Cone-beam reconstruction using backprojection of locally filtered projections and X-ray CT apparatus

1. A method of computing an intermediate function for use in cone-beam (CB) reconstruction used to reconstruct 3D images, the intermediate function evaluated at a selected point in three dimensions where image reconstruction will be performed, the method comprising:

obtaining CB projection data from a vertex path segment, wherein the vertex path segment comprises a path taken by an x-ray source in a reference frame of a target object to be reconstructed and wherein a chord connecting first and second endpoints of the vertex path segment does not include the selected point; and

determining a difference between first and second values;

the first value comprising a first Hilbert transform of a density function along a straight line passing through the first endpoint of the vertex path segment and the selected point, the first Hilbert transform evaluated at the selected point,

the second value comprising a second Hilbert transform of the density function along a straight line passing through the second endpoint of the vertex path segment and the selected point, the second Hilbert transform evaluated at the selected point; and

storing the difference values.

2. The method of claim 1 , wherein the method further comprises computing a differentiated backprojection (DBP) of the CB projection data on the vertex path segment.

3. The method of claim 2 , wherein the method further comprises determining a weighted average of first, second and third values;

wherein the first value comprises an integral of the density function along a ray originating at the first endpoint of the vertex path segment and passing through the selected point;

wherein the second value comprises a DBP of the CB projection data on the vertex path segment; and

wherein the third value comprises an integral of the density function along a ray originating at the second endpoint of the vertex path segment and passing through the selected point.

4. A method of reconstructing an image from cone-beam (CB) projections acquired along a vertex path, the method comprising:

selecting at least one image portion;

for each image portion:

selecting a series of points;

for each point of the selected series of points:

selecting at least one vertex path segment such that the point does not lie on a chord connecting a first endpoint and a second endpoint of any of the selected vertex path segments;

for each of the selected vertex path segments: computing an intermediate function from CB projection data from the selected vertex path segment;

using the computed intermediate functions to reconstruct a portion of the image; and

reconstructing the image from the reconstructed portions of the image.

5. The method according to claim 4 , further comprising storing the reconstructed image.

6. The method according to claim 5 , wherein storing the reconstructed image comprises storing the reconstructed image on a computer readable medium.

7. The method according to claim 4 , further comprising displaying the reconstructed image.

8. The method of claim 4 , wherein each of the selected image portions is a line segment and wherein using the computed intermediate functions to reconstruct the selected image portion for each point on the selected line segment comprises:

weighting the computed intermediate functions for each of the selected vertex path segments;

adding the weighted, computed intermediate functions to produce a Hilbert transform of a density function along the line segment evaluated at the point; and

inverting the Hilbert transform along the line segment to produce a reconstruction of the density function along the line segment.

9. A computed tomography (CT) scanner, comprising:

an image reconstruction processor for reconstructing an image of an object from cone-beam (CB) projections acquired along a vertex path, the image reconstruction processor programmed to select at least one image portion and to select a series of points for each of the at least one image portion;

for each point:

selecting at least one vertex path segment such that the point does not lie on a chord connecting a first endpoint and a second endpoint of any of the selected vertex path segments;

for each of the selected vertex path segments:

computing an intermediate function from CB projection data from the selected vertex path segment;

using the computed intermediate functions to reconstruct a portion of the image; and

reconstructing the image from the reconstructed portions of the image; and

a monitor in communication with the image reconstruction processor for viewing the image.

10. The CT scanner according to claim 9 , further comprising a computer readable medium in communication with the image reconstruction processor for storing the image.

11. The CT scanner according to claim 9 , wherein each of the image portions selected by the image reconstruction processor is a line segment and wherein the image reconstruction processor further processes each of the points by:

weighting the computed intermediate functions for each of the selected at least one vertex path segments;

adding the weighted, computed intermediate functions to produce a Hilbert transform of a density function of the object along the line segment evaluated at the point; and

inverting the Hilbert transform along the line segment to produce a reconstruction of the density function along the line segment.

12. The CT scanner according to claim 9 , wherein the image reconstruction processor computes the intermediate function by:

evaluating the intermediate function at a selected point in three dimensions where image reconstruction will be performed;

obtaining CB projection data from a vertex path segment, wherein the vertex path segment comprises a path taken by an x-ray source in a reference frame of a target object to be reconstructed and wherein a chord connecting first and second endpoints of the vertex path segment does not include the selected point; and

determining a difference between first and second values;

the first value comprising a first Hilbert transform of a density function along a straight line passing through the first endpoint of the vertex path segment and the selected point, the first Hilbert transform evaluated at the selected point, and

the second value comprising a second Hilbert transform of the density function along a straight line passing through the second endpoint of the vertex path segment and the selected point, the second Hilbert transform evaluated at the selected point.

13. The CT scanner according to claim 12 , wherein the image reconstruction processor further computes the intermediate function by computing a differentiated backprojection (DBP) of the CB projection data on the vertex path segment.

14. The CT scanner according to claim 13 , wherein the image reconstruction processor further computes the intermediate function by determining a weighted average of first, second and third values:

the first value comprising an integral of the density function along a ray originating at the first endpoint of the vertex path segment and passing through the selected point;

the second value comprising a DBP of the CB projection data on the vertex path segment; and

the third value comprising an integral of the density function along a ray originating at the second endpoint of the vertex path segment and passing through the selected point.