System and method for interactive contouring for 3D medical images

1. A method for segmenting an object in a volume image, the method executed at least in part on a computer and comprising:

rendering the volume image data to a two-dimensional display screen showing first, second, and third mutually orthogonal planes in the two-dimensional rendering;

accessing one or more operator instructions that identify a plurality of seed points on the rendered volume image, the identified plurality of seed points simultaneously covering the first, second, and third mutually orthogonal planes, and the identified plurality of seed points including geometry and intensity information of the object;

converting two-dimensional coordinates of the identified plurality of seed points displayed on the display screen to three-dimensional seed-point coordinates relating to the first, second, and third mutually orthogonal planes;

segmenting the volume image according to the converted three-dimensional seed-point coordinates and the geometry and intensity information;

while segmenting, displaying at least one intermediate segmentation result on the display screen, relative to the first, second, and third mutually orthogonal planes; and

displaying a final segmentation result on the display screen, relative to the first, second, and third mutually orthogonal planes.

2. The method according to claim 1 further comprising accessing one or more additional operator instructions that identify one or more additional seed points on the at least one intermediate displayed segmentation results and repeating the steps of:

converting two-dimensional coordinates of the identified seed points on the display screen to three-dimensional seed-point coordinates relating to the first, second, and third mutually orthogonal planes;

performing an additional segmentation operation on the volume image according to the converted three-dimensional seed-point coordinates; and

displaying the results of the additional segmentation on the display screen, relative to the first, second, and third mutually orthogonal planes.

3. The method according to claim 1 wherein the seed points identify a region of a tooth.

4. The method according to claim 1 wherein rendering the volume image data further comprises mapping the volume image data to a graphics processing unit intensity texture.

5. The method of claim 1 wherein accessing the one or more operator instructions comprises recording one or more of the seed points according to entries from a computer mouse or other pointer.

6. The method of claim 1 wherein the seed points include seed points for background and foreground image content and display in different colors.

7. The method of claim 1 further comprising using the identified seed points for a region-growing operation.

8. A method for segmenting an object in a volume image, the method executed at least in part on a computer and comprising:

mapping the volume image data to a graphics processing unit intensity texture;

rendering the mapped volume image data to a display screen showing first, second, and third mutually orthogonal planes;

accessing operator instructions that identify a plurality of seed points on the mapped volume image;

converting two-dimensional coordinates of the identified seed points to three-dimensional coordinates in volume space;

rendering the three-dimensional coordinates to a graphics processing unit label texture;

iteratively performing an iterative segmentation that combines the mapped volume image data in the intensity texture with the rendered label texture data to form a mixed texture; and

displaying the mixed texture of each iteratively performed segmentation on the display screen, showing the first, second, and third mutually orthogonal planes.

9. The method of claim 8 wherein the intensity texture is a two-dimensional texture object in the graphics processing unit.

10. The method of claim 8 wherein mapping the volume image data further comprises:

flattening the volume image to a plurality of two-dimensional image slices; and

uploading the two-dimensional image slices to a two-dimensional texture object in the graphics processing unit.

11. The method of claim 8 wherein accessing the operator instructions comprises recording one or more of the seed points according to entries from a computer mouse or other pointer.

12. The method of claim 8 wherein the seed points include seed points for background and foreground image content and display in different colors.

13. The method of claim 8 further comprising using the identified seed points for a region-growing operation.

14. The method of claim 8 further comprising updating the label texture one or more times, according to one or more segmentation iterations.

15. The method of claim 8 wherein the object is a tooth.

16. The method of claim 8 wherein the mixed texture is a two-dimensional texture object in the graphics processing unit.

17. A method for segmenting an object in a volume image, the method executed at least in part on a computer, comprising:

rendering the volume image data to a 2D display screen showing first, second, and third mutually orthogonal planes in the 2D rendering;

accessing one or more operator instructions that identify a plurality of seed points on the rendered volume image;

converting 2D coordinates of the identified plurality of seed points displayed on the display screen to 3D seed-point coordinates relating to the first, second, and third mutually orthogonal planes;

segmenting the volume image according to the converted 3D seed-point coordinates;

automatically displaying a plurality of intermediate segmentation results on the display screen, relative to the first, second, and third mutually orthogonal planes; and

automatically displaying a final segmentation result on the display screen, relative to the first, second, and third mutually orthogonal planes.