Key result
A gradient-boosted machine model improved the detection of pressure-induced sensor attenuations in continuous glucose monitors compared to a previously reported rules-based algorithm.
Why the study?
Continuous glucose monitors are prone to pressure-induced sensor attenuations that cause false low readings, leading to undesirable pump shutoffs and increased hyperglycemia risk.
Do machine learning methods improve the detection of pressure-induced sensor attenuations in continuous glucose monitors compared to a rules-based algorithm?
Do machine learning methods improve the detection of pressure-induced sensor attenuations in continuous glucose monitors compared to a rules-based algorithm?
Machine learning models, particularly gradient-boosted machines, can effectively detect pressure-induced sensor attenuations in continuous glucose monitors, potentially reducing false hypoglycemia alerts and undesirable pump shutoffs.
Machine learning improves PISA detection in CGMs; extends rules-based methods but leaves open effects on false alerts and clinical outcomes.
Continuous glucose monitors (CGMs) are prone to faults termed pressure-induced sensor attenuations (PISAs), particularly when the user rolls over on the sensor during sleep. PISAs result in false, low blood glucose readings, leading to undesirable pump shutoffs and an increased risk of hyperglycemia. Data from an outpatient trial with PISA glucose readings labeled for 1125 nights was used. Machine learning methods such as decision trees, support vector machines, neural networks, random forest, and gradient-boosted machine models are compared against each other and with a previously reported rules-based algorithm for PISA detection on the same data set. The best-performing gradient-boosted machine model is further improved using the developed start-PISA model. Model interpretation methods are used to confirm that PISA behavior is well-captured and provides insights into decision-making.
No takes yet. Share an insight, caveat, or question.
Navarathna et al. (2023) studied this question. Machine learning models vs. Previously reported rules-based algorithm was evaluated on Detection of pressure-induced sensor attenuations (PISAs). A gradient-boosted machine model improved the detection of pressure-induced sensor attenuations in continuous glucose monitors compared to a previously reported rules-based algorithm.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: