Image processing using measures of similarity

• US 7,260,248 B2
• Filed: 03/15/2002
• Issued: 08/21/2007
• Est. Priority Date: 12/15/1999
• Status: Expired due to Fees

• Current Assignee: Luma Imaging Corp. (SpectraScience, Inc.)
• Original Assignee: MediSpectra, Inc. (SpectraScience, Inc.)
• Inventors: Kaufman, Howard, Schmid, Philippe
• Primary Examiner(s): Ahmed; Samir
• Assistant Examiner(s): MISTRY, O NEAL RAJAN
• Application Number:
  US10/099,881
• Publication Number:
  US 20030144585A1
• Time in Patent Office: 1,985 Days
• Field of Search: 382/209, 382/173, 382/128, 382/133, 382/212, 600/407, 600/476, 600/431
• US Class Current: 382/128
• CPC Class Codes: A61B 5/0059 using light, e.g. diagnosis by transillumination, diascopy, fluorescence A61B5/0093, A61B5/1455, A61B5/02416 and A61B5/0261 take precedence
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/10064 Fluorescence image
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/20152 Watershed segmentation
• CPC Class Codes: G06T 2207/30016 Brain
• CPC Class Codes: G06T 2207/30024 Cell structures in vitro; Tissue sections in vitro
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/155 involving morphological operators
• CPC Class Codes: G06T 7/174 involving the use of two or more images
• CPC Class Codes: G06V 20/695 Preprocessing, e.g. image segmentation

Abstract

The invention provides methods of relating a plurality of images based on measures of similarity. The methods of the invention are useful in the segmentation of a sequence of colposcopic images of tissue, for example. The methods may be applied in the determination of tissue characteristics in acetowhitening testing of cervical tissue, for example.

Claims

1. A method of locating a portion of tissue with a characteristic of interest, the method comprising the steps of:

(a) characterizing an acetowhitening signal from a temporal sequence of images of a tissue following application of a chemical agent to the tissue, wherein the chemical agent comprises acetic acid;

(b) analyzing the acetowhitening signal to determine a measure of similarity between two selected regions of the tissue, the measure of similarity indicating how similarly tissue in each region responds to the chemical agent, and grouping the two selected regions if the measure of similarity is larger than a given threshold, wherein the step of determining the measure of similarity comprises, for each of the two selected regions, averaging data corresponding to pixels within the region at each of a plurality of time steps to obtain a mean signal for the region, then quantifying the similarity between the two resulting, mean signals;

(c) repeating step (b), thereby differentiating regions according to how tissue in each region responds to the chemical agent; and

(d) locating a portion of the tissue with a characteristic of interest, the located portion corresponding to at least one of the differentiated regions.

2. The method of claim 1, wherein step (b) further comprises merging the two selected regions into a single region if the measure of similarity satisfies a predetermined criterion.

3. The method of claim 2, wherein steps (b) and (c) together comprise iteratively segmenting an area of the tissue into regions according to evolution of mean intensity of each region following application of the chemical agent.

4. The method of claim 1, wherein step (c) comprises repeating step (b) for each of a plurality of pairs of selected regions.

5. The method of claim 1, wherein the measure of similarity indicates a similarity in evolution of mean intensity of each region following application of the chemical agent.

6. The method of claim 1, wherein step (b) comprises computing an N-dimensional dot product of mean signal intensities of the two selected regions.

7. The method of claim 1, wherein the chemical agent further comprises a member selected from the group consisting of formic acid, propionic acid, butyric acid, Lugol'"'"'s iodine, Shiller'"'"'s iodine, methylene blue, toluidine blue, and indigo carmine.

8. The method of claim 1, further comprising the step of:

(e) determining a condition of the located portion.

9. The method of claim 8, wherein the condition comprises a member selected from the group consisting of normal squamous tissue, metaplasia, CIN I, CIN II, CIN III, and CIN II/III.

10. The method of claim 8, wherein step (e) comprises determining a condition of the located portion based at least in part on evolution of mean intensity of the located portion.

11. The method of claim 8, wherein step (e) further comprises obtaining a biopsy specimen within the located portion prior to determining the condition of the located portion.

12. The method of claim 8, wherein the condition comprises a member selected from the group consisting of CIN II, CIN III, and CIN II/III.

13. The method of claim 1, wherein the characteristic of interest is a suspicion of pathology.

14. The method of claim 1, wherein the tissue comprises cervical tissue.

15. The method of claim 1, wherein the tissue comprises at least one member selected from the group consisting of epithelial tissue, colorectal tissue, skin, and uterine tissue.

16. The method of claim 1, further comprising the step of illuminating the tissue using a white light source, a UV light source, or both.

17. A method of differentiating regions of a tissue, the method comprising the steps of:

(a) accessing a temporal sequence of images of a tissue following application of a chemical agent to the tissue; and

(b) creating a segmentation mask that represents an image plane divided into regions according to how similarly tissue in each region responds to the chemical agent, wherein step (b) comprises analyzing an acetowhitening signal to determine a measure of similarity between two selected regions of the tissue, the measure of similarity indicating how similarly tissue in each region responds to the chemical agent, and grouping the two selected regions if the measure of similarity is larger than a given threshold, wherein the step of determining the measure of similarity comprises, for each of the two selected regions, averaging data corresponding to pixels within the region at each of a plurality of time steps to obtain a mean signal for the region, then quantifying the similarity between the two resulting mean signals.

18. The method of claim 17, wherein the chemical agent comprises a member selected from the group consisting of acetic acid, formic acid, propionic acid, butyric acid, Lugol'"'"'s iodine, Shiller'"'"'s iodine, methylene blue, toluidine blue, and indigo carmine.

19. The method of claim 17, wherein the chemical agent comprises acetic acid.

20. The method of claim 17, further comprising the step of:

(c) locating a portion of the tissue with a characteristic of interest, the located portion corresponding to at least one of the differentiated regions.

21. The method of claim 20, further comprising the step of:

(d) determining a condition of the located portion.

22. The method of claim 21, wherein the condition comprises a member selected from the group consisting of normal squamous tissue, metaplasia, CIN I, CIN II, CIN III, and CIN II/III.

23. The method of claim 21, wherein the condition comprises a member selected from the group consisting of CIN II, CIN III, and CIN II/III.

24. The method of claim 17, wherein the tissue comprises cervical tissue.

25. The method of claim 17, wherein the tissue comprises at least one member selected from the group consisting of epithelial tissue, colorectal tissue, skin, and uterine tissue.

26. The method of claim 17, wherein the step of creating the segmentation mask comprises analyzing an acetowhitening signal characterized from the temporal sequence of images, wherein the chemical agent comprises acetic acid.

27. A system for differentiating regions of a tissue, the system comprising:

a light source that illuminates a tissue;

a camera that obtains a temporal sequence of images of the tissue following application of a chemical agent to the tissue, the chemical agent comprising acetic acid; and

software that performs the steps of;

(i) characterizing an acetowhitening signal from temporal sequence of images;

(ii) analyzing the acetowhitening signal to determining a measure of similarity between two selected regions of the tissue, the measure of similarity indicating how similarly tissue in each region responds to the chemical agent, and grouping the two selected regions if the measure of similarity is larger than a given threshold, wherein the step of determining the measure of similarity comprises, for each of the two selected regions, averaging data corresponding to pixels within the region at each of a plurality of time steps to obtain a mean signal for the region, then quantifying the similarity between the two resulting mean signals; and

(iii) repeating step (ii), thereby differentiating regions according to how tissue in each region responds to the chemical agent.

28. The system of claim 27, wherein the chemical agent further comprises a member selected from the group consisting of formic acid, propionic acid, butyric acid, Lugol'"'"'s iodine, Shiller'"'"'s iodine, methylene blue, toluidine blue, and indigo carmine.

29. The system of claim 27, wherein the software further performs the step of:

(c) locating a portion of the tissue with a characteristic of interest, the located portion corresponding to at least one of the differentiated regions.

30. The system of claim 29, further comprising the step of:

(d) determining a condition of the located portion.

31. The system of claim 30, wherein the condition comprises a member selected from the group consisting of normal squamous tissue, metaplasia, CIN I, CIN II, CIN III, and CIN II/III.

32. The system of claim 30, wherein the condition comprises a member selected from the group consisting of CIN II, CIN III, and CIN II/III.

33. The system of claim 27, wherein the tissue comprises cervical tissue.

34. The system of claim 27, wherein the tissue comprises at least one member selected from the group consisting of epithelial tissue, colorectal tissue, skin, and uterine tissue.