System and/or method for glucose sensor calibration

Patent metadata

Field	Value
Publication number	US-9402569-B2
Application number	US-201113239265-A
Country	US
Kind code	B2
Filing date	Sep 21, 2011
Priority date	Oct 28, 2010
Publication date	Aug 2, 2016
Grant date	Aug 2, 2016

Abstract

The subject matter disclosed herein relates to systems, methods and/or devices for calibrating sensor data to be used in estimating a blood glucose concentration. A relationship between sensor signal values and reference readings may be used to estimate a relationship between sensor signal values and measurements of blood glucose concentration.

First claim

What is claimed is:

1. A method comprising:
obtaining a sequence of blood glucose reference measurements from a patient;
deriving, using an electronic computing device, a non-linear mapping between sensor signal values (ISIG) and blood glucose concentration in said patient to calibrate a glucose monitor for observing glucose characteristics of said patient, said non-linear mapping based, at least in part, on temporal pairings of at least some of said blood glucose reference measurements with said sensor signal values, wherein said deriving said non-linear mapping further comprises deriving a piece-wise function that maps sensor signal values to blood-glucose concentration, the piece-wise function comprising at least one linear portion and at least one non-linear portion, the at least one non-linear portion being derived at least in part by determining coefficients e, f, and g corresponding to the following function:
blood-glucose reference measurement=( I SIG++ e ) f ; and

with said electronic computing device, using, at least in part, said determined coefficients e, f and g in calibrating subsequent sensor signal values received by said glucose monitor to provide said patient one or more estimated blood glucose levels corresponding to said subsequent sensor signal values.

2. The method of claim 1 , and further comprising determining coefficients e, f, and g based, at least in part, on said temporal pairings.

3. The method of claim 1 , and further comprising, for values in said non-linear portion, obtaining a measurement of blood glucose concentration by applying an offset to a function defining said linear portion.

4. An apparatus comprising:
a sensor to generate signal values (ISIG) responsive to a blood glucose concentration in a body; and
a special purpose computing apparatus programmed to:
obtain a sequence of blood glucose reference measurements taken from said body; and

derive a non-linear mapping between said signal values and blood glucose concentration in said body to calibrate a glucose monitor for observing blood glucose characteristics in said body based, at least in part, on temporal pairings of at least some of said blood glucose reference measurements with said signal values, wherein said non-linear mapping is derived by deriving a piece-wise function that maps sensor signal values to blood glucose concentration, the piece-wise function comprising at least one linear portion and at least one non-linear portion, the at least one non-linear portion being derived at least in part by a determination of coefficients e, f, and g corresponding to the following function:
blood-glucose reference measurement=( I SIG+ e ) f +g; and

use, at least in part, said determined coefficients e, f and g in calibrating subsequent signal values received by said glucose monitor to provide one or more estimated blood glucose levels corresponding to said subsequent signal values.

5. The apparatus of claim 4 , wherein said special purpose computing apparatus is further programmed to obtain a measurement of blood glucose concentration by applying an offset to a function defining said linear portion for values in said non-linear portion.

6. The apparatus of claim 4 , wherein said determination of coefficients e, f, and g is based, at least in part, on said temporal pairings.

7. An article comprising:
a non-transitory storage medium having machine-readable instructions stored thereon which are executable by a special purpose computing apparatus to:
obtain a sequence of blood glucose reference measurements taken from a body; and

derive a non-linear mapping between sensor signal values (ISIG) and blood glucose concentration in said body to calibrate a glucose monitor for observing blood glucose characteristics in said body based, at least in part, on temporal pairings of at least some of said blood glucose reference measurements with said sensor signal values, wherein said non-linear mapping is derived by deriving a piece-wise function that maps sensor signal values to blood-glucose concentration, the piece-wise function comprising at least one linear portion and at least one non-linear portion, wherein the at least one non-linear portion is derived at least in part by a determination of parameters e, f, and g corresponding to the following function:
blood-glucose reference measurement=( I SIG+ e ) f +g; and

use, at least in part, said determined coefficients e, f, and g to calibrate subsequent sensor signal values received by said glucose monitor to provide one or more estimated blood glucose levels corresponding to said subsequent sensor signal values.

8. The article of claim 7 , wherein said instructions are further executable by said special purpose computing apparatus to obtain a measurement of blood glucose concentration by applying an offset to a function defining said linear portion for values in said non-linear portion.

9. The article of claim 7 , wherein a determination of parameters e, f, and g is based, at least in part, on said temporal pairings.

10. An apparatus comprising:
means for obtaining a sequence of blood glucose reference measurements from a body; and
means for deriving a non-linear mapping between sensor signal values (ISIG) and blood glucose concentration in said body to calibrate a glucose monitor for observation of blood glucose characteristics in said body based, at least in part, on temporal pairings of at least some of said blood glucose reference measurements with said sensor signal values, wherein said means for deriving said non-linear mapping further comprises means for deriving a piece-wise function that maps sensor signal values to blood-glucose concentration, the piece-wise function comprising at least one linear portion and at least one non-linear portion, wherein the at least one non-linear portion is derived at least in part by a determination of parameters e, f, and g corresponding to the following function:
blood-glucose reference measurement=( I SIG+ e ) f+g ; and

means for using, at least in part, said determined coefficients e, f, and g to calibrate subsequent sensor signal values received by said glucose monitor to provide one or more estimated blood glucose levels corresponding to said subsequent sensor signal values.

11. The apparatus of claim 10 , and further comprising means for determining parameters of said exponential expression based, at least in part, on said temporal pairings.

12. The apparatus of claim 10 , and further comprising, for values in said non-linear portion, means for obtaining a measurement of blood glucose concentration by applying an offset to a function defining said linear portion.

Assignees

• Liang Bradley
• Cooper Kenneth W
• Gautham Raghavendhar
• Shah Rajiv
• Nogueira Keith
• Medtronic Minimed Inc

Inventors

• Liang Bradley
• Cooper Kenneth W
• Gautham Raghavendhar
• Shah Rajiv
• Nogueira Keith

Classifications

• A61B5/7239

  using differentiation including higher order derivatives · CPC title

• A61B5/14865

  invasive, e.g. introduced into the body by a catheter or needle or using implanted sensors · CPC title

• A61B5/14503

  invasive, e.g. introduced into the body by a catheter or needle or using implanted sensors (A61B5/1459, A61B5/1464, A61B5/1473, A61B5/1482, A61B5/14865 take precedence) · CPC title

• A61B5/14532 Primary

  for measuring glucose, e.g. by tissue impedance measurement · CPC title

• A61B2560/0223

  of calibration, e.g. protocols for calibrating sensors · CPC title