Method and device for predicting physiological values

• US 6,180,416 B1
• Filed: 09/30/1998
• Issued: 01/30/2001
• Est. Priority Date: 09/30/1998
• Status: Expired due to Term

• Current Assignee: LifeScan IP Holdings, LLC
• Original Assignee: Cygnus, Inc. (Johnson & Johnson)
• Inventors: Kurnik, Ronald T., Potts, Russell O., Tamada, Janet A., Jayalakshmi, Yalia, Waterhouse, Steven Richard, Oliver, Jonathan James, Dunn, Timothy C., Lesho, Matthew J., Wei, Charles W.
• Primary Examiner(s): Chin, Christopher L.
• Application Number:
  US09/163,856
• Time in Patent Office: 853 Days
• Field of Search: 435/14, 435/25, 435/176, 435/287.1, 435/817, 436/518, 436/525, 436/149, 436/150, 436/151, 436/806, 422/82.01, 422/82.02, 204/400, 204/403
• US Class Current: 600/316
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: Y10S 435/817 Enzyme or microbe electrode
• CPC Class Codes: Y10S 436/806 Electrical property or magnetic property

Abstract

The invention relates generally to methods, systems, and devices for measuring the concentration of target analytes present in a biological system using a series of measurements obtained from a monitoring system and a Mixtures of Experts (MOE) algorithm. In one embodiment, the present invention describes a method for measuring blood glucose in a subject.

Claims

1. A method for continually or continuously measuring an analyte present in a biological system, said method comprising:

(a) transdermally extracting the analyte from the biological system using a sampling system that is in operative contact with a skin or mucosal surface of said biological system;

(b) obtaining a raw signal from the extracted analyte, wherein said raw signal is specifically related to the analyte;

(c) performing a calibration step which correlates the raw signal obtained in step (b) with a measurement value indicative of the concentration of analyte present in the biological system at the time of extraction;

(d) repeating steps (a)-(b) to obtain a series of measurement values at selected time intervals, wherein the sampling system is maintained in operative contact with the skin or mucosal surface of said biological system to provide for a continual or continuous analyte measurement; and

(e) predicting a measurement value based on the series of measurement values using the Mixtures of Experts algorithm, where the individual experts have a linear form $\begin{array}{cc}\mathrm{An}=\underset{i=1}{\overset{n}{\sum <br><br>}}<br><br>{\mathrm{An}}_i<br><br>w_i& \left(1\right)\end{array}$

wherein (An) is an analyte of interest, n is the number of experts, An_i is the analyte predicted by Expert i; and

w_i is a weighting value, and the individual experts An_i are further defined by the expression shown as Equation (2) $\begin{array}{cc}{\mathrm{An}}_i=\underset{j=1}{\overset{m}{\sum <br><br>}}<br><br>a_{\mathrm{ij}}<br><br>P_j+z_i& \left(2\right)\end{array}$

wherein, An_i is the analyte predicted by Expert i;

P_j is one of m parameters, m is typically less than 100;

a_ij are coefficients; and

z_i is a constant; and

further where the weighting value, w_i, is defined by the formula shown as Equation (3) $\begin{array}{cc}{w}_i=\frac{e^{d_i}}{\left[\underset{k=1}{\overset{n}{\sum <br><br>}}<br><br>e^{d_k}\right]}& \left(3\right)\end{array}$

where e refers to than exponential function and the d_k (note that the d_i in the numerator of Equation 3 is one of the d_k) are a parameter set analogous to Equation 2 that is used to determine the weights w_i, The d_k are given by Equation 4 $\begin{array}{cc}{d}_k=\underset{j=1}{\overset{m}{\sum <br><br>}}<br><br>{\mathrm{\alpha <br><br>}}_{\mathrm{jk}}<br><br>P_j+{\mathrm{\omega <br><br>}}_k& \left(4\right)\end{array}$

where α

_jk is a coefficient, P_j is one of m parameters, and where ω

_k is a constant.

2. The method of claim 1, wherein the analyte is extracted from the biological system in step (a) into a collection reservoir to obtain a concentration of the analyte in said reservoir.

3. The method of claim 2, wherein the collection reservoir is in contact with the skin or mucosal surface of the biological system and the analyte is extracted using an iontophoretic current applied to said skin or mucosal surface.

4. The method of claim 3, wherein the collection reservoir contains an enzyme that reacts with the extracted analyte to produce an electrochemically detectable signal.

5. The method of claim 4, wherein the analyte is glucose.

6. The method of claim 5, wherein the enzyme is glucose oxidase.

7. The method of claim 1, wherein the prediction of step (e) is carried out using said series of two or more measurement values in an algorithm represented by the Mixtures of Experts algorithm, where the individual experts have a linear form

8. The method of claim 7, which includes further parameters for measurement values selected from the group consisting of temperature, ionophoretic voltage, and skin conductivity.

9. A method for measuring blood glucose in a subject, said method comprising:

(a) obtaining a raw signal from a sensing apparatus, wherein said raw signal is specifically related to the glucose detected by the sensing apparatus;

(b) performing a calibration step which correlates the raw signal obtained in step (a) with a measurement value indicative of the subject'"'"'s blood glucose concentration;

(c) repeating step (a) to obtain a series of measurement values at selected time intervals; and

(d) predicting a measurement value using the Mixtures of Experts algorithm, where the individual experts have a linear form;

$\begin{array}{cc}\mathrm{An}=\underset{i=1}{\overset{n}{\sum <br><br>}}<br><br>{\mathrm{An}}_i<br><br>w_i& \left(1\right)\end{array}$

wherein (An) is blood glucose value, n is the number of experts, An_j is the blood glucose value predicted by Expert i; and

w_i is a weighting value, and the individual experts An_i are further defined by the expression shown as Equation (2) $\begin{array}{cc}{\mathrm{An}}_i=\underset{j=1}{\overset{m}{\sum <br><br>}}<br><br>a_{\mathrm{ij}}<br><br>P_j+z_i& \left(2\right)\end{array}$

wherein, An_i is the blood glucose value predicted by Expert i;

P_j is one of m parameters, m is typically less than 100;

a_ij are coefficients; and

z_i is a constant; and

further where the weighting value, w_i, is defined by the formula shown as Equation (3) $\begin{array}{cc}{w}_i=\frac{e^{d_i}}{\left[\underset{k=1}{\overset{n}{\sum <br><br>}}<br><br>e^{d_k}\right]}& \left(3\right)\end{array}$

where e refers to the exponential function and the d_k (note that the d_i in the numerator of Equation 3 is one of the d_k) are a parameter set analogous to Equation 2 that is used to determine the weights w_i, The d_k are given by Equation 4 $\begin{array}{cc}{d}_k=\underset{j=1}{\overset{m}{\sum <br><br>}}<br><br>{\mathrm{\alpha <br><br>}}_{\mathrm{jk}}<br><br>P_j+{\mathrm{\omega <br><br>}}_k& \left(4\right)\end{array}$

where α

_jk is a coefficient, P_j is one of m parameters, and where ω

_k is a constant.

10. The method of claim 9, where in said Mixtures of Experts algorithm, the individual experts have a linear form

11. The method of claim 10, wherein the sensing apparatus is a near-IR spectrometer.

12. The method of claim 10, wherein the sensing means comprises a biosensor having an electrochemical sensing element.

13. A monitoring system for continually or continuously measuring an analyte present in a biological system, said system comprising, in operative combination:

(a) sampling means for continually or continuously extracting the analyte from the biological system, wherein said sampling means is adapted for extracting the analyte across a skin or mucosal surface of said biological system;

(b) sensing means in operative contact with the analyte extracted by the sampling means, wherein said sensing means obtains a raw signal from the extracted analyte and said raw signal is specifically related to the analyte; and

(c) microprocessor means in operative communication with the sampling moans and the sensing means, wherein said microprocessor means (i) is used to control the sampling means and the sensing means to obtain a series of raw signals at selected time intervals during a continual or continuous measurement period, (ii) correlate the raw signals with measurement values indicative of the concentration of analyte present in the biological system, and (iii) predict a measurement value using the Mixtures of Experts algorithm, where the individual experts have a linear form $\begin{array}{cc}\mathrm{An}=\underset{i=1}{\overset{n}{\sum <br><br>}}<br><br>{\mathrm{An}}_i<br><br>w_i& \left(1\right)\end{array}$

wherein (An) is an analyte of interest, n is the number of experts, An_i is the analyte predicted by Expert i; and

w_i is a weighting value, and the individual experts An_i are further defined by the expression shown as Equation (2) $\begin{array}{cc}{\mathrm{An}}_i=\underset{j=1}{\overset{m}{\sum <br><br>}}<br><br>a_{\mathrm{ij}}<br><br>P_j+z_i& \left(2\right)\end{array}$

wherein, An_i is the analyte predicted by Expert i;

P_j is one of m parameters, m is typically less than 100;

a_ij are coefficients; and

z_i is a constant; and

further where the weighting value, w_i, is defined by the formula shown as Equation (3) $\begin{array}{cc}{w}_i=\frac{e^{d_i}}{\left[\underset{k=1}{\overset{n}{\sum <br><br>}}<br><br>e^{d_k}\right]}& \left(3\right)\end{array}$

where e refers to the exponential function and the d_k (note that the d_i in the numerator of Equation 3 is one of the d_k) are a parameter set analogous to Equation 2 that is used to determine the weights w_i, The d_k are given by Equation 4 $\begin{array}{cc}{d}_k=\underset{j=1}{\overset{m}{\sum <br><br>}}<br><br>{\mathrm{\alpha <br><br>}}_{\mathrm{jk}}<br><br>P_j+{\mathrm{\omega <br><br>}}_k& \left(4\right)\end{array}$

where α

_jk is a coefficient, P_j is one of m parameters, and where ω

_k is a constant.

14. The monitoring system of claim 13, wherein the sampling means includes one or more collection reservoirs for containing the extracted analyte.

15. The monitoring system of claim 14, wherein the sampling means uses an iontophoretic current to extract the analyte from the biological system.

16. The monitoring system of claim 15, wherein the collection reservoir contains an enzyme that reacts with the extracted analyte to produce an electrochemically detectable signal.

17. The monitoring system of claim 16, wherein the analyte is glucose and the enzyme is glucose oxidase.

18. A monitoring system for measuring blood glucose in a subject, said system comprising, in operative combination:

(a) sensing means in operative contact with the subject or with a glucose-containing sample extracted from the subject, wherein said sensing means obtains a raw signal specifically related to blood glucose in the subject; and

(b) microprocessor means in operative communication with the sensing means, wherein said microprocessor means (i) is used to control the sensing means to obtain a series of raw signals at selected time intervals, (ii) correlates the raw signals with measurement values indicative of the concentration of blood glucose present in the subject, and (iii) predicts a measurement value at a further time interval using the Mixtures of Experts algorithm, where the individual experts have a linear form

19. The monitoring system of claim 18, which includes further parameters for raw signals selected from the group consisting of temperature, ionophoretic voltage, and skin conductivity.

20. The monitoring system of claim 18, wherein the sensing means comprises a biosensor having an electrochemical sensing element.

21. The monitoring system of claim 18, wherein the sensing means comprises a near-IR spectrometer.