Concurrent reconstruction using multiple bed frames or continuous bed motion

• US 7,750,304 B2
• Filed: 12/05/2005
• Issued: 07/06/2010
• Est. Priority Date: 12/15/2004
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Wang, Wenli
• Primary Examiner(s): Porta; David P
• Assistant Examiner(s): BRYANT, MICHAEL C
• Application Number:
  US11/721,718
• Publication Number:
  US 20090238337A1
• Time in Patent Office: 1,674 Days
• Field of Search: 250361 R-36309, 382/131
• US Class Current: 250/363.04
• CPC Class Codes: A61B 6/037 Emission tomography
• 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: 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

In an imaging system (10), a short axial length 4D sinograms are loaded one at a time from a data memory (40). A portion of an image memory (44) that corresponds to a currently reconstructed sinogram subset (1112), is initialized. If a part of the object is already reconstructed, an iterative reconstruction is performed in which the previously reconstructed image (m1) is iteratively improved by using the data from the currently reconstructed overlapping image (m2) to converge on the final image.

Claims

1. An imaging system comprising:

a data memory into which an acquired sinogram of projection data of an entire object is loaded; and

an object reconstruction processor which reconstructs a volumetric image representation of the entire object in successive volumetric image portions, which reconstruction processor is programmed to;

extract sinogram subsets each corresponding to a prespecified axial length from the data memory one at a time, the prespecified axial lengths partially overlapping,

sequentially reconstruct sinogram subsets into the successive volume image portions with an iterative reconstruction algorithm using a trailing portion of a preceding image portion as an initial estimate for a leading edge portion of a currently reconstructed image portion of the image and reconstructs the trailing edge portion of the currently reconstructed image portion to be used as an initial estimate for a leading edge portion of a next subsequent image portion.

2. The system as set forth in claim 1, wherein the reconstruction processor is programmed to iteratively correct the initial estimates of image portion with the sinogram subsets.

3. The system as set forth in claim 1, wherein each image portion is a volume image portion including a plurality of slices.

4. The system as set forth in claim 1, further including:

a motor which repeatedly advances a subject support, which carries the object, a prespecified axial distance; and

a scanner which acquires sinogram data during idle periods between advances to generate a set of the sinogram subsets each subset corresponding to the prespecified axial length.

5. The system as set forth in claim 4, wherein the motor advances the subject support such that adjacent sinogram subsets overlap each other to account for a non-zero tilt sinogram.

6. The system as set forth in claim 4, wherein the sinogram subset axial length is equal to a field of view of the scanner.

7. The system as set forth in claim 1, further including:

a motor which continuously advances a subject support, which carries the object, from an initial position to a final position; and

a scanner which acquires sinogram data during the continuous motion of the subject support and generates a set of the sinogram subsets each subset corresponding to the prespecified axial length.

8. The system as set forth in claim 7, wherein adjacent sinogram subsets overlap each other by an overlap which accounts for non-zero tilt sinograms.

9. The system as set forth in claim 7, wherein the adjacent sinogram subsets overlap a distance which is equal to a field of view of the scanner.

10. The system as set forth in claim 1, further including:

a motor which advances a subject support, which carries the object, from an initial position to a final position; and

a scanner which acquires sinogram data from the subject to generate a set of the sinogram subsets which each has a selected fixed axial length equal to at least a third of a field of view of the scanner.

11. The system as set forth in claim 10, wherein the trailing portion and the leading edge portion each has an axial length equal to one half a field of view of the scanner.

12. The system as set forth in claim 10, wherein the scanner is a PET scanner.

13. An imaging method comprising:

extracting sinogram subsets corresponding to a prespecified axial length from a data memory one at a time which data memory holds sinogram data acquired over a region of an object;

applying an iterative reconstruction algorithm to each of the sinogram subsets in which an estimated image is iteratively improved to converge on a final image, including;

applying an iterative algorithm to each of the sinogram subsets in which an estimated image is iteratively improved to converge on a final image of the region of the object, including;

(a) applying the iterative reconstruction algorithm to a given one of the sinogram subsets to estimate a portion of a current image section using a portion of a preceding image section, and finishing the portion of the preceding image section; and

(b) repeating step (a) to apply the iterative reconstruction algorithm to the plurality of sinogram subsets to reconstruct the final image of the region of the object.

14. The method as set forth in claim 13, wherein the step of applying the iterative reconstruction algorithm further includes:

estimating a first image portion of an n^th image section and finalizing a second image portion of a an (n−

1)^th image section by applying the iterative reconstruction algorithm to an n^th of the sinogram subsets;

estimating the first image portion of an (n+1)^th image section and finalizing the second image portion of the n^th image section by applying the iterative reconstruction algorithm to an (n+1)^th of the sinogram subsets.

15. The method as set forth in claim 14, wherein the step of applying the iterative reconstruction algorithm further includes:

estimating the first image portion of an (n+2)^th image section and finalizing the second portion of the (n+1)^th image section by applying the iterative reconstruction algorithm to an (n+2)^th of the sinogram subsets.

16. The method as set forth in claim 13, further including:

continuously advancing a subject support, which carries the object, from an initial position to a final position;

with a scanner, acquiring the sinogram data during the continuous motion of the subject support; and

storing the sinogram data in the data memory.

17. The method as set forth in claim 16, wherein each image section corresponds to a length of a field of view of the scanner.

18. The method as set forth in claim 16, wherein the adjacent image sections overlap by 50%.

19. The method as set forth in claim 16, further including:

allocating memory space sections to each sinogram subset being acquired by the scanner including;

storing a first acquired sinogram subset into the data memory,

compressing the first acquired subset into a first section portion to release a second section portion, and

allocating the released second section portion for a storage of subsequently acquired second sinogram subset.

20. A diagnostic imaging system comprising:

a scanner which acquires the sinogram data; and

one or more processors programmed to perform the method of claim 13.

21. The method as set forth in claim 13, wherein each portion is a multi-slice volume.

22. An imaging method comprising:

continuously advancing a subject support;

with a scanner, acquiring sinogram data of an object during the continuous motion of the subject support;

storing the sinogram data in a data memory;

extracting sinogram subsets corresponding to a prespecified axial length from the data memory one at a time;

sequentially reconstructing each of the sinogram subsets one at a time into successive volume image sections corresponding to successive regions of the object, each region of the object having an axial length equal to at least a half of a field of view of the scanner;

wherein the step of reconstructing further includes;

initially estimating an image portion adjacent a leading edge one image section and finalizing an image portion adjacent a trailing edge of a preceding image section.

23. The method as set forth in claim 22, wherein the step of reconstructing includes:

applying an iterative reconstruction in which estimated volume image sections are iteratively improved to converge on a final image.