Computer aided detection (CAD) for 3D digital mammography

• US 7,218,766 B2
• Filed: 04/15/2002
• Issued: 05/15/2007
• Est. Priority Date: 04/15/2002
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Eberhard, Jeffrey Wayne, Kapur, Ajay, Alyassin, Abdalmajeid Musa
• Primary Examiner(s): Mehta; Shavesh M
• Assistant Examiner(s): LAVIN, CHRISTOPHER L
• Application Number:
  US10/121,866
• Publication Number:
  US 20030194121A1
• Time in Patent Office: 1,856 Days
• Field of Search: 382/128, 382/129, 382130-132, 382/181, 382/190, 382/195, 382/224, 382/225, 382/226, 382/227, 382/228, 382/278, 382/294, 382/298, 382/299
• US Class Current: 382/132
• CPC Class Codes: A61B 6/03 Computed tomography [CT] echo-tomography A61B8/14
• CPC Class Codes: A61B 6/463 characterised by displaying multiple images or images and diagnostic data on one display
• CPC Class Codes: A61B 6/466 adapted to display 3D data
• CPC Class Codes: A61B 6/502 for diagnosis of breast, i.e. mammography
• CPC Class Codes: A61B 6/5247 combining images from an ionising-radiation diagnostic technique and a non-ionising radiation diagnostic technique, e.g. X-ray and ultrasound
• CPC Class Codes: A61B 8/463 characterised by displaying multiple images or images and diagnostic data on one display composite display in devices for short-range imaging with acoustic waves G01S7/52074
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: Y10S 128/922 including image analysis

Abstract

There is provided a method of analyzing a plurality of views of an object, the object including an edge portion partially extending from a surface of the object into an internal volume of the object, comprising the step of analyzing each acquired view. The step of analyzing each acquired view includes analysis of the edge portion. Preferably, the object comprises breast tissue.

Claims

1. A method of analyzing a plurality of angularly displaced views of an object, comprising the steps of:

analyzing each of at least three angularly displaced views including a plurality of x-ray image data collected at different angular positions of the radiation source with respect to the object, wherein the step of analyzing each acquired view comprises detecting at least one region of concern; and

classifying said detected region of concern;

correlating a detected region of concern in a first view of said plurality of angularly displaced views with a corresponding region in a second view of said plurality of angularly displaced views; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

2. The method of claim 1, wherein said object comprises breast tissue, and wherein the step of classifying said detected region of concern classifies pathologies within said breast tissue.

3. The method of claim 1, wherein the step of correlating comprises at least one of:

determining that a corresponding region does not exist; and

comparing at least one of;

a shape of the detected region of concern;

a size of the detected region of concern;

a contrast of the detected region of concern; and

a contrast distribution of the detected region of concern.

4. The method of claim 1, further comprising a step of reconstructing a plurality of reconstructed planes from said plurality of angularly displaced views.

5. The method of claim 4, further comprising the steps of:

detecting at least one region of concern in a reconstructed plane; and

classifying said detected region of concern in the reconstructed plane.

6. The method of claim 5, further comprising the steps of:

correlating a detected region of concern in a first reconstructed plane of said plurality of reconstructed planes with at least one of;

a corresponding region in a second reconstructed plane of said plurality of reconstructed planes;

a corresponding region in at least one of said plurality of angularly displaced views; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

7. The method of claim 1, further comprising a step of reconstructing a three dimensional (3D) image of said object from said plurality of angularly displaced views.

8. The method of claim 7, further comprising the steps of:

detecting regions of concern in said reconstructed 3D image; and

classifying said detected regions of concern in said reconstructed 3D image.

9. The method of claim 8, further comprising the steps of:

correlating a detected region of concern in said 3D image with a corresponding region in at least one view of said plurality of angularly displaced views; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

10. The method of claim 8, further comprising at least one of the steps of:

summing image data along a line to simulate a ray projection along that line;

determining maximum intensity projections within a detected region of concern;

determining a minimum intensity projection within a detected region of concern;

determining a mean intensity projection within a detected region of concern; and

determining median intensity projection amongst said detected regions of concern.

11. The method of claim 1 wherein said plurality of angularly displaced views are acquired with an x-ray modality, further comprising the steps of:

correlating said detected region of concern with a corresponding region in an image acquired by one of an ultrasound modality and a nuclear medicine modality; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

12. The method of claim 11, wherein the step of correlating comprises at least one of:

determining that a corresponding region does not exist; and

comparing at least one of;

a shape of the detected region of concern;

a size of the detected region of concern;

a contrast of the detected region of concern; and

a contrast distribution of the detected region of concern.

13. The method of claim 1, further comprising the steps of:

angularly displacing a radiation source about a pivot point to a position corresponding to said view;

radiating said object via said radiation source; and

detecting radiation from said object.

14. A program product recorded on a computer readable medium for causing a machine to analyze a plurality of angularly displaced views from a tomosynthesis system, said tomosynthesis system imaging an object, said program product causing the machine to perform the steps of:

analyzing each of at least three angularly displaced views including a plurality of x-ray image data collected at different angular positions of the radiation source with respect to the object wherein the step of analyzing each acquired view comprises detecting at least one region of concern; and

classifying said detected region of concern;

correlating a detected region of concern in a first view of said plurality of angularly displaced views with a corresponding region in a second view of said plurality of angularly displaced views; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

15. The program product of claim 14, further causing the machine to perform the step of:

reconstructing a plurality of reconstructed planes from said plurality of angularly displaced views.

16. The program product of claim 15, further causing the machine to perform the steps of:

detecting at least one region of concern in a reconstructed plane; and

classifying said detected region of concern in the reconstructed plane.

17. The program product of claim 16, further causing the machine to perform the steps of:

correlating a detected region of concern in a first reconstructed plane of said plurality of reconstructed planes with at least one of;

a corresponding region in a second reconstructed plane of said plurality of reconstructed planes;

a corresponding region in at least one of said plurality of angularly displaced views; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

18. The program product of claim 16, wherein said plurality of angularly displaced views are acquired with an x-ray modality, further causing the machine to perform the steps of:

correlating said detected region of concern with a corresponding region in an image acquired by one of an ultrasound modality and a nuclear medicine modality; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

19. The program product of claim 14, further causing the machine to perform the step of reconstructing a three dimensional (3D) image of said object from said plurality of angularly displaced views.

20. The program product of claim 19, further causing the machine to perform the steps of:

detecting regions of concern in said reconstructed 3D image; and

classifying said detected regions of concern in said reconstructed 3D image.

21. The program product of claim 20, further causing the machine to perform the steps of:

correlating a detected region of concern in said 3D image with a corresponding region in at least one view of said plurality of angularly displaced views; and

weighting said classification with a weighting factor corresponding to a degree of correlation.

22. The program product of claim 14, further causing the machine to perform the steps of:

angularly displacing a radiation source about a pivot point to a position corresponding to said view;

radiating said object via said radiation source; and

detecting radiation from said object.

23. A tissue imaging device, comprising:

a radiation source for emitting radiation through tissue to be imaged, said radiation source being angularly displaceable through a plurality of emission positions corresponding to a plurality of angularly displaced views;

a detector positioned to detect radiation emitted through said tissue, said detector generating a signal representing an angularly displaced view of said tissue, wherein the angularly displaced view includes x-ray image data collected at a different angular positions of the radiation source with respect to the tissue; and

a processor electrically coupled to said detector for analyzing said signal, wherein said processor analyzes at least three of said plurality of angularly displaced views by detecting at least one region of concern classifying, said detected region of concern, correlating a detected region of concern in a first view of said plurality of angularly displaced views with a corresponding region in a second view of said plurality of angularly displaced views and weighting said classification with a weighting factor corresponding to a degree of correlation.

24. The tissue imaging device of claim 23, wherein said processor further:

reconstructs a plurality of reconstructed planes from said plurality of angularly displaced views.

25. The tissue imaging device of claim 24, wherein said processor further:

detects at least one region of concern in a reconstructed plane; and

classifies said detected region of concern in the reconstructed plane.

26. The tissue imaging device of claim 25, wherein said processor further:

correlates a detected region of concern in a first reconstructed plane of said plurality of reconstructed planes with at least one of;

a corresponding region in a second reconstructed plane of said plurality of reconstructed planes;

a corresponding region in at least one of said plurality of angularly displaced views; and

weights said classification with a weighting factor corresponding to a degree of correlation.

27. The tissue imaging device of claim 23, wherein said processor further:

reconstructs a three dimensional (3D) image of said object from said plurality of angularly displaced views.

28. The tissue imaging device of claim 27, wherein said processor:

detects regions of concern in said reconstructed 3D image; and

classifies said detected regions of concern in said reconstructed 3D image.

29. The tissue imaging device of claim 28, wherein said processor further:

correlates a detected region of concern in said 3D image with a corresponding region in at least one view of said plurality of angularly displaced views; and

weights said classification with a weighting factor corresponding to a degree of correlation.

30. The tissue imaging device of claim 23, wherein the radiation source is angularly displaceable through less than 360°

about said tissue.

31. The tissue imaging device of claim 23, further comprising one of:

an ultrasound imager comprising;

an ultrasound source for emitting ultrasound through tissue to be imaged;

an ultrasound detector positioned to detect ultrasound emitted through said tissue to generate an ultrasound image;

a nuclear medicine detector to generate a nuclear medicine image, wherein said processor further;

detects a region of concern in said plurality of angularly displaced views;

correlates said detected region of concern with a corresponding region in one of said ultrasound image and said nuclear medicine image; and

weights said classification with a weighting factor corresponding to a degree of correlation.

32. A method of analyzing an object with a multi-modality imaging system, comprising the steps of:

generating a plurality of angularly displaced views of said object by a first modality, wherein each of the plurality of angularly displaced views includes x-ray image data collected at different angular positions of a radiation source with respect to said object;

detecting a region of concern in at least three of the plurality of angularly displaced views of said object generated by the first modality and a second image of said object generated by a second modality;

classifying said detected region of concern;

correlating said region of concern with a corresponding region in other of said angularly displaced views and said second image; and

weighting said classification with a weighting factor corresponding to a degree of correlation, wherein said first modality is different from said second modality.

33. The method of claim 32, wherein said first modality and said second modality each comprises one of x-ray, ultrasound, and nuclear medicine.

34. The method of claim 32, wherein at least one of said angularly displaced view and said second image comprise a three dimensional (3D) reconstructed image of said object.

35. The method of claim 32, further comprising at least one of:

determining that a corresponding region does not exist; and

comparing at least one of;

a shape of the detected region of concern;

a size of the detected region of concern;

a contrast of the detected region of concern; and

a contrast distribution of the detected region of concern.

36. The method of claim 32, further comprising the step of:

fusing said angularly displaced view with said second image.

37. The method of claim 32, further comprising the steps of:

registering said angularly displaced view with said second image; and

correcting differences in spatial resolution between said angularly displaced view and said second image.

38. An imaging system for imaging an object, comprising:

means for generating at least one of a plurality of angularly displaced views of said object from x-ray radiation;

means for generating a second image of said object from ultrasound;

means for detecting a region of concern in at least three of said angularly displaced views and said second image;

means for classifying said detected region of concern;

means for correlating said detected region of concern with a corresponding region in other of said angularly displaced views and said second image;

means for at least one of;

determining whether said abnormality is present in said corresponding region in said other of said angularly displaced view and said second image; and

comparing at least one of;

a shape of the detected region of concern;

a size of the detected region of concern;

a contrast of the detected region of concern; and

a contrast distribution of the detected region of concern;

means for classifying said abnormality; and

means for weighting said classification in relation to a degree of correlation.