Method and device for predicting physiological values
The invention relates generally to methods, systems, and devices for measuring the concentration of target analytes present in a biological system using a series of measurements obtained from a monitoring system and a Mixtures of Experts (MOE) algorithm. In one embodiment, the present invention describes a method for measuring blood glucose in a subject.
1. A method of calibrating an analyte monitoring device for use in measuring analyte amount or concentration in a biological system, and said method comprising
determining a calibration ratio (CalRatio) value, wherein
CalRatio
=
BG
cp
(
active
cp
+
offset
)
wherein BG cp is a blood glucose concentration at the calibration point, active cp is an active signal that corresponds to an electrochemical sensor signal at the calibration point, and offset was a constant value;
providing two or more ranges of CalRatio values;
identifying a range in which said determined CalRatio value falls;
employing an algorithm to predict further measurement values, the algorithm selected from one of a first algorithm and a second algorithm, the first algorithm comprising:
BG=w 1 BG 1 +w 2 BG 2 +w 3 BG 3
where
BG 1 =p 1 (time)+ q 1 (active)+ r 1 (signal)+ s 1 ( BG|cp )+ t 1
BG 2 =p 2 (time)+ q 2 (active)+ r 2 (signal)+ s 2 ( BG|cp )+ t 2
BG 3 =p 3 (time)+ q 3 (active)+ r 3 (signal)+ s 3 ( BG|cp )+ t 3
w
1
=
ⅇ
d
1
ⅇ
d
1
+
ⅇ
d
2
+
ⅇ
d
3
w
2
=
ⅇ
d
2
ⅇ
d
1
+
ⅇ
d
2
+
ⅇ
d
3
w
3
=
ⅇ
d
3
ⅇ
d
1
+
ⅇ
d
2
+
ⅇ
d
3
d 1 τ 1 (time)+β 1 (active)+γ 1 (signal)+δ 1 ( BG|cp )+ε 1
d 2 τ 2 (time)+β 2 (active)+γ 2 (signal)+δ 2 ( BG|cp )+ε 2
d 3 τ 3 (time)+β 3 (active)+γ 3 (signal)+δ 3 ( BG|cp )+ε 3
in which BG i is the analyte predicted, BG/cp is the blood glucose value at a calibration point, time is the elapsed time, ative is the active signal, signal being the calibrated signal, p i , q i , r i are coefficients, t i is a constant, e indicates an exponential function, d i is a parameter set usable to determine weightings w i , with τ i , β i , γ i , δ i , and ε i are constants, and the second equation comprising:
BG=w 1 BG 1 +w 2 BG 2 +w 3 BG 3
where
BG 1 =p 1 (time c )+ q 1 (active)+ r 1 (signal)+ s 1 ( BG|cp )+ t 1
BG 2 =p 2 (time c )+ q 2 (active)+ r 2 (signal)+ s 2 ( BG|cp )+ t 2
BG 3 =p 3 (time c )+ q 3 (active)+ r 3 (signal)+ s 3 ( BG|cp )+ t 3
w
1
=
ⅇ
d
1
ⅇ
d
1
+
ⅇ
d
2
+
ⅇ
d
3
w
2
=
ⅇ
d
2
ⅇ
d
1
+
ⅇ
d
2
+
ⅇ
d
3
w
3
=
ⅇ
d
3
ⅇ
d
1
+
ⅇ
d
2
+
ⅇ
d
3
d 1 τ 1 (time c )+β 1 (active)+γ 1 (signal)+δ 1 ( BG|cp )+ε 1
d 2 τ 2 (time c )+β 2 (active)+γ 2 (signal)+δ 2 ( BG|cp )+ε 2
d 3 τ 3 (time c )+β 3 (active)+γ 3 (signal)+δ 3 ( BG|cp )+ε 3
in which BG i is the analyte predicted, time c is the elapsed time since calibration, active is the active signal, signal is the calibrated signal, BG/cp is the blood glucose value at a calibration point, p i , q i , r i are coefficients, t i is a constant, e indicates an exponential function, d i is a parameter set usable to determine weightings w i , with τ i , β i , γ i , δ i , and ε i are constants, wherein each of said algorithms is optimized for performance in the identified range; and
generating further measurement values indicative of amount or concentration of analyte present in the biological system, said generating comprising obtaining a raw signal specifically related to analyte amount or concentration in the biological system and using said algorithm to correlate the raw signal with a measurement value.