Display for computer-aided evaluation of medical images and for establishing clinical recommendation therefrom

• US 20040184644A1
• Filed: 01/12/2004
• Published: 09/23/2004
• Est. Priority Date: 10/31/2002
• Status: Active Grant

• Current Assignee: Siemens Computer Aided Diagnosis Ltd. (Siemens AG)
• Original Assignee: Cadvision Medical Technologies Ltd.
• Inventors: Lederman, Richard, Bamberger, Philippe, Leichter, Isaac
• Granted Patent: US 7,418,119 B2
• Time in Patent Office: Days
• US Class Current: 382/128
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06V 10/987 with the intervention of an operator
• CPC Class Codes: G06V 2201/03 Recognition of patterns in medical or anatomical images
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A method for displaying a computer-generated determination of the overall likelihood of malignancy in a mammogram lesion. The method requires providing a digitized image of a mammogram, displaying the digitized image, and selecting a region of interest directly on the displayed digitized image. The digitized image is then processed so that classifier data of the lesion in the user-selected region of interest are generated and displayed. The overall likelihood of malignancy is generated from the classifier data and also displayed. A clinical recommendation for further assessing the lesion is derived from the overall evaluation of malignancy and presented to the user. A system for displaying a determination of the overall likelihood of malignancy in a mammogram lesion and for presenting a clinical recommendation to the user is also provided. A method for integrating the overall likelihood of malignancy with a physician'"'"'s independent score for the lesion is discussed.

Claims

1. A method for providing to a user a clinical recommendation for assessing the pathology of a lesion in living tissue, said method including the steps of:

providing a digitized image of the living tissue;

processing the digitized image and computing an overall evaluation of the likelihood of disease;

based on the overall evaluation of the likelihood of disease and correlated therewith, evaluating a clinical recommendation for further assessing the lesion; and

presenting to the user the clinical recommendation.

2. A method according to claim 1 wherein said step of processing includes processing the digitized image so that classifier data of a characterization feature of the lesion in the imaged tissue are quantified, and an overall evaluation of the likelihood of disease is computed therefrom.

3. A method according to claim 2 further including the step of selecting a characterization feature for which to compute classifier data.

4. A method according to claim 1, wherein the clinical recommendation is presented as a verbal message in the form of printed output in said step of presenting.

5. A method according to claim 1, wherein the clinical recommendation is presented as a numerical score in said step of presenting.

6. A method according to claim 1, wherein the clinical recommendation is presented as an audible signal in said step of presenting.

7. A method according to claim 1 wherein said step of presenting includes the step of displaying the clinical recommendation on a display.

8. A method according to claim 7, wherein the clinical recommendation is displayed as a verbal message.

9. A method according to claim 7, wherein the clinical recommendation is displayed on a part of a display separate from a displayed digitized image.

10. A method according to claim 9, wherein the clinical recommendation is displayed in a graphical format.

11. A method according to claim 10, wherein the displayed clinical recommendation is displayed as a color chosen from a set of colors, each of the colors in the set representing a different clinical recommendation from the other colors in the set.

12. A method according to claim 7, wherein the displayed clinical recommendation is displayed on a display of the digitized image.

13. A method according to claim 12, wherein the displayed clinical recommendation is displayed directly on the displayed digitized image as one element of a predefined set of indicators.

14. A method according to claim 13, wherein the indicators are chosen from a group consisting of a set of different colors, a set of different sizes, and a set of different shapes, each element of a set representing a different clinical recommendation from the other elements in that set.

15. A method according to claim 2, wherein said method further includes the step of selecting a region of interest directly on a displayed digitized image which contains a suspected lesion and wherein said step of processing includes quantifying classifier data for the suspected lesion.

16. A method according to claim 15, wherein the lesion in the user-selected region of interest is displayed in the same color as the overall evaluation of the likelihood of disease when the overall evaluation is displayed in a color chosen from colors in a predetermined multicolor format.

17. A method according to claim 15, wherein said step of processing includes processing the digitized image using a computer processor so that all classifier data of all characterization features in the digitized image are quantified and stored for later retrieval,

and further including the step of retrieving from storage the quantified classifier data related to a characterization feature in the selected region of interest.

18. A method according to claim 17, wherein said step of retrieving includes retrieving the stored classifier data from storage and returning it to the processor before the classifier data is transferred from the processor to a display.

19. A method according to claim 1 wherein said step of presenting includes the step of providing the clinical recommendation over a communications network to at least one distant user.

20. A method according to claim 1, wherein the tissue being evaluated is breast tissue.

21. A method according to claim 20, wherein said step of providing includes the step of directly providing a digitized image of the breast.

22. A method according to claim 21, wherein said method further includes the step of scoring the lesion found in the digitized image, the scoring effected according to a predefined scoring system intended to indicate the overall likelihood of disease of a lesion found in the digitized image.

23. A method according to claim 20, wherein said step of providing includes the step of scanning an analogue film mammogram so as to produce a digitized image and wherein said method further includes the step of reading the film mammogram on a light box.

24. A method according to claim 23, wherein said method further includes the step of scoring the lesion found in the film mammogram, the scoring effected according to a predefined scoring system intended to indicate the overall likelihood of disease of a lesion found in the mammogram.

25. A system for providing to a user a clinical recommendation for assessing a lesion in imaged living tissue, said system including:

an image provider for providing a digital image of the living tissue;

a processor operative to receive the digital image from said image provider, said processor computing an overall evaluation of the likelihood of disease in the lesion in the digitized image and said processor also deriving a clinical recommendation correlated with the computed overall evaluation; and

providing means arranged in data communication with said processor for receiving therefrom the clinical recommendation and for presenting the recommendation to the user.

26. A system according to claim 25, wherein said image provider is a digitizer for scanning and digitally representing an analogue image.

27. A system according to claim 25, wherein said providing means is a printer and the clinical recommendation is presented as a message on printed output.

28. A system according to claim 25, wherein said providing means for presenting the clinical recommendation is an audible device.

29. A system according to claim 25, wherein said providing means for presenting the clinical recommendation is a set of lights, each light representing a different clinical recommendation.

30. A system according to claim 25, wherein said providing means for presenting the clinical recommendation is a means for presenting a numerical score.

31. A system according to claim 25, wherein said providing means for presenting the clinical recommendation is a display, said display also displaying the digitized image.

32. A system according to claim 31, said system further including an input device, in communication with said processor and said display, for selectably indicating a region of interest on the displayed digitized image, wherein the clinical recommendation is computed for a lesion in the selected region of interest.

33. A system according to claim 31, wherein the clinical recommendation is presented as a verbal message on said display.

34. A system according to claim 31, wherein the clinical recommendation is presented as a numerical score on said display.

35. A system according to claim 31, wherein the clinical recommendation is displayed directly on the digitized image as one element of a pre-selected set of indicators.

36. A system according to claim 35, wherein said set of indicators is chosen from a group consisting of a set of different colors, a set of different sizes, and a set of different shapes, each element in a set representing a different clinical recommendation from the other elements in that set.

37. A system according to claim 31, wherein the displayed clinical recommendation is displayed on a part of said display separate from the digitized image.

38. A system according to claim 37, wherein the clinical recommendation is presented as a verbal message on said display.

39. A system according to claim 37, wherein the clinical recommendation is presented as a numerical score on said display.

40. A system according to claim 25, wherein said system is in communication with a communications network which provides the clinical recommendation to at least one distant user.

41. A system according to claim 25, said system further including a means for entering a score determined by the user, the score indicating the user'"'"'s evaluation of the expected abnormality of the lesion according to a predetermined scoring system, the score correlated with a specific clinical recommendation, wherein the user-determined clinical recommendation is modified by the overall evaluation of the likelihood of disease generated by said system, the modified recommendation provided to the user by said providing means.

42. A method for determining a clinical recommendation for assessing the pathology of a lesion in imaged living tissue, said method including the steps of:

having the user generate a score for the lesion according to a predefined scoring system, the score indicating the expected abnormality of the lesion and suggesting a clinical recommendation for further assessing the pathology of the lesion;

providing a digitized image of the imaged living tissue to a processor;

processing the digitized image in a processor so that an overall evaluation of the likelihood of disease is computed for the lesion, a clinical recommendation for further assessing the pathology of the lesion being derived from the overall evaluation and correlated therewith;

conveying to the user the overall evaluation; and

comparing the user-determined score indicating expected abnormality with the processor-computed overall evaluation of the likelihood of disease and using the overall evaluation to modify, according to predetermined criteria, the clinical recommendation suggested by the user-determined score.

43. A method according to claim 42, wherein said step of processing includes processing the digitized image so that classifier data of a characterization feature of the lesion in the imaged tissue are quantified, and an overall evaluation of the likelihood of disease is computed therefrom.

44. A method according to claim 43, wherein said method further includes the step of selecting a characterization feature for which to compute classifier data.

45. A method according to claim 42, wherein said step of conveying includes the step of displaying the overall evaluation on a display, the display also displaying a digitized image.

46. A method according to claim 45, wherein the overall evaluation is displayed on a part of a display separate from a displayed digitized image in said step of displaying.

47. A method according to claim 46, wherein the overall evaluation is displayed as part of a graphical display.

48. A method according to claim 47, wherein the displayed overall evaluation is displayed as a color chosen from a set of colors, each of the colors in the set representing a different overall evaluation.

49. A method according to claim 45, where the displayed overall evaluation is displayed on the display of the digitized image.

50. A method according to claim 49, wherein the displayed overall evaluation is displayed directly on the displayed digitized image as an element from a pre-selected set of indicators.

51. A method according to claim 50, wherein the indicators are chosen from a group consisting of sets of different colors, sets of different sizes, and sets of different shapes, each element of a set representing a clinical recommendation different from the other elements in that set.

52. A method according to claim 42, where the overall evaluation is conveyed as a numeric score in said step of conveying.

53. A method according to claim 42, wherein said method further includes the step of displaying a digitized image and the step of selecting a region of interest directly on the displayed digitized image, there being a suspected lesion in the user-selected region of interest, and quantified classifier data being computed for the lesion.

54. A method according to claim 53, wherein the lesion in the user-selected region of interest is displayed in the same color as the overall evaluation of the likelihood of disease when the overall evaluation is displayed in a color chosen from colors in a predetermined multicolor format.

55. A method according to claim 53, wherein the step of processing includes processing the digitized image using a computer processor so that all classifier data of all characterization features in the digitized image are quantified and stored for later retrieval,

and said method further includes the step of retrieving from storage the quantified classifier data related to a characterization feature in the selected region of interest.

56. A method according to claim 55, wherein said step of retrieving includes retrieving the stored classifier data from storage and returning it to the processor before the classifier data is transferred from the processor to a display in said step of displaying.

57. A method according to claim 42, wherein said step of conveying includes the step of providing the clinical recommendation over a communications network to at least one distant user.

58. A method according to claim 42, wherein the tissue being evaluated is breast tissue and the digitized image is a mammogram.

59. A method according to claim 58, wherein said step of providing includes the step of directly providing a digitized image of the breast.

60. A method according to claim 58, wherein said step of providing includes the step of scanning an analogue film mammogram so as to produce a digitized image and wherein said method further includes the step of reading the film mammogram on a light box prior to generating a score based on said step of reading.