System and method for diagnosis of living tissue diseases using digital image processing

• US 5,854,851 A
• Filed: 08/11/1994
• Issued: 12/29/1998
• Est. Priority Date: 08/13/1993
• Status: Expired due to Term

• Current Assignee: Siemens Computer Aided Diagnosis Ltd. (Siemens AG)
• Original Assignee: Sophisview Technologies Ltd.
• Inventors: Bamberger, Philippe, Leichter, Isaac, Fields, Scott
• Primary Examiner(s): JOHNS, ANDREW W
• Application Number:
  US08/289,381
• Time in Patent Office: 1,601 Days
• Field of Search: 382/128, 382/132, 382/169, 382/254, 382/270, 382/274, 378/62, 364/413.13, 364/413.26, 128/653.1
• US Class Current: 382/132
• CPC Class Codes: G06T 2207/30068 Mammography; Breast
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement

Abstract

A system and method for diagnosis of living tissue diseases is described. The system includes a computer device for controlling its operation. An operator control device is coupled to the computer device. A viewing screen is coupled to the computer device for displaying digitized images of the living tissue. The operator, using the control device, selects desired portions of the digitized image for further image enhancement according to a desired image enhancement feature selectable from a plurality of image enhancement features. The image enhancement features include any combination of grey scale stretching, contrast enhancement based on logarithmic histogram equalization, spot enhancement and magnification. The system further includes means for visualization and quantification of micro-calcifications, and means for visualization and quantification of mass spiculations.

Claims

1. A system for diagnosis of living tissue diseases, comprising:

a computer device for controlling the operation of said system;

an operator control device coupled to said computer device;

a viewing screen coupled to said computer device for displaying digitized images of said living tissue, said operator control device selecting desired portions of said digitized image for further image enhancement according to a desired image enhancement feature selectable from a plurality of image enhancement features;

said plurality of image enhancement features including any combination of grey scale stretching using an S-shaped grey scale conversion curve, contrast enhancement based on logarithmic histogram equalization, and spot enhancement.

2. The system according to claim 1 further comprising means for visualization and quantification of micro-calcifications.

3. The system according to claim 2 further comprising means for visualization and quantification of mass spiculations.

4. The system as claimed in claim 1, further comprising means for displaying the grey level values of at least one pixel on the digital image.

5. The system as claimed in claim 1, further comprising a network terminal for transmitting information to other locations.

6. A method for diagnosis of living tissue diseases, comprising the steps of:

providing a digital image of said living tissue to a computer;

displaying said digital image on a viewing screen;

selecting a portion of said digital image displayed on said viewing screen;

enhancing the image contained within the selected portion according to one or more options from a plurality of image enhancement features including grey scale stretching using an S-shaped grey scale conversion curve, contrast enhancement based on logarithmic histogram equalization, and spot enhancement.

7. The method according to claim 6 further comprising the steps of visualizing micro-calcifications and quantifying said micro-calcifications.

8. The method according to claim 7 further comprising the steps of visualizing mass spiculations and quantifying said mass spiculations.

9. The method as claimed in claim 8, wherein said enhancement of the displayed image is accomplished in real time.

10. The method as claimed in claim 6, wherein said grey scale stretching further comprises the step of converting original grey levels of the image into new values according to a curve including exponential and logarithmic portions.

11. The method as claimed in claim 10, wherein said conversion curve yields images with adjustable contrast and brightness.

12. The method as claimed in claim 6, wherein said digital image is enhanced by utilizing a logarithmic histogram equalization.

13. The method as claimed in claim 6, wherein said spot enhancement step further comprises the steps of checking the grey level difference between pixels of the image and their surroundings, diminishing the differences above a certain threshold of grey levels and accentuating the differences below a certain threshold.

14. A system for diagnosis of living tissue diseases, comprising:

a microcomputer for controlling the operation of said system;

a memory coupled to said microcomputer for storing information as directed by said microcomputer;

a scanning device coupled to said microcomputer for scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

a viewing screen coupled to said microcomputer wherein said digital image of said mammogram is displayed on said viewing screen;

said microcomputer including circuit means enabling on-line display of selected portions of said scanned mammogram; and

means for generating a plurality of grey scale conversion curves having a desired center and a desired slope at said center, said conversion curves converting the grey scale of said digital image into a new grey scale of a converted digital image, said desired center being a grey level value corresponding to said digital image such that for all said plurality of grey scale conversion curves said value of desired center is converted into a fixed grey level value on the new converted scale.

15. The system according to claim 14 wherein said grey scale conversion curve includes an exponential curve from grey level value "0" to said desired center and a logarithmic curve from said desired center to grey level value "255."

16. The system according to claim 15 wherein said fixed grey level value on the new converted scale is "127."

17. The system according to claim 16 further comprising means for generating a look up table representing said grey scale conversion curve.

18. The system according to claim 17 further comprising:

means for selecting a desired grey scale conversion curve for yielding a new digital image which is either darker and brighter than the original digital image; and

means for selecting a desired slope at said center for yielding a digital image with a variable contrast.

19. The system according to claim 18 further comprising means for generating a color look up table for converting said new digital image into a color image with pixels having different colors according to different combinations of RGB components.

20. The system according to claim 19 further comprising:

means for generating a plurality of conversion curves with center values ranging from a low grey level value to a high grey level value; and

means for displaying converted digital images based on said plurality of conversion curves, such that a changing color image appears on said viewing screen.

21. The system according to claim 20 further comprising:

means for determining an initial grey scale conversion curve including,

means for generating a histogram of said digital image,

means for selecting two grey level values g1 and g2 such that a given K percent of pixels in said digital image have original grey level values below g1 and a given N percent of pixels in said digital image have original grey level values above g2,

means for selecting said center and said slope such that a given (100-(K+N)) percent of new grey level values scale to original grey level values between g1 and g2.

22. A system for diagnosis of living tissue diseases, comprising:

a microcomputer for controlling the operation of said system;

a memory coupled to said microcomputer for storing information as directed by said microcomputer;

a scanning device coupled to said microcomputer for scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

a viewing screen coupled to said microcomputer wherein said digital image of said mammogram is displayed on said viewing screen;

said microcomputer including circuit means enabling on-line display of selected portions of said scanned mammogram; and

means for generating a plurality of grey scale conversion curves for converting the grey scale of said digital image into a new grey scale said grey scale conversion curves including an exponential curve from grey level value "0" to a desired grey level and a logarithmic curve from said desired grey level to grey level value "255."

23. The system according to claim 22 wherein the slope of said exponential curve and said logarithmic curves are equal at said desired grey level.

24. The system according to claim 23 wherein said slope is variable.

25. The system according to claim 24 further comprising means for generating a look up table representing said grey scale conversion curve.

26. A system for diagnosis of living tissue diseases comprising:

a microcomputer for controlling the operation of said system;

a scanning device couple to said microcomputer for scanning a mammogram and providing a digital image of said scanned mammogram;

means for generating a cumulative logarithmic histogram of said digital image; and

means for generating a look-up table for converting said digital image with a grey level scale into a new digital image with a new grey level scale such that the cumulative logarithmic histogram of said new digital image is equal to said generated cumulative logarithmic histogram.

27. The system according to claim 26 wherein said means for generating a cumulative logarithmic histogram of said digital image provides a plurality of cumulative logarithmic histogram curves CH(X), following the relationship ##EQU16## where "x" is a grey level value of said digital image between 0 and 255, "q" is a constant parameter for each of said cumulative logarithmic histogram curves, CH(255) is total amount of pixels in the digital image with a grey level value equal to maximum grey level in said digital image and CH(0) is the total amount of pixels in the digital image with a grey level value equal to minimum grey level in said digital image.

28. The system according to claim 27 further comprising means for generating the cumulative histogram of said digital image, wherein said lookup table is an exponential function

29. The system according to claim 28 further comprising means for varying said factor "q" for generating a plurality of said lookup tables.

30. The system according to claim 29 further comprising a pointing device coupled to said microcomputer wherein sliding movement of said pointing device back and forth along one direction causes the change in said constant "q".

31. The system according to claim 30 wherein said movement of said pointing device along an x axis changes a grey level value mx from 0 to 255 and said constant q is calculated by relationship ##EQU18##

32. A system for diagnosis of living tissue diseases comprising:

a microcomputer for controlling the operation of said system;

a scanning device coupled to said microcomputer for scanning a mammogram and providing a digital image of said scanned mammogram;

means for comparing grey level values between pixels of the image and their surroundings; and

means for generating a new digital image wherein differences above a certain threshold of grey levels are diminished and differences below a certain threshold of grey levels are accentuated.

33. The system according to claim 32 wherein said means for comparing the grey level values between pixels of the image and their surroundings further comprises:

means for generating a contrast table having a plurality of pairs of grey level values the first element of each pair being grey level value of each pixel in the digital image and the second element of each pair being the average value of grey levels of a desired number of pixels surrounding said pixel in the digital image;

means for comparing said plurality of pairs of grey level values; and

means for reducing said plurality of pairs when elements in one pair overlap with elements in another pair until none of the remaining pairs overlap.

34. The system according to claim 33 wherein said means for generating a new digital image further comprises:

means for generating a lookup table having a plurality of incrementing steps, the input of which is the grey level values of said digital image, each step representing an interval defined by each pair among said remaining pairs, said lookup table converting the grey level values within each interval into new grey level values which are linearly distributed within a fixed increment along the new grey scale.

35. A system for diagnosis of living tissue diseases, comprising:

a microcomputer for controlling the operation of said system;

a memory coupled to said microcomputer for storing information as directed by said microcomputer;

a scanning device coupled to said microcomputer for scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

a viewing screen coupled to said microcomputer wherein said digital image of said mammogram is displayed on said viewing screen;

said microcomputer including circuit means enabling on-line display of selected portions of said scanned mammogram;

means for user definition of micro-calcifications by interactively trimming a grey level threshold such that all pixels above said grey level threshold are highlighted as micro-calcifications; and

a region growing means for estimating the contour of said micro-calcifications, said region growing means comprising means for selecting and defining a current set of calcium pixels for each said micro-calcification.

36. The system according to claim 35 further comprising:

a background suppression means for suppressing the background image; and

means for producing an image with micro-calcifications having estimated contours which closely approximate the actual contour of said micro-calcifications.

37. The system according to claim 36 further comprising means for estimating a plurality of parameters of each micro-calcification corresponding to brightness, area, length, shape factor and distance to the nearest micro-calcification.

38. The system according to claim 37 further comprising means for identifying all pairs of micro-calcifications within a cluster that are considered to be adjacent using a triangulation technique.

39. The system according to claim 38 further comprising means for generating a plurality of line segments connecting said pairs of adjacent micro-calcifications.

40. The system according to claim 39 further comprising means for calculating a plurality of parameters based on said line segments including the length of each line segment, the number of neighbors of each micro-calcification and the density of a micro-calcification cluster.

41. A system for diagnosis of living tissue diseases, comprising:

a microcomputer for controlling the operation of said system;

a memory coupled to said microcomputer for storing information as directed by said microcomputer;

a scanning device coupled to said microcomputer for scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

a viewing screen coupled to said microcomputer wherein said digital image of said mammogram is displayed on said viewing screen;

said microcomputer including circuit means enabling on-line display of selected portions of said scanned mammogram; and

means for measuring the degree of spiculation of a mass including,

means for generating a binary gradient image of said digital image said binary gradient image having identified spiculation pixels, and

means for measuring spiculation ratio based on said binary gradient image.

42. The system according to claim 41 further comprising means for generating histogram of spiculation pixels.

43. The system according to claim 42 further comprising means for calculating the shift of said histogram of spiculation pixels.

44. The system according to claim 43 further comprising means for calculating the slope of said histogram of spiculation pixels.

45. The system according to claim 44 further comprising means for calculating the height of said histogram of spiculation pixels.

46. In a computer based system, a method for diagnosis of living tissue diseases, comprising the steps of:

scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

displaying said digital image of said mammogram on a viewing screen;

selecting a desired portion of said digital image; and

generating a plurality of grey scale conversion curves having a desired center and a desired slope at said center, each one of said conversion curves converting the grey scale of said digital image into a new grey scale of a converted digital image, said desired center being a grey level value corresponding to said digital image such that for all said plurality of grey scale conversion curves said value of desired center is converted into a fixed grey level value on the new converted scale.

47. The method according to claim 46 wherein said grey scale conversion curve includes an exponential curve from grey level value "0" to said desired center and a logarithmic curve from said desired center to grey level value "255".

48. The method according to claim 47 further comprising the steps of:

selecting a desired grey scale conversion curve for yielding a new digital image which is either darker and brighter than the original digital image; and

selecting a desired slope at said center for yielding a digital image with a variable contrast.

49. The method according to claim 48 further comprising the step of generating a color look up table for converting said new digital image into a color image with pixels having different colors according to different combinations of RGB components.

50. The method according to claim 49 further comprising the steps of generating a plurality of images resulting from a plurality of conversion curves with center values ranging from a low grey level value to a high grey level value;

displaying said plurality of images at a given rate creating an appearance of a moving image with changing colors.

51. In a computer based system, a method for diagnosis of living tissue diseases comprising the steps of:

scanning a mammogram and providing a digital image of said scanned mammogram to said computer;

generating a cumulative logarithmic histogram of said digital image; and

generating a look-up table for converting said digital image with a grey level scale into a new digital image with a new grey level scale such that the cumulative logarithmic histogram of said new digital image closely approximates said generated cumulative logarithmic histogram.

52. The method according to claim 51 wherein said step of generating a cumulative logarithmic histogram of said digital image provides a plurality of cumulative logarithmic histogram curves CH(X), following the relationship ##EQU19## where "x" is a grey level value of said digital image between 0 and 255, "q" is a constant parameter for each of said cumulative logarithmic histogram curves, CH(255) is total amount of pixels in the digital image with a grey level value equal to maximum grey level in said digital image and CH(0) is the total amount of pixels in the digital image with a grey level value equal to minimum grey level in said digital image.

53. The method according to claim 52 further comprising the step of generating the cumulative histogram of said digital image, wherein said lookup table is an exponential function

54. The method according to claim 53 further comprising the step of varying said factor "q" for generating a plurality of said lookup tables.

55. In a computer based system, a method for diagnosis of living tissue diseases comprising the steps of:

scanning a mammogram and providing a digital image of said scanned mammogram;

comparing grey level values between pixels of the image and their surroundings; and

generating a new digital image wherein differences above a certain threshold of grey levels are diminished and differences below a certain threshold are accentuated.

56. The method according to claim 55 wherein said step for comparing the grey level values between pixels of the image and their surroundings further comprises:

generating a contrast table having a plurality of pairs of grey level values the first element of each pair being grey level value of each pixel in the digital image and the second element of each pair being the average value of grey levels of a desired number of pixels surrounding said pixel in the digital image;

comparing said plurality of pairs of grey level values;

reducing said plurality of pairs when elements in one pair overlap with elements in another pair until none of the remaining pairs overlap;

generating a lookup table having a plurality of incrementing steps, the input of which is the grey level values of said digital image, each step representing an interval defined by each pair among said remaining pairs, said lookup table converting the grey level values within each interval into new grey level values which are linearly distributed within a fixed increment along the new grey scale.

57. In a computer based system a method for diagnosis of living tissue diseases, comprising the steps of:

scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

displaying said digital image on a viewing screen;

selecting a desired portion of said scanned mammogram;

displaying micro-calcifications which are user-defined by interactively trimming a grey level threshold such that all pixels above said grey level threshold are highlighted as micro-calcifications;

estimating the contour of said micro-calcifications using a region growing process, said region growing process comprising selecting and defining a current set of calcium pixels for each said micro-calcification; and

producing an image with micro-calcifications which closely approximates the actual contour of said micro-calcifications.

58. The method according to claim 57 further comprising the steps of:

suppressing the background image prior to said step of interactive trimming;

estimating a plurality of parameters of each micro-calcification region corresponding to brightness, area, length, shape factor and distance to the nearest micro-calcification region after said step of producing an image with micro-calcifications.

59. A method according to claim 58 further comprising the steps of:

identifying all pairs of micro-calcifications within a cluster that are considered to be adjacent using a triangulation technique;

generating a plurality of line segments connecting said pairs of adjacent micro-calcifications; and

calculating a plurality of parameters based on said line segments including the length of each line segment, the number of neighbors of each micro-calcification and the density of a micro-calcification cluster.

60. In a computer based system, a method for diagnosis of living tissue diseases, comprising the steps of:

scanning a mammogram to provide a digital image of said mammogram for storing, viewing and further processing;

displaying said digital image on a viewing screen;

selecting a desired portion of said digital image; and

measuring the degree of spiculation of a mass by, generating a binary gradient image of said digital image said binary gradient image having identified spiculation pixels.

61. The method according to claim 60 further comprising the steps of:

measuring spiculation ratio based on said binary gradient image;

generating histogram of spiculation pixels;

calculating the shift, the slope and the height of said histogram of spiculation pixels.