Indirect measurement of tissue analytes through tissue properties

• US 7,039,446 B2
• Filed: 01/22/2003
• Issued: 05/02/2006
• Est. Priority Date: 01/26/2001
• Status: Expired due to Term

• Current Assignee: GLT Acquisition Corporation, Regina Miracle International Limited
• Original Assignee: Sensys Medical, Inc. (Sensys Medical Limited)
• Inventors: Ruchti, Timothy L., Monfre, Stephen L., Hazen, Kevin H., Blank, Thomas B., Lorenz, Alexander D., Thennadil, Suresh N.
• Primary Examiner(s): WINAKUR, ERIC FRANK
• Application Number:
  US10/349,573
• Publication Number:
  US 20040127777A1
• Time in Patent Office: 1,196 Days
• Field of Search: 600/310, 600/316, 600/317, 600/322, 600/323, 600/407, 600/437, 600/438, 600/473, 600/476
• US Class Current: 600/310
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• CPC Class Codes: A61B 5/0071 by measuring fluorescence emission
• 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 5/14532 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 21/274 Calibration, base line adjustment, drift correction
• 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 21/33 using ultraviolet light G01N21/39 takes precedence
• CPC Class Codes: G01N 21/359 using near infrared light
• CPC Class Codes: G01N 21/65 Raman scattering

Abstract

Methods and system for noninvasive determination of tissue analytes utilize tissue properties as reflected in key features of an analytical signal to improve measurement accuracy and precision. Physiological conditions such as changes in water distribution among tissue compartments lead to complex alterations in the measured analytical signal of skin, leading to a biased noninvasive analyte measurement. Changes in the tissue properties are detected by identifying key features in the analytical signal responsive to physiological variations. Conditions not conducive to the noninvasive measurement are detected. Noninvasive measurements that are biased by physiological changes in tissue are compensated. In an alternate embodiment, the analyte is measured indirectly based on natural physiological response of tissue to changes in analyte concentration. A system capable of such measurements is provided.

Claims

1. A method for noninvasive measurement of a target analyte in a tissue, comprising the steps of:

collecting an analytical signal from the tissue, said collected signal comprising a tissue measurement;

extracting at least one spectral feature from the analytical signal indicative of the effect of the target analyte on the probed tissue; and

either correcting a direct analyte measurement based on said at least one feature;

or

calculating concentration of said analyte indirectly by application of a calibration model to said at least one feature.

2. The method of claim 1, wherein said analytical signal is generated using any of:

fluorescence spectroscopy;

ultraviolet spectroscopy from 200 to 400 nm;

visible spectroscopy from 400 to 700 nm;

infrared and Fourier transform infrared (FTIR) spectroscopy;

passive infrared spectroscopy;

mid infrared spectroscopy from 4000–

700 cm<sup>−

1</sup> in any of diffuse reflectance and transmission;

attenuated total reflectance (ATR) spectroscopy;

far infrared radiation spectroscopy;

Kromoscopy in reflection or transmission;

emission spectroscopy;

Raman spectroscopy;

photoacoustic and pulse laser photoacoustic spectroscopy;

photon scattering from 400 to 2500 nm;

bioelectrical response spectroscopy;

oscillating thermal gradient spectroscopy;

polarimetry;

ultrasound spectroscopy;

near infrared spectroscopy from 700 to 2500 nm in any of diffuse reflectance, transflectance, and transmission mode; and

radio wave spectroscopy.

3. The method of claim 2, further comprising the step of:

preprocessing said tissue measurement.

4. The method of claim 3, wherein said step of preprocessing said tissue measurement comprises any of the steps of:

correcting said signal utilizing a reference;

filtering said signal;

calculating any of a first and second derivative of said signal;

normalizing said signal;

selecting portions of said signal;

scatter correcting said signal; and

translating said signal.

5. The method of claim 3, further comprising the step of:

determining difference between a tissue template and either the preprocessed tissue measurement or the at least one extracted feature according to;

z=x−

(cx_t+d);

wherein x comprises either the pre-processed measurement or a set of at least one extracted features, x_t comprises a tissue template associated with a measurement period, and c and d are slope and intercept adjustments to the tissue template.

6. The method of claim 5, wherein said tissue template is determined through one or more tissue measurements combined according to a predetermined data selection criterion during each measurement period.

7. The method of claim 6, wherein a measurement period comprises a time period over which accuracy of the tissue measurement remains within desired specifications.

8. The method of claim 6, further comprising the step of:

providing an associated set of reference values combined according to said predetermined data selection criterion to form a measurement bias adjustment.

9. The method of claim 5, wherein the tissue template comprises any set of features from a given subject or calibration set that future tissue measurements will be compared with, wherein c and d are determined through least-square fit of the tissue template over a particular wavelength range to the tissue measurement.

10. The method of claim 5, wherein the step of correcting a direct analyte measurement based on said at least one feature comprises:

supplementing a second calibration model based on direct effect of glucose on said analytical signal with said at least one selected feature according to;

ŷ

=f(x_p,z)+b;

where ŷ

is an estimated analyte concentration, x_pε

^N is a processed tissue measurement, zε

^M is a set of features representative of the physiological state or optical properties of the tissue, f;

^N,M→

¹ is a model used to measure the analyte on the basis of a preprocessed measurement and at least one extracted feature, and b is a baseline adjustment for analyte measurement associated with both a tissue template and said second calibration model.

11. The method of claim 10, wherein measurement site comprises any of:

a finger;

palmar region;

hand;

forearm;

upper arm;

eye;

earlobe;

torso;

abdominal region;

leg;

plantar region;

a foot; and

toes.

12. The method of claim 5, wherein the step of correcting a direct analyte measurement based on said at least one feature comprises:

supplementing a second calibration model based on direct effect of glucose on said analytical signal with said at least one selected feature according to;

ŷ

=f(x_p)−

(m,g(z)+m_i)+b;

where ŷ

is an estimated analyte concentration, x_pε

^N is a processed tissue measurement, zε

^M is a set of features representative of any of the structural, chemical, physiological and optical properties of the tissue, wherein x_p and z are independent, where f;

^N→

¹ is a model used to measure the analyte in the absence of physiological or other tissue variation, g;

^M→

¹ is a model used to map said at least one feature to a variable correlated to error in analyte measurement caused by a change in the properties of the tissue, m_s and m_i are slope and intercepts used to convert g(z) to correct units, and b is a baseline adjustment for analyte measurement associated with both a tissue template and said calibration model.

13. The method of claim 12, wherein f(·

) and g(·

) are separately determined experimentally, wherein f(·

) is determined by manipulating analyte concentration while tissue properties remain constant, and wherein the properties of tissue are allowed to fluctuate and g(·

), m_s and m_i are determined on the basis of the error in analyte measurement where target value for g(·

) is given by;

r=y−

f(x_p)−

b;

where y is a reference analyte concentration.

14. The method of claim 13, wherein said step of correcting a direct analyte measurement on the basis of said detected changes comprises supplementing said second model with at least one selected features according to:

ŷ

=x_pF−

(m_szG+m_i)+b;

wherein f(·

) and g(·

) are determined to be linear over range of measurement and where Fε

^Nx1 and Gε

^Mx1.

15. The method of claim 12, wherein said y values are determined from samples of blood, serum, plasma or interstitial fluid taken from a fingertip, a site near the measurement site or an alternate site.

16. The method of claim 5, wherein said calibration model is determined from a calibration set of exemplary paired data points each consisting of a preprocessed spectral measurement, x, and an associated reference analyte value, y.

17. The method of claim 16, wherein said step of measuring said analyte indirectly on the basis of said at least one spectral feature comprises using at least one extracted features to measure glucose indirectly according to:

ŷ

=(m_sg(z)+m_i)+b;

where g;

^M→

¹ comprises said model, said model used to map set of features z to a variable correlated to a reference glucose level and m_s and m_i are slope and intercepts used to convert g(z) to the correct units and b is a baseline adjustment for glucose measurement.

18. The method of claim 17, wherein said at least one feature is selected based on their combined correlation to the reference analyte concentration.

19. The method of claim 18, wherein said at least one feature is selected based on any of:

a priori knowledge;

trial-and-error;

stepwise regression;

random search techniques;

genetic algorithms; and

evolutionary programming.

20. The method of claim 18, wherein g(·

) is determined according to;

ŷ

=(m_szG+m_i)+b;

where Gε

^Mx1.

21. The method of claim 2, wherein feature extraction comprises any mathematical transformation that enhances a quality or aspect of said tissue measurement to concisely represent tissue state, wherein tissue state comprises any of structural, chemical, physiological, and optical properties of the tissue that are indirectly related to the target analyte.

22. The method of claim 21, wherein said step of extracting at least one feature comprises the step of:

developing a set of features that represents tissue state based on distinct patterns that change according to changes in said structural, chemical, physiological and optical properties, wherein changes in tissue state are indirectly related to changes in target analyte concentration.

23. The method of claim 22, wherein said changes in tissue state comprise any of:

alteration of water distribution among body compartments;

changes to thickness of various skin layers; and

changes in distance from skin surface to adipose tissue layer.

24. The method of claim 23, wherein said changes in tissue state result in alterations of skin properties, said skin properties comprising any of:

localized scattering;

localized refractive index; and

skin thickness.

25. The method of claim 21, wherein a features includes any of:

a simple feature;

a derived features;

an abstract feature;

a normalization point;

a fat band point;

a protein band point; and

a water band point.

26. The method of claim 25, wherein said simple feature is derived directly from the tissue measurement.

27. The method of claim 25, wherein said derived feature comprises at least one mathematical combinations of simples features.

28. The method of claim 25, wherein said abstract feature is derived through at least one transformation of the analytical signal, wherein said at least one transformation is either linear or nonlinear.

29. The method of claim 2, further comprising any of the steps of:

detecting conditions not conducive to analyte measurement; and

detecting outliers.

30. The method of claim 29, wherein said step of performing outlier detection comprises:

performing Mahalanobis distance outlier detection.

31. The method of claim 1, wherein at least one abstract feature that reflects changes in tissue properties is used as independent variables for said calibration model and wherein said step of measuring said analyte indirectly comprises:

preprocessing said tissue measurement; and

decomposing said preprocessed tissue measurement according to;

z=xP;

where xε

^1xN is the preprocessed tissue measurement, N is number of wavelengths selected for calibration, Pε

^1xM is a set of M eigenvectors or loadings obtained from a principal components analysis of a calibration set and zε

^1xM is a set of abstract features used to measure glucose through application of said calibration model, wherein said model is either linear or nonlinear.

32. The method of claim 31, wherein analyte measurement associated with the tissue measurement is determined according to:

ŷ

=xG+b;

where Gε

^Mx1 is a linear transformation, derived from partial least-squares regression that represents both the feature extraction step and the calibration model.

33. The method of claim 1, wherein said analyte comprises any of:

water;

fat;

protein; and

glucose.

34. The method of claim 1, wherein said step of collecting said analytical signal comprises making repeated tissue measurements at predetermined time intervals.

35. The method of claim 1, wherein said analyte is glucose.

36. The method of claim 35, wherein said calibration model uses multivariate analysis.

37. The method of claim 36, wherein said analytical signal is generated using:

near infrared spectroscopy from 700 to 2500 nm in any of diffuse reflectance, transflectance, and transmission mode.

38. The method of claim 37, further comprising steps of:

correcting said signal;

subsequently calculating a derivative of said signal; and

subsequently selecting portions of said signal.

39. A system for noninvasive measurement of a target analyte in a tissue, comprising:

means for noninvasively collecting an analytical signal from said tissue, said collected signal comprising a tissue measurement; and

means for measuring concentration of said analyte based on at least one spectral feature extracted from the analytical signal indicative of the effect of the target analyte on the probed tissue.

40. The system of claim 39, wherein said means for collecting an analytical signal comprises:

means for detecting said analytical signal; and

means for digitizing said detected analytical signal.

41. The system of claim 40, wherein said means for measuring concentration of said analyte comprises:

a processing element in communication with said collection means, wherein said collection means passes said digitized signal to said processing element; and

computer readable code embodied on a tangible medium, wherein said processor executes said code, said code comprising code means for executing a method for noninvasive measurement of said target analyte, said method comprising the steps of;

collecting an analytical signal from the tissue, said collected signal comprising a tissue measurement;

extracting at least one feature from the analytical signal indicative of the effect of the target analyte on the probed tissue; and

either correcting a direct analyte measurement based on said at least one feature;

or

calculating concentration of said analyte indirectly by application of a calibration model to said at least one feature.

42. The system of claim 41, said system further comprising the step of:

preprocessing said tissue measurement.

43. The system of claim 42, wherein said step of preprocessing said tissue measurement comprises any of the steps of:

correcting said signal utilizing a reference;

filtering said signal;

calculating any of a first and second derivative of said signal;

selecting portions of said signal;

normalizing said signal;

scatter correcting said signal; and

translating said signal.

44. The system of claim 41, wherein feature extraction comprises any mathematical transformation that enhances a quality or aspect of said tissue measurement to concisely represent tissue state, wherein tissue state comprises any of structural, chemical, physiological, and optical properties of the tissue that are indirectly related to the target analyte.

45. The system of claim 44, wherein said step of extracting at least one feature comprises the step of:

developing a set of features that represents tissue state based on distinct patterns that change according to changes in said structural, chemical, physiological and optical properties, wherein changes in tissue state are indirectly related to changes in target analyte concentration.

46. The system of claim 45, wherein physiological changes comprise any of:

alteration of water distribution among body compartments;

changes to thickness of various skin layers; and

changes in distance from skin surface to adipose tissue layer.

47. The system of claim 46, wherein said physiological changes result in alterations of skin properties, said skin properties comprising any of:

localized scattering;

localized refractive index; and

skin thickness.

48. The methed system of claim 46, wherein a feature includes any of:

a simple feature;

a derived feature;

an abstract feature;

a normalization point;

a fat band point;

a protein band point; and

a water band point.

49. The system of claim 48, wherein said simple feature is derived directly from the tissue measurement.

50. The system of claim 48, wherein said derived feature comprises at least one mathematical combination of simple features.

51. The system of claim 48, wherein said abstract feature is derived through at least one transformation of the analytical signal, said at least one transformation is either linear or nonlinear.

52. The system of claim 41, said system further comprising the step of:

determining difference between a tissue template and either the preprocessed tissue measurement or the at least one extracted feature according to;

z=x−

(cx_t+d);

wherein x comprises either the pre-processed measurement or a set of at least one extracted feature, x_t comprises a tissue template associated with a measurement period, and c and d are slope and intercept adjustments to the tissue template.

53. The system of claim 52, wherein said tissue template is determined through one or more tissue measurements combined according to a predetermined data selection criterion during each measurement period.

54. The system of claim 53, wherein a measurement period comprises a time period over which accuracy of the tissue measurement remains within desired specifications.

55. The system of claim 53, said system further comprising the step of:

providing an associated set of reference values combined according to said predetermined data selection criterion to form a measurement bias adjustment.

56. The system of claim 53, wherein the tissue template comprises any set of features from a given subject or calibration set that future tissue measurements will be compared with, wherein c and d are determined through least-square fit of the tissue template over a particular wavelength range to the tissue measurement.

57. The system of claim 52, wherein the step of correcting a direct analyte measurement based on said at least one feature comprises:

supplementing a second calibration model based on direct effect of glucose on said analytical signal with said at least one selected feature according to;

ŷ

=f(x_p,z)+b;

where ŷ

is an estimated analyte concentration, x_pε

^N is a processed tissue measurement, zε

^M is a set of features representative of the physiological state or optical properties of the tissue, f;

^N,M→

¹ is a model used to measure the analyte on the basis of a preprocessed measurement and at least one extracted feature, and b is a baseline adjustment for analyte measurement associated with both a tissue template and said second calibration model.

58. The system of claim 52, wherein the step of correcting a direct analyte measurement based on said at least one feature comprises:

supplementing a second calibration model based on direct effect of glucose on said analytical signal with said at least one selected feature according to;

ŷ

=f(x_p)−

(m_sg(z)+m_i)+b;

where ŷ

is an estimated analyte concentration, x_pε

^N is a processed tissue measurement, zε

^M is a set of features representative of any of the structural, chemical, physiological and optical properties of the tissue, wherein x_p and z are independent, where F;

ⁿ→

¹ is a model used to measure the analyte in the absence of physiological or other tissue variation, g;

^M→

¹ is a model used to map the at least one feature to a variable correlated to error in analyte measurement caused by a change in the properties of the tissue, m_s and m_i are slope and intercepts used to convert g(z) to correct units, and b is a baseline adjustment for analyte measurement associated with both a tissue template and said calibration model.

59. The system of claim 58, wherein f(·

) and g(·

) are separately determined experimentally, wherein f(·

) is determined by manipulating analyte concentration while tissue properties remain constant, and wherein the properties of tissue are allowed to fluctuate and g(·

), m_s and m_i are determined on the basis of the error in analyte measurement where target value for g(·

) is given by;

r=y−

f(x_p)−

b;

where y is a reference analyte concentration.

60. The system of claim 52, wherein said step of correcting a direct analyte measurement on the basis of said detected changes comprises supplementing said second model with at least one selected feature according to:

ŷ

=x_pF−

(m_szG+m_i)+b;

wherein f(·

) and g(·

) are determined to be linear over range of measurement and where Fε

^Nx1 and Gε

^Mx1.

61. The system of claim 52, wherein said calibration model is determined from a calibration set of exemplary paired data points each consisting of a preprocessed spectral measurement, x and an associated reference analyte value, y.

62. The system of claim 52, wherein said step of measuring said analyte indirectly on the basis of at least one spectral feature comprises using at least one extracted feature to measure glucose indirectly according to:

ŷ

=(m_sg(z)+m_i)+b;

where g;

^M→

¹ comprises said model, said model used to map a set of features z to a variable correlated to a reference glucose level and m_s and m_i are slope and intercepts used to convert g(z) to the correct units and b is a baseline adjustment for glucose measurement.

63. The system of claim 41, said system further comprising any of the steps of:

detecting conditions not conducive to analyte measurement; and

detecting outliers.

64. The system of claim 39, wherein at least a portion of said system is implanted in body of a subject, said system adapted to measure said analyte in a manner that is noninvasive to tissue probed.

65. The system of claim 64, wherein site of implantation comprises peritoneal cavity.

66. The system of claim 64, wherein said measurement means is located remotely from said body.

67. The system of claim 66, wherein said measurement system and said collection system are in communication via telemetry.

68. The system of claim 39, further comprising means for generating a probing signal, wherein said probing signal is directed toward said tissue.

69. The system of claim 39, wherein said tissue measurement comprises an in vivo measurement from a human subject and wherein said target analyte comprises glucose.

70. The system of claim 69, wherein said analytical signal is derived from use of any of:

fluorescence spectroscopy;

ultraviolet spectroscopy from 200 to 400 nm;

visible spectroscopy from 400 to 700 nm;

infrared and Fourier transform infrared (FTIR) spectroscopy;

passive infrared spectroscopy;

mid infrared spectroscopy from 4000–

700 cm<sup>−

1</sup> in any of diffuse reflectance and transmission;

attenuated total reflectance (ATR) spectroscopy;

far infrared radiation spectroscopy;

Kromoscopy in reflection or transmission;

emission spectroscopy;

Raman spectroscopy;

photoacoustic and pulse laser photoacoustic spectroscopy;

photon scattering from 400 to 2500 nm;

bioelectrical response spectroscopy;

oscillating thermal gradient spectroscopy;

polarimetry;

ultrasound spectroscopy;

near infrared spectroscopy from 700 to 2500 nm in any of diffuse reflectance, transflectance, and transmission mode; and

radio wave spectroscopy.

71. The system of claim 69, wherein said analytical signal is derived from use of light scattering.

72. The system of claim 69, wherein said at least one feature comprises any of:

one or more water absorbance bands;

one or more fat absorbance bands; and

one or more protein absorbance bands.

73. The system of claim 72, wherein said one or more water absorbance bands are centered at any of the wavelengths:

approximately 1450 nm;

approximately 1900 nm; and

approximately 2600 nm.

74. The system of claim 73, wherein said one or more fat absorbance bands are centered at any of the wavelengths:

approximately 1675 nm;

approximately 1715 nm;

approximately 1760 nm;

approximately 2130 nm;

approximately 2250 nm; and

approximately 2320 nm.

75. The system of claim 73, wherein said one or more protein absorbance bands are centered at any of the wavelengths:

approximately 1180 nm;

approximately 1280 nm;

approximately 1690 nm;

approximately 1730 nm;

approximately 2170 nm; and

approximately 2285 nm.

76. The system of claim 39, wherein collecting said analytical signal from said tissue comprises taking repeated tissue measurements at predetermined time intervals.

77. A method for noninvasive measurement of a target analyte in a tissue, comprising the steps of:

collecting an analytical signal from the tissue, said collected signal comprising tissue measurement;

extracting at least one feature from the analytical signal indicative of the effect of the target analyte on the probed tissue, wherein said effect comprises alteration of water distribution among body compartments; and

either correcting a direct analyte measurement based on said at least one feature;

or

calculating concentration of said analyte indirectly by application of a calibration model to said at least one feature.

78. The method of claim 77, further comprising the step of:

preprocessing said tissue measurement.

79. The method of claim 77, further comprising any of the steps of:

detecting conditions not conducive to analyte measurement; and

detecting outliers.

80. The method of claim 77, wherein said analyte comprises any of:

water;

fat;

protein; and

glucose.

81. The system of claim 77, wherein said at least one feature comprises any of:

one or more water absorbance bands;

one or more fat absorbance bands; and

one or more protein absorbance bands.

82. A method for noninvasive measurement of a target analyte in a tissue, comprising the steps of:

collecting an analytical signal from the tissue, said collected signal comprising a tissue measurement;

extracting at least one feature from the analytical signal indicative of the effect of the target analyte on the probed tissue wherein said at least one feature comprises at least a portion of a water band, a portion of a fat band, and a portion of a protein band;

either correcting a direct analyte measurement based on said at least one feature;

or

calculating concentration of said analyte indirectly by application of a calibration model to said at least one feature.

83. The method of claim 82, further comprising the step of:

preprocessing said tissue measurement.

84. The method of claim 82, further comprising any of the steps of:

detecting conditions not conducive to analyte measurement; and

detecting outliers.

85. The method of claim 82, wherein said analyte comprises any of:

water;

fat;

protein; and

glucose.

86. The method claim 82, wherein said effect comprises alteration of water distribution among body compartments.

87. A method for noninvasive measurement of a target analyte in a tissue, comprising the steps of:

collecting an analytical signal from the tissue, said collected signal comprising a tissue measurement;

preprocessing said tissue measurement;

extracting at least one feature from the analytical signal indicative of the effect of the target analyte on the probed tissue; and

either correcting a direct analyte measurement based on said at least one feature;

or

calculating concentration of said analyte indirectly by application of a calibration model to said at least one feature.

88. The method of claim 87, wherein said step of preprocessing said tissue measurement comprises any of the steps of:

correcting said signal utilizing a reference;

filtering said signal;

calculating any of a first and second derivative of said signal;

normalizing said signal;

selecting portions of said signal;

scatter correcting said signal; and

translating said signal.

89. The method of claim 87, further comprising any of the steps of:

detecting conditions not conducive to analyte measurement; and

detecting outliers.

90. The method of claim 87, wherein said analyte comprises any of:

water;

fat;

protein; and

glucose.

91. The method of claim 87, wherein said effect comprises alteration of water distribution among body compartments.

92. The method of claim 87, wherein said at least one feature comprise any of:

at least a portion of a water absorbance band;

at least a portion of a fat absorbance band; and

at least a portion of a protein absorbance band.