Automated separation of binary overlapping trees

• US 10,360,672 B2
• Filed: 03/14/2014
• Issued: 07/23/2019
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Abramoff D. Michael, Department of Veterans Affairs The United States of Government As Represented By
• Original Assignee: University of Iowa Research Foundation
• Inventors: Abramoff, Michael D., Garvin, Mona K., Qiao, Hu
• Primary Examiner(s): Park, Soo Jin
• Application Number:
  US14/774,801
• Publication Number:
  US 20160035088A1
• Time in Patent Office: 1,957 Days
• Field of Search: None
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/20044 Skeletonization; Medial axis transform
• CPC Class Codes: G06T 2207/20072 Graph-based image processing
• CPC Class Codes: G06T 2207/20076 Probabilistic image processing
• CPC Class Codes: G06T 2207/30041 Eye; Retina; Ophthalmic
• 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 7/136 involving thresholding
• CPC Class Codes: G06T 7/162 involving graph-based methods

Abstract

Provided are systems and methods for analyzing images. An exemplary method can comprise receiving at least one image having one or more annotations indicating a feature. The method can comprise generating training images from the at least one image. Each training image can be based on a respective section of the at least one image. The training images can comprise positive images having the feature and negative images without the feature. The method can comprise generating a feature space based on the positive images and the negative images. The method can further comprise identifying the feature in one or more unclassified images based upon the feature space.

Claims

1. A method for image analysis, comprising:

receiving an image comprising a plurality of overlapping tree structures;

generating a potential connectivity map (PCM) based on the plurality of overlapping tree structures, wherein the PCM comprises segments;

determining potential connectivity between the segments and a corresponding cost for each potential connectivity by artificially extending the segments until one or more of;

a respective neighbor is reached or a respective maximum extension is reached, wherein the maximum extension comprises a circular sector based on a radius proportional to a segment length measured from a head of the segment to a tail of the segment;

processing the PCM to separate the plurality of overlapping tree structures into separate tree structures with a minimal cost; and

outputting the separate tree structures having minimal cost.

2. The method of claim 1, wherein the plurality of overlapping tree structures comprises a plurality of overlapping anatomical vascular trees.

3. The method of claim 2, wherein the segments are vessel segments.

4. The method of claim 3, wherein generating the PCM based on the plurality overlapping of tree structures comprises separating vessel centerlines into separated disambiguated longest vessel segments which exhibit no landmarks except at each end.

5. The method of claim 4, wherein separating vessel centerlines into separated disambiguated longest vessel segments comprises:

classifying branch and crossing points for each vessel segment; and

separating each vessel segment into a set of disambiguated longest vessel segments.

6. The method of claim 1, wherein determining the potential connectivity between the segments and the corresponding cost for each potential connectivity comprises determining neighbor segments from among the segments and determining the potential connectivity between neighbor segments.

7. The method of claim 1, wherein processing the PCM to separate the plurality of overlapping tree structures into separate tree structures with the minimal cost comprises transferring the PCM into a graph model.

8. A system for image analysis, comprising:

a memory having computer-executable instructions encoded thereon; and

at least one processor functionally coupled to the memory and configured, by the computer-executable instructions, for;

receiving an image comprising a plurality of overlapping tree structures;

generating a potential connectivity map (PCM) based on the plurality of overlapping tree structures, wherein the PCM comprises segments;

determining potential connectivity between the segments and a corresponding cost for each potential connectivity by artificially extending the segments until one or more of;

a respective neighbor is reached or a respective maximum extension is reached, wherein the maximum extension comprises a circular sector based on a radius proportional to a segment length measured from a head of the segment to a tail of the segment;

processing the PCM to separate the plurality of overlapping tree structures into separate tree structures with a minimal cost; and

outputting the separate tree structures having minimal cost.

9. The system of claim 8, wherein the plurality of overlapping tree structures comprises a plurality of overlapping anatomical vascular trees.

10. The system of claim 9, wherein the segments are vessel segments.

11. The system of claim 10, wherein generating the PCM based on the plurality of overlapping tree structures comprises separating vessel centerlines into separated disambiguated longest vessel segments which exhibit no landmarks except at each end.

12. The system of claim 11, wherein separating vessel centerlines into separated disambiguated longest vessel segments comprises:

classifying branch and crossing points for each vessel segment and

separating each vessel segment into a set of disambiguated longest vessel segments.

13. The system of claim 8, wherein determining the potential connectivity between the segments and the corresponding cost for each potential connectivity comprises determining neighbor segments from among the segments and determining the potential connectivity between neighbor segments.

14. The system of claim 8, wherein processing the PCM to separate the plurality of overlapping tree structures into separate tree structures with the minimal cost comprises transferring the PCM into a graph model.

15. A non-transitory computer readable medium having computer-executable instructions encoded thereon for image analysis, said computer-executable instructions adapted to be executed to implement a method comprising:

receiving an image comprising a plurality of overlapping tree structures;

generating a potential connectivity map (PCM) based on the plurality of overlapping tree structures, wherein the PCM comprises segments;

determining potential connectivity between the segments and a corresponding cost for each potential connectivity by artificially extending the segments until one or more of;

a respective neighbor is reached or a respective maximum extension is reached, wherein the maximum extension comprises a circular sector based on a radius proportional to a segment length measured from a head of the segment to a tail of the segment;

processing the PCM to separate the plurality of overlapping tree structures into separate tree structures with a minimal cost; and

outputting the separate tree structures having minimal cost.

16. The non-transitory computer readable medium of claim 15, wherein the plurality of overlapping tree structures comprises a plurality of overlapping anatomical vascular trees.

17. The non-transitory computer readable medium of claim 16, wherein the segments are vessel segments.

18. The non-transitory computer readable medium of claim 17, wherein generating the PCM based on the plurality of overlapping tree structures comprises separating vessel centerlines into separated disambiguated longest vessel segments which exhibit no landmarks except at each end.

19. The non-transitory computer readable medium of claim 15, wherein determining potential connectivity between the segments and the corresponding cost for each potential connectivity comprises determining neighbor segments from among the segments and determining the potential connectivity between neighbor segments.

20. The non-transitory computer readable medium of claim 15, wherein processing the PCM to separate the plurality of overlapping tree structures into separate tree structures with the minimal cost comprises transferring the PCM into a graph model.