Marking system for computer-aided detection of breast abnormalities

• US 9,256,799 B2
• Filed: 07/06/2011
• Issued: 02/09/2016
• Est. Priority Date: 07/07/2010
• Status: Active Grant

• Current Assignee: iCAD Incorporated
• Original Assignee: VuCOMP, Inc.
• Inventors: Wehnes, Jeffrey C., Wang, Shujun
• Primary Examiner(s): Mehta, Bhavesh
• Assistant Examiner(s): PATEL, PINALBEN V
• Application Number:
  US13/808,229
• Publication Number:
  US 20130109953A1
• Time in Patent Office: 1,679 Days
• Field of Search: 382/132, 382/128, 382/163, 382/173, 382/242, 382/257, 382/266, 382/286, 382/287, 382/305, 382/203, 382/199, 382/141, 600/442, 600/431, 345/441
• US Class Current: 1/1
• CPC Class Codes: A61B 6/5211 involving processing of medical diagnostic data
• CPC Class Codes: G06T 11/00 2D [Two Dimensional] image generation
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 40/63 for local operation

Abstract

An embodiment method for marking an anomaly in an image comprises generating an initial boundary description representing a size, a shape and a location of the anomaly in the image, dilating the initial boundary description to generate a dilated boundary description representing the shape, the location and an enlarged size of the initial boundary description, and saving, on a non-transitory computer-readable medium, the dilated boundary description as an overlay plane object in an output format compliant with a industry standard digital image format.

Claims

1. A method for marking an anomaly in an image comprising pixels, the method comprising:

generating an initial boundary description representing a size, a shape and a location of the anomaly in the image;

dilating the initial boundary description, without dilating or eroding the anomaly in the image, to generate a dilated boundary description representing the shape, the location and an enlarged size of the initial boundary description, wherein the dilated boundary description marks but does not obscure and does not touch the anomaly, and wherein the dilated boundary description does not obscure an image region between the dilated boundary description and the anomaly; and

saving, on a non-transitory computer-readable medium, the dilated boundary description as an overlay plane object in an output format compliant with an industry standard digital image format.

2. The method of claim 1, wherein the industry standard is Digital Imaging and Communication in Medicine (DICOM) standard.

3. The method of claim 1, wherein the dilated boundary description is pixel data, and the method further comprises, before the saving, converting the pixel data to vectorized data.

4. The method of claim 1, wherein the dilated boundary description is vectorized, and the method further comprises, before the saving, iteratively merging adjacent vectors in the dilated boundary description that differ in direction from each other by less than a first angle.

5. The method of claim 1, wherein the image is a mammogram, and wherein the anomaly is a medically-suspicious mass.

6. The method of claim 1, wherein the dilated boundary description is initial vectorized data, and wherein the method further comprises generating, before the saving, a second outer boundary description representing the shape, the location and a second dilated size of the dilated boundary description, wherein the second dilated size is larger than the enlarged size;

and

wherein the saving further comprises saving the second outer boundary description with the dilated boundary description.

7. The method of claim 1, wherein the dilated boundary description is vectorized, and wherein the method further comprises, before the saving, inserting vectorized boundary tokens into the vectorized dilated boundary description.

8. The method of claim 7, wherein the boundary tokens each comprise a first shape selected from the group consisting of:

full-diamond, half-diamond, split-diamond, circle and arc.

9. The method of claim 7, wherein the inserting further comprises inserting N boundary tokens, wherein N is determined by rounding up a result of a total path length of the dilated boundary description divided by a boundary token length D1.

10. The method of claim 9, wherein a boundary token spacing D2 is determined to be the total path length divided by N, minus the boundary token length D1, and wherein the inserting the boundary tokens further comprises, for each of the boundary tokens:

moving along the dilated boundary description from an initial point a distance D2 to set a token return point;

moving along the dilated boundary description from the return point a distance D1 to set a token end point; and

adding token vectorization between the token end point and the token return point.

11. The method of claim 7, wherein the image is a mammogram, and wherein the anomaly is a medically-suspicious microcalcification cluster.

12. A system for marking an anomaly in an image comprising pixels, the system comprising:

a processor; and

a non-transitory computer-readable storage medium storing programming for execution by the processor, the programming including instructions for;

generating an initial boundary description representing a size, a shape and a location of the anomaly in the image; and

dilating the initial boundary description, without dilating or eroding the anomaly in the image, to generate a dilated boundary description representing the shape, the location and an enlarged size of the initial boundary description, wherein the dilated boundary description marks but does not obscure and does not touch the anomaly, and wherein the dilated boundary description does not obscure an image region between the dilated boundary description and the anomaly;

the non-transitory computer-readable storage medium further storing the dilated boundary description as an overlay plane object in an output format compliant with an industry standard digital image format.

24. The system of claim 12, wherein the industry standard is Digital Imaging and Communication in Medicine (DICOM) standard.

25. The system of claim 12, wherein the dilated boundary description is pixel data, and the programming further includes instructions for converting the pixel data to vectorized data.

26. The system of claim 12, wherein the dilated boundary description is vectorized, and the programming further includes instructions for iteratively merging adjacent vectors in the dilated boundary description that differ in direction from each other by less than a first angle.

27. The system of claim 12, wherein the image is a mammogram, and wherein the anomaly is a medically-suspicious mass.

28. The system of claim 12, wherein the dilated boundary description is initial vectorized data, and wherein the programming further includes instructions for generating a second outer boundary description representing the shape, the location and a second dilated size of the dilated boundary description, wherein the second dilated size is larger than the enlarged size;

and

wherein the instructions for saving further include instructions for saving the second outer boundary description with the dilated boundary description.

29. The system of claim 12, wherein the dilated boundary description is vectorized, and wherein the programming further includes instructions for inserting vectorized boundary tokens into the vectorized dilated boundary description.

30. The system of claim 29, wherein the boundary tokens each comprise a first shape selected from the group consisting of:

full-diamond, half-diamond, split-diamond, circle and arc.

31. The system of claim 29, wherein the instructions for inserting further include instructions for inserting N boundary tokens, wherein N is determined by rounding up a result of a total path length of the dilated boundary description divided by a boundary token length D1.

32. The system of claim 31, wherein a boundary token spacing D2 is determined to be the total path length divided by N, minus the boundary token length D1, and wherein the instructions for inserting the boundary tokens further include, for each of the boundary tokens:

instructions for moving along the dilated boundary description from an initial point a distance D2 to set a token return point;

instructions for moving along the dilated boundary description from the return point a distance D1 to set a token end point; and

instructions for adding token vectorization between the token end point and the token return point.

33. The system of claim 29, wherein the image is a mammogram, and wherein the anomaly is a medically-suspicious microcalcification cluster.

13. A computer program product for marking an anomaly an image, the computer program product comprising:

a non-transitory computer-readable medium with a computer program embodied thereon, the computer program comprising;

computer program code for generating an initial boundary description representing a size, a shape and a location of the anomaly in the image;

computer program code for dilating the initial boundary description, without dilating or eroding the anomaly in the image, to generate a dilated boundary description representing the shape, the location and an enlarged size of the initial boundary description, wherein the dilated boundary description marks but does not obscure and does not touch the anomaly, and wherein the dilated boundary description does not obscure an image region between the dilated boundary description and the anomaly; and

computer program code for saving the dilated boundary description as an overlay plane object in an output format compliant with an industry standard digital image format.

14. The computer program product of claim 13, wherein the industry standard is Digital Imaging and Communication in Medicine (DICOM) standard.

15. The computer program product of claim 13, wherein the dilated boundary description is pixel data, and the computer program product further comprises computer program code for converting the pixel data to vectorized data.

16. The computer program product of claim 13, wherein the dilated boundary description is vectorized, and the computer program product further comprises computer program code for iteratively merging adjacent vectors in the dilated boundary description that differ in direction from each other by less than a first angle.

17. The computer program product of claim 13, wherein the image is a mammogram, and wherein the anomaly is a medically-suspicious mass.

18. The computer program product of claim 13, wherein the dilated boundary description is initial vectorized data, and wherein the computer program product further comprises computer program code for generating a second outer boundary description representing the shape, the location and a second dilated size of the dilated boundary description, wherein the second dilated size is larger than the enlarged size;

and

wherein the computer program code for saving further comprises computer program code for saving the second outer boundary description with the dilated boundary description.

19. The computer program product of claim 13, wherein the dilated boundary description is vectorized, and wherein the computer program product further comprises computer program code for inserting vectorized boundary tokens into the vectorized dilated boundary description.

20. The computer program product of claim 19, wherein the boundary tokens each comprise a first shape selected from the group consisting of:

full-diamond, half-diamond, split-diamond, circle and arc.

21. The computer program product of claim 19, wherein the computer program code for inserting further comprises computer program code for inserting N boundary tokens, wherein N is determined by rounding up a result of a total path length of the dilated boundary description divided by a boundary token length D1.

22. The computer program product of claim 21, wherein a boundary token spacing D2 is determined to be the total path length divided by N, minus the boundary token length D1, and wherein the computer program code for inserting the boundary tokens further comprises, for each of the boundary tokens:

computer program code for moving along the dilated boundary description from an initial point a distance D2 to set a token return point;

computer program code for moving along the dilated boundary description from the return point a distance D1 to set a token end point; and

computer program code for adding token vectorization between the token end point and the token return point.

23. The computer program product of claim 19, wherein the image is a mammogram, and wherein the anomaly is a medically-suspicious microcalcification cluster.