Method and apparatus for producing a three-dimensional computerized tomography image of an object with improved conversion of cone beam data to radon data

• US 5,341,460 A
• Filed: 09/13/1993
• Issued: 08/23/1994
• Est. Priority Date: 08/28/1992
• Status: Expired due to Term

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Tam, Kwok C.
• Primary Examiner(s): Herndon, Heather R.
• Assistant Examiner(s): JANKUS, ALMIS R
• Application Number:
  US08/119,803
• Time in Patent Office: 344 Days
• Field of Search: 395/119, 395/124, 395/125, 395/126, 395/132, 395/133, 395/161, 364/413.13, 364/413.14, 364/413.15, 378/5, 378/901
• US Class Current: 345/419
• 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: G06T 2211/421 Filtered back projection [FBP]
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

Cone beam geometry imaging uses an area or two-dimensional array detector and a cone beam x-ray source. Image reconstruction by inverse Radon transformation is used following the calculation of planar integrals. Specifically, the integral is calculated by first changing it to a form wherein fast Fourier transforms can be used to minimize the number of operations in the calculations of the integral.

Claims

1. A method of imaging comprising the steps of:

(a) using a computed tomography cone beam source to apply cone beam energy, dependent upon an object of interest, to an area detector;

(b) detecting the cone beam energy to define a cone beam data image X(i,j) based upon the cone beam energy detected at points (i,j) on the area detector, each point (i,j) having a corresponding detector element;

(c) determining the value of line integral J on the detector where ##EQU7## and where X(t) is cone beam data image at point t along a line of integration, SC'"'"' is the distance from a source S of cone beam energy to a rotation center C'"'"' on the detector, SC is the distance from the source S to a point C, wherein point C is the closest point to the origin on the line of integration and Δ

C is the displacement of C'"'"' from C, the value of J being determined by the substeps of;

(c1) calculating a modified image F(i,j)=X(i,j)/R(i,j) where R(i,j) is the distance from source S to point (i,j) on the area detector;

(c2) calculating a two-dimensional Fourier transform g(m,n) in (m,n) space using a fast Fourier transform and dependent on F(i,j);

(c3) interpolating from g(m,n) a line of Fourier components g.sub.α

(k) where α

is a desired projection angle, k are inputs to g.sub.α

along the line of Fourier components passing through an origin in (m,n) space and oriented perpendicular to the desired projection angle;

(c4) performing a one-dimensional fast Fourier transform on g.sub.α

(k) to obtain a one-dimensional projection p.sub.α

(1);

(c5) obtaining p.sub.α

(s) by one-dimensional interpolation from p.sub.α

(1) where p.sub.α

(s) is a line integral on a particular line at a location s in the desired projection angle α

;

(d) using a Radon inversion process on p.sub.α

(s) to produce a reconstructed three-dimensional image of the object; and

(e) displaying the reconstructed three-dimensional image of the object.

2. The method of imaging of claim 1 wherein steps (c3) through (c5) are repeated for various values of α

.

3. The method of imaging of claim 1 wherein steps (c3) and (c4) take into account at least all points in Radon space which are on or within the object.

4. The method of imaging of claim 3 further comprising causing movement of the source relative to the object and repeating at least steps (b) and (c) for different relative positions of the source.

5. The method of imaging of claim 1 wherein said cone beam energy is x-ray radiation from the source and further comprising the step of causing movement of the source relative to the object.

6. The method of imaging of claim 5 wherein g(m,n) is calculated by calculating where y'"'"'(ij) is the Y displacement of the point (ij) from C'"'"' and taking the 2D fast Fourier transform of g(m,n) being the Fourier transform of

7. The method of imaging of claim 6 wherein p.sub.α

(1) is the one-dimensional projection of G(i,j) in the desired projection angle

8. The method of imaging of claim 7 wherein steps (c3) and (c4) take into account all points in Radon space.

9. The method of imaging of claim 7 further comprising causing movement of the source relative to the object and repeating at least steps (b) and (c) for different relative positions of the source.

10. A method of imaging comprising the steps of:

(a) using a computed tomography cone beam source to apply cone beam energy, dependent upon an object of interest, to an area detector;

(b) detecting the cone beam energy to define a cone beam data image X(i,j) based upon the cone beam energy detected at points (i,j) on the area detector, each point (i,j) having a corresponding detector element;

(c) determining the value of line integral J on the detector where ##EQU8## and where X(t) is cone beam data image at point t along a line of integration, SC'"'"' is the distance from a source S of cone beam energy to a rotation center C'"'"' on the detector, SC is the distance from the source S to a point C, wherein point C is the closest point to the origin on the line of integration and Δ

C is the displacement of C'"'"' from C, the value of J being determined by the substeps of;

(c1) calculating a modified image F(i,j)=X(i,j)/R(i,j) where R(i,j) is the distance from source S to point (i,j) on the area detector;

(c2) calculating a two-dimensional Fourier transform g(m,n) in (m,n) space using a fast Fourier transform and dependent on F(i,j);

(c3) interpolating from g(m,n) a line of Fourier components g.sub.α

(k) where α

is a desired projection angle, k are inputs to g.sub.α

along the line of Fourier components passing through an origin in (m,n) space and oriented perpendicular to the desired projection angle;

(c4) performing a one-dimensional fast Fourier transform on g.sub.α

(k) to obtain a one-dimensional projection p.sub.α

(1);

(c5) obtaining p.sub.α

(s) by one-dimensional interpolation from p.sub.α

(1) where p.sub.α

(s) is a line integral on a particular line at a location s in the desired projection angle α

; and

(d) using a Radon inversion process on p.sub.α

(s) to produce a reconstructed three-dimensional image of the object.

11. The method of imaging of claim 10 wherein g(m,n) is calculated by calculating G(i,j)=F(k,j)/y'"'"'(i,j) where y'"'"'(i,j) is the Y displacement of point (ij) from C'"'"' and taking the 2D fast Fourier transform of G(i,j), g(m,n) being the Fourier transform of G(ij).

12. The method of imaging of claim 11 wherein p.sub.α

(1) is the one-dimensional projection of G(ij) in the desired projection angle α

.

13. The method of imaging of claim 12 further comprising causing movement of the source relative to the object and repeating at least steps (b) and (c) for different relative positions of the source and further comprising displaying the image of the object.

14. An imaging system comprising:

a computed tomography source of cone beam energy;

an area detector for detecting cone beam energy from the source to provide a cone beam data image X(i,j) of an object of interest;

a processor operably connected to the area detector to receive the image S(i,j) from the area detector, said processor having means for determining the value of line integral J on the detector where ##EQU9## and where X(t) is cone beam data image at point t along a line of integration, SC'"'"' is the distance from a source S of cone beam energy to a rotation center C'"'"' on the detector, SC is the distance from the source S to a point C, wherein point C is the closest point to the origin on the line of integration and Δ

C is the displacement of C'"'"' from C, the value of J being determined by the substeps of;

means for calculating a modified image F(i,j)=X(i,j)/R(i,j) where R(i,j) is the distance from source S to point (i,j) on the area detector;

means for calculating a two-dimensional Fourier transform g(m,n) in (m,n) space using a fast Fourier transform and dependent on F(i,j);

means for interpolating from g(m,n) a line of Fourier components g.sub.α

(k) where α

is a desired projection angle, k are inputs to g.sub.α

along the line of Fourier components passing through an origin in (m,n) space and oriented perpendicular to the desired projection angle;

means for performing a one-dimensional fast Fourier transform on g.sub.α

(k) to obtain a one-dimensional projection p.sub.α

(1);

means for obtaining p.sub.α

(s) by one-dimensional interpolation from p.sub.α

(1) where p.sub.α

(s) is a line integral on a particular line at a location s in the desired projection angle α

; and

means for reconstructing a three-dimensional image of the object from p.sub.α

(S) using a Radon inversion process; and

a display operably connected to said processor to display the reconstructed three-dimensional image of the object.

15. The imaging system of claim 14 wherein said means for calculating g(m,n) includes means for calculating G(i,j)=F(i,j)/y'"'"'(i,j) where y'"'"'(ij) is the y displacement of the point (ij) from C'"'"' and means for taking the 2D fast Fourier transform of G(ij), g(m,n) being the Fourier transform of G(ij).

16. The imaging system of claim 15 wherein p.sub.α

(1) is the one-dimensional projection of G(ij) in the desired projection angle α

.

17. The imaging system of claim 16 wherein the source provides x-ray energy.

18. The imaging system of claim 14 wherein the source provides x-ray energy.