Reprojection-based three-dimensional image reconstruction

• US 6,850,587 B1
• Filed: 10/24/2002
• Issued: 02/01/2005
• Est. Priority Date: 10/24/2001
• Status: Expired due to Fees

• Current Assignee: Analogic Corporation
• Original Assignee: Analogic Corporation
• Inventors: Crawford, Carl R., Karimi, Seemeen S.
• Primary Examiner(s): BRUCE, DAVID VERNON
• Application Number:
  US10/279,412
• Time in Patent Office: 831 Days
• Field of Search: 378/4, 378/8, 378/15, 378/901
• US Class Current: 378/15
• CPC Class Codes: A61B 6/027 characterised by the use of a particular data acquisition trajectory, e.g. helical or spiral
• CPC Class Codes: A61B 6/032 Transmission computed tomography [CT]
• CPC Class Codes: A61B 6/488 involving pre-scan acquisition
• 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 2211/432 Truncation
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

A method of reconstructing a three-dimensional image from a set of X-ray projections generated by a scanning system, where the image corresponds to a first volume segment that is a portion of a longer object, includes reconstructing a first estimate image of a second volume segment larger than the first volume segment. The second volume segment includes the first volume segment. The method also includes applying a weighting function to the first estimate image for masking the first volume segment, so as to produce a hypothetical object image. The method further includes re-projecting the hypothetical object image so as to produce a set of simulated hybrid projections corresponding to the first volume segment. The method also includes reconstructing the hybrid projections so as to produce a three-dimensional image corresponding to the first volume segment.

Claims

1. A method of reconstructing a three dimensional image from a set of X-ray projections generated by a scanning system, the image corresponding to a first volume segment that is a portion of a longer object, comprising:

reconstructing, from the set of X-ray projections, a first estimate image of a second volume segment larger than the first volume segment, wherein the second volume segment includes the first volume segment;

applying a weighting function to the first estimate image for masking the first volume segment, so as to produce a hypothetical object image;

re-projecting the hypothetical object image so as to produce a set of simulated projections;

combining the projections generated by the scanning system and simulated re-projections to give hybrid projections corresponding to the first volume segment only;

reconstructing the hybrid projections so as to produce a three dimensional image corresponding to the first volume segment.

2. A method according to claim 1, further including reconstructing a first estimate image of the second volume segment via an approximate reconstruction algorithm including a Helical Feldkamp algorithm.

3. A method according to claim 1, further including reconstructing a first estimate image of the second volume segment via an approximate reconstruction algorithm including a nutating slice reconstruction algorithm.

4. A method according to claim 1, further including reconstructing a first estimate image of a second volume segment larger than the first volume segment, wherein (i) the first volume segment is disposed in a middle portion of the second volume segment, such that a first region ROI includes an intersection of the first volume segment and the second volume segment, and a second region {overscore (ROI)} includes portions of the second volume that do not intersect the first volume.

5. A method according to claim 4, further including extending and tapering the first volume into the second region {overscore (ROI)}.

6. A method according to claim 4, further including extending the second region {overscore (ROI)} to a predetermined distance from the first region ROI, the predetermined distance chosen such that the hybrid projections are reduced to zero at the end of the second region {overscore (ROI)}.

7. A method according to claim 1, further including applying a set of binary coefficients to the first estimate image, so as to produce a hypothetical object image.

8. A method according to claim 7, further including applying binary coefficients substantially equal to zero to portions of the first estimate image corresponding to the first volume segment, and applying binary coefficients substantially equal to one elsewhere.

9. A method according to claim 1, further including re-projecting the hypothetical object via a computer simulation using parameters corresponding to the geometry of the scanning system.

10. A method according to claim 9, further including simulating the scanning system so as to reproduce one or more relationships among the hypothetical object, a simulated X-ray source, and a simulated detector array, wherein the one or more relationships derive from the scanning system.

11. A method according to claim 1, further including subtracting the simulated re-projections from the projections generated by the scanning system to produce hybrid projections and masking the hybrid projections beyond those projection values substantially equal to zero.

12. A method according to claim 1, further including reconstructing the hybrid projections by executing a Radon transform inversion.

13. A method according to claim 1, further including reconstructing the hybrid projections by executing a filtered back-projection algorithm.

14. A system for reconstructing a three dimensional image from a set of X-ray projections generated by a scanning system, the image corresponding to a first volume segment that is a portion of a longer object, comprising:

a first reconstruction processor for reconstructing, from the set of X-ray projections, a first estimate image of a second volume segment larger than the first volume segment, wherein the second volume segment includes the first volume segment;

a masking processor for applying a weighting function to the first estimate image for masking the first volume segment, so as to produce a hypothetical object image;

a re-projection processor for re-projecting the hypothetical object image so as to produce a set of simulated projections;

a combination processor for combining the projections generated by the scanning system and the simulated projections to produce a set of hybrid projections corresponding to the first volume segment;

a second reconstruction processor for reconstructing the hybrid projections so as to produce a three dimensional image corresponding to the first volume segment.

15. A system according to claim 14, wherein the first reconstruction processor reconstructs a first estimate image of the second volume segment via an approximate reconstruction algorithm including a Helical Feldkamp algorithm.

16. A system according to claim 14, wherein the first reconstruction processor reconstructs a first estimate image of the second volume segment via an approximate reconstruction algorithm including a nutating slice reconstruction algorithm.

17. A system according to claim 14, wherein the first reconstruction processor reconstructs a first estimate image of a second volume segment larger than the first volume segment, and (i) the first volume segment is disposed in a middle portion of the second volume segment, such that a first region ROI includes an intersection of the first volume segment and the second volume segment, and a second region {overscore (ROI)} includes portions of the second volume that do not intersect the first volume.

18. A system according to claim 17, wherein the first volume extends and tapers into the second region {overscore (ROI)}.

19. A system according to claim 17, wherein the second region {overscore (ROI)} extends from the first region ROI by a predetermined distance chosen such that the hybrid projections are reduced to zero at the end of the second region {overscore (ROI)}.

20. A system according to claim 14, wherein the masking processor applies a set of binary coefficients to the first estimate image, so as to produce a hypothetical object image.

21. A system according to claim 20, wherein the masking processor applies binary coefficient s substantially equal to zero to portions of the first estimate image corresponding to the first volume segment, and applying binary coefficients substantially equal to one elsewhere.

22. A system according to claim 14, wherein the re-projection processor re-projects the hypothetical object via a computer simulation using parameters corresponding to the geometry of the scanning system.

23. A system according to claim 22, wherein the re-projection processor simulates the scanning system so as to reproduce one or more relationships among the hypothetical object, a simulated X-ray source, and a simulated detector array, where in the one or more relationships derive from the scanning system.

24. A system according to claim 14, wherein the combination processor subtracts the simulated projections from the original projections generated by the scanning system to produce hybrid projections, and masks the hybrid projections beyond those projection values substantially equal to zero.

25. A system according to claim 14, wherein the second reconstruction processor reconstructs the hybrid projections by executing a Radon transform inversion.

26. A system according to claim 14, wherein the second reconstruction processor reconstructs the hybrid projections by executing a filtered back-projection algorithm.

27. A system according to claim 14, wherein the second reconstruction processor is substantially more complex as compared to the first reconstruction processor and provides a substantially higher quality image as compared to an image generated by first reconstruction algorithm processor.

28. A system for reconstructing a three dimensional image from a set of X-ray projections generated by a scanning system, the image corresponding to a first volume segment that is a portion of a longer object, comprising:

means for reconstructing, from the set of X-ray projections, a first estimate image of a second volume segment larger than the first volume segment, wherein the second volume segment includes the first volume segment;

means for applying a weighting function to the first estimate image for masking the first volume segment, so as to produce a hypothetical object image;

means for re-projecting the hypothetical object image so as to produce a set of simulated hybrid projections corresponding to the first volume segment;

means for reconstructing the hybrid projections so as to produce a three dimensional image corresponding to the first volume segment.

29. A method of reconstructing a three dimensional image from a set of X-ray projections generated by a scanning system, the image corresponding to a first volume segment that is a portion of a longer object, comprising:

reconstructing, via an approximate reconstruction algorithm including a Helical Feldkamp algorithm, a first estimate image of a second volume segment larger than the first volume segment, wherein the second volume segment includes the first volume segment, the second volume segment is larger than the first volume segment, and (i) the first volume segment is disposed in a middle portion of the second volume segment, such that a first region ROI includes an intersection of the first volume segment and the second volume segment, and a second region {overscore (ROI)} include those portions of the second volume that do not intersect the first volume;

applying a weighting function to the first estimate image for masking the first volume segment, so as to produce a hypothetical object image, by applying binary coefficients substantially equal to zero to portions of the first estimate image corresponding to the first volume segment, and applying binary coefficients substantially equal to one elsewhere;

re-projecting the hypothetical object image, via a computer simulation using parameters corresponding to the geometry of the scanning system, so as to produce a set of simulated hybrid projections corresponding to the first volume segment;

reconstructing the hybrid projections by executing a Radon transform inversion or a filtered back-projection algorithm so as to produce a three dimensional image corresponding to the first volume segment.

30. A system for reconstructing a three dimensional image from a set of X-ray projections generated by a scanning system, the image corresponding to a first volume segment that is a portion of a longer object, comprising:

a first reconstruction processor for reconstructing, from the set of X-ray projections and via an approximate reconstruction algorithm including a Helical Feldkamp algorithm, a first estimate image of a second volume segment larger than the first volume segment, wherein the second volume segment includes the first volume segment, the second volume segment is larger than the first volume segment, and (i) the first volume segment is disposed in a middle portion of the second volume segment, such that a first region ROI includes an intersection of the first volume segment and the second volume segment, and a second region {overscore (ROI)} include those portions of the second volume that do not intersect the first volume;

a masking processor for applying a weighting function to the first estimate image for masking the first volume segment, so as to produce a hypothetical object image, by applying binary coefficients substantially equal to zero to portions of the first estimate image corresponding to the first volume segment, and applying binary coefficients substantially equal to one elsewhere;

a re-projection processor for re-projecting the hypothetical object image, via a computer simulation using parameters corresponding to the geometry of the scanning system, so as to produce a set of simulated hybrid projections corresponding to the first volume segment;

a second reconstruction processor for reconstructing the hybrid projections by executing a Radon transform inversion or a filtered back-projection algorithm so as to produce a three dimensional image corresponding to the first volume segment.