METHOD AND SYSTEM FOR IMAGE RECONSTRUCTION

• US 20090190814A1
• Filed: 01/25/2008
• Published: 07/30/2009
• Est. Priority Date: 01/25/2008
• Status: Active Grant

• Current Assignee: General Electric Company, Purdue Research Foundation (Purdue University), University of Notre DAME Du Lac
• Original Assignee: University of Notre DAME Du Lac
• Inventors: Yu, Zhou, Thibault, Jean-Baptiste, Bouman, Charles A., Sauer, Ken David
• Granted Patent: US 8,135,186 B2
• Time in Patent Office: Days
• US Class Current: 382/131
• 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 2211/424 Iterative

Abstract

Disclosed is an imaging system including a source generating an x-ray beam, a detector array receiving the x-ray beam and generating projection data, a translatable table configured for disposal of an object thereon and operable to translate in relation to the source and the detector array, the source and the detector array rotating about the translatable table to scan the object, and an image reconstructor electrically coupled to the detector array and configured to reconstruct an image in response to the projection data via an iterative reconstruction technique configured to perform a targeted statistical reconstruction.

Claims

1. An imaging system comprising:

a source generating an x-ray beam;

a detector array receiving said x-ray beam and generating projection data;

a translatable table configured for disposal of an object thereon and operable to translate in relation to said source and said detector array;

said source and said detector array rotating about said translatable table to scan said object; and

an image reconstructor electrically coupled to said detector array and configured to reconstruct an image in response to said projection data via an iterative reconstruction technique configured to perform a targeted statistical reconstruction.

2. The system of claim 1, wherein said reconstruction technique, in performing said targeted statistical reconstruction, includes application of a weighting function that weights each of a plurality of pixels in said image according to a contribution of each of said pixels to a region of interest, wherein said reconstruction technique, in performing said targeted statistical reconstruction, also includes an update in low resolution of each of said plurality of pixels that said weighting function determines to be most contributory to said region of interest.

3. The system of claim 1, wherein said reconstruction technique, in performing said targeted statistical reconstruction, includes performance of a high resolution forward projection in a region of interest and performance of a low resolution forward projection over a remaining area of said image, wherein said reconstruction technique, in performing said targeted statistical reconstruction, also includes correction for a resulting cumulative forward projection that is not scale-invariant.

4. The system of claim 1, wherein said reconstruction technique, in performing said targeted statistical reconstruction, includes an update in high resolution of high frequency components disposed outside of region of interest.

5. The system of claim 1, wherein, the imaging system comprises a computed tomography (CT) imaging system.

6. A method of reconstructing an image of an object comprising:

providing projection data generated by a scanning device; and

performing an iterative reconstruction technique to reconstruct the image in response to said projection data, wherein said reconstruction technique performs a targeted statistical reconstruction.

7. The method of claim 6, wherein said performing said reconstruction technique, in performing said targeted statistical reconstruction, includes weighting each of a plurality of pixels in said image according to a contribution of each of said pixels to a region of interest, wherein said performing said reconstruction technique, in performing said targeted statistical reconstruction, also includes updating in low resolution of each of said plurality of pixels that said weighting determines to be most contributory to said region of interest.

8. The method of claim 7, wherein a high individual weight is assigned to each of said plurality of pixels most closely related to said region of interest, said updating occurring for those of said plurality of pixels that are assigned said high individual weight.

9. The method of claim 7, further including implementing said weighting and said updating via a weighted non-homogeneous coordinate descent algorithm.

10. The method of claim 8, wherein said reconstruction technique includes using said weighting and said updating to increase convergence speed in said region of interest.

11. The method of claim 6, wherein said performing said reconstruction technique, in performing said targeted statistical reconstruction, includes performing a high resolution forward projection in a region of interest and performance of a low resolution forward projection over a remaining area of said image, wherein said performing said reconstruction technique, in performing said targeted statistical reconstruction, also includes correcting for a resulting cumulative forward projection that is not scale-invariant.

12. The method of claim 11, wherein said performing and said correcting include running a low resolution reconstruction and saving final low resolution synthesized projections derived from said low resolution reconstruction, subtracting a region of interest portion of said low resolution forward projection said from final low resolution synthesized projections, adding said high resolution forward projection of said region of interest to said final low resolution synthesized projections, said adding and said subtracting creating a set of corrected synthesized projections, and running a high resolution reconstructing using said set of corrected synthesized projections.

13. The method of claim 11, further including implementing said performing and said correcting via a corrected error sinogram.

14. The method of claim 11, wherein said high resolution forward construction and said low resolution forward construction are not scale-invariant, and said reconstruction technique includes using said performing and said correcting to correct for the non-scale-invariance of said high resolution forward construction and said low resolution forward construction.

15. The method of claim 6, wherein said performing said reconstruction technique, in performing said targeted statistical reconstruction, includes an updating in high resolution of high frequency components disposed outside of a region of interest and an updating of the elements disposed inside of the region of interest.

16. The method of claim 15, further including performing an edge detection algorithm at an end of a low resolution reconstruction to identify said high frequency components.

17. The method of claim 16, wherein said performing occurs via a Sobel edge detector to detect edges.

18. The method of claim 17, further including thresholding an image map resulting from said running of said Sobel edge detector.

19. The method of claim 18, further including combining said image map with pixels comprising said region of interest.

20. The method of claim 15, wherein said reconstruction technique includes using said updating to improve reconstruction quality of the image.

21. A method of reconstructing an image comprising:

scanning an object to acquire projection data;

performing an iterative reconstruction technique to reconstruct the image in response to said projection data, wherein said reconstruction technique includes,

weighting each of a plurality of pixels in said image according to a contribution of each of said pixels to a region of interest,

updating in low resolution each of said plurality of pixels that said weighting determines to be most contributory to said region of interest,

performing a high resolution forward projection in said region of interest and performing a low resolution forward projection over the remaining area of said image,

correcting for a resulting cumulative forward projection that is not scale-invariant, and

updating in high resolution high frequency components disposed outside of said region of interest; and

performing a targeted statistical reconstruction via said reconstruction technique.

22. The method of claim 21, wherein said reconstruction technique employs a weighted non-homogenous iterative coordinate descent algorithm, which incorporates said weighting, said updating, said performing, said correcting, and said updating, in order to perform said targeted statistical reconstruction