This model uses a regression model to describe the deterministic measurement error and an autoregressive stochastic process model to describe sensor noise.
Parameters
- alpha_w1, alpha_w2, alpha_cc1, alpha_cc2: Coefficients of autoregressive model for noise and “common component”, respectively.
- sigma_2_w, sigma_2_cc: squared standard deviation (variance) of noise and “common component”, respectively.
- a0, a1, a2: Polynomial coefficients describing deviation of relative sensor error over time.
- b0, b1, b2: Polynomial coefficients describing deviation absolute sensor drift over time.
Source code
https://github.com/hpeuscher/loopinsight1/blob/master/src/core/sensors/Facchinetti2014.ts
Reference
Facchinetti, A.; Del Favero, S.; Sparacino, G.; Castle, J.R.; Ward, W.K.; Cobelli, C.: “Modeling the Glucose Sensor Error”
IEEE Transactions on Biomedical Engineering, Volume 61, No. 3, 2014
DOI: 10.1109/TBME.2013.2284023
See also
Dexcom G5 (Vettoretti et al. 2018), Dexcom G6 (Vettoretti et al. 2019)