Voting in mammography processing

• US 8,311,296 B2
• Filed: 11/21/2007
• Issued: 11/13/2012
• Est. Priority Date: 11/21/2007
• Status: Active Grant

• Current Assignee: Parascript, LLC
• Original Assignee: Parascript, LLC
• Inventors: Filatov, Alexander, Nikitin, Vadim
• Primary Examiner(s): Johnston, Phillip A
• Application Number:
  US11/943,957
• Publication Number:
  US 20090129656A1
• Time in Patent Office: 1,819 Days
• Field of Search: 382/128
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/10116 X-ray image
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 7/0012 Biomedical image inspection

Abstract

Methods and systems are disclosed to aid in the detection of areas of interest in an image. Multiple image recognition processes analyze the image and identify areas of interest. The identified areas of interest are compared to determine confidence values for each identified area of interest using a voting process. The confidence values may be used in determining areas of increased interest which are highlighted on the image. In embodiments, identified areas of interest meeting a certain threshold requirement are selected as areas of increased interest. In other embodiments, new areas of increased interest are created by combining areas of interest. Embodiments of the disclosed methods and system may be used to aid in the detection of cancer in mammogram images.

Claims

1. A computer implemented method for determining at least one confidence value for recognizing areas of interest on an image, the computer implemented method comprising:

identifying a first identified area of interest based on output from a first image recognition process, the first identified area of interest having a first location on the image;

identifying a second area of interest, wherein at least part of the second area of interest is different from the first identified area of interest, the second identified area of interest having a second location on the image;

comparing the first identified area of interest on the image to the second identified area of interest on the image to calculate at least one confidence value for at least part of the at least first identified area of interest, wherein the comparison comprises a voting process that determines the confidence value based at least upon a proximity between the first location and second location; and

providing the at least one confidence value.

2. The computer implemented method of claim 1, wherein the second area of interest is identified by output from a second image recognition process that is different from the first image recognition process.

3. The computer implemented method of claim 1, further comprising:

selecting at least one identified area of interest based upon the at least one confidence value; and

outputting an indication of the selected areas of interest.

4. The computer implemented method of claim 1, further comprising:

combining the identified areas of interest to produce one or more new of areas of interest; and

outputting an indication of the one or more new locations of areas of interest on the image.

5. The computer implemented method of claim 1, wherein the voting process further comprises:

determining a plurality of boundaries for the first and second identified areas of interest;

comparing the boundaries to determine whether the one or more of the plurality of determined boundaries intersect; and

based upon the comparison, producing the at least one confidence value.

6. The computer implemented method of claim 1, wherein the voting process further comprises:

comparing the first and second identified areas of interest to determine whether the first and second identified areas of interest overlap on the image; and

based upon the comparison, producing the at least one confidence value.

7. The computer implemented method of claim 1, wherein the voting process further comprises:

defining a relative location measure between the first and second locations;

computing the relative location measure between the first and second locations; and

based upon the computed measure, determining the at least one confidence value.

8. The computer implemented method of claim 7, wherein the each of the first and second locations comprises a focal point.

9. A computer storage media encoding a computer program of instructions for executing a computer process for a method of determining at least one confidence value for recognizing a areas of interest on an image, the method comprising:

identifying a first identified area of interest based on output from a first image recognition process on the image, the first identified area of interest having a first location on the image;

identifying a second identified area of interest based on output from a second image recognition process on the image, wherein at least part of the second identified area of interest is different from the first area of interest, the second identified area of interest having a second location on the image;

comparing the at least first identified area of interest identified on the image to the second identified area of interest identified on the image to calculate the at least one confidence value for at least part of the at least first identified area of interest, wherein the comparison comprises a voting process that determines the confidence value based at least upon a proximity between the first location and second location; and

providing the at least one confidence value.

10. The method of claim 9, wherein the first and second image recognition processes do not share information used to identify the identified areas of interest on the image.

11. The method of claim 9, further comprising:

selecting at least one identified area of interest based upon the at least one confidence value; and

outputting an indication of the selected areas of interest.

12. The method of claim 9, further comprising:

combining the identified areas of interest to produce one or more new locations of areas of interest; and

outputting an indication of the one or more new locations of areas of interest on the image.

13. The computer implemented method of claim 9, wherein the voting process further comprises:

determining a plurality of boundaries for the first and second identified areas of interest;

comparing the boundaries to determine whether the one or more of the plurality of determined boundaries intersect; and

based upon the comparison, producing the at least one confidence value.

14. The computer implemented method of claim 9, wherein the voting process further comprises:

comparing the first and second identified areas of interest to determine whether the first and second identified areas of interest overlap on the image; and

based upon the comparison, producing the at least one confidence value.

15. The computer implemented method of claim 9, wherein the voting process further comprises:

defining a relative location measure between the first and second locations;

computing the relative location measure between the first and second locations; and

based upon the computed measure, determining the at least one confidence value.

16. The computer implemented method of claim 15, wherein the each of the first and second locations comprises a focal point.

17. A computer implemented method for determining at least one confidence value for recognizing areas of interest on an image, the computer implemented method comprising:

identifying a first area of interest on the image based on output from the first image recognition process on the image, the first area of interest having a first location;

comparing the first area of interest identified on the image to output of the second image recognition process on the image, the output of the second image recognition process identifying a second area of interest having a second location, to produce at least one confidence value for at least the first area of interest and the second area of interest identified on the image, wherein the comparison comprises a voting process that determines the at least one confidence value based at least upon a proximity between the first location and the second location; and

determining one or more areas of increased interest, wherein determining the one or more areas of increased interest comprises at least one of;

a. selecting at least one of the first and second areas of interest as an area of increased interest based upon the at least one confidence value; and

b. combining the first and second identified areas of interest to produce at least one new area of increased interest; and

outputting an indication of the one or more areas of increased interest on the image.

18. The computer implemented method of claim 17, wherein voting process further comprises:

determining a first boundary for the first area of interest and a second boundary for the second area of interest;

comparing the first and second boundaries to determine whether the first and second boundaries intersect; and

based upon the comparison, producing the at least one confidence value.

19. The computer implemented method of claim 17, wherein voting process further comprises:

comparing the first area and the second area to determine whether the the first and second areas overlap on the image; and

based upon the comparison, producing the at least one confidence value.

20. The computer implemented method of claim 17, wherein voting process further comprises:

determining a first focal point for the first area of interest and a second focal point for the second area of interest;

computing the distance between the first and second focal points; and

based upon the computed distance, determining the at least one confidence value.

21. A computer implemented method for determining at least one confidence value for recognizing areas of interest on single image, the computer implemented method comprising:

identifying at least a first identified area of interest having a first location on the single image based on output from a first image recognition process on the single image, the first identified area of interest having a first confidence value;

identifying a second result for the single image based on output from a second image recognition process on the single image, the second result comprising one of;

(a) no areas of interest; and

(b) a second area of interest different from the first area of interest and having a second location;

comparing the first identified area of interest to the second result, wherein, when the second result comprises a second area of interest having a second location, the step of comparing comprises a voting process that determines the confidence value based at least upon a proximity between the first location and the second location; and

adjusting the first confidence value based on the comparison of the first area of interest to the second result.