Background and Aims

The prediction of glycemic profiles based on continuous glucose monitoring contributes to enhance diabetes management, helping to prevent hyper and hypoglycemic events or integrated in artificial pancreas. This paper proposes a new set of parameters to assess glucose predictors by integrating accuracy, delay and noise parameters.

Methods

The new metrics complement the frequently used root mean square error (RMSE), with the following: the delay-compensated RMSE (RMSE_DC), the delay-compensated RMSE into the 25-75% of the extremes (RMSE_DC-25/75), the overshooting/undershooting (OS/US); the mean delay (MD_25/75), the peak delay (PD); and the noise energy ratio (ENOISE_norm). We have applied the parameters to 5 synthetic test-profiles (TP) that emulate prediction profiles. TP_1-2 are built by adding a constant delay (15&10 min respectively) to an original CGM profile, and artificial deformations at the local maxima (TP_2). TP_3-4-5 correspond to a 10 min delayed CGM profile, adding differet types of noise (20dBW WGN in TP_3, plus extra high&low frequency noise in TP_4-5 respectively).

Figure. Synthetic test profiles

Results

The RMSE is not able to discriminate among TP_1-2-3, having clearly distinct performance (Figure 1) (RMSE:13.0-12.2-13.9), while the proposed metrics reveal their differences (RMSE_DC:0-7.0-8.9, RMSE_DC-25/75:0-0-9.4). MD_25/75 detects the delay in 4 out of 5 TPs and PD reveals if that delay is homogeneous. OS/US analysis allows a critical insight about the absence of false-alarms/false-negatives. ENOISE_norm reveals the presence of an extra medium-low-frequency noise in TP_4-5 vs. TP_3 (ENOISE_norm:4.27-8.09-9.85).

Conclusions

The proposed metrics allow an in-depth characterization of glucose predictors, helpint to select the optimum prediction technique depending on the specific application.