Predictive calibration

• US 9,486,171 B2
• Filed: 03/15/2013
• Issued: 11/08/2016
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: Tandem Diabetes Care Incorporated
• Original Assignee: Tandem Diabetes Care Incorporated
• Inventors: Saint, Sean
• Primary Examiner(s): Lin, Jerry
• Application Number:
  US13/841,028
• Publication Number:
  US 20140273042A1
• Time in Patent Office: 1,334 Days
• Field of Search: 702/23
• US Class Current: 1/1
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/1495 Calibrating or testing of in-vivo probes
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: G16H 40/40 for the management of medical equipment or devices, e.g. scheduling maintenance or upgrades

Abstract

A portable medical monitor system generates a current calibration curve for generating the estimate of the level of an analyte, such as glucose, being monitored. The current calibration curve is based on at least two measured data values of the level being monitored. The system determines a transformation function based on the calibration curve and at least one preceding calibration curve such that the transformation function produces a predictive calibration curve, and generates an estimated level value of the level being monitored, based on sensor output from a sensor associated with the portable medical monitor system, in accordance with the predictive calibration curve.

Claims

1. A method of operating a portable medical monitor system that generates an estimate of an analyte level of a patient being monitored with an analyte sensor, the method comprising:

generating a current calibration curve for generating the estimate of the analyte level being monitored, wherein the current calibration curve is based on at least two measured data values of the level being monitored;

determining a transformation function based on the current calibration curve and at least one preceding calibration curve such that the transformation function produces a predictive calibration curve that predicts changes to the current calibration curve resulting from sensor drift since the acquisition of the at least two measured data values used to generate the current calibration curve;

generating an estimated level value of the analyte level being monitored, based on sensor output from the analyte sensor, in accordance with the predictive calibration curve; and

causing medicament to be delivered to the patient with a portable infusion pump based on the estimated level value of the analyte being monitored.

2. The method as in claim 1, wherein the transformation function is determined based on a plurality of all calibration curves utilized by the portable medical monitor system beginning at a system start time.

3. The method as in claim 2, wherein the transformation function is determined based on less than all calibration curves utilized by the portable medical monitor system from the system start time through a system time.

4. The method as in claim 2, wherein determining comprises determining a transformation function based on the current calibration curve and an immediately preceding calibration curve such that the determined transformation function substantially produces the current calibration curve when applied to the immediately preceding calibration curve.

5. The method as in claim 1, wherein the transformation function produces a linear regression function of the current calibration curve and the at least one preceding calibration curve.

6. The method as in claim 1, wherein the transformation function produces a polynomial regression function of the current calibration curve and the at least one preceding calibration curve.

7. The method as in claim 1, wherein the at least one preceding calibration curve comprises a default calibration curve that is not based on measured data values of the analyte level being monitored.

8. The method as in claim 1, wherein the portable medical monitor system and portable infusion pump comprise a portable insulin pump, and wherein causing medicament to be delivered to the patient includes actuating a pump mechanism.

9. The method as in claim 1, wherein the analyte being monitored is blood glucose.

10. The method as in claim 1, wherein causing medicament to be delivered to the patient includes transmitting the estimated level value to the portable infusion pump.

11. The method as in claim 1, wherein causing medicament to be delivered to the patient includes transmitting program instructions from the server computer to the portable medical monitor system to transmit the estimated level value to the portable infusion pump.

15. The portable medical monitor system as in claim 1, wherein the control processor determines the transformation function based on the current calibration curve and an immediately preceding calibration curve such that the determined transformation function substantially produces the current calibration curve when applied to the immediately preceding calibration curve.

12. A portable medical monitor system comprising:

an analyte sensor that provides a sensor output value;

a control processor that receives the sensor output value, wherein the processor is configured to execute program instructions stored in memory of the portable medical monitor system to generate a current calibration curve based on at least two measured values of an analyte level of a patient being monitored, to determine a transformation function based on the current calibration curve and at least one preceding calibration curve such that the transformation function produces a predictive calibration curve that predicts changes to the current calibration curve resulting from sensor drift since the acquisition of the at least two measured values used to generate the current calibration curve, and to generate an estimated analyte level value of the analyte being monitored, based on the sensor output value, in accordance with the predictive calibration curve, and

wherein the controller processor is further configured to cause medicament to be delivered to the patient with a portable infusion pump based on the estimated level value of the analyte being monitored.

13. The portable medical monitor system as in claim 12, wherein the control processor determines the transformation function based on a plurality of all calibration curves utilized by the portable medical monitor system beginning at a system start time.

14. The portable medical monitor system as in claim 13, wherein the transformation function is determined based on less than all calibration curves utilized by the portable medical monitor system from the system start time through a current system time.

16. The portable medical monitor system as in claim 12, wherein the transformation function produces a linear regression function of the current calibration curve and the at least one preceding calibration curve.

17. The portable medical monitor system as in claim 12, wherein the transformation function produces a polynomial regression function of the current calibration curve and the at least one preceding calibration curve.

18. The portable medical monitor system as in claim 12, wherein the at least one preceding calibration curve comprises a default calibration curve that is not based on measured data values of the glucose level being monitored.

19. The portable medical monitor system as in claim 12, wherein the portable medical monitor system and the portable infusion pump comprise an insulin pump, and wherein causing medicament to be delivered to the patient includes actuating a pump mechanism.

20. The portable medical monitor system as in claim 12, wherein the analyte being monitored is blood glucose.

21. The portable medical monitor system of claim 12, wherein causing medicament to be delivered to the patient includes transmitting the estimated level value to the portable infusion pump.

22. A method of configuring a portable medical monitor system, the method comprising:

establishing a network communication link at a server computer to the portable medical monitor system;

transmitting program instructions from the server computer to the portable medical monitor system over the network communication link, wherein the transmitted program instructions are configured to be stored in memory of the portable medical monitor system and executed to receive a sensor output value of an analyte sensor and generate a current calibration curve based on at least two measured values of an analyte level of a patient being monitored, to determine a transformation function based on the current calibration curve and at least one preceding calibration curve such that the transformation function produces a predictive calibration curve that predicts changes to the current calibration curve resulting from sensor drift since the acquisition of the at least two measured data values used to generate the current calibration curve, and to generate an estimated analyte level value of the analyte being monitored, based on the sensor output level, in accordance with the predictive calibration curve; and

causing medicament to be delivered to the patient with a portable infusion pump based on the estimated level value of the analyte being monitored.

23. The method as in claim 22, wherein the transformation function is determined based on a plurality of all calibration curves utilized by the portable medical monitor system beginning at a predetermined system start time.

24. The method as in claim 23, wherein the transformation function is determined based on less than all calibration curves utilized by the portable medical monitor system from the system start time through a system time.

25. The method as in claim 23, wherein determining comprises determining a transformation function based on the current calibration curve and an immediately preceding calibration curve such that the determined transformation function substantially produces the current calibration curve when applied to the immediately preceding calibration curve.

26. The method as in claim 22, wherein the transformation function produces a linear regression function of the current calibration curve and the at least one preceding calibration curve.

27. The method as in claim 22, wherein the transformation function produces a polynomial regression function of the current calibration curve and the at least one preceding calibration curve.

28. The method as in claim 22, wherein the at least one preceding calibration curve comprises a default calibration curve that is not based on measured data values of the glucose level being monitored.

29. The method as in claim 22, wherein the portable medical monitor system and the portable infusion pump comprise an insulin pump, and wherein causing medicament to be delivered to the patient includes actuating a pump mechanism.

30. The method as in claim 22, wherein the analyte being monitored is blood glucose.