Multi-tier method of developing localized calibration models for non-invasive blood analyte prediction

• US 6,512,937 B2
• Filed: 04/03/2001
• Issued: 01/28/2003
• Est. Priority Date: 07/22/1999
• Status: Expired

• Current Assignee: GLT Acquisition Corporation
• Original Assignee: Sensys Medical, Inc. (Sensys Medical Limited)
• Inventors: Ruchti, Timothy L., Monfre, Stephen L., Blank, Thomas B., Thennadil, Suresh
• Primary Examiner(s): WINAKUR, ERIC FRANK
• Application Number:
  US09/825,687
• Publication Number:
  US 20010021803A1
• Time in Patent Office: 665 Days
• Field of Search: 600/310, 600/322, 600/323, 600/330, 600/336, 600/473, 128/920, 128/923, 128/924, 128/925, 356/402, 250/340, 250/341.1, 250/339.01
• US Class Current: 600/322
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• CPC Class Codes: A61B 2560/0233 Optical standards
• 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 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: A61B 5/1495 Calibrating or testing of in-vivo probes
• CPC Class Codes: 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 21/274 Calibration, base line adjustment, drift correction
• 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: G01N 21/49 within a body or fluid
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics

Abstract

A method of multi-tier classification and calibration in noninvasive blood analyte prediction minimizes prediction error by limiting co-varying spectral interferents. Tissue samples are categorized based on subject demographic and instrumental skin measurements, including in vivo near-IR spectral measurements. A multi-tier intelligent pattern classification sequence organizes spectral data into clusters having a high degree of internal consistency in tissue properties. In each tier, categories are successively refined using subject demographics, spectral measurement information and other device measurements suitable for developing tissue classifications.

The multi-tier classification approach to calibration utilizes multivariate statistical arguments and multi-tiered classification using spectral features. Variables used in the multi-tiered classification can be skin surface hydration, skin surface temperature, tissue volume hydration, and an assessment of relative optical thickness of the dermis by the near-IR fat band. All tissue parameters are evaluated using the NIR spectrum signal along key wavelength segments.

Claims

1. A method of developing a multi-tiered calibration model for estimating concentration of a target blood analyte from measured tissue spectra, comprising the steps of:

providing a calibration set, wherein said calibration set comprises a data set of exemplar spectral measurements from a representative sampling of a subject population;

initially, classifying said exemplar measurements into previously defined classes based on a priori information pertaining to a corresponding subject;

further classifying said exemplar measurements into previously defined classes based on at least one instrumental measurement at a tissue measurement site;

extracting at least one feature from said exemplar measurements for still further classification, wherein a decision rule makes class assignments; and

calculating at least one localized calibration model based on said classified measurements and an associated set of reference values.

2. The method of claim 1, wherein said initial classification step comprises the steps of:

in a first tier, classifying said measured spectrum into previously defined classes based on subject'"'"'s age; and

in a second tier, further classifying said measured spectrum into previously defined classes based on subject'"'"'s sex.

3. The method of claim 1, wherein said further classification step further comprises the steps of:

in a third tier further classifying said exemplar measurements into previously defined classes based on an estimation of stratum corneum hydration at said tissue measurement site; and

in a fourth tier, further classifying said exemplar measurements into previously defined classes based on skin temperature at said tissue measurement site.

4. The method of claim 3, wherein said stratum corneum hydration estimate is based on a measurement of ambient humidity at said tissue measurement site.

5. The method of claim 1, wherein said feature extraction step comprises any mathematical transformation that enhances a quality or aspect of sample measurement for interpretation to represent concisely structural properties and physiological state of a tissue measurement site, wherein a resulting set of features is used to classify a subject and determine a calibration model that is most useful for blood analyte prediction.

6. The method of claim 5, wherein said features are represented in a vector, zΣ

^M that is determined from a preprocessed measurement through;

7. The method of claim 6, wherein individual features are divided into categories, said categories comprising:

abstract features that do not necessarily have a specific interpretation related to a physical system; and

simple features that are derived from an a priori understanding of a sample and that can be related directly to a physical phenomenon.

8. The method of claim 7, wherein said simple features can be calculated from NIR spectral absorbance measurements, said simple features including any of:

thickness of adipose tissue;

hematocrit level;

tissue hydration;

magnitude of protein absorbance;

scattering properties of said tissue;

skin thickness;

temperature related effects;

age related effects;

spectral characteristics;

pathlength estimates;

volume fraction of blood in tissue; and

spectral characteristics related to environmental influences.

9. The method of claim 1, further comprising the step of:

employing spectral decomposition to determine features related to a known spectral absorbance pattern.

10. The method of claim 1, further comprising the step of:

employing factor-based methods to build a model capable of representing variation in a measured absorbance spectrum related to a demographic variable;

wherein projection of a measured absorption onto said model constitutes a feature that represents spectral variation related to said demographic variable.

11. The method of claim 1, wherein said feature extraction step assigns a measurement to one of many predefined classes.

12. The method of claim 1, further comprising the steps of;

measuring the similarity of a feature to predefined classes; and

assigning class membership.

13. The method of claim 1, further comprising the step of;

using measurements and class assignments to determine a mapping from features to class assignments.

14. The method of claim 13, further comprising the steps of:

defining classes from said features in a supervised manner, wherein each set of features is divided into two or more regions, and wherein classes are defined by combination of feature divisions;

performing a cluster analysis on the spectral data to determine groups of said defined classes that can be combined, wherein the final number of class definitions is significantly reduced;

designing a classifier subsequent to class definition through supervised pattern recognition by determining an optimal mapping or transformation from the feature space to a class estimate that minimizes the number of misclassifications; and

creating a model based on class definitions that transforms a measured set of features to an estimated classification, wherein said class definitions are optimized to satisfy specifications of a measurement system used to take said measurements.

15. The method of claim 14, wherein said optimized classes comprise groups of measurements wherein similarity between measurements within a group is greater than similarity between groups.

16. The method of claim 15, said step of calculating at least one localized calibration model comprising:

calculating weights, w, for said exemplar measurements according to;

17. The method of claim 16, wherein a vector of weights of spectral measurements within one of said groups comprises a regression vector for said group;

wherein said regression vector comprises a calibration model for said group.

18. A method of developing a multi-tiered calibration model for estimating concentration of a target blood analyte from measured tissue spectra, comprising the steps of:

providing a calibration set, wherein said calibration set comprises a data set of exemplar spectral measurements from a representative sampling of a subject population;

in at least one tier, classifying said exemplar measurements into previously defined classes; and

extracting at least one feature from said exemplar measurements for still further classification; and

calculating at least one localized calibration model based on said classified exemplar measurements and a set of associated reference values.

19. The method of claim 18, wherein said classifying step is based on any of:

abstract and simple features.

20. The method of claim 18, further comprising the step of mapping said exemplar measurements to estimates of said analyte based on either a linear or a nonlinear model.

21. The method of claim 18, wherein said classifying step is based on any of:

a priori information; and

at least one instrumental measurement at a tissue measurement site at which optical samples were taken for said spectral measurements.

22. The method of claim 18, wherein said classifying step comprises multiple tiers.

23. The pattern classification method of claim 22, wherein said classifying step comprises any of the steps of:

classifying said exemplar measurements into previously defined classes based on subject'"'"'s age;

classifying said exemplar measurements into previously defined classes based on subject'"'"'s sex;

classifying said exemplar measurements into previously defined classes based on an estimation of stratum corneum hydration at said tissue measurement site; and

classifying said exemplar measurements into previously defined classes based on skin temperature at said tissue measurement site.