Prediction of Insulin Requirements by Explainable Machine Learning for Individuals With Type 1 Diabetes
Authors' conclusion
We here developed machine-learning models that predict the TDD of insulin from clinical information. Such models could contribute to the treatment of many individuals undergoing insulin therapy, with further developments being warranted.
Study context
- Population
- Individuals with T1D using CGM with CSII, Kobe University Hospital, April 1, 2010 to February 29, 2024
- Participants
- 110 individuals
- Outcome
- Performance of machine-learning models predicting total daily dose (TDD) of insulin, by mean absolute percentage error (MAPE)
Methodological detailssingle-centre, machine learning, explainable artificial intelligence
- Single-centre
- Machine learning
- Explainable artificial intelligence
Main finding
As reported in the abstract
MAPE for TDD of insulin prediction
| Group | Value | Mean absolute percentage error (MAPE) |
|---|---|---|
| Random forest | 19.8% | 19.8% |
Random forest
Value19.8%
Mean absolute percentage error (MAPE)19.8%
Observational data: associations, not effects.
Original publication
Prediction of Insulin Requirements by Explainable Machine Learning for Individuals With Type 1 Diabetes. J Clin Endocrinol Metab. Published 11 December 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →