Use of electrochemical impedance spectroscopy (EIS) in gross failure analysis
Electrochemical Impedance Spectroscopy (EIS) is used in conjunction with continuous glucose monitors and continuous glucose monitoring (CGM) to enable in-vivo sensor calibration, gross (sensor) failure analysis, and intelligent sensor diagnostics and fault detection. An equivalent circuit model is defined, and circuit elements are used to characterize sensor behavior.
1. A method for determining a permanent sensitivity loss in a glucose sensor having sensor electronics, a microcontroller, at least one working electrode, and a counter electrode, the method comprising:
causing, by the microcontroller, an AC potential signal to be applied to the glucose sensor;
measuring, by the sensor electronics, an AC current signal at the at least one working electrode;
performing, by the microcontroller, an electrochemical impedance spectroscopy (EIS) procedure for the at least one working electrode based on the AC potential signal and the AC current signal;
generating, by the microcontroller, a Nyquist plot based on output of the EIS procedure;
monitoring, by the microcontroller, a voltage (Vcntr) across the counter electrode;
detecting, by the microcontroller, that the Vcntr reaches a rail voltage over a predetermined time duration; and
in a case where the Vcntr is detected to reach the rail voltage over the predetermined time duration, determining whether or not there is a permanent sensitivity loss based on a parallel shift in the Nyquist plot.
2. The method of claim 1 , wherein the parallel shift in the Nyquist plot is detected based on one or more of membrane resistance (Rmem), solution resistance (Rsol), and membrane capacitance (Cmem).
3. The method of claim 2 , wherein the Nyquist plot includes a semicircle joined with a straight line at an inflection point.
4. The method of claim 3 , wherein the semicircle is an indicative of a change in Rmem.
5. The method of claim 3 , wherein a sum of Rmem and Rsol corresponds to the inflection point.
6. The method of claim 5 , wherein the sum is estimated by localizing the inflection point by detecting changes in directionality of a Nyquist slope of the Nyquist plot.
7. The method of claim 3 , wherein a change in Cmem is detected by a presence of the semicircle.
8. The method of claim 3 , wherein Cmem is estimated by tracking an imaginary component of impedance in a high frequency range.
9. The method of claim 8 , wherein the high frequency range is from 1 kHz to 8 kHz.
10. The method of claim 3 , wherein Cmem is estimated by tracking a highest point in the semicircle in a high frequency region.
11. The method of claim 3 , wherein Cmem is estimated by measuring a Cartesian distance between two higher-frequency points in the Nyquist plot.
12. A glucose sensor comprising:
at least one working electrode;
a counter electrode;
sensor electronics configured measure an AC current signal at the at least one working electrode and monitor a voltage (Vcntr) across the counter electrode; and
a microcontroller configured to:
cause an AC potential signal to be applied the glucose sensor;
perform an electrochemical impedance spectroscopy (EIS) procedure for the at least one working electrode based on the AC potential signal and the AC current signal;
generate a Nyquist plot based on output of the EIS procedure;
detect that the Vcntr reaches a rail voltage over a predetermined time duration; and
in a case where the Vcntr is detected to reach the rail voltage over the predetermined time duration, determine whether or not there is a permanent sensitivity loss based on a parallel shift in the Nyquist plot.
13. The glucose sensor of claim 12 , wherein the parallel shift in the Nyquist plot is detected based on one or more of membrane resistance (Rmem), solution resistance (Rsol), or membrane capacitance (Cmem).
14. The glucose sensor of claim 13 , wherein the Nyquist plot includes a semicircle joined with a straight line at an inflection point.
15. The glucose sensor of claim 14 , wherein the semicircle is an indicative of a change in Rmem.
16. The glucose sensor of claim 14 , wherein a sum of Rmem and Rsol corresponds to the inflection point.
17. The glucose sensor of claim 16 , wherein the microprocessor is further configured to estimate the sum by localizing the inflection point by detecting changes in directionality of a Nyquist slope of the Nyquist plot.
18. The glucose sensor of claim 14 , wherein a change in Cmem is detected by a presence of the semicircle.
19. The glucose sensor of claim 14 , wherein Cmem is estimated by tracking a higher-frequency imaginary component of impedance.
20. The glucose sensor of claim 14 , wherein Cmem is estimated by tracking a highest point in the semicircle in a high frequency region.