Multispectral image processing method, apparatus and computer program product for determining the blood oxygen saturation in a vessel

• US 20030088165A1
• Filed: 08/13/2002
• Published: 05/08/2003
• Est. Priority Date: 08/14/2001
• Status: Active Grant

• Current Assignee: University of Alabama (University of Alabama System)
• Original Assignee: University of Alabama (University of Alabama System)
• Inventors: Denninghoff, Kurt R., Hillman, Lloyd W., Manchenahalli, Raghunandan, Smith, Matthew H.
• Granted Patent: US 6,728,561 B2
• Time in Patent Office: Days
• US Class Current: 600/323
• CPC Class Codes: A61B 3/0025 characterised by electronic signal processing, e.g. eye models
• CPC Class Codes: A61B 3/12 for looking at the eye fundus, e.g. ophthalmoscopes A61B3/13 takes precedence
• CPC Class Codes: A61B 3/1233 for measuring blood flow, e.g. at the retina
• CPC Class Codes: A61B 5/14555 specially adapted for the eye fundus
• CPC Class Codes: G01N 21/359 using near infrared light

Abstract

Methods, apparatus and computer program products are provided for more accurately determining the percent transmittance of a vessel, such as a retinal vessel, at each of a number of different wavelengths and, in turn, for more accurately determining the blood oxygen saturation within the vessel. The blood oxygen saturation is determined based upon a plurality of images of the vessel obtained with illumination of different wavelengths. Background images are generated based upon respective images of the vessel in order to approximate the tissue bed underlying the vessel, such as the background fundus underlying the retinal vessel. Thereafter, a plurality of transmittance images are determined based upon respective pairs of the background images and the images of the vessel, such as by dividing the image of the vessel by the respective background image. Based upon a plurality of transmittance images, the blood oxygen saturation in the vessel may be determined.

Claims

1. A method for determining blood oxygen saturation in a vessel based upon a plurality of images of the vessel obtained in response to illumination of the vessel at different wavelengths, the method comprising:

generating a plurality of background images based upon respective ones of the plurality of images of the vessel, wherein generating the background images comprises approximating an image of a tissue bed underlying the vessel;

determining a plurality of transmittance images based upon respective pairs of the background images and the images of the vessel; and

determining the blood oxygen saturation in the vessel based upon the plurality of transmittance images.

2. A method according to claim 1 further comprising identifying at least one vessel in each of the images of the vessel prior to generating the background images.

3. A method according to claim 1 further comprising registering the vessel in each of the images of the vessel prior to generating the background images.

4. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein the method further comprises processing the plurality of images of the vessel to separate, within each image, pixels representative of the vessel from pixels representative of the tissue bed.

5. A method according to claim 4 wherein processing the plurality of images of the vessel comprises separately analyzing the pixels of each image along respective lines extending across the vessel to separate the pixels along the respective lines into pixels representative of the vessel and pixels representative of the tissue bed.

6. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein the method further comprises processing each image to identify pixels representative of the vessel that are at least partially surrounded by pixels representative of the tissue bed and to redefine any pixels that are so identified to be representative of the tissue bed.

7. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein the method further comprises processing the plurality of images to identify a pixel in one image that differs from a corresponding pixel in another image in its representation of that portion of the image as either the vessel or the tissue bed and to redefine one of the pixels to be consistent with the other pixel.

8. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein the method further comprises processing the plurality of images to identify a group of adjacent pixels that each represent the vessel and to redefine any pixels that are also initially representative of the vessel but are remote from the group of adjacent pixels to be representative of the tissue bed.

9. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein generating the plurality of background images comprises scaling the pixels representative of the vessel to have the same mean and standard deviation as the pixels representative of the tissue bed.

10. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein generating the plurality of background images comprises scaling the pixels representative of the vessel based upon a shape of the vessel.

11. A method according to claim 1 wherein each image is comprised of a plurality of pixels, and wherein generating the plurality of background images comprises redefining the pixels that formerly represented the vessel in accordance with curves that are fit based upon values of the pixels representative of the tissue bed that are on opposite sides of the pixels representative of the vessel.

12. A method according to claim 1 wherein determining a plurality of transmittance images comprises dividing each image of the vessel by the respective background image.

13. A method according to claim 1 wherein determining the blood oxygen saturation in the vessel comprises generating an image of the blood oxygen saturation in the vessel.

14. An apparatus for determining blood oxygen saturation in a vessel based upon a plurality of images of the vessel obtained in response to illumination of the vessel at different wavelengths, the apparatus comprising a processing element capable of generating a plurality of background images based upon respective ones of the plurality of images of the vessel, wherein said processing element is capable of generating the background images by approximating an image of a tissue bed underlying the vessel, wherein said processing element is also capable of determining a plurality of transmittance images based upon respective pairs of the background images and the images of the vessel, and wherein said processing element is further capable of determining the blood oxygen saturation in the vessel based upon the plurality of transmittance images.

15. An apparatus according to claim 14 wherein said processing element is further capable of registering the vessel in each of the images of the vessel prior to generating the background images.

16. An apparatus according to claim 14 wherein each image is comprised of a plurality of pixels, and wherein said processing element is further capable of processing the plurality of images of the vessel to separate, within each image, pixels representative of the vessel from pixels representative of the tissue bed.

17. An apparatus according to claim 16 wherein said processing element is capable of processing the plurality of images of the vessel by separately analyzing the pixels of each image along respective lines extending across the vessel to separate the pixels along the respective lines into pixels representative of the vessel and pixels representative of the tissue bed.

18. An apparatus according to claim 14 wherein each image is comprised of a plurality of pixels, and wherein said processing element is further capable of processing each image to identify pixels representative of the vessel that are at least partially surrounded by pixels representative of the tissue bed and to redefine any pixels that are so identified to be representative of the tissue bed.

19. An apparatus according to claim 16 wherein each image is comprised of a plurality of pixels, and wherein said processing element is further capable of processing the plurality of images to identify a pixel in one image that differs from a corresponding pixel in another image in its representation of that portion of the image as either the vessel or the tissue bed and to redefine one of the pixels to be consistent with the other pixel.

20. An apparatus according to claim 14 wherein each image is comprised of a plurality of pixels, and wherein said processing element is further capable of processing the plurality of images to identify a group of adjacent pixels that each represent the vessel and to redefine any pixels that are also initially representative of the vessel but are remote from the group of adjacent pixels to be representative of the tissue bed.

21. An apparatus according to claim 14 wherein each image is comprised of a plurality of pixels, and wherein said processing element is capable of generating the plurality of background images by scaling the pixels representative of the vessel to have the same mean and standard deviation as the pixels representative of the tissue bed.

22. An apparatus according to claim 14 wherein each image is comprised of a plurality of pixels, and wherein said processing element is capable of generating the plurality of background images by scaling the pixels representative of the vessel based upon a shape of the vessel.

23. An apparatus according to claim 14 wherein each image is comprised of a plurality of pixels, and wherein said processing element is capable of generating the plurality of background images by redefining the pixels that formerly represented the vessel in accordance with curves that are fit based upon values of the pixels representative of the tissue bed that are on opposite sides of the pixels representative of the vessel.

24. An apparatus according to claim 14 wherein said processing element is capable of determining a plurality of transmittance images by dividing each image of the vessel by the respective background image.

25. An apparatus according to claim 14 wherein said processing element is further capable of generating an image of the blood oxygen saturation in the vessel.

26. A computer program product for determining blood oxygen saturation in a vessel based upon a plurality of images of the vessel obtained in response to illumination of the vessel at different wavelengths, the computer program product comprising a computer-readable storage medium having computer-readable instructions embodied therein, the computer-readable instructions comprising:

first computer-readable instructions capable of generating a plurality of background images based upon respective ones of the plurality of images of the vessel, wherein said first computer-readable instructions are capable of generating the background images by approximating an image of a tissue bed underlying the vessel;

second computer-readable instructions capable of determining a plurality of transmittance images based upon respective pairs of the background images and the images of the vessel; and

third computer-readable instructions capable of determining the blood oxygen saturation in the vessel based upon the plurality of transmittance images.

27. A computer program product according to claim 26 further comprising fourth computer-readable instructions capable of registering the vessel in each of the images of the vessel prior to generation of the background images by said first computer-readable instructions.

28. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein the computer program product further comprises fifth computer-readable instructions capable of processing the plurality of images of the vessel to separate, within each image, pixels representative of the vessel from pixels representative of the tissue bed.

29. A computer program product according to claim 28 wherein said fifth computer-readable instructions are capable of separately analyzing the pixels of each image along respective lines extending across the vessel to separate the pixels along the respective lines into pixels representative of the vessel and pixels representative of the tissue bed.

30. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein the computer program product further comprises fifth computer-readable instructions capable of processing each image to identify pixels representative of the vessel that are at least partially surrounded by pixels representative of the tissue bed and to redefine any pixels that are so identified to be representative of the tissue bed.

31. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein the computer program product further comprises fifth computer-readable instructions capable of processing the plurality of images to identify a pixel in one image that differs from a corresponding pixel in another image in its representation of that portion of the image as either the vessel or the tissue bed and to redefine one of the pixels to be consistent with the other pixel.

32. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein the computer program product further comprises fifth computer-readable instructions capable of processing the plurality of images to identify a group of adjacent pixels that each represent the vessel and to redefine any pixels that are also initially representative of the vessel but are remote from the group of adjacent pixels to be representative of the tissue bed.

33. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein said first computer-readable instructions are further capable of scaling the pixels representative of the vessel to have the same mean and standard deviation as the pixels representative of the tissue bed.

34. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein said first computer-readable instructions are further capable of scaling the pixels representative of the vessel based upon a shape of the vessel.

35. A computer program product according to claim 26 wherein each image is comprised of a plurality of pixels, and wherein said first computer-readable instructions are further capable of redefining the pixels that formerly represented the vessel in accordance with curves that are fit based upon values of the pixels representative of the tissue bed that are on opposite sides of the pixels representative of the vessel.

36. A computer program product according to claim 26 wherein said second computer-readable instructions are further capable of dividing each image of the vessel by the respective background image.

37. A computer program product according to claim 26 wherein said third computer-readable instructions are further capable of generating an image of the blood oxygen saturation in the vessel.

38. A method for generating a background image based upon an image of the vessel obtained in response to illumination of the vessel at a respective wavelength, the method comprising:

processing the image of the vessel to separate pixels representative of the vessel from pixels representative of a tissue bed; and

redefining the pixels that formerly represented the vessel based upon respective values of at least some of the pixels representative of the tissue bed to create the background image, wherein the background image at least partially represents the tissue bed that underlies the vessel.

39. A method according to claim 38 wherein redefining the pixels that formerly represented the vessel comprises scaling the pixels representative of the vessel to have the same mean and standard deviation as the pixels representative of the tissue bed.

40. A method according to claim 38 wherein redefining the pixels that formerly represented the vessel comprises scaling the pixels representative of the vessel based upon a shape of the vessel.

41. A method according to claim 38 wherein redefining the pixels that formerly represented the vessel comprises redefining the pixels that formerly represented the vessel in accordance with curves that are fit based upon values of the pixels representative of the tissue bed that are on opposite sides of the pixels representative of the vessel.

42. A method according to claim 38 wherein processing the plurality of images of the vessel comprises separately analyzing the pixels along respective lines extending across the vessel to separate the pixels along the respective lines into pixels representative of the vessel and pixels representative of the tissue bed.

43. A method according to claim 38 further comprising processing the image to identify pixels representative of the vessel that are at least partially surrounded by pixels representative of the tissue bed and to redefine any pixels that are so identified to be representative of the tissue bed, prior to the redefining step.

44. A method according to claim 38 further comprising processing the image to identify a group of adjacent pixels that each represent the vessel and to redefine any pixels that are also initially representative of the vessel but are remote from the group of adjacent pixels to be representative of the tissue bed.

45. A method according to claim 38 wherein a plurality of background images of the vessel are generated based upon a plurality of images of the vessel obtained in response to illumination of the vessel at different wavelengths, and wherein the method further comprises registering the vessel in each of the images of the vessel prior to processing the plurality of images of the vessels.

46. A method according to claim 45 further comprising processing each image to identify a pixel in one image that differs from a corresponding pixel in another image in its representation of that portion of the image as either the vessel or the tissue bed and to redefine one of the pixels to be consistent with the other pixel.

47. A method of separating pixels representative of a vessel from pixels representative of a tissue bed in an image comprised of a plurality of pixels, wherein the method comprises:

examining the pixels along a plurality of lines that extend across the vessel;

determining, for each line of pixels, a threshold based upon values of the pixels along the respective line; and

separating the pixels along each line into pixels representative of the vessel and pixels representative of the tissue bed depending upon a relationship of the threshold for the respective line to values of the pixels along the line.

48. A method according to claim 47 wherein examining the pixels comprises examining the pixels along a plurality of lines that extend perpendicular to the vessel.

49. A method according to claim 47 wherein determining the threshold for each line of pixels comprises determining the threshold based upon at least one of a mean and a standard deviation of the values of the pixels along the respective line.

50. A method according to claim 49 wherein determining the threshold for a respective line further comprises determining the threshold x as follows: