Prospective Detection of Hypoglycemia and Near-Hypoglycemia Inducing Pressure-Induced Sensor Attenuation Onset in Continuous Glucose Monitoring Time Series
Authors' conclusion
The results indicate that the PA could feasibly be used to inform other decision-making processes, for example, canceling of hypoglycemia alarms that would otherwise be triggered.
Study context
- Population
- Continuous glucose monitoring (CGM) time series with pressure-induced sensor attenuations (PISAs); testing set of 325 PISAs (26,890 hours of data, 33 patients)testing set
- Participants
- 33 participants
- Outcome
- Area under the precision-recall (PR) curve for detection of PISA onset
Methodological detailsalgorithm development with separate training, validation and testing data sets, probability estimation module (PEM) trained on 846 PISAs (59 122 hours of data, 67 patients), Optimal prospective algorithm (PA) version selected on 89 PISAs (8189 hours of data, 11 patients)
- Algorithm development with separate training, validation and testing data sets
- PEM trained on 846 PISAs (59 122 hours of data, 67 patients)
- Optimal PA version selected on 89 PISAs (8189 hours of data, 11 patients)
Performance
As reported in the abstract
Area under the PR curve (testing set)
| Test or model | AUC (precision-recall (PR) curve) |
|---|---|
| PA | 0.590 (95% CI 0.509, 0.672) |
PA
AUC (precision-recall (PR) curve)0.590 (95% CI 0.509, 0.672)
- Baseline prevalence of PISA onset in the testing set
- 1.2%
Original publication
Prospective Detection of Hypoglycemia and Near-Hypoglycemia Inducing Pressure-Induced Sensor Attenuation Onset in Continuous Glucose Monitoring Time Series. J Diabetes Sci Technol. Published 18 August 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →