Granted Patent B2

US 8,577,104 · App. 12/823,346 · Granted Nov 5, 2013

Abstract

Claims (35)

1. A method for segmenting a first object from a bounding object within a digital image, comprising

identifying, from within the digital image, a plurality of pixels between at least a first endpoint pixel and a second endpoint pixel, wherein the first and second endpoint pixels are identified by a user using a display;

defining a volume of interest spanning at least a portion of the object according to the identified plurality of pixels;

partitioning pixels within the volume of interest into a first subset of pixels that are neither the first object nor bounding object pixels and second subset of pixels that are not contained in the first subset;

executing, by a computer processor, a spatially varying two-phase segmentation process that segments the first object from the bounding object within the second subset of pixels according to the location of each subject pixel within the second subset relative to pixels in the first subset of pixels;

refining the first object segmentation according to the location of the subject pixel relative to the segmented bounding object and to the first subset of pixels;

forming a segmented image according to the refined first object segmentation; and

displaying, storing, or transmitting the segmented image.

2. The method of claim 1 wherein partitioning pixels into first and second subsets comprises thresholding.

3. The method of claim 2 wherein thresholding is performed according to predefined intensity values.

4. The method of claim 2 wherein thresholding is performed according to values of pixels within the volume of interest.

5. The method of claim 1 wherein partitioning pixels into first and second subsets comprises determining a boundary of an anatomical feature.

6. The method of claim 5 wherein determining the boundary comprises using segmentation of a bodily organ.

7. The method of claim 5 wherein the anatomical feature comprises a rib cage.

8. The method of claim 1 further comprising correcting for non-uniformity in the second subset prior to executing the spatially varying two-phase segmentation process.

9. The method of claim 1 further comprising correcting for non-uniformity in the second subset during execution of the spatially varying two-phase segmentation process.

10. The method of claim 1 further comprising correcting for non-uniformity in the second subset using a linear polynomial correction term.

11. The method of claim 1 wherein executing the spatially varying two-phase segmentation process comprises using a region-based, variational-segmentation model with spatially varying weights between a regularizer and a data term.

12. The method of claim 1 wherein executing the spatially varying two-phase segmentation process further comprises applying a weighting that varies according to the distance between the subject pixel and pixels in the first subset of pixels.

13. The method of claim 1 wherein defining the volume of interest comprises defining a spherical or elliptical region.

14. The method of claim 1 wherein refining the first object comprises using a predefined shape prior.

15. The method of claim 1 further comprising forming a class map according to the segmented image, wherein the class map defines two or more different portions of the digital image.

16. The method of claim 1 further comprising determining if the first object touches a surface defined by pixels in the first subset.

17. The method of claim 1 wherein refining comprises generating a parabolic boundary.

18. The method of claim 1 further comprising performing morphological processing using the class map.

19. A method for segmenting a lesion from bounding tissue in a 3-D volume image, comprising:

defining a volume of interest spanning at least a portion of the lesion according to user input that identifies, from within the volume image, at least a first endpoint voxel and a second endpoint voxel;

partitioning voxels within the region of interest into a first subset of voxels that are neither lesion nor bounding tissue voxels and second subset of voxels that are not contained in the first subset;

segmenting, by a computer processor, the lesion of the volume image from the bounding tissue by partitioning the second subset of voxels according to a spatially varying two-phase segmentation process, wherein, for each subject voxel within the second subset, one or more parameters of the spatially varying two phase segmentation process varies according to the location of the subject voxel relative to voxels in the first subset of voxels;

forming a segmented volume image according to the segmentation process; and

displaying, storing, or transmitting the segmented volume image.

20. The method of claim 19 further comprising characterizing the lesion as either hyperdense or hypodense.

21. The method of claim 19 further comprising determining a boundary that limits leakage of the lesion in the image according to anatomy features.

22. The method of claim 19 further comprising using a predefined shape to constrain leakage of the lesion in the image.

23. The method of claim 19 further comprising refining the lesion segmentation according to voxel location within the segmented lesion relative to the segmented background and first subset of voxels.

Assignments (9)