Classification system for sex determination and tissue characterization

• US 6,493,566 B1
• Filed: 01/19/2000
• Issued: 12/10/2002
• Est. Priority Date: 01/22/1999
• Status: Expired due to Fees

• Current Assignee: GLT Acquisition Corporation
• Original Assignee: Instrumentation Metrics Inc.
• Inventors: Ruchti, Timothy L., Thennadil, Suresh, Malin, Stephen F., Kees, Glenn Aaron, Rennert, Jessica
• Primary Examiner(s): Winakur, Eric F.
• Assistant Examiner(s): KREMER, MATTHEW J
• Application Number:
  US09/487,733
• Time in Patent Office: 1,056 Days
• Field of Search: 209/576-590, 600/300, 600/309, 600/310, 600/322, 356/39-41, 250/339-1-3, 250/339.12
• US Class Current: 600/310
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• 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 5/0075 by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy A61B5/0071 takes precedence
• CPC Class Codes: A61B 5/1075 for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer A61B8/0858 takes precedence
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/14546 for measuring analytes not otherwise provided for, e.g. ions, cytochromes
• CPC Class Codes: A61B 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: A61B 5/441 Skin evaluation, e.g. for skin disorder diagnosis
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• 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 21/359 using near infrared light
• CPC Class Codes: G01N 21/4785 Standardising light scatter apparatus; Standards therefor
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Instrumentation and procedures for noninvasively determining the sex of human and animal subjects in vivo have been developed based on the irradiation of skin tissue with near infrared light. The method of sex determination provides additional information about primary sources of systematic tissue variability, namely, the thickness of the dermis and the subcutaneous fat. Categorization of subjects on the basis of the determination is therefore suitable for further spectral analysis and the measurement of biological and chemical compounds, such as blood analytes.

Claims

1. A method for determining the sex of animal and/or human subjects based on NIR measurements of a skin tissue, comprising the steps of:

measuring tissue properties and characteristics that are manifested spectrally and that vary systematically according to said subject'"'"'s sex;

applying a calibration model that is empirically derived from a set of exemplary samples consisting of tissue measurements and the actual sex of a population of subjects; and

determining said subject'"'"'s sex.

2. The method of claim 1, wherein said measuring step comprises the step of:

using a spectroscopic apparatus in conjunction with an optical interface.

3. The method of claim 1, wherein said measurements are made in a spectrum denoted by the vector mε

^N of absorbance values pertaining to a set of N wavelengths λ

ε

^N that span the near infrared (700 to 2500 nm).

4. The method of claim 1, wherein said method further comprises the step of:

determining tissue properties of a sampled tissue volume.

5. The method of claim 4, wherein said properties comprise any of thickness of the dermis, collagen content, skin density, and the amount of subcutaneous fat at a measurement site.

6. The method of claim 1, wherein said determining step further comprises any of the steps of:

performing an outlier detection step to detect invalid measurements through spectral variations that result from any of the following problems in the instrument, poor sampling of the subject, and a subject outside the calibration set;

performing a preprocessing step;

performing a feature extraction step to represent concisely and enhance the properties and characteristics of a tissue measurement site for sex determination; and

performing a classification step.

7. The method of claim 6, wherein said outlier detection step comprises the step of:

performing a principal components analysis and an analysis of the residuals.

8. The method of claim 6, wherein said preprocessing step comprises the step of:

performing transformations that attenuate the noise and instrumental variation without affecting the signal of interest.

9. The method of claim 8, wherein said step of performing transformations comprises any of the steps of performing scaling, normalization, smoothing, derivatives and filtering.

10. The method of claim 6, wherein said feature extraction step comprises the step of:

applying a mathematical transformation to enhance a quality and/or aspect of a sample measurement for interpretation.

11. The method of claim 10, wherein said mathematical transformation comprises any of:

a scores from factor analysis;

location of critical points on a measured spectrum; and

a determination of relative absorption of water and fat.

12. The method of claim 6, wherein said classification step comprises the steps of:

performing a mapping; and

making a decision.

13. The method of claim 12, wherein mapping and decision limits are determined from a calibration set of exemplary features and corresponding sexes through a classification calibration procedure.

14. The method of claim 13, wherein said calibration procedure comprises any of linear discriminant analysis, k nearest-neighbor, fuzzy classification, and the use of artificial neural networks.

15. A method for determining the sex of animal and/or human subjects based on measurements of a skin tissue, comprising the steps of:

measuring an NIR spectrum of said subject'"'"'s skin tissue; and

predicting sex based on spectral shift observed in said measured NIR spectrum;

wherein systematic sex related differences in characteristics and properties of said subject'"'"'s subcutaneous fat and dermis layers causes systematic variation of absorbance bands of specific analytes while other background analytes tend to remain constant; and

wherein variation in one absorbance band among a particular background is manifested through an apparent shift in peak and valley locations.

16. The method of claim 15, further comprising the step of:

preprocessing said measured spectrum via a 15-point Savisky-Golay smoother in the form of a finite impulse response filter.

17. The method of claim 16, further comprising the step of:

measuring an apparent shift at a critical points with reference to a processed standard absorbance spectrum.

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

using a statistical model to identify critical points that are applied using a decision rule, wherein a location of said critical points is determined over a finite window in the vicinity of known critical points and wherein locations for critical points that vary according to subject sex.

19. The method of claim 18, further comprising the steps of:

determining a wavelength position of each critical point by fitting a second-order polynomial function to the measured spectrum in the vicinity of said known critical point locations;

determining a derivative of said polynomial; and

calculating a root.

20. The method of claim 19, further comprising the step of:

extracting spectral features.

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

using linear discriminant analysis or Mahalanobis distance to determine a classification model.

23. The method of claim 21, further comprising the steps of:

sub-dividing said spectrum into one or more regions according to wavelength (wavelength selection);

preprocessing said regions; and

normalizing said regions to enhance spectral variation related to sex.

24. The method of claim 23, further comprising the step of:

projecting measurements of said regions onto one or more sets of previously determined factors to determine scores, wherein said scores are extracted features.

25. The method of claim 24, further comprising the step of:

subjecting said scores to a classification procedure that, optionally, comprises any of linear discriminant analysis, SIMCA, k nearest-neighbor, and a form of artificial neural networks, to predict the sex of the subject.

26. The method of claim 25, further comprising the steps of:

spectral preprocessing;

decomposition through principal components analysis; and

classification through linear discriminant analysis.

22. A method for determining the sex of animal and/or human subjects based on measurements of a skin tissue, comprising the steps of:

measuring an NIR spectrum of said subject'"'"'s skin tissue; and

using factor analysis to develop a set of abstract features capable of representing the spectral variation related to sex.

27. A method for determining the sex of animal and/or human subjects based on measurements of a skin tissue, comprising the steps of:

measuring an NIR spectrum of said subject'"'"'s skin tissue; and

projecting a basis set of pure component spectra onto a measured noninvasive absorbance spectrum to generate key features.

28. The method of claim 27, further comprising the step of:

preprocessing and wavelength selection of said measured spectrum.

29. The method of claim 27, wherein said basis set is comprised of key constituents which are related to the systematic spectral variation caused by the sex of a subject.

30. The method of claim 27, further comprising the step of:

applying a discriminant function to classify subjects.

31. A method for determining the sex of animal and/or human subjects based on measurements of a skin tissue, comprising the steps of:

measuring an NIR spectrum of said subject'"'"'s skin tissue; and

using either of a general linear or nonlinear mapping to directly determine the sex of a subject.

32. The method of claim 31, further comprising the step of:

subjecting said NIR spectral measurement to preprocessing and normalization to reduce background interference and enhance variation related to sex.

33. An apparatus for determining the sex of animal and/or human subjects based on NIR measurements of skin tissue, comprising:

means for measuring tissue properties and characteristics that are manifested spectrally and that vary systematically according to said subject'"'"'s sex;

means for applying a calibration model that is empirically derived from a set of exemplary samples consisting of tissue measurements and the actual sex of a population of subjects; and

means for determining said subject'"'"'s sex.