said radiographic image to produce a digitized image, said digitized image being composed of a plurality of pixels; and

means for backprojecting each pixel of said digitized image through said reconstruction volume to said point source to form raylines.

14. The apparatus of claim 12, wherein said means for determining an intersection comprises:

means for defining a reconstruction volume in three dimensional space, said reconstruction volume having a plurality of voxels; and

means for determining, for each backprojected image, the intersection of said raylines with said plurality of voxels to provide a plurality of intersected voxels, wherein each intersected voxel defines a portion of the three dimensional, reconstructed image of the object.

15. The apparatus of claim 14, further comprising:

means for displaying the three, dimensional, reconstructed image of the object, wherein each intersected voxel is displayed using a first intensity and each voxel that is not intersected by a rayline from each of said plurality of backprojected images is displayed using a second intensity.

16. The apparatus of claim 14, further comprising:

means for displaying the three, dimensional, reconstructed image of the object, wherein each intersected voxel is displayed using a first color and each voxel that is not intersected by a rayline from each of said plurality of backprojected images is displayed using a second color.