Page 201 - Glucose Monitoring Devices

204 CHAPTER 10 CGM filtering and denoising techniques

FIGURE 10.1
Two representative CGM time series measured for 24 h, at 3 min sampling rate, in two
diabetic subjects by Glucoday. Top: low SNR condition. Bottom: high SNR condition.
Taken from Facchinetti A, Sparacino G, Cobelli C. An online self-tunable method to denoise CGM sensor data.
IEEE Transactions on Biomedical Engineering 2010;57(3):634e641.

The denoising problem
In formal terms, if we consider the following equation

yðtÞ¼ uðtÞþ vðtÞ (10.1)
where y(t) is the glucose level measured at time t by the CGM sensor, u(t) is the true,
unknown, glucose level, and v(t) is the random noise affecting it, supposed to be