Explainable Machine Learning for Real-Time Hypoglycemia and Hyperglycemia Prediction and Personalized Control Recommendations
Authors' conclusion
Maximizing model performance for glucose risk prediction and management is crucial to reduce the burden of alarm fatigue on CGM users. Machine learning enables more precise and timely predictions in comparison with baseline models. SHAP helps identify what about a CGM user's glucose control has led to predictions of risk which can be used to reduce their long-term risk of complications.
Study context
- Population
- People living with T1D in the CITY survey; Continuous glucose monitoring (CGM) data totaling more than 28,000 days of usage>28,000 days of CGM
- Participants
- 153 people
- Comparison
- XGBoost models vs baseline heuristic and logistic regression model
- Outcome
- Prediction of hypoglycemia (<70 mg/dL) and hyperglycemia (>270 mg/dL)
- Follow-up
- Up to 60 minutes ahead of time
Methodological detailsexplainable machine learning (SHAP), XGBoost
- Explainable machine learning (SHAP)
- XGBoost
Performance
As reported in the abstract
XGBoost hypoglycemia AUROC
0.998
XGBoost hypoglycemia average precision
0.953
XGBoost hyperglycemia AUROC
0.989
XGBoost hyperglycemia average precision
0.931
Original publication
Explainable Machine Learning for Real-Time Hypoglycemia and Hyperglycemia Prediction and Personalized Control Recommendations. J Diabetes Sci Technol. Published 13 June 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →