Advanced calibration for analyte sensors

• US 10,335,075 B2
• Filed: 03/14/2013
• Issued: 07/02/2019
• Est. Priority Date: 03/14/2013
• Status: Active Grant

• Current Assignee: DexCom Incorporated
• Original Assignee: DexCom Incorporated
• Inventors: Vanslyke, Stephen J., Bhavaraju, Naresh C., Bohnett, Lucas, Garcia, Arturo, Kamath, Apurv Ullas, Pryor, Jack
• Primary Examiner(s): Quigley, Kyle R
• Application Number:
  US13/827,119
• Publication Number:
  US 20140278189A1
• Time in Patent Office: 2,301 Days
• CPC Class Codes: A61B 5/1495 Calibrating or testing of in-vivo probes
• CPC Class Codes: G16H 40/40 for the management of medical equipment or devices, e.g. scheduling maintenance or upgrades

Abstract

Systems and methods for processing sensor data and calibration of the sensors are provided. In some embodiments, the method for calibrating at least one sensor data point from an analyte sensor comprises receiving a priori calibration distribution information; receiving one or more real-time inputs that may influence calibration of the analyte sensor; forming a posteriori calibration distribution information based on the one or more real-time inputs; and converting, in real-time, at least one sensor data point calibrated sensor data based on the a posteriori calibration distribution information.

Claims

1. A method for improving blood glucose sensing using a glucose sensor, the method comprising:

receiving a priori calibration distribution information obtained prior to a sensor session, the a priori calibration distribution information including probability information about combinations of sensitivity and baseline;

receiving blood glucose sensor data from the glucose sensor during a sensing session for the glucose sensor, the blood glucose sensor data including at least one sensor data point; and

calibrating, during the sensor session, the at least one sensor data point without relying exclusively on factory calibration, assumptions of blood glucose accuracy, or reference data, wherein calibrating includes;

receiving one or more real-time inputs during the sensor session, each of the one or more real-time inputs being capable of influencing calibration of the glucose sensor;

forming a posteriori calibration distribution information using a Bayesian approach and based on the one or more real-time inputs;

converting, in real-time during the sensor session, the at least one sensor data point from the glucose sensor to calibrated sensor data based on the a posteriori calibration distribution information; and

displaying an indication of the calibrated sensor data for the glucose sensor.

2. The method of claim 1, wherein the a priori calibration distribution information comprises information from previous calibrations of a particular sensor session and/or information obtained prior to sensor insertion.

3. The method of claim 1, wherein the a priori calibration distribution information comprises probability distributions for sensitivity (m), sensitivity-related information, baseline (b), or baseline-related information.

4. The method of claim 1, wherein the a priori calibration distribution information comprises a priori guidance or validation ranges.

5. The method of claim 1, wherein the one or more real-time inputs comprise data received or determined since a previous calibration process.

6. The method of claim 5, wherein the one or more real-time inputs comprises at least one of:

internally-derived real-time data, externally-derived real-time data, and combinations of internally- and externally-derived real-time data.

7. The method of claim 1, wherein the one or more real-time inputs includes at least one type of information selected from the group consisting of:

stimulus signal output of sensor;

sensor data measured by the sensor indicative of a glucose concentration;

sensor data indicative of glucose rate-of-change;

temperature measurements;

sensor data from multi-electrode sensors;

sensor data generated by redundant sensors;

sensor data generated by one or more auxiliary sensors;

data representative of a pressure on sensor;

data generated by an accelerometer;

sensor diagnostic information;

impedance; and

certainty level.

8. The method of claim 1, wherein the one or more real-time inputs includes at least one type of information selected from the group consisting of:

glucose concentration information obtained from a reference monitor;

information related to meal;

insulin dosing time and amounts;

insulin estimates;

exercise;

sleep;

illness;

stress;

hydration; and

hormonal conditions.

9. The method of claim 1, wherein the one or more real-time inputs includes combinations of internally- and externally-derived real-time data including at least one type of information selected from the group consisting of:

information gathered from population based data;

glucose concentration of the host;

error at calibration or error in matched data pair;

site of sensor implantation specific relationships;

time since sensor manufacture;

exposure of sensor, while on shelf, to temperature, humidity, or external factors;

a measure of noise in a glucose concentration signal; and

a level of certainty.

10. The method of claim 1, further comprising determining a level of certainty associated with calibration information and/or calibrated sensor data.

11. The method of claim 1, wherein forming a posteriori calibration distribution information comprises at least one of:

an adjustment of the a priori calibration distribution information or a creation of a new range or distribution information based on the one or more real-time inputs.

12. The method of claim 11, wherein an adjustment of the a priori calibration distribution information comprises shifting, tightening, or loosening the a priori calibration distribution.

13. The method of claim 1, wherein the calibration distribution information is selected from the group consisting of:

sensitivity;

change in sensitivity;

rate of change of sensitivity;

baseline;

change in baseline, rate of change of baseline, baseline profile associated with the sensor;

sensitivity profile associated with the sensor;

linearity;

response time;

relationships between properties of the sensor;

relationships between particular stimulus signal output; and

patient specific relationships between sensor and sensitivity, baseline, drift, impedance, impedance/temperature relationship, site of sensor implantation.

14. The method of claim 1, further comprising providing output of calibrated sensor data.

15. The method of claim 1, wherein the method is implemented using a processor and a memory coupled to the processor.

16. A system for improving blood glucose sensing using a continuous glucose sensor, the system comprising sensor electronics configured to be operably connected to the continuous glucose sensor, the sensor electronics configured to:

receive a priori calibration distribution information obtained prior to a sensor session, the a priori calibration distribution information including probability information about combinations of sensitivity and baseline;

receive blood glucose sensor data from the glucose sensor during the sensing session for the glucose sensor, the blood glucose sensor data including at least one sensor data point; and

calibrate, during the sensor session, the at least one sensor data point without relying exclusively on factory calibration, assumption of blood glucose accuracy or reference data, wherein calibrate includes;

receive one or more real-time inputs during the sensing session, each of the one or more real-time inputs being capable of influencing calibration of the glucose sensor;

form a posteriori calibration distribution information using a Bayesian approach and using the one or more real-time inputs;

convert, in real-time during the sensor session, the at least one sensor data point to calibrated sensor data based on the a posteriori calibration distribution information; and

display an indication of the calibrated sensor data for the glucose sensor.

17. The system of claim 16, wherein the sensor electronics comprise a processor module, the processor module comprising instructions stored in computer memory, wherein the instructions, when executed by the processor module, cause the sensor electronics to form the a posteriori calibration distribution information.

19. The system of claim 17, wherein the instructions, when executed by the processor module, cause the sensor electronics to adjust the a priori calibration distribution information to form the a posteriori calibration distribution information.

20. The system of claim 17, wherein the instructions, when executed by the processor module, cause the sensor electronics to form the a posteriori calibration distribution information by creating a new range or distribution information based on the one or more real-time inputs.

18. A system for improving blood glucose sensing using a glucose sensor, the system comprising:

means for receiving a priori calibration distribution information obtained prior to a sensor session, the a priori calibration distribution information including probability information about combinations of sensitivity and baseline;

means for receiving blood glucose sensor data from the glucose sensor during the sensing session for the glucose sensor, the blood glucose sensor data including at least one sensor data point; and

means for calibrating, during the sensor session, the at least one sensor data point without relying exclusively on factory calibration, assumption of blood glucose accuracy or reference data, wherein the means for calibrating includes;

means for receiving one or more real-time inputs during the sensing session, each of the one or more real-time inputs being capable of influencing calibration of the glucose sensor;

means for forming a posteriori calibration distribution information using a Bayesian approach and based on the one or more real-time inputs;

means for converting, in real-time during the sensor session, the at least one sensor data point to calibrated sensor data based on the a posteriori calibration distribution information; and

means for displaying an indication of the calibrated sensor data for the glucose sensor.