METHODS AND SYSTEMS FOR TOMOGRAPHIC RECONSTRUCTION

• US 20140363067A1
• Filed: 01/10/2013
• Published: 12/11/2014
• Est. Priority Date: 01/10/2012
• Status: Active Grant

• Current Assignee: Johns Hopkins University
• Original Assignee: Johns Hopkins University
• Inventors: Stayman, Joseph Webster, Siewerdsen, Jeffrey H.
• Granted Patent: US 9,305,379 B2
• Time in Patent Office: Days
• US Class Current: 382/131
• 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 11/008 Specific post-processing after tomographic reconstruction, e.g. voxelisation, metal artifact correction
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 7/38 Registration of image sequences

Abstract

A method for processing an image of a series of images includes receiving first data representing a first previously reconstructed image and receiving second data representing a second image. A second image is reconstructed in accordance with the first data, the second data and a noise model. The noise model is a likelihood estimation. The second image is reconstructed in accordance with a penalty function. The penalty function is a roughness penalty function. The penalty function is updated by iteratively adjusting an image volume estimate. The penalty function is updated by iteratively adjusting a registration term. The penalty function is a prior image penalty function and the prior image penalty function and a registration term are jointly optimized. The penalty function is determined in accordance with a noise model. The function is a p-norm penalty function.

Claims

1. A method for processing an image of a series of images, comprising:

receiving first data representing a first previously reconstructed image;

receiving second data representing a second image; and

reconstructing the second image in accordance with the first data, the second data and a noise model.

2. The method for processing an image of claim 1, wherein the noise model comprises a likelihood estimation.

3. The method for processing an image of claim 1, further comprising reconstructing the second image in accordance with a penalty function.

4. The method for processing an image of claim 3, wherein the penalty function comprises a roughness penalty function.

5. The method for processing an image of claim 3, further comprising updating the penalty function by iteratively adjusting an image volume estimate.

6. The method for processing an image of claim 3, further comprising updating the penalty function by iteratively adjusting a registration term.

7. The method for processing data of claim 3, wherein the penalty function is a prior image penalty function further comprising jointly optimizing the prior image penalty function and a registration term.

8. The method for processing data of claim 3, wherein the penalty function is determined in accordance with a noise model.

9. The method for processing an image of claim 3, wherein the penalty function comprises a p-norm penalty function.

10. The method for processing an image of claim 1, further comprising reconstructing the second image in accordance with a plurality of penalty functions.

11. The method for processing an image of claim 10, wherein the plurality of penalty functions comprises a first penalty function determined in accordance with the first data and a second penalty function determined in accordance with a roughness function of the second data.

12. The method for processing an image of claim 11, wherein the first penalty function comprises a penalty function determined in accordance with a difference between the first data and the second data.

13. The method for processing an image of claim 1, further comprising registering the first and second images in accordance with a registration parameter and an image parameter.

14. The method for processing an image of claim 13, wherein the image parameter is an image volume.

15. The method for processing an image of claim 13, further comprising jointly optimizing the registration parameter and an image parameter.

16. The method for processing an image of claim 15, wherein jointly optimizing the registration parameter and the image parameter further comprises:

(a) performing an optimization over the registration term with the image parameter fixed;

(b) performing an optimization over the image parameter with the registration term fixed; and

(c) repeating steps (a) and (b) until an objective function is maximized.

17. The method for processing data of claim 1, further comprising registering the first image and the second image jointly with reconstructing the second image.

18. The method for processing data of claim 17, further comprising registering the first image and the second image jointly with reconstructing the second image by adjusting a registration term and an image volume estimate.

19. A system for processing an image of a series of images, comprising:

first data representing a first previously reconstructed image;

second data representing a second image; and

a reconstructed second image provided in accordance with the first data, the second data and a noise model.

20. The system for processing an image of claim 19, wherein the reconstructed second image is reconstructed in accordance with a likelihood.

21. The system for processing an image of claim 19, wherein the reconstructed second image is reconstructed jointly with an adjustment of an image registration term.