Method for generating a net analyte signal calibration model and uses thereof

• US 20060167348A1
• Filed: 01/24/2005
• Published: 07/27/2006
• Est. Priority Date: 01/24/2005
• Status: Active Grant

• Current Assignee: University of Iowa Research Foundation
• Original Assignee: University of Iowa Research Foundation
• Inventors: Arnold, Mark A., Olesberg, Jonathon T.
• Granted Patent: US 7,460,895 B2
• Time in Patent Office: Days
• US Class Current: 600/310
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: A61B 5/1495 Calibrating or testing of in-vivo probes
• CPC Class Codes: G01N 2021/3595 using FTIR
• CPC Class Codes: G01N 21/1702 with opto-acoustic detection, e.g. for gases or analysing solids
• CPC Class Codes: G01N 21/274 Calibration, base line adjustment, drift correction
• CPC Class Codes: G01N 21/35 using infrared light G01N21/39 takes precedence
• CPC Class Codes: G01N 21/4738 Diffuse reflection precedence is given to G01N21/55 - G01N21/57 if specular component is taken into consideration, e.g. also for testing fluids, fibrous materials
• CPC Class Codes: G01N 21/552 Attenuated total reflection
• CPC Class Codes: G01N 2201/129 Using chemometrical methods

Abstract

A method for generating a net analyte signal calibration model for use in detecting and/or quantifying the amount of an analyte in a test subject. The net analyte signal can be generated by providing a set of in vivo infrared spectra for a test subject during a period in which an analyte concentration is essentially constant; calculating an optimal subspace of spectra that at least substantially describes all non-analyte dependent spectral variance in the in vivo spectra; providing a pure component infrared spectrum for the analyte; and calculating a net analyte signal spectrum from a data set comprising the optimal subspace spectra and the pure analyte spectrum. The net analyte signal calibration model can be used, for example, in measuring the concentration of analyte in a test subject, and/or for evaluating the analytical significance of an in vivo multivariate calibration model.

Claims

1. A method for generating a net analyte signal calibration model for use in detecting an analyte in a test subject, comprising:

a) providing a set of in vivo infrared spectra for a test subject during a period in which an analyte concentration is essentially constant;

b) calculating an optimal subspace of spectra that at least substantially describes all non-analyte dependent spectral variance in the in vivo spectra of step a);

c) providing a pure component infrared spectrum for the analyte; and

d) calculating a net analyte signal spectrum from a data set comprising the optimal subspace spectra of b) and the pure analyte spectrum of c),

wherein the net analyte signal spectrum identifies one or more in vivo spectral features specific to the analyte.

2. The method of claim 1, wherein the spectra are absorption spectra.

3. The method of claim 1, wherein the spectra are reflectance spectra.

4. The method of claim 1, wherein the spectra are single-beam spectra.

5. The method of claim 2, wherein the absorption spectra are near infrared absorption spectra in the range of from approximately 4000 cm⁻

1 to approximately 5000 cm<sup>−

1</sup>.

6. The method of claim 2, wherein the absorption spectra are near infrared absorption spectra in the range of from approximately 5500 cm⁻

1 to approximately 6500 cm<sup>−

1</sup>.

7. The method of claim 1, wherein the infrared spectra are absorption spectra in the mid infrared spectral range.

8. The method of claim 7, wherein the absorption spectra are in the range of from approximately 1200 cm⁻

1 to approximately 900 cm<sup>−

1</sup>.

9. The method of claim 1, wherein the analyte is a physiological chemical.

10. The method of claim 9, wherein the analyte is glucose, urea, lactate, triglyceride, total protein, cholesterol, or ethanol.

11. The method of claim 9, wherein the physiological chemical comprises at least one C—

H, N—

H, or O—

H molecular bond.

12. The method of claim 9, wherein the analyte is glucose.

13. The method of claim 1, wherein the test subject is a living organism.

14. The method of claim 13, wherein the test subject is a plant.

15. The method of claim 13, wherein the test subject is an animal.

16. The method of claim 15, wherein the animal is non-mammalian.

17. The method of claim 15, wherein the animal is mammalian.

18. The method of claim 13, wherein the test subject is a microbial species.

19. The method of claim 13, wherein the test subject is a human.

20. The method of claim 1, wherein step b) comprises a principle component analysis.

21. The method of claim 1, further comprising reporting the net analyte signal calibration spectrum on a display device.

22. The method of claim 1, further comprising storing the net analyte signal spectrum on a recordable medium.

24. The method of claim 23, wherein the in vivo net analyte signal calibration model of b) is generated by the method of claim 1.

35. The method of claim 34, wherein the in vivo net analyte signal calibration vector of b) is generated by the method of claim 1.

23. A method for non-invasively measuring the concentration of an analyte in a test subject, comprising:

a) identifying a test subject in need of having an analyte concentration measured;

b) providing an in vivo net analyte signal calibration model for the test subject;

c) providing an in vivo infrared spectrum of the test subject; and

d) calculating a predicted concentration of the analyte in the test subject from a data set comprising the net analyte signal calibration model and the in vivo infrared spectrum of the test subject.

25. The method of claim 23, wherein the analyte is a physiological chemical.

26. The method of claim 25, wherein the analyte is glucose, urea, lactate, triglyceride, total protein, cholesterol, or ethanol.

27. The method of claim 26, wherein the analyte is glucose.

28. The method of claim 23, wherein the test subject is any living organism.

29. The method of claim 28, wherein the test subject is a plant.

30. The method of claim 28, wherein the test subject is a mammal.

31. The method of claim 28, wherein the test subject is a human.

32. The method of claim 23, further comprising reporting the predicted concentration on a display device.

33. The method of claim 23, further comprising storing the predicted concentration on a recordable medium

34. A method for evaluating the analytical significance of an in vivo multivariate calibration model, comprising:

a) providing an in vivo multivariate calibration vector for an analyte in a test subject;

b) providing an in vivo net analyte signal calibration vector for the test subject; and

c) comparing the in vivo multivariate calibration vector to the in vivo net analyte signal calibration vector for an analytically significant similarity in at least one spectral feature.

36. The method of claim 34, wherein the multivariate calibration vector is generated by a partial least squares regression method.

37. The method of claim 34, wherein the multivariate calibration vector is generated by a principle component regression method.

38. The method of claim 34, wherein the analyte is a physiological chemical.

39. The method of claim 38, wherein the analyte is glucose, urea, lactate, triglyceride, total protein, cholesterol, or ethanol.

40. The method of claim 39, wherein the analyte is glucose.

41. The method of claim 34, wherein the test subject is any living organism.

42. The method of claim 41, wherein the test subject is a plant.

43. The method of claim 41, wherein the test subject is a mammal.

44. The method of claim 41, wherein the test subject is a human.

45. The method of claim 34, further comprising quantifying the analytical significance of the similarity in the least one spectral feature.

46. The method of claim 45, further comprising reporting the quantified analytical significance on a display device.

47. The method of claim 45, further comprising storing the quantified analytical significance on a recordable medium.