System and method for feature score mapping and visualization of medical images

• US 8,150,192 B2
• Filed: 11/21/2007
• Issued: 04/03/2012
• Est. Priority Date: 11/27/2006
• Status: Active Grant

• Current Assignee: Merative US LP
• Original Assignee: Merge CAD, Inc. (International Business Machines Corporation)
• Inventors: Niemeyer, Tanya L., Budak, Paul V., Epps, Brian W., Boisseranc, James J.
• Primary Examiner(s): Kassa, Yosef
• Application Number:
  US11/944,338
• Publication Number:
  US 20080130970A1
• Time in Patent Office: 1,595 Days
• Field of Search: 382/128, 382/130, 382/131, 382/132, 382/209, 382/219
• US Class Current: 382/256
• CPC Class Codes: G06T 2200/24 involving graphical user interfaces [GUIs]
• CPC Class Codes: G06T 2207/10088 Magnetic resonance imaging [MRI]
• 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

A system to collect and analyze medical image data is applicable for multi-modality medical imaging systems, such as x-ray, MRI, and the like. Medical image data is collected and analyzed to determine one or more regions of interest. A selected region of interest is further analyzed to determine morphological characteristics. A feature library, which may be in the form of a data base, is used to analyze the image on a pixel-by-pixel basis to determine the degree to which the region of interest matches selected morphological characteristics, such as shape or margin. The resultant data is used to generate a map indicating a continuum over which the region of interest matches the morphological characteristics. A display receives the map data and applies it to the video image to thereby provide the user with a visualization of the degree to which the region of interest matches the morphological characteristics.

Claims

1. A system comprising:

an input configured to receive medical image data, the medical image data including a region of interest;

a data storage structure to store the received medical image data;

a feature mapping processor configured to analyze the region of interest with respect to a plurality of morphological characteristics in each of a plurality of morphological categories, and to generate a morphological score along a continuum that maps substantially the entire region of interest to a score range indicative of a degree to which the region of interest matches the plurality of morphological characteristics in each of the plurality of morphological categories from a lesser degree of match to a greater degree of match; and

a display to display the medical image data, including the region of interest, the display including visual data indicative of the morphological score range across the continuum from less suspicious to more suspicious.

2. The system of claim 1 wherein the feature mapping processor is further configured to analyze the region of interest with respect to one of the plurality of morphological characteristics and to generate a morphological score on a continuum indicating the degree to which the region of interest matches the plurality of morphological characteristics from a lesser degree of match to a greater degree of match, the display being configured to provide visual data indicative of the continuum of morphological scores.

3. The system of claim 1 wherein the medical image data is a two-dimensional image represented by a plurality of pixels and the feature mapping processor is further configured generate a map having map elements corresponding to the pixels of at least a portion of the medical image data with each of the map elements having a data value representative of the morphological score for the corresponding pixel, the display being configured to display visual data based on the generated map.

4. The system of claim 1 wherein the medical image data is a three-dimensional image represented by a plurality of voxels and the feature mapping processor is further configured generate a map having map elements corresponding to the voxels of at least a portion of the medical image data with each of the map elements having a data value representative of the morphological score for the corresponding voxel, the display being configured to display visual data based on the generated map.

5. The system of claim 1 wherein the region of interest has a border portion and the feature mapping processor is further configured generate morphological score for the border portion.

6. The system of claim 1 wherein the region of interest has a two-dimensional area and the feature mapping processor is further configured generate morphological score for the region area.

7. The system of claim 1 wherein the region of interest has a three-dimensional volume and the feature mapping processor is further configured generate morphological score for the region volume.

8. The system of claim 1 wherein the region of interest comprises a portion of the medical image data less than the entire medical image and the feature mapping processor is further configured generate the morphological score for the entire medical image.

9. The system of claim 1 wherein the display is configured to show at least a portion of the visual data in color with different colors indicating the morphological score.

10. The system of claim 1 wherein the display is configured to show at least a portion of the visual data in color with different color intensity indicating the morphological score.

11. The system of claim 1 wherein the display is configured to show at least a portion of the visual data with a time-varying display brightness indicating the morphological score.

12. The system of claim 1, further comprising an input device operable by a user to select one of the plurality of morphological categories, the feature mapping processor being further configured to analyze the region of interest with respect to the plurality of morphological characteristics for the selected morphological category wherein the display displays the medical image data, including the region of interest, the display including visual data indicative of the morphological score for the selected morphological category.

13. The system of claim 1, further comprising a library of morphological characteristics wherein the feature mapping processor is further configured to for pattern recognition using the library of morphological characteristics to generate the morphological score.

14. The system of claim 1 wherein the medical image data is breast image data and the morphological characteristics correspond to ACR BI-RADS classifications.

15. The system of claim 1 wherein the morphological characteristics correspond to ACR classifications corresponding a tissue type for the region of interest.

16. A method for scoring features in medical image data comprising:

receiving medical image data, the medical image data including a region of interest;

analyzing the region of interest with respect to a plurality of morphological characteristics in each of a plurality of morphological categories;

generating a morphological score along a continuum that maps substantially the entire region of interest to a score range indicative of a degree to which the region of interest matches the plurality of morphological characteristics in each of the plurality of morphological categories from a lesser degree of match to a greater degree of match; and

displaying the medical image data, including visual data indicative of the morphological score range across the continuum from less suspicious to more suspicious.

17. The method of claim 16 wherein generating a morphological score comprises generating a morphological score on a continuum indicating the degree to which the region of interest matches the plurality of morphological characteristics from a lesser degree of match to a greater degree of match, wherein displaying the medical image data comprises generating visual data indicative of the continuum of morphological scores.

18. The method of claim 16 wherein the medical image data is a two-dimensional image represented by a plurality of pixels and generating a morphological score comprises generating a map having map elements corresponding to the pixels of at least a portion of the medical image data with each of the map elements having a data value representative of the morphological score for the corresponding pixel.

19. The method of claim 16 wherein the medical image data is a three-dimensional image represented by a plurality of voxels and generating a morphological score comprises generating a map having map elements corresponding to the voxels of at least a portion of the medical image data with each of the map elements having a data value representative of the morphological score for the corresponding voxel.

20. The method of claim 16 wherein the region of interest has a border portion and generating a morphological score comprises generating a morphological score for the border portion.

21. The method of claim 16 wherein the region of interest has a two-dimensional area and generating a morphological score comprises generating a morphological score for the region area.

22. The method of claim 16 wherein the region of interest has a three-dimensional volume and generating a morphological score comprises generating a morphological score for the region volume.

23. The method of claim 16 wherein the region of interest comprises a portion of the medical image data less than the entire medical image and generating a morphological score comprises generating a morphological score for the entire medical image.

24. The method of claim 16 wherein generating a morphological score comprises using pattern recognition to compare the region of interest to a feature library of morphological characteristics.

25. The method of claim 16 wherein the medical image data is breast image data and the morphological characteristics correspond to ACR BI-RADS classifications.

26. The method of claim 16 wherein the morphological characteristics correspond to industry-standard classifications corresponding a tissue type for the region of interest.

27. The method of claim 16 wherein the morphological characteristics correspond to user-selectable classifications corresponding a tissue type for the region of interest.

28. A non-transitory computer-readable medium containing instructions for scoring features in medical image data by causing a computer system to:

receive medical image data, the medical image data including a region of interest;

analyze the region of interest with respect to a plurality of morphological characteristics in each of a plurality of morphological categories;

generate a morphological score along a continuum that maps substantially the entire region of interest to a score range indicative of a degree to which the region of interest matches the plurality of morphological characteristics in each of the plurality of morphological categories from a lesser degree of match to a greater degree of match; and

display the medical image data, including visual data indicative of the morphological score range across the continuum from less suspicious to more suspicious.

29. The computer-readable medium of claim 28 wherein instructions to generate a morphological score comprises generating a morphological score on a continuum indicating the degree to which the region of interest matches the plurality of morphological characteristics from a lesser degree of match to a greater degree of match, and displaying the medical image data comprises instructions to generate visual data indicative of the continuum of morphological scores.

30. The computer-readable medium of claim 28 wherein the medical image data is a two-dimensional image represented by a plurality of pixels and instructions to generate a morphological score comprises generating a map having map elements corresponding to the pixels of at least a portion of the medical image data with each of the map elements having a data value representative of the morphological score for the corresponding pixel.

31. The computer-readable medium of claim 28 wherein the medical image data is a three-dimensional image represented by a plurality of voxels and instructions to generate a morphological score comprises generating a map having map elements corresponding to the voxels of at least a portion of the medical image data with each of the map elements having a data value representative of the morphological score for the corresponding voxel.

32. The computer-readable medium of claim 28 wherein the region of interest has a border portion and instructions to generate a morphological score comprises generating a morphological score for the border portion.

33. The computer-readable medium of claim 28 wherein the region of interest has a two-dimensional area and instructions to generate a morphological score comprises generating a morphological score for the region area.

34. The computer-readable medium of claim 28 wherein the region of interest has a three-dimensional volume and instructions to generate a morphological score comprises generating a morphological score for the region volume.

35. The computer-readable medium of claim 28 wherein the region of interest comprises a portion of the medical image data less than the entire medical image and instructions to generate a morphological score comprises generating a morphological score for the entire medical image.

36. The computer-readable medium of claim 28 wherein instructions to generate a morphological score comprises using pattern recognition to compare the region of interest to a feature library of morphological characteristics.

37. The computer-readable medium of claim 28 wherein the medical image data is breast image data and the morphological characteristics correspond to ACR BI-RADS classifications.

38. The computer-readable medium of claim 28 wherein the morphological characteristics correspond to industry-standard classifications corresponding a tissue type for the region of interest.

39. The computer-readable medium of claim 28 wherein the morphological characteristics correspond to user-selectable classifications corresponding a tissue type for the region of interest.