Simplified cone beam image reconstruction using 3D backprojection

• US 5,881,123 A
• Filed: 06/29/1998
• Issued: 03/09/1999
• Est. Priority Date: 03/31/1998
• Status: Expired due to Term

• Current Assignee: Siemens Medical Solutions USA Incorporated (Siemens AG)
• Original Assignee: Siemens Corporate Research Incorporated (Siemens AG)
• Inventors: Tam, Kwok
• Primary Examiner(s): Porta, David P.
• Assistant Examiner(s): Bruce, David Vernon
• Application Number:
  US09/106,537
• Time in Patent Office: 253 Days
• Field of Search: 378/4, 378/15, 378/901
• US Class Current: 378/4
• 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: 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 2211/421 Filtered back projection [FBP]
• CPC Class Codes: Y10S 378/901 Computer tomography program or processor

Abstract

A scanning and data acquisition method and apparatus for three dimensional (3D) computerized tomography (CT) imaging of a region-of-interest (ROI) in an object, wherein image reconstruction processing is applied to a plurality of sets of cone beam projection data, each set being acquired on a detector at a corresponding plurality of scan path source positions. A first image reconstruction processing step comprises applying a mask to each set of the projection data to form masked data sets. The mask for each data set is formed by cone beam projections onto the plane of the detector of portions of the source scan path that are above and below the source position that acquired the data set being masked. The next image reconstruction processing step comprises ramp filtering of each masked data set along a plurality of parallel lines formed therein that are parallel with and in the direction of a parallel projection into the data set of a line that is tangent to the scan path at the source position that acquired that data set and in the direction of the next source position, to generate filtered data sets. Each filtered data set is then subjected to a weighted 3D backprojection into a 3D space, thereby reconstructing in a piecewise manner a 3D image of the ROI in the object.

Claims

1. A scanning and data acquisition method for three dimensional (3D) computerized tomography (CT) imaging of a region-of-interest (ROI) in an object, comprising the steps of:

applying radiation energy from a cone beam source to at least the ROI of the object;

defining a source scanning trajectory as a scan path traversed by the source;

using the cone beam source, fixed relative to an area detector with both source and detector movably positioned relative to the object, to scan about the ROI in the object along the scan path;

scanning at a plurality of source positions in a direction along the scan path to cause said area detector to acquire a set of cone beam projection data at each of said source positions;

applying a mask to each set of the projection data to form masked data sets;

convolution processing each masked data set along a plurality of parallel lines formed therein that are parallel with and in the direction of a parallel projection into the data set of a line that is tangent to the scan path at the source position that acquired that data set and in the direction of scanning, to generate filtered data sets;

subjecting the data of each filtered data set to a weighted 3D backprojection into a 3D space, thereby reconstructing in a piecewise manner a 3D image of the ROI in the object.

2. The method of claim 1, wherein said applying step includes a forming step for forming the mask for each data set by cone beam projecting onto the plane of the detector portions of the source scan path that are above and below the source position that acquired the data set being masked.

3. The method of claim 2, wherein the forming step forms a mask for each source position that has a top curve and a bottom curve, and all of the projection data outside the top and bottom curves of the mask are set to zero, for forming the masked data sets.

4. The method of claim 3, wherein the source scanning trajectory is defined as a spiral scan path that surrounds the ROI in the object, and is connected to a first scan path circle at a top portion of the ROI in the object, and a second scan path circle at a bottom portion of the ROI in the object.

5. The method according to claim 4, wherein at source positions of the scan path at the top of the ROI in the object, the top curve of the mask includes a horizontal line portion formed from a projection of the first scan path circle, and the bottom curve of the mask is formed from a projection of the spiral scan path.

6. The method according to claim 4, wherein at source positions of the scan path at the bottom of the ROI in the object, the top curve of the mask is formed from a projection of the spiral scan path, and the bottom curve of the mask includes a horizontal line portion formed from a projection of the second scan path circle.

7. The method according to claim 5, wherein the top curve is formed from an intersection of a projection of the first scan path circle and a projection of the spiral scan path.

8. The method according to claim 6, wherein the bottom curve is formed from an intersection of a projection of the second scan path circle and a projection of the spiral scan path.

9. The method of claim 1, wherein the convolution processing step comprises a ramp filtering of the masked data set.

10. The method of claim 9, wherein the convolving processing comprises the function:

##EQU4##

11. The method of claim 1, wherein the weighted 3D backprojection step comprises performing a weighted 3D backprojection of the data of each of the filtered data sets onto a plurality of sample points in a 3D object volume.

12. The method of claim I 1, wherein the 3D object volume corresponds to the ROI volume in the object that is scanned by the cone beam source.

13. The method of claim 12, wherein the weight assigned to the filtered data that is backprojected to each sample point in the 3D object volume, is proportional to the inverse of the square of the distance between each sample point and the location of the cone beam source that acquired that data set being backprojected.

14. Apparatus for three dimensional (3D) computerized tomography (CT) imaging of a region-of-interest (ROI) in an object, comprising:

a cone beam source for applying radiation energy to at least the ROI of the object;

an area detector for detecting radiation energy;

means for defining a source scanning trajectory as a scan path traversed by the source;

a manipulator for causing the cone beam source, fixed relative to an area detector with both source and detector movably positioned relative to the object, to scan about the ROI in the object at a plurality of source positions in a direction along the scan path to cause said area detector to acquire a set of cone beam projection data at each of said source positions;

a masking means for applying a mask to each set of the projection data to generate masked data sets;

a convolution processing means for convolution processing each masked data set along a plurality of parallel lines formed therein that are parallel with and in the direction of a parallel projection into the data set of a line that is tangent to the scan path at the source position that acquired that data set and in the direction of scanning, to generate filtered data sets; and

3D backprojection means for subjecting the data of each filtered data set to a weighted 3D backprojection into a 3D space, thereby reconstructing in a piecewise manner a 3D image of the ROI in the object.

15. The apparatus of claim 14, wherein said masking means forms the mask for each data set by cone beam projecting onto the plane of the detector portions of the source scan path that are above and below the source position that acquired the data set being masked.

16. The apparatus of claim 14, wherein the masking means forms a mask for each source position that has a top curve and a bottom curve, and all of the projection data outside the top and bottom curves of the mask are set to zero, for forming the masked data sets.

17. The apparatus of claim 16, wherein the means defining the source scanning trajectory defines it as a spiral scan path that surrounds the ROI in the object, which path is connected to a first scan path circle at a top portion of the ROI in the object, and a second scan path circle at a bottom portion of the ROI in the object.

18. The apparatus of claim 17, wherein at source positions of the scan path at the top of the ROI in the object, the top curve of the mask includes a horizontal line portion formed from a projection of the first scan path circle, and the bottom curve of the mask is formed from a projection of the spiral scan path.

19. The apparatus of claim 17, wherein at source positions of the scan path at the bottom of the ROI in the object, the top curve of the mask is formed from a projection of the spiral scan path, and the bottom curve of the mask includes a horizontal line portion formed from a projection of the second scan path circle.

20. The apparatus of claim 18, wherein the top curve is formed from an intersection of a projection of the first scan path circle and a projection of the spiral scan path.

21. The apparatus of claim 19, wherein the bottom curve is formed from an intersection of a projection of the second scan path circle and a projection of the spiral scan path.

22. The apparatus of claim 14, wherein the convolution processing means performs a ramp filtering of the masked data set.

23. The apparatus of claim 14, wherein the weighted 3D backprojection means performs a weighted 3D backprojection of the data of each of the filtered data sets onto a plurality of sample points in a 3D object volume.

24. The apparatus of claim 23, wherein the 3D object volume corresponds to the ROI volume in the object that is scanned by the cone beam source.

25. The apparatus of claim 24, wherein the weight assigned to the filtered data that is backprojected to each sample point in the 3D object volume, is proportional to the inverse of the square of the distance between each sample point and the location of the cone beam source that acquired that data set being backprojected.