System and method of iterative image reconstruction for computed tomography

• US 8,787,521 B2
• Filed: 12/23/2009
• Issued: 07/22/2014
• Est. Priority Date: 12/23/2009
• Status: Active Grant

• Current Assignee: General Electric Company, Purdue Research Foundation (Purdue University), University of Notre DAME Du Lac
• Original Assignee: General Electric Company, Purdue Research Foundation (Purdue University), University of Notre DAME Du Lac
• Inventors: Thibault, Jean-Baptiste, De Man, Bruno Kristiaan Bernard, Zeng, Kai, Yu, Zhou, Bouman, Charles Addison Jr., Sauer, Ken David
• Primary Examiner(s): SONG, HOON K
• Application Number:
  US12/646,653
• Publication Number:
  US 20110150308A1
• Time in Patent Office: 1,672 Days
• Field of Search: 382/131, 378/98.9, 378 4- 20
• US Class Current: 378/20
• 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

Abstract

A system and method include acquisition of a set of image data corresponding to a time period of data acquisition, the set of image data corresponding to a plurality of voxels, wherein each of the plurality of voxels corresponds to a distinct acquisition time within the time period of data acquisition. The system and method further include the modeling of the plurality of voxels as a function of time based on a plurality of kinetic parameters associated therewith and reconstruction of an image from the set of image data based on the modeled plurality of voxels.

Claims

1. A non-transitory computer readable storage medium having stored thereon a computer program comprising instructions, which, when executed by a computer, cause the computer to:

acquire a set of projection data corresponding to a set of image data over a time period of data acquisition, the set of image data corresponding to a plurality of voxels, wherein each of the plurality of voxels corresponds to a distinct acquisition time within the time period of data acquisition;

develop a mathematical model representing the plurality of voxels as a function of time based on a plurality of kinetic parameters associated therewith, the plurality of kinetic parameters corresponding to a kinetic motion of the plurality of voxels;

calculate the plurality of kinetic parameters of the plurality of voxels from the mathematical model; and

reconstruct an image from the set of projection data based on the calculated kinetic parameters of the plurality of voxels.

2. The computer readable storage medium of claim 1 wherein the instructions cause the computer to define a parametric function to model the plurality of voxels as a function of time.

3. The computer readable storage medium of claim 2 wherein the instructions cause the computer to:

define a first parametric function to model a first voxel of the plurality of voxels; and

define a second parametric function to model a second voxel of the plurality of voxels.

4. The computer readable storage medium of claim 2 wherein the instructions cause the computer to define a first linear combination of basis functions having a plurality of estimated coefficients associated therewith corresponding to the plurality of kinetic parameters.

5. The computer readable storage medium of claim 4 wherein the instructions further cause the computer to:

define a second linear combination of basis functions different from the first linear combination of basis functions;

model a first portion of the plurality of voxels based on the first linear combination of basis functions; and

model a second portion of the plurality of voxels based on the second linear combination of basis functions.

6. The computer readable storage medium of claim 4 wherein the instructions cause the computer to:

define the first linear combination of basis functions as a combination of basis functions that remain fixed throughout the reconstruction; and

define the first linear combination of basis functions as one of a combination of polynomial functions, a combination of cosine functions, and a combination of spline functions.

7. The computer readable storage medium of claim 4 wherein the instructions cause the computer to estimate the first linear combination of basis functions during reconstruction using one of acquired data and external information.

8. The computer readable storage medium of claim 2 wherein the instructions cause the computer to define a parametric function having non-linear parameters.

9. The computer readable storage medium of claim 2 wherein the instructions cause the computer to regularize the parametric function to penalize variations in the model.

10. The computer readable storage medium of claim 9 wherein the instructions cause the computer to regularize the plurality of kinetic parameters associated with a first of the plurality of voxels independent of the plurality of kinetic parameters associated with a second of the plurality of voxels.

11. The computer readable storage medium of claim 9 wherein the instructions cause the computer to penalize a spatial variation during reconstruction.

12. The computer readable storage medium of claim 9 wherein the instructions cause the computer to independently regularize each of the plurality of kinetic parameters associated with a voxel.

13. The computer readable storage medium of claim 9 wherein the instructions cause the computer to regularize one of the plurality of kinetic parameters associated with one of the plurality of voxels based on another of the plurality of kinetic parameters associated with the one of the plurality of voxels.

14. The computer readable storage medium of claim 9 wherein the instructions cause the computer to regularize the plurality of kinetic parameters based on a temporal variation in the image.

15. The computer readable storage medium of claim 2 wherein the instructions cause the computer to estimate the plurality of kinetic parameters from the set of acquired data.

16. The computer readable storage medium of claim 15 wherein the instructions further cause the computer to sequentially update the plurality of kinetic parameters using an iterative coordinate descent algorithm.

17. The computer readable storage medium of claim 16 wherein the instructions further cause the computer to define a basis function for each of the plurality of voxels centered at a given time, wherein the given time is determined as a function of a local edge direction and a scan geometry.

18. The computer readable storage medium of claim 2 wherein the instructions cause the computer to:

identify a number of the plurality of voxels having motion associated therewith; and

estimate a plurality of kinetic parameters corresponding to the number of the plurality of voxels.

19. The computer readable storage medium of claim 2 wherein the instructions cause the computer to detect temporal changes in the reconstructed image based on the plurality of kinetic parameters.

32. The computer readable storage medium of claim 1 wherein the instructions cause the computer to:

iteratively calculate the plurality of kinetic parameters; and

iteratively reconstruct the image.

33. The computer readable storage medium of claim 1 wherein the instructions cause the computer to calculate the plurality of kinetic parameters while reconstructing the image.

20. A method comprising:

acquiring a set of projection data corresponding to a plurality of image voxels of a set of CT image data, wherein each of the plurality of image voxels corresponds to a distinct time of data acquisition;

selecting a kinetic parameter model to estimate a plurality of sets of kinetic parameters for the plurality of image voxels, each set of kinetic parameters representing a motion of an object at the distinct time of data acquisition of the associated image voxel;

calculating motion of the plurality of image voxels using the kinetic parameter model; and

reconstructing an image of the object based on the calculated motion; and

wherein the kinetic parameter model estimates the plurality of sets of kinetic parameters for the plurality of image voxels independently from the set of projection data.

21. The method of claim 20 further comprising:

directly reconstructing motion of the object; and

translating the reconstructed motion into voxel intensity changes as a function of time using spatial information acquired in a first-pass reconstruction.

22. The method of claim 21 comprising modeling motion in an x-direction, a y-direction, and a z-direction using x-derivatives, y-derivatives, and z-derivatives of the first-pass reconstruction.

23. The method of claim 21 comprising modeling motion in a direction of a local gradient of the first-pass reconstruction based on one of a size of the local gradient and a change in voxel values in the first-pass reconstruction along the direction of the local gradient.

24. The method of claim 21 comprising modeling motion at a resolution different from a resolution of the image.

25. The method of claim 21 comprising modeling the plurality of voxels based on at least one of a smoothness criteria and a sparseness criteria.

26. An imaging system comprising:

a rotatable gantry having an opening for receiving an object to be scanned;

an x-ray source coupled to the rotatable gantry and configured to project x-rays through the opening;

a generator configured to energize the x-ray source to an energy level to generate x-rays corresponding to the energy level;

a detector having pixels therein, the detector attached to the rotatable gantry and positioned to receive x-rays projected from the x-ray source; and

a computer programmed to;

access a CT projection dataset corresponding to a plurality of image voxels acquired over a period of data acquisition;

model a plurality of kinetic parameters corresponding to the plurality of image voxels, the model representing motion of the plurality of image voxels;

calculate a motion of the object from the model; and

iteratively reconstruct an image from the CT projection dataset based on the calculated motion.

27. The system of claim 26 wherein the computer is programmed to model the plurality of voxels based on a plurality of voxel values estimated directly from the set of acquired data.

28. The system of claim 27 wherein the computer is programmed to:

divide the CT projection dataset into a plurality of data subsets corresponding to intervals of the period of data acquisition; and

estimate the plurality of kinetic parameters corresponding to the plurality of data subsets.

29. The system of claim 26 wherein the computer is programmed to model the plurality of kinetic parameters based on parameters of the imaging system.

30. The system of claim 26 wherein the computer is further programmed to generate a snapshot image of the object frozen at a sample time.

31. The system of claim 26 wherein the computer is further programmed to generate a four-dimensional reconstruction of the object.