Sensor data calibration based on weighted values
Disclosed herein are techniques related to sensor data calibration based on weighted values. In some embodiments, the techniques involve obtaining a plurality of sample values of an electrical signal generated by a sensor. Each sample value may be indicative of a concentration of an analyte in a first type of bodily fluid. Additionally, the techniques may involve obtaining a plurality of reference measurement values (RMVs). Each RMV may correspond to a concentration of the analyte in a second type of bodily fluid. The techniques may also involve temporally associating a first sample value to a first RMV and a second sample value to a second RMV, which is greater than the first RMV. The techniques may further involve applying a first weighting to the first sample value and a second weighting to the second sample value. The first weighting may be greater than the second weighting.
1 . A processor-implemented method for calibration, the method comprising:
obtaining a plurality of sample values of an electrical signal generated by a sensor, each sample value of the plurality of sample values being indicative of a concentration of an analyte in a first type of bodily fluid of a patient;
obtaining a plurality of reference measurement values, each reference measurement value of the plurality of reference measurement values corresponding to a concentration of the analyte in a second type of bodily fluid of the patient;
temporally associating a first sample value of the plurality of sample values to a first reference measurement value of the plurality of reference measurement values;
temporally associating a second sample value of the plurality of sample values to a second reference measurement value of the plurality of reference measurement values, the second reference measurement value being greater than the first reference measurement value;
applying a first weighting to the first sample value to generate a first weighted sample value;
applying a second weighting to the second sample value to generate a second weighted sample value, wherein the first weighting is greater than the second weighting;
determining a calibration factor based, at least in part, on the first weighted sample value and the second weighted sample value; and
calibrating a value of the electrical signal to correspond to an estimated concentration of the analyte in the second type of bodily fluid based on applying the calibration factor to the value of the electrical signal.
2 . The method of claim 1 , wherein the first type of bodily fluid is interstitial fluid.
3 . The method of claim 1 , wherein the second type of bodily fluid is blood.
4 . The method of claim 1 , wherein the analyte is glucose.
5 . The method of claim 1 , wherein determining the calibration factor comprises determining a linear regression based on at least the first weighted sample value, the second weighted sample value, the first reference measurement value, and the second reference measurement value.
6 . The method of claim 1 , wherein determining the calibration factor comprises:
determining a linear regression sensitivity ratio based, at least in part, on the first weighted sample value, the second weighted sample value, the first reference measurement value and the second reference measurement value;
selecting an offset based, at least in part, on the determined linear regression sensitivity ratio; and
determining a modified linear regression sensitivity ratio based, at least in part, on the selected offset, the first weighted sample value, the second weighted sample value, the first reference measurement value, and the second reference measurement value.
7 . The method of claim 1 , wherein determining the calibration factor comprises determining a linear relationship based on at least the first weighted sample value and the second weighted sample value.
8 . A system comprising:
one or more processors; and
one or more processor-readable media storing instructions which, when executed by the one or more processors, cause performance of:
obtaining a plurality of sample values of an electrical signal generated by a sensor, each sample value of the plurality of sample values being indicative of a concentration of an analyte in a first type of bodily fluid of a patient;
obtaining a plurality of reference measurement values, each reference measurement value of the plurality of reference measurement values corresponding to a concentration of the analyte in a second type of bodily fluid of the patient;
temporally associating a first sample value of the plurality of sample values to a first reference measurement value of the plurality of reference measurement values;
temporally associating a second sample value of the plurality of sample values to a second reference measurement value of the plurality of reference measurement values, the second reference measurement value being greater than the first reference measurement value;
applying a first weighting to the first sample value to generate a first weighted sample value;
applying a second weighting to the second sample value to generate a second weighted sample value, wherein the first weighting is greater than the second weighting;
determining a calibration factor based, at least in part, on the first weighted sample value and the second weighted sample value; and
calibrating a value of the electrical signal to correspond to an estimated concentration of the analyte in the second type of bodily fluid based on applying the calibration factor to the value of the electrical signal.
9 . The system of claim 8 , wherein the first type of bodily fluid is interstitial fluid.
10 . The system of claim 8 , wherein the second type of bodily fluid is blood.
11 . The system of claim 8 , wherein the analyte is glucose.
12 . The system of claim 8 , wherein determining the calibration factor comprises determining a linear regression of at least the first weighted sample value, the second weighted sample value, the first reference measurement value, and the second reference measurement value.
13 . The system of claim 8 , wherein determining the calibration factor comprises:
determining a linear regression sensitivity ratio based, at least in part, on the first weighted sample value, the second weighted sample value, the first reference measurement value and the second reference measurement value;
selecting an offset based, at least in part, on the determined linear regression sensitivity ratio; and
determining a modified linear regression sensitivity ratio based, at least in part, on the selected offset, the first weighted sample value, the second weighted sample value, the first reference measurement value, and the second reference measurement value.
14 . The system of claim 8 , wherein determining the calibration factor comprises determining a linear relationship based on at least the first weighted sample value and the second weighted sample value.
15 . One or more non-transitory processor-readable media storing instructions, which, when executed by one or more processors, cause performance of:
obtaining a plurality of sample values of an electrical signal generated by a sensor, each sample value of the plurality of sample values being indicative of a concentration of an analyte in a first type of bodily fluid of a patient;
obtaining a plurality of reference measurement values, each reference measurement value of the plurality of reference measurement values corresponding to a concentration of the analyte in a second type of bodily fluid of the patient;
temporally associating a first sample value of the plurality of sample values to a first reference measurement value of the plurality of reference measurement values;
temporally associating a second sample value of the plurality of sample values to a second reference measurement value of the plurality of reference measurement values, the second reference measurement value being greater than the first reference measurement value;
applying a first weighting to the first sample value to generate a first weighted sample value;
applying a second weighting to the second sample value to generate a second weighted sample value, wherein the first weighting is greater than the second weighting;
determining a calibration factor based, at least in part, on the first weighted sample value and the second weighted sample value; and
calibrating a value of the electrical signal to correspond to an estimated concentration of the analyte in the second type of bodily fluid based on applying the calibration factor to the value of the electrical signal.
16 . The one or more non-transitory processor-readable media of claim 15 , wherein the first type of bodily fluid is interstitial fluid.
17 . The one or more non-transitory processor-readable media of claim 15 , wherein the second type of bodily fluid is blood.
18 . The one or more non-transitory processor-readable media of claim 15 , wherein determining the calibration factor comprises determining a linear regression of at least the first weighted sample value, the second weighted sample value, the first reference measurement value, and the second reference measurement value.
19 . The one or more non-transitory processor-readable media of claim 15 , wherein determining the calibration factor comprises:
determining a linear regression sensitivity ratio based, at least in part, on the first weighted sample value, the second weighted sample value, the first reference measurement value and the second reference measurement value;
selecting an offset based, at least in part, on the determined linear regression sensitivity ratio; and
determining a modified linear regression sensitivity ratio based, at least in part, on the selected offset, the first weighted sample value, the second weighted sample value, the first reference measurement value, and the second reference measurement value.
20 . The one or more non-transitory processor-readable media of claim 15 , wherein determining the calibration factor comprises determining a linear relationship based on at least the first weighted sample value and the second weighted sample value.