Methods and apparatus for characterization of tissue samples

• US 7,309,867 B2
• Filed: 04/18/2003
• Issued: 12/18/2007
• Est. Priority Date: 04/18/2003
• Status: Expired due to Fees

• Current Assignee: Luma Imaging Corp. (SpectraScience, Inc.)
• Original Assignee: MediSpectra, Inc. (SpectraScience, Inc.)
• Inventors: Costa, Peter J., Flewelling, Ross F., Schomacker, Kevin T., Schott, Jean-Pierre, Draayer, Bret F., Sum, Stephen T.
• Primary Examiner(s): Porta, David
• Assistant Examiner(s): Sung, Christine
• Application Number:
  US10/418,668
• Publication Number:
  US 20040206913A1
• Time in Patent Office: 1,705 Days
• Field of Search: 250/458.1, 250/461.1
• US Class Current: 250/458.1
• 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: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: A61B 5/7267 involving training the classification device
• 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: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

The invention provides methods for determining the probability that a given region of a tissue sample contains tissue of a given category, such as CIN 1 (cervical intraepithelial neoplasia, grade 1), CIN 2/3 (cervical intraepithelial neoplasia grades 2 and/or 3), normal squamous, normal columnar, and metaplasia, for example. The invention provides increased diagnostic accuracy by combining a plurality of statistical classification techniques. Furthermore, in one embodiment, the invention comprises combining one or more statistical techniques with one or more non-statistical classification techniques.

Claims

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

(a) illuminating a region of tissue and obtaining reflectance spectral data from said region;

(b) processing said reflectance spectral data to determine a probability that said region comprises tissue within a predefined tissue class, said probability determined using training data indicative of said predefined tissue class;

(c) weighting said probability according to a likelihood that spectral data obtained from said region are affected by an obstruction; and

(d) determining a condition of said region using said weighted probability.

2. The method of claim 1, wherein said predefined tissue class is a member selected from the group consisting of CIN 1, CIN 2, CIN 3, CIN 2/3, NED, normal squamous, normal columnar, metaplasia, and cancer.

3. The method of claim 1, wherein said condition is a member selected from the group consisting of CIN 2/3, NED, indeterminate, and necrotic.

4. The method of claim 1, wherein said probability in step (b) is an overall probability determined using a first measure of probability and a second measure of probability.

5. The method of claim 4, wherein said first measure of probability is determined using a principal component analysis method and wherein said second measure of probability is determined using a feature coordinate extraction method.

6. The method of claim 1, wherein step (c) further comprises weighting said probability according to a likelihood that said region lies within a zone of interest.

7. The method of claim 1, further comprising the step of displaying a representation of said weighted probability.

8. A method of determining a condition of a region of tissue, the method comprising the steps of:

(a) illuminating a region of tissue and obtaining reflectance spectral data from said region;

(b) for each of a plurality of predefined tissue classes, processing said reflectance spectral data to determine a probability that said region comprises tissue within said class, said probability determined using training data indicative of said class;

(c) weighting at least one of said probabilities according to a likelihood that spectral data obtained from said region are affected by an obstruction; and

(d) determining a condition of said region using said probabilities.

9. The method of claim 8, further comprising the step of displaying a representation of at least one of said probabilities.

10. A method of characterizing a condition of a region of tissue, the method comprising the steps of:

(a) illuminating a region of tissue and obtaining spectral data from said region;

(b) processing said spectral data to determine, for each member of a plurality of predefined tissue classes, a probability that said region comprises tissue within said member, said probability determined using training data indicative of said member;

(c) weighting at least one of said probabilities according to a likelihood that spectral data obtained from said region are affected by an obstruction;

(d) evaluating a classification metric using spectral data obtained from said region;

(e) if said classification metric is satisfied, characterizing a condition of said region according to said classification metric; and

(f) if said classification metric is not satisfied, characterizing a condition of said region according to said probabilities.

11. The method of claim 10, wherein said evaluating step comprises using fluorescence spectral data.

12. The method of claim 10, wherein said processing step comprises processing reflectance spectral data.

13. The method of claim 10, wherein said evaluating step comprises using fluorescence spectral data and said processing step comprises processing reflectance spectral data.

14. The method of claim 10, wherein said processing step comprises applying one or more statistical methods to a set of reflectance spectral data obtained from said tissue.

15. The method of claim 10, wherein said classification metric comprises a non-statistically-based component.

16. The method of claim 15, wherein said non-statistically-based component is indicative of a substance present in tissue within at least one of said predefined tissue classes.

17. The method of claim 16, wherein said substance is selected from the group consisting of collagen, porphyrin, FAD, and NADH.

18. The method of claim 10, wherein said classification metric comprises one or more statistically-based components and one or more non-statistically-based components.

19. The method of claim 10, further comprising the step of displaying a representation of at least one of said probabilities.

20. A system for characterizing one or more regions of a tissue, the system comprising:

(a) an optical device adapted to illuminate one or more regions of a tissue and obtain reflectance spectral data from said one or more regions;

(b) a memory that stores code;

(c) a processor that executes said code to;

(i) process reflectance spectral data obtained from a region of said tissue by said optical device to determine a probability that said region comprises tissue within a predefined tissue class, said probability determined using training data indicative of said predefined tissue class; and

(ii) weight said probability according to a likelihood that spectral data obtained from said region are affected by an obstruction.

21. The system of claim 20, wherein said predefined tissue class is a member selected from the group consisting of CIN 1, CIN 2, CIN 3, CIN 2/3, NED, normal squamous, normal columnar, metaplasia, and cancer.

22. The system of claim 20, wherein said probability in step (i) is an overall probability determined using a first measure of probability and a second measure of probability.

23. The system of claim 22, wherein said first measure of probability is determined using a principal component analysis method and wherein said second measure of probability is determined using a feature coordinate extraction method.

24. The system of claim 20, wherein step (ii) further comprises weighting said probability according to a likelihood that said region lies within a zone of interest.

25. The system of claim 20, further comprising a display for displaying a representation of said weighted probability.

26. A method of determining the condition of a region of tissue, the method comprising:

(a) for each of one or more predefined tissue classes, processing reflectance spectral data obtained from a region of tissue to determine a first measure of probability and a second measure of probability that said region comprises tissue within said class, said first and second measures of probability determined using training data indicative of said class; and

(b) using said first and second measure of probability to determine and overall probability that said region comprises tissue within said class, said overall probability weighted according to a likelihood that spectra data obtained from said regions are affected by an obstruction.

27. The method of claim 26, wherein said first measure of probability is determined using a principal component analysis method, and wherein said second measure of probability is determined using a feature coordinate extraction method.

28. The method of claim 27, wherein said first measure of probability is determined by transforming said reflectance spectral data into a lower dimensional feature space along orthogonal directions of maximal variance, and wherein said second measure of probability is determined by transforming said reflectance spectral data into a lower dimensional feature space along orthogonal directions of maximal discrimination.

29. The method of claim 26, wherein said one or more tissue classes comprises at least one member selected from the group consisting of CIN 1, CIN 2, CIN 3, CIN 2/3, NED, normal squamous, normal columnar, metaplasia, cancer, indeterminate and necrotic.

30. The method of claim 26, wherein said one ore more tissue classes comprises CIN 2/3.

31. The method of claim 26, wherein step (c) comprises determining an overall probability that said region comprises tissue within said class, said overall probability weighted according to a likelihood that said region lies within a zone of interest.

32. The method of claim 26, further comprising the step of displaying a representation of said weighted probability.

33. A system for determining the condition of one or more regions of a tissue, the system comprising:

(a) an optical device adapted to illuminate one or more regions of a tissue and obtain reflectance spectral data from said one or more regions;

(b) a memory that stores code;

(c) a processor that executes said code to, for each of one or more predefined tissue classes;

(i) process reflectance spectral data obtained from a region of tissue to determine a first measure of probability and a second measure of probability that said region comprises tissue within said class, said first and second measures of probability determined using training data indicative of said class; and

(ii) use said first and said second measures of probability to determine an overall probability that said region comprises tissue within said class, said overall probability weighted according to a likelihood that spectral data obtained from said region are affected by an obstruction.

34. The system of claim 33, wherein said first measure of probability is determined using a principal component analysis method, and wherein said second measure of probability is determined using a feature coordinate extraction method.

35. The system of claim 34, wherein said first measure of probability is determined by transforming said reflectance spectral data into a lower dimensional feature space along orthogonal directions of maximal variance, and wherein said second measure of probability is determined by transforming said reflectance spectral data into a lower dimensional feature space along orthogonal directions of maximal discrimination.

36. The system of claim 33, wherein said one or more tissue classes comprises at least one member selected from the group consisting of CIN 1, CIN 2, CIN 3, CIN 2/3, NED, normal squamous, normal columnar, metaplasia, cancer, indeterminate, and necrotic.

37. The system of claim 33, wherein said one or more tissue classes comprises CIN 2/3.

38. The system of claim 33, wherein said overall probability is weighted according to a likelihood that said region lies within a zone of interest.

39. The system of claim 33, further comprising a display for displaying a representation of said weighted probability.