Method for predicting a user's future glycemic state

• US 7,731,659 B2
• Filed: 10/18/2007
• Issued: 06/08/2010
• Est. Priority Date: 10/18/2007
• Status: Active Grant

• Current Assignee: Cilag Gmbh International (Johnson & Johnson)
• Original Assignee: LifeScan Scotland Limited (Johnson & Johnson)
• Inventors: Malecha, Michael
• Primary Examiner(s): Marmor, II; Charles A
• Assistant Examiner(s): JANG, CHRISTIAN YONGKYUN
• Application Number:
  US11/874,803
• Publication Number:
  US 20090105572A1
• Time in Patent Office: 964 Days
• Field of Search: 600/316, 600/319, 600/347, 600/365
• US Class Current: 600/365
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• 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: G16H 50/50 for simulation or modelling of medical disorders
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

A method for predicting a user'"'"'s future glycemic state includes measuring a user'"'"'s glucose concentration at intervals over a time duration, thereby generating a plurality of glucose concentrations as a function of time. First and second glucose prediction equations that are fits to the plurality of glucose concentrations based on first and second non-identical mathematical models, respectively, are then derived. The method also includes calculating first and second predicted glucose concentrations at a future time using the first and second glucose prediction equations, respectively. Thereafter, an average predicted glucose concentration and a merit index are calculated based on the first and second predicted glucose calculations. The plurality of glucose concentrations as a function of time, the merit index and average predicted glucose concentration are input into a trained model (for example, a Hidden Markov Model) that outputs a set of glucose concentration probabilities. The user'"'"'s future glycemic state is then predicted based on the set of glucose concentration probabilities.

Claims

1. A method for predicting a user'"'"'s future glycemic state using a continuous glucose sensor, the method comprising:

measuring a user'"'"'s glucose concentration with the continuous glucose sensor at intervals over a predetermined time duration, thereby generating a plurality of glucose concentrations as a function of time;

deriving a first glucose prediction equation that is a fit to the plurality of glucose concentrations as a function of time, the fit being based on a first mathematical model;

deriving a second glucose prediction equation that is a fit to the plurality of glucose concentrations as a function of time, the fit being based on a second mathematical model;

calculating a first predicted glucose concentration g1(t_f) at a predetermined future time (t_f) using the first glucose prediction equation;

calculating a second predicted glucose concentration g2(t_f) at the predetermined future time (t_f) using the second glucose prediction equation;

calculating an average predicted glucose concentration and a merit index M based on the first and second predicted glucose calculations;

wherein the merit index M is calculated using the following algorithms;

M=g2(t_f)−

g1(t_f)/g2(t_f), if g2(t_f)≧

g1(t_f)

M=g1(t_f)−

g2(t_f)/g1(t_f), if g2(t_f)<

g1(t_f);

inputting the plurality of glucose concentrations as a function of time, the average predicted glucose concentration and the merit index into a trained model that outputs a set of glucose concentration probabilities, and

predicting the user'"'"'s future glycemic state based on the set of glucose concentration probabilities, wherein the first mathematical model and the second mathematical model are non-identical to one another.

2. The method of claim 1 wherein the predicting step predicts the probability that the user will be in a hypoglycemic state at the predetermined future time based on the set of glucose concentration probabilities.

3. The method of claim 2 further including the step of:

alerting the user if the probability of the user being in a hypoglycemic state is greater than a predetermined threshold probability.

4. The method of claim 1 wherein the plurality of glucose concentrations as a function of time is in the range of seven to thirty glucose concentrations as a function of time.

5. The method of claim 1, wherein the predetermined future time point is a time in the range of about 10 minutes to about 30 minutes after the predetermined time duration.

6. The method of claim 1 wherein the interval is in the range of about 0.2 minutes to about 1 minute.

7. The method of claim 1 wherein the set of glucose concentration probabilities is centered on the average predicted glucose concentration and has a range that is dependent on the merit index.