Methods and apparatus for CT scout image processing

• US 6,366,638 B1
• Filed: 03/16/2001
• Issued: 04/02/2002
• Est. Priority Date: 03/16/2001
• Status: Expired due to Fees

• Current Assignee: GE Medical Systems Global Technology Company LLC (GE Healthcare Technologies, Inc.)
• Original Assignee: GE Medical Systems Global Technology Company LLC (GE Healthcare Technologies, Inc.)
• Inventors: Hsieh, Jiang, Fox, Stanley H.
• Primary Examiner(s): DUNN, DREW A
• Application Number:
  US09/810,922
• Time in Patent Office: 382 Days
• Field of Search: 378/4, 378/19, 378/15, 378/901, 382/131, 382/132
• US Class Current: 378/19
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/488 involving pre-scan acquisition
• CPC Class Codes: G06T 11/005 Specific pre-processing for tomographic reconstruction, e.g. calibration, source positioning, rebinning, scatter correction, retrospective gating
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

Methods and Apparatus for reducing image artifacts when reconstructing an image with a multislice computed tomographic (CT) imaging scanner are provided. Scout images are generated by obtaining a plurality of projection views of an object, modifying the projection data utilizing a deconvolution kernel, generating a horizontal gradient and a vertical gradient based on the modified projection data, applying helical weights to the horizontal gradient and vertical gradient, and applying a desired level of enhancement to the weighted horizontal and vertical gradients.

Claims

1. A method for reconstructing scout images using a computed tomography imaging system, said method comprising the steps of:

scanning an object to obtain projection data for at least one scout scan;

modifying projection data utilizing a deconvolution kernel;

determining a horizontal gradient and a vertical gradient using the modified projection data;

applying desired levels of enhancement based on the horizontal and vertical gradients; and

generating a plurality of scout images based on the modified projection data and horizontal and vertical gradients.

2. A method in accordance with claim 1 wherein said step of modifying projection data comprises the step of modifying projection data according to the relationship:

3. A method in accordance with claim 2 wherein the deconvolution kernel comprises a five-point deconvolution kernel.

4. A method in accordance with claim 2 wherein the horizontal gradient comprises the relationship:

5. A method in accordance with claim 4 wherein the average of a sample in a horizontal direction, ψ

_i(i,j), comprises the relationship;

${\mathrm{\psi <br><br>}}_i<br><br>\left(i,j\right)=\frac{1}{2<br><br>N+1}<br><br>\underset{k=i-N}{\overset{1+N}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(k,j\right),$

where p′

(j,k) is a sample of projection data of a view j and a channel k, and N is the number of points over which the sample is averaged in the horizontal direction.

6. A method in accordance with claim 2 wherein the vertical gradient comprises the relationship:

7. A method in accordance with claim 6 wherein the average of a sample in a vertical direction, ψ

_j(i,j), comprises the relationship;

${\mathrm{\psi <br><br>}}_j<br><br>\left(i,j\right)=\frac{1}{2<br><br>M+1}<br><br>\underset{k=j-M}{\overset{j+M}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,k\right),$

where p′

(i,k) is a sample of projection data of a channel i and a view k, and M is the number of points over which the sample is averaged in the vertical direction.

8. A method in accordance with claim 1 wherein said step of applying a relative weight comprises the step of determining a relative weight in the horizontal direction according to the relationship:

$\frac{{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ξ

_z(i,j) is the vertical gradient and ε

_x(i,j) is the horizontal gradient.

9. A method in accordance with claim 1 wherein said step of applying a relative weight comprises the step of determining a relative weight in the vertical direction according to the relationship:

$\frac{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ξ

_z(i,j) is a vertical gradient and ξ

_x(i,j) is a horizontal gradient.

10. A method in accordance with claim 5 wherein said step of applying relative weights comprises the step of applying relative weights in the horizontal direction to the average sample ψ

_i(i,j), according to the relationship;

$\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_i<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ψ

_i(i,j) is the average of a sample in the horizontal direction, ξ

_z(i,j) is the vertical gradient, ξ

_x(i,j) is the horizontal gradient, and α

is a parameter that adjusts the amount of enhancement.

11. A method in accordance with claim 7 wherein said step of applying relative weights comprises the step of applying relative weights in the vertical direction to the average sample, ψ

_j(i,j), according to the relationship;

$\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_j<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ψ

_j(i,j) is an average of a sample in the vertical direction, ξ

_z(i,j) is the vertical gradient, ξ

_x(i,j) is the horizontal gradient, and α

is a parameter that adjusts the amount of enhancement.

12. A method according to claim 2 wherein the scout image comprises the relationship:

$\mathrm{\varepsilon <br><br>}<br><br>\left(i,j\right)=p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_i<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_j<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where α

is a parameter that adjusts the amount of enhancement, ψ

_i(i,j) is an average sample in the horizontal direction, ψ

_j(i,j) is an average sample in the vertical direction, where $\frac{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{and}<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}$

and are a relative weights,

wherein ξ

_x(i,j) is the horizontal gradient, and ξ

_z(i,j) is the vertical gradient.

13. A method according to claim 12 wherein said step of applying a desired level of enhancement comprises the step of enhancing the scout image in a horizontal direction according to the relationship:

14. A method according to claim 13 wherein said step of applying a desired level of enhancement comprises the step of enhancing the scout image in a vertical direction according to the relationship:

15. An imaging system for generating scout images of an object, said computed tomography system comprising a computer, a gantry having a detector array, an x-ray source for radiating an x-ray beam along an imaging plane toward a detector array including a plurality of detector cells, the computer coupled to the x-ray source and the gantry, said system configured to:

scan an object to obtain projection data for at least one scout scan;

modify said projection data utilizing a deconvolution kernel;

determine a horizontal gradient and a vertical gradient using said modified projection data;

apply desired levels of enhancement to said horizontal and vertical gradients; and

generate a plurality of scout images based on said modified projection data and horizontal and vertical gradients.

16. An imaging system in accordance with claim 15 wherein projection data is modified according to the relationship:

17. An imaging system in accordance with claim 16 wherein said deconvolution kernel comprises a five-point deconvolution kernel.

18. An imaging system in accordance with claim 16 wherein said horizontal gradient comprises the relationship:

19. An imaging system in accordance with claim 18 wherein said average of the sample in a horizontal direction, ψ

_i(i,j) , comprises the relationship;

${\mathrm{\psi <br><br>}}_i<br><br>\left(i,j\right)=\frac{1}{2<br><br>N+1}<br><br>\underset{k=i-N}{\overset{1+N}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(k,j\right),$

where p′

(j,k) is said sample of projection data of a view j and a channel k, and N is the number of points over which the sample is averaged in the horizontal direction.

20. An imaging system in accordance with claim 16 wherein said vertical gradient comprises the relationship:

21. An imaging system in accordance with claim 16 wherein said average of the sample in a vertical direction, ψ

_j(i,j), comprises the relationship;

${\mathrm{\psi <br><br>}}_j<br><br>\left(i,j\right)=\frac{1}{2<br><br>M+1}<br><br>\underset{k=j-M}{\overset{j+M}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,k\right),$

where p′

(i,k) is a sample of projection data of a channel i and a view k, and M is the number of points over which the sample is averaged in the vertical direction.

22. An imaging system in accordance with claim 15 wherein said relatives weights in a horizontal direction comprises the relationship:

$\frac{{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ξ

_z(i,j) is said vertical gradient and ξ

_x(i,j) is said horizontal gradient.

23. An imaging system in accordance with claim 15 wherein said relative weights in the vertical direction comprises the relationship:

$\frac{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ξ

_z(i,j) is said vertical gradient and ξ

_x(i,j) is said horizontal gradient.

24. An imaging system in accordance with claim 19 wherein said average sample weighted in the horizontal direction comprises the relationship:

$\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_i<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ψ

_i(i,j) is an average of a sample in the horizontal direction, ξ

_z(i,j) represents the vertical gradient, ξ

_x(i,j) is represents the horizontal gradient, and α

is a parameter that adjusts the amount of enhancement.

25. An imaging system in accordance with claim 21 wherein said average sample weighted in the vertical direction comprises the relationship:

$\frac{\mathrm{\alpha <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>{\mathrm{\psi <br><br>}}_j<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where ψ

_j(i,j) is an average of a sample in the vertical direction, ξ

_z(i,j) represents the vertical gradient, ξ

_x(i,j) is represents the horizontal gradient, and α

is a parameter that adjusts the amount of enhancement.

26. An imaging system in accordance with claim 16 wherein said scout image comprises the relationship:

$\mathrm{\varepsilon <br><br>}<br><br>\left(i,j\right)=p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_i<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_j<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where α

is a parameter that adjusts the amount of enhancement, ψ

_i(i,j) is said average sample in the horizontal direction, ψ

_j(i,j) is said average sample in the vertical direction, where $\frac{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{and}<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}$

are said relative weights, wherein ξ

_x(i,j) is said horizontal gradient and ξ

_z(i,j) is said vertical gradient.

27. An imaging system in accordance with claim 26 wherein said scout image enhanced in the horizontal direction comprises the relationship:

28. An imaging system in accordance with claim 27 wherein said scout image enhanced in the horizontal and vertical directions comprises the relationship:

29. A processor programmed to reconstruct scout images in a computed tomography system, said processor configured to:

modify projection data utilizing a five-point deconvolution kernel;

determine a horizontal gradient and a vertical gradient using said modified projection data;

apply desired levels of enhancement to said horizontal and vertical gradients; and

generate a plurality of scout images based on said modified projection data and horizontal and vertical gradients.

30. A processor programmed according to claim 29 wherein said projection data is modified according to the relationship:

31. A processor programmed according to claim 30 wherein said horizontal gradient comprises the relationship:

${\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)=<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{1}{2<br><br>N+1}<br><br>\underset{k=i-N}{\overset{1+N}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(k,j\right)<br><br>,$

where p′

(i,j) is a sample of projection data of a channel i and a view j, p′

(j,k) is a sample of projection data of a view j and a channel k, and N is the number of points over which the sample is averaged in the horizontal direction.

32. A processor programmed according to claim 30 wherein said vertical gradient comprises the relationship:

${\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)=<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{1}{2<br><br>M+1}<br><br>\underset{k=j-M}{\overset{j+M}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,k\right)<br><br>,$

where p′

(i,j) is a sample of projection data of channel i and view j, p′

(i,k) is a sample of projection data of a channel i and a view k, and M is the number of points over which the sample is averaged in the vertical direction.

33. A processor programmed according to claim 30 wherein said scout image comprises the relationship:

$\mathrm{\varepsilon <br><br>}<br><br>\left(i,j\right)=p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_i<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_j<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where α

is a parameter that adjusts the amount of enhancement, ψ

_i(i,j) is an average sample in the horizontal direction, ψ

_j(i,j) is an average sample in the vertical direction, where $\frac{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{and}<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}$

are said relative weights, wherein ξ

_x(i,j) is horizontal gradient, and ξ

_z(i,j) is said vertical gradient.

34. A processor programmed according to claim 33 wherein said scout image enhanced in the horizontal direction comprises the relationship:

35. A processor programmed according to claim 34 wherein said scout image enhanced in the horizontal and vertical direction comprises the relationship:

36. A computer-readable medium in an imaging system, said computer-readable medium comprising:

records of projection data for a plurality of projection views;

records of modified projection data based on said records of projection data modified by a five-point deconvolution kernel;

records of horizontal and vertical gradients based on records of modified projection data;

a plurality of rules to apply desired levels of enhancement to said records of horizontal and vertical gradients; and

a plurality of rules to reconstruct records of scout images based on said modified projection data and horizontal and vertical gradients.

37. A computer-readable medium according to claim 36 wherein said projection data is modified according to the relationship:

38. A computer-readable medium according to claim 37 wherein said horizontal gradient comprises the relationship:

${\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)=<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{1}{2<br><br>N+1}<br><br>\underset{k=i-N}{\overset{1+N}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(k,j\right)<br><br>,$

where p′

(i,j) is a sample of projection data of a channel i and a view j, p′

(j,k) is a sample of projection data of a view j and a channel k, and N is the number of points over which the sample is averaged in the horizontal direction.

39. A computer-readable medium according to claim 37 wherein said vertical gradient comprises the relationship:

${\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)=<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{1}{2<br><br>M+1}<br><br>\underset{k=j-M}{\overset{j+M}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>p^{\mathrm{\prime <br><br>}}<br><br>\left(i,k\right)<br><br>,$

where p′

(i,j) is a sample of projection data of channel i and view j, p′

(i,k) is a sample of projection data of a channel i and a view k, and M is the number of points over which the sample is averaged in the vertical direction.

40. A computer-readable medium according to claim 37 wherein said scout image comprises the relationship:

$\mathrm{\varepsilon <br><br>}<br><br>\left(i,j\right)=p^{\mathrm{\prime <br><br>}}<br><br>\left(i,j\right)-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_i<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}-\frac{{\mathrm{\alpha <br><br>\psi <br><br>}}_j<br><br>\left(i,j\right)<br><br>{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)},$

where α

is a parameter that adjusts the amount of enhancement, ψ

_i(i,j) is an average sample in the horizontal direction, ψ

_j(i,j) is an average sample in the vertical direction, where $\frac{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{and}<br><br>\text{\hspace{1em}<br><br>}<br><br>\frac{{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}{{\mathrm{\xi <br><br>}}_x<br><br>\left(i,j\right)+{\mathrm{\xi <br><br>}}_z<br><br>\left(i,j\right)}$

are said relative weights, wherein ξ

_x(i,j) is horizontal gradient, and ξ

_z(i,j) is said vertical gradient.

41. A computer-readable medium according to claim 40 wherein said scout image enhanced in the horizontal direction comprises the relationship:

42. A computer-readable medium according to claim 41 wherein said scout image enhanced in the horizontal and vertical direction comprises the relationship: