Detection of intervertebral disk orientation in spine images using curve evolution

1. A computer-implemented method for vertebra segmentation comprising:

providing an image of a plurality of vertebrae;

determining a seed in each of at least two adjacent vertebrae in the image;

mapping a unit square to the seeds in the image as corresponding shape constraints on a segmentation;

evolving the shape constraints to determine the segmentation of the adjacent vertebrae, wherein evolutions of the shape constraints interact; and

outputting a segmented image indicating a location of the vertebra.

2. The computer-implemented method of claim 1 , further comprising initializing values of a translation and a rotation of the shape constraints.

3. The computer-implemented method of claim 1 , further comprising weighting edges of the shape constraints during the evolutions.

4. The computer-implemented method of claim 3 , wherein the weighting is a constant factor.

5. The computer-implemented method of claim 1 , further comprising applying an interaction force between adjacent shape constraints on adjacent vertebrae.

6. The computer-implemented method of claim 5 , wherein the interaction force penalizes increasing differences in orientation and scale between adjacent shape constraints.

7. The computer-implemented method of claim 1 , further comprising determining an orientation of a disk disposed between adjacent located vertebrae.

8. The computer-implemented method of claim 1 , wherein the unit square is initialized having a position substantially matching the seed point.

9. A program storage device readable by machine, tangibly embodying a program of instructions executable by the machine to perform method steps for vertebra segmentation, the method steps comprising:

providing an image of a plurality of vertebrae;

determining a seed in each of at least two adjacent vertebrae in the image;

mapping a unit square to the seeds in the image as corresponding shape constraints on a segmentation;

evolving the shape constraints to determine the segmentation of the adjacent vertebrae, wherein evolutions of the shape constraints interact; and

outputting a segmented image indicating a location of the vertebra.

10. The method of claim 9 , further comprising initializing values of a translation and a rotation of the shape constraints.

11. The method of claim 9 , further comprising weighting edges of the shape constraints during the evolutions.

12. The method of claim 11 , wherein the weighting is a constant factor.

13. The method of claim 9 , further comprising applying an interaction force between adjacent shape constraints on adjacent vertebrae.

14. The method of claim 13 , wherein the interaction force penalizes increasing differences in orientation and scale between adjacent shape constraints.

15. The method of claim 9 , further comprising determining an orientation of a disk disposed between adjacent located vertebrae.

16. The method of claim 9 , wherein the unit square is initialized having a position substantially matching the seed point.

17. A computer-implemented method for vertebrae segmentation comprising:

providing an image of a plurality of vertebra;

determining a seed point in adjacent vertebra;

mapping a unit square to the image, the mapping includes registration parameters represented by a rotation angle of a mapped unit square, non-uniform scale parameters of the mapped unit square and displacement parameters of the mapped unit square;

performing a semi-affine transformation of the mapped unit square;

evolving the mapped unit square to determine a segmentation of corresponding vertebra;

weighting edges of the shape constraint during an evolution;

applying an interaction force between adjacent shape constraints on adjacent vertebrae, wherein the interaction force penalizes increasing differences in orientation and scale between adjacent shape constraints;

determining an orientation of a disk disposed between vertebra located by the segmentation; and

outputting a segmented image comprising located vertebra and an orientation of the disk.

18. The method of claim 17 , wherein the orientation of the disk is identified by determining a line that bisects a box connecting segmentation of the adjacent vertebra.