Automatic generation of optimal views for computed tomography thoracic diagnosis

1. A scanning system for generating at least one view of a portion of a computed tomography image comprising:

a controller having a plurality of interconnected devices including:

a device for selecting a seed point for a structure of interest within the image;

a device for preprocessing a region of interest surrounding the seed point;

a device for estimating at least one inertia axis for the region of interest, wherein the device for estimating at least one inertia axis:

assigns density values to voxels in the region of interest depending on each voxel's intensity and distance to the seed point; and

utilizes the density values to define a main axis of inertia and two other axes of inertia orthogonal to the main axis of inertia; and

a device for generating the at least one view from one or more of three planes, defined by one or more of the three axes, that include the seed point and are orthogonal to the at least one axis of inertia.

2. The system of claim 1 , wherein the device for selecting a seed point performs a local search to refine the seed point.

3. The system of claim 2 , wherein the local search includes analyzing voxels proximate the seed point within the structure of interest by identifying a local maximum or minimum centered within the structure of interest.

4. The system of claim 1 , wherein the device for preprocessing a region of interest auto-correlates the region of interest.

5. The system of claim 1 , wherein the device for preprocessing a region of interest defines the region of interest as voxels surrounding the seed point and extending to boundaries of the structure of interest.

6. The system of claim 1 , wherein the device for estimating at least one inertia axis further defines the main axis of inertia by: generating a plurality of 2 dimensional orthogonal slices along the main inertial axis; locating a center point of the structure of interest within each slice; and fitting a line to the center points of interest for the plurality of slices.

7. A method of generating at least one view of a portion of a computed tomography image comprising:

selecting seed point for a structure of interest within the image;

pre-processing a region of interest surrounding the seed point;

estimating at least one inertia axis for the region of interest, wherein estimating at least one inertia axis includes:

assigning density values to voxels in the region of interest depending on each voxel's intensity and distance to the seed point; and

utilizing the density values to define a main axis of inertia and two other axes of inertia orthogonal to the main axis of inertia; and

generating the at least one view from one or more of three planes that include the seed point and are orthogonal to each axis of inertia.

8. The method of claim 7 , wherein selecting a seed point includes performing a local search to refine the seed point.

9. The method of claim 8 , wherein the local search includes analyzing voxels proximate the seed point within the structure of interest by identifying a local maximum or minimum centered within the structure of interest.

10. The method of claim 7 , wherein pre-processing a region of interest includes auto-correlating the region of interest.

11. The method of claim 7 , wherein pre-processing a region of interest includes defining the region of interest as voxels surrounding the seed point and extending to boundaries of the structure of interest.

12. The method of claim 7 , wherein the main axis of inertia is further defined by: generating a plurality of 2 dimensional orthogonal slices along the main inertial axis; locating a center point of the structure of interest within each slice; and fitting a line to the center points of interest for the plurality of slices.

13. A computer program product comprising a computer useable medium having a computer readable program, wherein the computer readable program when executed on a computer causes the computer to generate at least one view of a portion of a computed tomography image by:

selecting seed point for a structure of interest within the image; pre-processing region of interest surrounding the seed point;

estimating at least one inertia axis for the region of interest, wherein estimating at least one inertia axis includes:

assigning density values to voxels in the region of interest depending on each voxel's intensity and distance to the seed point; and

utilizing the density values to define a main axis of inertia and two other axes of inertia orthogonal to the main axis of inertia; and

generating the at least one view from one or more of three planes that include the seed point and are orthogonal to each axis of inertia.

14. The computer program product of claim 13 , wherein selecting a seed point includes performing a local search to refine the seed point.

15. The computer program product of claim 14 , wherein the local search includes analyzing voxels proximate the seed point within the structure of interest by identifying a local maximum or minimum centered within the structure of interest.

16. The computer program product of claim 13 , wherein auto-correlating a region of interest includes defining the region of interest as voxels surrounding the seed point and extending to boundaries of the structure of interest.

17. The computer program product of claim 13 , wherein the main axis of inertia is further defined by: generating a plurality of 2 dimensional orthogonal slices along the main inertial axis; locating a center point of the structure of interest within each slice; and fitting a line to the center points of interest for the plurality of slices.