Systems and methods for processing sensor data
Systems and methods for processing sensor data are provided. In some embodiments, systems and methods are provided for calibration of a continuous analyte sensor. In some embodiments, systems and methods are provided for classification of a level of noise on a sensor signal. In some embodiments, systems and methods are provided for determining a rate of change for analyte concentration based on a continuous sensor signal. In some embodiments, systems and methods for alerting or alarming a patient based on prediction of glucose concentration are provided.
1. A system for processing continuous analyte sensor data, comprising:
a continuous analyte sensor configured to continuously measure a concentration of an analyte in a host;
a computer system configured to:
receive sensor data from the continuous analyte sensor;
calibrate the sensor data using calibration information, derived from an existing calibration set that includes a plurality of matched data pairs, each matched data pair including an analyte sensor signal and a corresponding analyte concentration;
determine a predictive accuracy of calibrated sensor data, the predictive accuracy measuring an accuracy of a prediction of a newly received matched data pair based on the calibrated sensor data;
compare a determined predictive accuracy with one or more accuracy level thresholds to produce an accuracy output; and
an insulin delivery device configured to control insulin delivery according to instructions that are based at least in part on the accuracy output.
2. The system of claim 1 , wherein the one or more accuracy level thresholds correspond to at least two predetermined ranges of accuracy.
3. The system of claim 1 , wherein the one or more accuracy level thresholds correspond to at least three predetermined ranges of accuracy.
4. The system of claim 1 , wherein the computer system is further configured to determine an additional predictive accuracy based on an evaluation of the calibration information.
5. The system of claim 1 , wherein the computer system is further configured to determine an additional predictive accuracy based on drift of the analyte sensor signal.
6. The system of claim 1 , wherein the computer system is further configured to determine an additional predictive accuracy based on a difference between the calibrated sensor data and corresponding reference data.
7. The system of claim 1 , wherein the computer system is further configured to control communication of the instructions that control the insulin delivery to a user interface for display based on the accuracy output.
8. The system of claim 1 , wherein the computer system is further configured to control transmission, to the insulin delivery device, of the instructions that control the insulin delivery based on the accuracy output.
9. The system of claim 1 , wherein the computer system is further configured to control communication, to the insulin delivery device, of the instructions that control the insulin delivery by determining when to trigger a fail-safe in a closed loop insulin delivery system based on the accuracy output.
10. The system of claim 1 , wherein the computer system is further configured to control communication, to the insulin delivery device, of the instructions that control the insulin delivery by requesting user interaction in a closed loop insulin delivery system based on the accuracy output.
11. The system of claim 1 , wherein the computer system is further configured to determine an additional predictive accuracy based on evaluating an accordance or discordance of the calibration information.
12. A method for processing continuous analyte sensor data, comprising:
receiving sensor data from a continuous analyte sensor;
calibrating the sensor data using calibration information, derived from a first calibration set that includes a plurality of matched data pairs, each matched data pair including an analyte sensor signal and a corresponding analyte concentration;
determining a predictive accuracy of calibrated sensor data, by evaluating a change from a first calibration line to a second calibration line, the first calibration line formed from the first calibration set and the second calibration line formed from a second calibration set that includes a newly received matched data pair not included in the first calibration set;
comparing a determined predictive accuracy with one or more accuracy level thresholds to produce an accuracy output; and
controlling an insulin delivery device using instructions that are based at least in part on the accuracy output.
13. The method of claim 12 , wherein the one or more accuracy level thresholds correspond to at least two predetermined ranges of accuracy.
14. The method of claim 12 , wherein the one or more accuracy level thresholds correspond to at least three predetermined ranges of accuracy.
15. The method of claim 12 , further comprising determining an additional predictive accuracy based on evaluating the calibration information.
16. The method of claim 12 , further comprising determining an additional predictive accuracy based on a drift of the analyte sensor signal.
17. The method of claim 12 , further comprising determining an additional predictive accuracy based on a difference between the calibrated sensor data and corresponding reference data.
18. The method of claim 12 , wherein determining the predictive accuracy is performed on the calibrated sensor data in units of blood glucose concentration.
19. The method of claim 12 , wherein the insulin delivery device is an insulin pump.
20. The method of claim 12 , wherein controlling the insulin delivery device using the instructions includes displaying, on a user interface, the instructions.
21. The method of claim 12 , wherein controlling the insulin delivery device using the instructions includes transmitting the instructions to the insulin delivery device.
22. The method of claim 12 , wherein controlling the insulin delivery device using the instructions includes determining when to trigger a fail-safe in a closed loop insulin delivery system.
23. The method of claim 12 , wherein controlling the insulin delivery device using the instructions includes requesting user interaction in a closed loop insulin delivery system.
24. The method of claim 12 , further comprising determining an additional predictive accuracy based on evaluating an accordance or discordance of the calibration information.