Region growing in anatomical images

• US 7,397,937 B2
• Filed: 09/30/2002
• Issued: 07/08/2008
• Est. Priority Date: 11/23/2001
• Status: Active Grant

• Current Assignee: MeVis Medical Solutions AG (Varex Imaging Corporation)
• Original Assignee: R2 Technology, Inc. (Hologic Incorporated)
• Inventors: Schneider, Alexander C., Wood, Susan A.
• Primary Examiner(s): Desire; Gregory M
• Application Number:
  US10/261,182
• Publication Number:
  US 20030099386A1
• Time in Patent Office: 2,108 Days
• Field of Search: 128/920, 128/922, 128/925, 324/307, 324/309, 345/419, 345/420, 345/424, 345/606, 345/611, 345/653, 378/4, 378/21, 378/163, 382/128, 382/130, 382/131, 382/132, 382/154, 382/173, 382/254, 382/286, 382/107, 382/133, 382/171, 600/407, 600/415, 600/419, 600/420, 600/425, 600/431, 600/481
• US Class Current: 382/130
• CPC Class Codes: G06T 2207/10081 Computed x-ray tomography [CT]
• CPC Class Codes: G06T 2207/20156 Automatic seed setting
• CPC Class Codes: G06T 2207/30061 Lung
• CPC Class Codes: G06T 2207/30064 Lung nodule
• CPC Class Codes: G06T 2207/30101 Blood vessel; Artery; Vein; Vascular
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/187 involving region growing; involving region merging; involving connected component labelling
• CPC Class Codes: G06V 10/267 by performing operations on regions, e.g. growing, shrinking or watersheds
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images
• CPC Class Codes: G06V 40/14 Vascular patterns
• CPC Class Codes: G16H 15/00 ICT specially adapted for medical reports, e.g. generation or transmission thereof
• CPC Class Codes: G16H 30/20 for handling medical images, e.g. DICOM, HL7 or PACS
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A region-growing method for identifying nodules in an anatomical volume segments a 3-D image volume by controlled voxel growth from seed points. The process is based on creation and use of a distance map for tracking the distance of vessel voxels from a predetermined location. A volume map is created that identifies the largest sphere that can pass between a voxel and a predetermined location without touching a non-vessel voxel. The ratio between the distance map and the volume map is analyzed to find regions more likely to contain nodules, the features of which can be extracted or otherwise highlighted.

Claims

1. A method for detecting a candidate region of interest from anatomical images comprising the steps of:

selecting one or more seed voxels from a volume created from the anatomical images;

growing a region from the seed voxels, said region defined by a volume made up of voxels contiguous to the seed voxel and satisfying growth criteria;

creating one or more arrays at least one of which relates each voxel of the region to a first voxel distance; and

processing a function of the created arrays to determine a candidate region of interest.

2. The method of claim 1 wherein the seed voxel is part of an image of a blood vessel.

3. The method of claim 1 wherein the seed voxel is part of an image of an airway.

4. The method of claim 1 wherein the seed voxel is part of an image of a lumen of an anatomical structure.

5. The method of claim 1 wherein the seed voxel is part of an image of one end of a linear anatomical structure.

6. The method of claim 1 wherein the seed voxel is part of an image of a blood vessel and at least one voxel face is associated with a heart or heart region.

7. The method of claim 1 wherein the seed voxel exceeds a threshold intensity level.

8. The method of claim 1 wherein the seed voxel is less than a threshold intensity level.

9. The method of claim 1 wherein the seed voxel is selected within an image of a geometric volume containing roots of blood vessel trees.

10. The method of claim 9 wherein the geometric volume is an ellipsoid.

11. The method of claim 1 wherein the growth criteria for adding voxels to the region is whether a voxel has an intensity level greater than a threshold.

12. The method of claim 1 wherein the growth criteria for adding voxels to the region is whether a voxel has an intensity level less than a threshold.

13. The method of claim 1 wherein the growing of a region stops once a predetermined number of voxels is added to the region.

14. The method of claim 1 wherein the growth criteria for adding voxels to the region is whether a voxel has an intensity level greater than a threshold in a Laplacian-of-Gaussian volume.

15. The method of claim 1 further including the step of performing noise reduction on the volume created from the anatomical images.

16. The method of claim 1 further including the step of creating a distance map which tracks the distance of each voxel in the region from a predetermined location.

17. The method of claim 16 wherein the predetermined location is the seed voxel.

18. The method of claim 16 wherein the predetermined location is a heart.

19. The method of claim 16 wherein the predetermined location is the end or lumen of an anatomical structure.

20. The method of claim 1 further including the step of detecting variations in a width of the region.

21. The method of claim 1 wherein the region is voxels in an image of a vessel volume.

22. The method of claim 1 wherein the region is voxels in an image of a blood vessel.

23. The method of claim 1 wherein the region is voxels in an image of an airway.

24. The method of claim 1 wherein the region is voxels in an image of a lumen of an anatomical structure.

25. The method of claim 1 wherein the region is a linear anatomical structure.

26. The method of claim 1 wherein the voxel distances are distances of the voxels in the region to the seed voxel.

27. The method of claim 1 wherein the voxel distance is a distance of a voxel in the region to a heart or heart region.

28. The method of claim 1 wherein the voxel distance is the distance from a voxel in the region to nearest voxels outside the region.

29. The method of claim 1 wherein the step of processing a function further includes identifying as candidate regions of interest voxels resulting in a function greater than a predetermined value.

30. A computer-readable medium storing a program for identifying a candidate region of interest in a series of digital images of the lung, said program comprising:

logic code for selecting a seed voxel from a volume created from the digital images;

logic code for growing a region from the seed voxel, said region defined by a volume made up of voxels contiguous to the seed voxel and satisfying growth criteria;

logic code for creating one or more arrays at least one of which relates each voxel of the region to a first voxel distance; and

logic code for processing a function of the created arrays to determine a candidate region of interest.

31. The computer-readable medium of claim 30 wherein the region is a blood vessel tree.

32. A method for detecting a candidate nodule from anatomical images comprising the steps of:

selecting a seed voxel from a volume created from the anatomical images;

growing a region from the seed voxel, said region defined by a volume made up of voxels contiguous across a plurality of anatomical images to the seed voxel and satisfying growth criteria;

creating a first array having first array values associated with widths of a vessel; and

processing the first array values by identifying local maxima or minima of the widths of a vessel to determine a candidate nodule location.

33. A method for detecting from anatomical images a candidate region of interest comprising the steps of:

selecting one or more seed voxels from a volume created from the anatomical images;

growing a first region from a seed voxel, said region defined by a volume made up of voxels contiguous to the seed voxel and satisfying a first set of growth criteria;

growing within the first region a second region, said second region defined by a volume made up of voxels contiguous with the seed voxel and satisfying a second set of growth criteria; and

analyzing one or more portions of the first region that are not included in the second region based on the voxel intensities for voxels in the one or more portions.

34. A method for detecting a candidate region of interest from anatomical images of a vessel structure comprising the steps of:

selecting a seed voxel from a volume created from the anatomical images;

growing a region from the seed voxel, said region defined by a volume made up of voxels contiguous to the seed voxel and satisfying growth criteria;

creating a first array having first array values associated with widths of a vessel at increasing distance from the seed voxel;

creating a second array having second array values that decrease monotonically with increasing distance from the seed voxel from a maximum width value to a minimum width value;

forming ratios of the second array values to the first array values; and

identifying a region of interest at a distance from the seed voxel where the ratio is substantially different from other ratios.