Method and apparatus for identifying spectral artifacts

• US 20040007674A1
• Filed: 09/13/2002
• Published: 01/15/2004
• Est. Priority Date: 07/09/2002
• Status: Active Grant

• Current Assignee: Luma Imaging Corp. (SpectraScience, Inc.)
• Original Assignee: MediSpectra, Inc. (SpectraScience, Inc.)
• Inventors: Flanagan, John, Gao, Harry, Meese, Thomas, Ouradnik, Michael, Schomacker, Kevin T.
• Granted Patent: US 6,818,903 B2
• Time in Patent Office: Days
• US Class Current: 250/458.1
• CPC Class Codes: A61B 1/043 for fluorescence imaging
• CPC Class Codes: A61B 1/0607 for annular illumination
• CPC Class Codes: A61B 1/0638 providing two or more wavelengths
• CPC Class Codes: A61B 1/0655 Control therefor
• CPC Class Codes: A61B 1/303 for the vagina, i.e. vaginoscopes
• 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: G01N 2021/6417 Spectrofluorimetric devices
• CPC Class Codes: G01N 21/31 Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry G01N21/72 takes precedence
• CPC Class Codes: G01N 21/64 Fluorescence; Phosphorescence

Abstract

The invention provides an apparatus and methods for determining whether spectral data obtained from a region of a tissue sample are affected by an artifact. Artifacts include, for example, lighting artifacts such as glare and shadow and obstruction artifacts, such as blood, a speculum, a smoke tube, or other obstruction. Additionally, the invention provides an apparatus and methods for obtaining redundant spectral data of a given region of a sample. A redundant set of spectral data is useful where one or more artifacts affect some but not all sets of the spectral data, such that the redundant set of data is unaffected by the artifact and is representative of the tissue. An embodiment of the invention comprises using representative spectral data in diagnosing a condition of a region of tissue.

Claims

1. A method of determining a condition of a region of a tissue sample, said method comprising the steps of:

(a) obtaining a first set of spectral data corresponding to a region of a tissue sample using light incident to said region at a first angle;

(b) obtaining a second set of spectral data corresponding to said region using light incident to said region at a second angle;

(c) selecting at least one of said first set of spectral data and said second set of spectral data that is representative of said region of said tissue sample; and

(d) determining a condition of said region of said tissue sample based at least in part on a subset of said at least one set of spectral data selected in step (c).

2. The method of claim 1, wherein said first set of spectral data comprises reflectance spectral data and said second set of spectral data comprises reflectance spectral data.

3. The method of claim 1, wherein said at least one of said first set of spectral data and said second set of spectral data comprises fluorescence spectral data.

4. The method of claim 1, further comprising obtaining a third set of spectral data corresponding to said region using light incident to said region at a third angle.

5. The method of claim 1, further comprising obtaining each of a plurality of sets of spectral data in addition to said first set and said second set using light incident to said region at a unique angle.

6. The method of claim 1, wherein said condition is a state of health.

7. The method of claim 6, wherein said state of health comprises at least one of the conditions of normal squamous tissue, metaplasia, CIN I, CIN II, CIN III, CIS, and cancer.

8. A method of determining whether spectral data obtained from a region of a tissue sample are affected by an artifact, said method comprising the steps of:

obtaining a first set of spectral data corresponding to a region of a tissue sample using light incident to said region at a first angle;

obtaining a second set of spectral data corresponding to said region using light incident to said region at a second angle; and

determining whether said first set of data is affected by an artifact based at least in part on a subset of said first set of data and a subset of said second set of data.

9. The method of claim 8, wherein said first set of spectral data comprises reflectance spectral data and said second set of spectral data comprises reflectance spectral data.

10. The method of claim 8, further comprising obtaining a third set of spectral data, where said third set of spectral data comprises fluorescence spectral data.

11. The method of claim 8, wherein said determining step comprises computing a difference between R₁, a member of said first set of spectral data, and R₂, a member of said second set of spectral data, and comparing said difference to a constant, where R₁ and R₂ correspond to at least approximately identical wavelengths.

12. The method of claim 11, wherein said difference is a percent difference.

13. The method of claim 8, wherein said determining step comprises computing N differences, |R₁(X_i)−

R₂(X_i)|, optionally weighting each of said differences using at least one of R₁(X_i) and R₂(X_i), defining a maximum of a subset of said N optionally-weighted differences, and comparing said maximum to a first constant, where i=1 to N, N is an integer, X_i is a wavelength between about 360 nm and about 720 nm, R₁(X_i) is a member of said first set of data corresponding to said wavelength X_i, and R₂(X_i) is a member of said second set of data corresponding to said wavelength X_i.

14. The method of claim 13, wherein said determining step further comprises comparing R₁(X₁) to a second constant, where R₁(X₁) is a member of said first set of data corresponding to a wavelength X₁ between about 409 nm and about 429 nm.

15. The method of claim 8, wherein said determining step comprises comparing R₁(X₁) to a constant, where R₁(X₁) is a member of said first set of data corresponding to a wavelength X₁ between about 409 nm and about 429 nm.

16. The method of claim 8, wherein said determining step comprises comparing the quotient R₁(X₁)/R₁(X₂) to a constant, where R₁(X₁) is a member of said first set of data corresponding to a wavelength X₁ between about 360 nm and about 720 nm, and R₁(X₂) is a member of said first set of data corresponding to a wavelength X₂ between about 360 nm and about 720 nm.

17. The method of claim 16 wherein X₁ is a wavelength between about 489 nm and 509 nm and X₂ is a wavelength between about 533 nm and about 553 nm.

18. The method of claim 13, wherein said determining step further comprises comparing the quotient {(R₁(X₁)/R₂(X₁))/(R₁(X₂)/R₂(X₂))} to a second constant, where X₁ is a wavelength between about 360 nm and about 720 nm, X₂ is a wavelength between about 360 nm and about 720 nm, R₁(X₁) is a member of said first set of data corresponding to said wavelength X₁, R₂(X₁) is a member of said second set of data corresponding to said wavelength X₁, R₁(X₂) is a member of said first set of data corresponding to said wavelength X₂, R₂(X₂) is a member of said second set of data corresponding to said wavelength X₂.

19. The method of claim 18 wherein X₁ is a wavelength between about 566 nm and about 586 nm, and X₂ is a wavelength between about 589 nm and about 609 nm.

20. The method of claim 19, wherein said determining step further comprises comparing R₁(X₃) to a third constant, where R₁(X₃) is a member of said first set of data corresponding to a wavelength X₃ between about 689 and about 709 nm.

21. The method of claim 13, wherein said determining step further comprises comparing a value Q to a second constant, where Q is an approximate slope of a plot of {R₁(X_i)/R₂(X_i)} with respect to wavelength, over a subset of a wavelength range of about 360 nm to about 720 nm.

22. The method of claim 15, wherein said determining step further comprises comparing a value Q to a second constant, where said value Q is an approximate slope of a plot of {R₁(X_i)/R₂(X_i)} with respect to wavelength, over a subset of a wavelength range of about 576 nm to about 599 nm.

23. The method of claim 13, wherein said determining step further comprises comparing R₁(X₁) to a second constant and comparing R₁(X₁) to R₂(X₁), where R₁(X₁) is a member of said first set of data corresponding to a wavelength X₁ between about 360 nm and about 720 nm, and R₂(X₁) is a member of said second set of data corresponding to said wavelength X₁.

24. The method of claim 13, wherein said determining step further comprises comparing R₁(X₁) to a second constant and comparing R₁(X₁) to R₂(X₁), where R₁(X₁) is a member of said first set of data corresponding to a wavelength X₁ between about 489 nm and about 509 nm, and R₂(X₁) is a member of said second set of data corresponding to said wavelength X₁.

25. The method of claim 18, wherein said determining step further comprises comparing R₁(X₃) to a third constant, where R₁(X₃) is a member of said first set of data corresponding to a wavelength X₃ between about 360 nm and about 720 nm.

26. The method of claim 18, wherein said determining step further comprises comparing R₁(X₃) to a third constant, where R₁(X₃) is a member of said first set of data corresponding to a wavelength X₃ between about 409 nm and about 429 nm.

27. The method of claim 8, wherein said determining step comprises comparing R₁ to a first constant and comparing R₂ to a second constant, where R₁ is a member of said first set of data corresponding to a wavelength between about 489 nm and about 509 nm and R₂ is a member of said second set of data corresponding to a wavelength between about 489 nm and about 509 nm.

28. The method of claim 8, wherein said artifact comprises a lighting artifact.

29. The method of claim 28, wherein said lighting artifact comprises glare.

30. The method of claim 28, wherein said lighting artifact comprises shadow.

31. The method of claim 8, wherein said artifact comprises an obstruction.

32. The method of claim 31, wherein said obstruction comprises blood.

33. The method of claim 31, wherein said obstruction comprises a portion of at least one of a group consisting of a speculum and a smoke tube.

34. The method of claim 31, wherein said obstruction comprises mucus.

35. The method of claim 8, wherein said tissue sample comprises cervical tissue.

36. The method of claim 8, wherein said tissue sample comprises epithelial cells.

37. The method of claim 8, wherein said tissue sample comprises at least one of a group consisting of colorectal, gastroesophageal, urinary bladder, lung, and skin tissue.

38. A method of determining whether spectral data corresponding to a region of a tissue sample is affected by an artifact, said method comprising the steps of:

obtaining a first set of reflectance spectral data corresponding to a region of a tissue sample using light incident to said region at a first angle;

obtaining a second set of reflectance spectral data corresponding to said region using light incident to said region at a second angle;

obtaining a set of fluorescence spectral data corresponding to said region; and

determining whether any of said first set of reflectance spectral data, said second set of reflectance spectral data and said set of fluorescence spectral data are affected by an artifact based at least in part on at least one of the following;

a subset of said first set of reflectance spectral data, a subset of said second set of reflectance spectral data, and a subset of said set of fluorescence spectral data.

39. The method of claim 38, wherein said determining step comprises comparing F to a constant, where F is a member of said set of fluorescence spectral data corresponding to a wavelength between about 469 nm and about 489 nm.

40. A method of determining a spectral characteristic of an artifact, said method comprising the steps of:

(a) at each of a first plurality of regions of tissue, obtaining a first set of reflectance spectral data affected by a known artifact;

(b) at each of a second plurality of regions of tissue, obtaining a second set of reflectance spectral data not affected by said known artifact; and

(c) determining a spectral characteristic of said known artifact based at least in part on said first set of spectral data and said second set of spectral data.

41. The method of claim 40, wherein said determining step comprises locating a wavelength at which there is a maximum difference between a mean of one or more members of said first set corresponding to said wavelength and a mean of one or more members of said second set corresponding to said wavelength, relative to a variation measure.

42. The method of claim 40, wherein said determining step comprises computing N differences, |μ

_i(A_j(X₁))−

μ

_i(B_k(X_i))|, and defining a maximum of a subset of said N differences, where i=1 to N, N is an integer, X₁ is a wavelength between about 360 nm and about 720 nm, j=1 to M1, M1 is an integer, A_j(X_i) represents one of M1 members of said first set of reflectance spectral data corresponding to said wavelength X_i, k=1 to M2, M2 is an integer, B_k(X_i) represents one of M2 members of said second set of reflectance spectral data corresponding to said wavelength X_i, μ

_i(A_j(X_i)) is a mean of said M1 members of said first set of data corresponding to said wavelength X_i, and μ

_i (B_k(X_i)) is a mean of said M2 members of said second set of data corresponding to said wavelength X_i.

43. The method of claim 40, wherein said determining step comprises computing N quotients, [|μ

_i(A_j(X_i))−

μ

_i(B_k(X_i))|/{σ

_i²(A_j(X_i))+σ

_i²(B_k(X_i))}^0.5], and defining a maximum of a subset of said N quotients, where i=1 to N, N is an integer, X_i is a wavelength between about 360 nm and about 720 nm, j=1 to M1, M1 is an integer, A_j(X_i) represents one of M1 members of said first set of reflectance spectral data corresponding to said wavelength X_i, k=1 to M2, M2 is an integer, B_k(X_i) represents one of M2 members of said second set of reflectance spectral data corresponding to said wavelength X_i, μ

(A_j(X_i)) is a mean of said M1 members of said first set of data corresponding to said wavelength X_i, μ

_i(B_k(X_i)) is a mean of said M2 members of said second set of data corresponding to said wavelength X_i, σ

_i(A_j(X_i)) represents a standard deviation of said M1 members of said first set of data corresponding to said wavelength X_i, and σ

_i(B_k(X_i)) represents a standard deviation of said M2 members of said second set of data corresponding to said wavelength X_i.

44. The method of claim 40, wherein said determining step comprises computing N quotients, [|μ

_i(A_j(X1_i)/A_j(X2_i))−

μ

_i(B_k(X1_i)/B_k(X2_i))|/{σ

_i(A_j(X1_i)/A_j(X2_i))+σ

_i²(B_k(X1_i)/B_k(X2_i))}⁵], and defining a maximum of a subset of said N quotients, where i=1 to N, N is an integer, X1_i is a wavelength between about 360 nm and about 720 nm, X2_i is a wavelength between about 360 nm and about 720 nm, j=1 to M1, M1 is an integer, A_j(X1_i) represents one of M1 members of said first set of reflectance spectral data corresponding to said wavelength X1_i, A_j(X2_i) represents one of M1 members of said first set of reflectance spectral data corresponding to said wavelength X2_i, k=1 to M2, M2 is an integer, B_k(X1_i) represents one of M2 members of said second set of reflectance spectral data corresponding to said wavelength X1_i, B_k(X2_i) represents one of M2 members of said second set of reflectance spectral data corresponding to said wavelength X2_i, μ

_i(A_j(X1_i)/A_j(X2_i)) is a mean of M1 quotients A_j(X1_i)/A_j(X2_i) for j=1 to M1, μ

_i(B_k(X1_i)/B_k(X2_i)) is a mean of M2 quotients B_k(X1_i)/B_k(X2_i) for k=1 to M2, σ

_i(A_j(X1_i)/A_j(X2_i)) represents a standard deviation of said M1 quotients A_j(X1_i)/A_j(X2_i), and σ

_i(B_k(X1_i)/B_k(X2_i)) represents a standard deviation of said M2 quotients B_k(X1_i,)B_k(X2_i).

45. A method of determining a characteristic of a region of a tissue sample, said method comprising the steps of:

(a) obtaining a first set of reflectance spectral data corresponding to a region of a tissue sample using light incident to said region at a first angle;

(b) obtaining a second set of reflectance spectral data corresponding to said region using light incident to said region at a second angle;

(c) determining whether at least one of said first set of reflectance data and said second set of reflectance data is affected by an artifact based at least in part on a subset of said first set of reflectance data and a subset of said second set of reflectance data;

(d) rejecting at least one member of at least one of said first set of reflectance data and said second set of reflectance data determined in step (c) to be affected by said artifact; and

(e) determining a characteristic of said region of said tissue sample based at least in part on at least one member of at least one of said first set of reflectance data and said second set of reflectance data not rejected in step (d).

46. The method of claim 45, further comprising obtaining a set of fluorescence spectral data corresponding to said region, and wherein step (e) comprises determining said condition of said region of said tissue sample based at least in part on at least one member of at least one of said first set of reflectance data and said second set of reflectance data and at least one member of said set of fluorescence spectral data.