Combined Use of Glucose-Specific Model Identification and Alarm Strategy Based on Prediction-Funnel to Improve Online Forecasting of Hypoglycemic Events
Authors' conclusion
The combined use of a glucose-specific metric and an alarm-raising strategy based on the prediction-funnel allows achieving a more effective and reliable hypoglycemia prediction algorithm.
Study context
- Population
- Individuals with T1D (real dataset)
- Participants
- 11 individuals
- Comparison
- gMSE cost function plus prediction-funnel alarm strategy vs conventional approach (mean square error (MSE) and single prediction horizon)
- Outcome
- Hypoglycemia prediction performance
Methodological detailsARIMAX models, real dataset
- ARIMAX models
- Real dataset
Main finding
As reported in the abstract
| Group | Precision | Recall | F1-score | False positives per day | Mean time gain |
|---|---|---|---|---|---|
| gMSE + prediction-funnel | 65% | 88% | 75% | 0.29 | 15 minutes |
gMSE + prediction-funnel
Precision65%
Recall88%
F1-score75%
False positives per day0.29
Mean time gain15 minutes
Original publication
Combined Use of Glucose-Specific Model Identification and Alarm Strategy Based on Prediction-Funnel to Improve Online Forecasting of Hypoglycemic Events. J Diabetes Sci Technol. Published 24 May 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →