Object centric data reformation with application to rib visualization

1. A computer-implemented method for detecting abnormalities, comprising:

performing a segmentation on a 3D image to generate a first segmented structure and a second segmented structure that is distinct from the first;

generating a first distance map from the first segmented structure and a second distance map from the second segmented structure;

generating a first iso-surface from the first distance map and a second iso-surface from the second distance map,

wherein the first iso-surface entirely surrounds a first boundary contour that bounds the first segmented structure, and space is present between the first iso-surface and the first boundary contour,

wherein the second iso-surface entirely surrounds a second boundary contour that bounds the second segmented structure, and space is present between the second iso-surface and the second boundary contour;

generating a connected surface by connecting closest respective points of the first iso-surface and the second iso-surface together;

generating a series of 2D images using the connected surface; and

using the 2D images to detect abnormalities.

2. The method of claim 1 , wherein each of the 2D images is generated by sampling points on the connected surface and deformations of the connected surface within the 3D image.

3. The method of claim 1 , further comprising using the 2D images to generate segmentations of additional anatomic structures.

4. The method of claim 1 , wherein performing a segmentation on the 3D image to generate each segmented structure comprises applying a level set function to the 3D image.

5. The method of claim 1 , wherein performing a segmentation on the 3D image to generate each segmented structure comprises performing a seeded region growing operation on the 3D image.

6. The method of claim 1 , wherein each distance map is obtained by applying a fast marching method to the corresponding segmented structure.

7. The method of claim 1 , wherein the first segmented structure is representative of a first lung of a pair of lungs and the second segmented structure is representative of a second lung of the pair and the 2D images forms an image of ribs.

8. The method of claim 1 , wherein the 3D image comprises, a computed tomography (CT) image.

9. The method of claim 1 , further comprising, stacking the series of 2D images to form a new 3D volume.

10. The method of claim 1 , wherein the first and second iso-surfaces do not surround one another.

11. A computer implemented-method for generating a visualization of a series of 2D images of ribs, comprising:

performing a segmentation on a 3D image of a pair of lungs to generate a first segmented structure and a second segmented structure;

generating a first distance map from the first segmented structure and a second distance map from the second segmented structure;

selecting a first iso-surface from the first distance map and a second iso-surface from the second distance map,

wherein the first iso-surface entirely surrounds a first boundary contour that bounds the first segmented structure, each point of the first iso-surface is spaced apart a first distance away from a corresponding point of the first boundary contour, and

wherein the second iso-surface entirely surrounds a second boundary contour that bounds the second segmented structure, and each point of the second iso-surface is spaced apart a second distance away from a corresponding point of the second boundary contour;

generating a connected surface by connecting the first iso-surface to the second iso-surface; and

generating 2D images of the ribs using points derived from the connected surface.

12. The method of claim 11 , wherein the visualization is used for a computer aided diagnosis application to detect abnormalities of the ribs.

13. The method of claim 11 , wherein the visualization is used as a basis to perform a segmentation of the ribs.

14. The method of claim 11 , wherein generating a first distance map from the first segmented structure and a second distance map from the second segmented structure comprises, applying a fast marching method to the first segmented structure and the second segmented structure.

15. The method of claim 11 , wherein the 3D image comprises, a computed tomography (CT) image.

16. The method of claim 11 , wherein the first iso-surface is connected to the second iso-surface by connecting closest respective points of the first iso-surface and the second iso-surface together.

17. The method of claim 11 , wherein the first and second distances are substantially the same.

18. The method of claim 11 , wherein the first and second iso-surfaces do not surround one another.

19. A computer-implemented method for generating a visualization of a series of 2D images, comprising:

performing a segmentation on a 3D image to generate a first segmented structure and a second segmented structure that is distinct from the first;

generating a first distance map from the first segmented structure and a second distance map from the second segmented structure;

generating a first iso-surface from the first distance map and a second iso-surface from the second distance map,

wherein the first iso-surface entirely surrounds a first boundary contour that bounds the first segmented structure, and space is present between the first iso-surface and the first boundary contour,

wherein the second iso-surface entirely surrounds a second boundary contour that bounds the second segmented structure, and space is present between the second iso-surface and the second boundary contour;

generating a connected surface by connecting the first and the second iso-surface together; and

generating a series of 2D images from the connected surface.

20. The method of claim 19 , wherein the first and second iso-surfaces do not surround one another.