Backprojection reconstruction method for CT imaging

• US 7,545,901 B2
• Filed: 07/07/2006
• Issued: 06/09/2009
• Est. Priority Date: 07/08/2005
• Status: Active Grant

• Current Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Original Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Inventors: Mistretta, Charles A.
• Primary Examiner(s): Glick; Edward J
• Assistant Examiner(s): Corbett; John M
• Application Number:
  US11/482,858
• Publication Number:
  US 20070009080A1
• Time in Patent Office: 1,068 Days
• Field of Search: 378/4, 378/9, 378/210, 378/901
• US Class Current: 378/4
• CPC Class Codes: A61B 6/507 for determination of haemodynamic parameters, e.g. perfusion CT
• 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 2211/412 Dynamic
• CPC Class Codes: G06T 2211/424 Iterative
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

Two-dimensional or three-dimensional, time-resolved CT frame images are acquired during a dynamic study of a subject. A composite image is produced and this is used to reconstruct each CT frame image by weighting the backprojection of each projection view acquired for that image frame by the corresponding value in the composite image. This weighted backprojection enables artifact-free image frames to be produced with far fewer projection views of the subject. The composite image may be reconstructed from views acquired separately, or it may be produced by combining views acquired during the course of the dynamic study.

Claims

1. A method for producing an image of a subject positioned in the field of view (FOV) of a computed tomography (CT) system, the steps comprising:

a) acquiring with the CT system a set of projection views of the subject positioned in the FOV;

b) producing a composite image with the CT system which indicates an attenuation value at each composite image pixel of the subject positioned in the FOV; and

c) reconstructing an image of the subject by;

c)i) backprojecting each projection view in the set into the FOV and weighting the value backprojected into each image pixel by the attenuation value of the corresponding pixel in the composite image; and

c)ii) summing the backprojected values for each image pixel to produce the image.

2. The method as recited in claim 1 in which each image pixel backprojected value μ

_n is calculated in step c)i) as

3. The method as recited in claim 1 in which step b) includes editing the composite image to remove an object therein and to thereby substantially reduce the appearance of that object in the reconstructed image.

4. The method as recited in claim 1 in which the weighting step c)i) includes normalizing each projection view using a corresponding projection view from the composite image and multiplying the backprojected value by the corresponding pixel in the composite image.

5. The method as recited in claim 1 which further includes:

d) repeating steps a) and c) to reconstruct a series of images that depict the subject during an examination; and

e) periodically updating the composite image during the reconstruction of the series of images to depict therein changes that occur in the subject during the examination.

6. The method as recited in claim 5 in which the updating of the composite image includes reconstructing the composite image using projection views acquired in step a).

7. The method as recited in claim 1 in Which the composite image is produced by acquiring data with the CT system in response to a gating signal indicative of a selected physiological event in the subject, and step a) is performed in response to a gating signal indicative of the selected physiological event.

8. The method as recited in claim 7 in which the selected physiological event is cardiac phase of the subject'"'"'s heart.

9. The method as recited in claim 1 in which steps a) and c) are repeated to acquire a plurality of sets of projection views and reconstruct a corresponding plurality of images in which each set of projection views acquired in step a) are interleaved with the projection views in other sets of acquired projection views and the composite image is produced by combining sets of interleaved projection views acquired in step a).

10. The method as recited in claim 1 which includes detecting the location of the subject when performing step a) and registering the composite image with the detected subject location.

11. The method as recited in claim 1 in which the FOV is three-dimensional, a three-dimensional image is produced, and the image I_(x,y,z) reconstructed in step c) is:

I(x,y,z)=Σ

(P(r,θ

,φ

)*C(x,y,z)<sub>(r,θ

,φ

)</sub>/P_c(r,θ

,φ

))

where the summation (Σ

) is over all projection views in the acquired set;

I_(x,y,z) is the image value at FOV pixel location x,y,z;

P<sub>(r,θ

,φ

)</sub> is the back projected value from the view acquired at a view angle θ

, φ

;

C_(x,y,z) is the composite image value at the pixel location x,y,z; and

P_c(r,θ

,φ

) is the profile value projected from the composite image at the view angle θ

, φ

.

12. The method as recited in claim 1 which includes:

d) acquiring a mask image that depicts at each of its image pixels the subject positioned in the FOV prior to the administration of a contrast agent;

e) administering a contrast agent to the subject prior to performing steps a) and b); and

f) subtracting the mask image from the composite image prior to performing step c).

13. The method as recited in claim 12 in which both the mask image and the composite image are acquired as sets of projection views and step f) is performed by subtracting projection views in the mask image set from corresponding projection views in the composite image set.

14. The method as recited in claim 12 in which step f) is performed by subtracting pixels in the mask image from corresponding pixels in the composite image.

15. The method as recited in claim 1 which includes:

d) acquiring a mask image as a set of projection views;

e) administering a contrast agent to the subject prior to performing steps a) and b); and

f) subtracting from each projection view in the set a corresponding projection view acquired in step d) before performing step c).

16. The method as recited in claim 15 which includes:

g) subtracting the mask image from the composite image prior to performing step c).

17. The method as recited in claim 1 which further includes:

d) repeating steps a) and c) to reconstruct a series of image frames that depict the subject during an examination; and

e) registering the composite image with each set of image frame projection views prior to performing step c).

18. The method as recited in claim 1 which includes:

d) reprojecting the composite image at view angles used to perform step a); and

in which the weighting in step c)i) includes normalizing each projection view by dividing values therein by corresponding values in the projection view of the composite image at the same view angle.

19. A method for producing an image of a subject positioned in the field of view (FOV) of a computed tomography (CT) system, the steps comprising:

a) acquiring with the CT system a set of projection views of the subject positioned in the FOV;

b) repeating step a) to acquire additional sets of projection views, wherein the sets of acquired projection views are interleaved with each other;

c) reconstructing a composite image from acquired projection views which indicates an attenuation value at each composite image pixel of the subject positioned in the FOV; and

d) reconstructing a frame image of the subject by;

d)i) normalizing each projection view in one of said sets of projection views;

d)ii) backprojecting each normalized projection view into the FOV;

d)iii) summing the backprojected pixel values for the set of projection views; and

d)iv) multiplying the summed pixel values by the corresponding pixel value in the composite image.

20. The method as recited in claim 19 in which steps c) and d) are repeated to produce additional frame images, wherein the acquired projection views used to reconstruct the composite image in step c) are updated to reflect changes occurring over time in the subject.

21. The method as recited in claim 19 which includes acquiring and reconstructing a mask image of the subject;

injecting a contrast agent into the subject prior to performing step a); and

subtracting the mask image from the composite image produced in step c).

22. A method for producing an image of a subject positioned in the field of view (FOV) of a computed tomography (CT) system having a plurality of x-ray sources disposed along an axis of rotation, the steps comprising:

a) acquiring a plurality of image frames with each x-ray source, each image frame including a set of projection views of the subject and the projection views in each set being interleaved with the projection views in the other sets;

b) reconstructing a region composite image for each x-ray source from interleaved projection views in a plurality of said sets of projection views;

c) reconstructing an image frame for one region using the projection views corresponding to that region and the composite image for that region which includes;

c)i) normalizing each projection view corresponding to that region;

c)ii) backprojection each normalized projection view into the region of the FOV;

c)iii) summing backprojected values produced in step c)ii) at each pixel in the region; and

c)iv) weighting the backprojected values with corresponding pixel values in the composite image for the region;

d) repeating step c) to reconstruct additional region image frames; and

e) combining the region image frames to produce the image frame.

23. The method as recited in claim 22 in which includes:

acquiring a mask image for each region;

administering a contrast agent to the subject prior to performing step a); and

step b) includes subtracting from each region composite image the corresponding region mask image.

24. A method for producing a plurality of image frames of a subject in the field of view (FOV) of a computed tomography (CT) system the steps comprising:

a) acquiring a plurality of image frames, each image frame including a set of projection views of the subject;

b) reconstructing a composite image by combining the projection views from the acquired image frames; and

c) reconstructing each image frame by

c)i) normalizing each image frame projection view using information derived from the composite image;

c)ii) backprojecting each normalized projection view;

c)iii) summing the backprojected values; and

c)iv) weighting the backprojected values with the values in the composite image at corresponding locations in the FOV.

25. The method as recited in claim 24 in which each image frame is comprised of projection views that under sample Radon space, the projection views in each acquired image frame are interleaved with the projection views in other acquired image frames, and the composite image is reconstructed in step b) by using substantially all the interleaved projection views.

26. The method as recited in claim 24 in which each image frame is comprised of projection views acquired with an x-ray beam intensity that is substantially less than the x-ray beam intensity used to acquired an image frame of comparable quality and the composite image is reconstructed in step b) by averaging substantially all the projection views acquired at the same view angle.

27. A method for acquiring perfusion images of tissues positioned in the field of view (FOV) of a computed tomography (CT) system, the steps comprising:

a) acquiring a fully sampled, mask image of the tissues with the CT system;

b) administering a contrast agent;

c) acquiring a series of low dose image frames of the tissues with the CT system;

d) subtracting mask image projection views acquired in step a) from image frame projection views acquired in step c) to produce difference image frame projection views;

e) combining difference image frame projection views from a plurality of said image frames and reconstructing a composite image therefrom; and

f) reconstructing perfusion images by;

f)i) backprojecting each difference image frame projection view from an acquired image frame into the FOV and weighting the value backprojected into each image pixel by the value of a corresponding pixel in the composite image; and

f)ii) summing the backprojected values for each image pixel.

28. The method as recited in claim 27 in which each image pixel backprojected value μ

_n is calculated in step f)i) as

29. The method as recited in claim 27 in which the weighting in step f)i) includes normalizing each image frame difference projection view using a corresponding projection view from the composite image and multiplying the backprojected value by the corresponding pixel in the composite image.

30. The method as recited in claim 27 in which step c) is performed by acquiring each image frame as an undersampled set of projection views which are interleaved with the projection views acquired for the other image frames.

31. The method as recited in claim 30 which in step e) is performed by combining into a data set image frame difference projection views acquired at different view angles and reconstructing the composite image using the combined data set.

32. The method as recited in claim 30 in which step e) is performed by averaging image frame difference projection views acquired at the same view angles and reconstructing the composite image using the averaged image frame difference projection views.

33. The method as recited in claim 27 in which step c) is performed by acquiring each image frame with an x-ray beam intensity that is substantially less than the x-ray beam intensity used to acquired an image frame of comparable quality.

34. A method for producing an image of a subject positioned in the field of view (FOV) of a computed tomography (CT) system, the steps comprising:

a) acquiring with a CT system a composite image of the subject;

b) acquiring with the CT system a set of projection views of the subject;

c) measuring the motion of the subject with respect to the composite image using projection views acquired in step b);

d) registering the composite image with the current subject position using the motion measured in step c); and

e) reconstructing a frame image of the subject by;

e)i) backprojecting each projection view in the set into the FOV and weighting the value backprojected into each image pixel by the value of a corresponding pixel in the registered composite image; and

e)ii) summing the backprojected values for each image pixel to produce the frame image.

35. The method as recited in claim 34 in which the weighting step e)i) includes normalizing each projection view using a corresponding projection view of the registered composite image and multiplying the backprojected value by the corresponding pixel in the registered composite image.