Proof-of-concept Study of Using Supervised Machine Learning Algorithms to Predict Self-care and Glycemic Control in Type 1 Diabetes Patients on Insulin Pump Therapy

Elsevier

Available online 8 March 2023

Endocrine PracticeAuthor links open overlay panel, , , , , , , Highlights•

This proof-of-concept study attempted to illustrate how two predictive models (insulin pump candidacy and glycemic control within six months) can be developed using supervised machine learning algorithms instead of more classical methods.

Important predictors for meeting insulin pump self-management behavioral (IPSMB) criteria were baseline HbA1c, sex, and prior continuous glucose monitoring use.

Predictors for achieving good glycemic control were baseline HbA1c, entering calculated carbohydrate into the pump, and following insulin bolus dose calculated by the pump.

The non-linear RF model outperformed the other algorithms for both outcomes.

Abstract:Objective

Using supervised machine learning algorithms (SMLAs), we built models to predict the probability of type 1 diabetes mellitus (T1DM) patients on insulin pump therapy for meeting insulin pump self-management behavioral (IPSMB) criteria and achieving good glycemic response within six months.

Methods

This was a single-center retrospective chart review of 100 adult T1DM patients on insulin pump therapy (>6 months). Three SMLAs were deployed: multivariable logistic regression (LR), random forest (RF), and K-nearest neighbor (k-NN); validated using repeated three-fold cross-validation. Performance metrics included AUC-ROC for discrimination and Brier scores for calibration.

Results

Variables predictive of adherence with IPSMB criteria were baseline HbA1c, continuous glucose monitoring (CGM), and sex. The models had comparable discriminatory power (LR=0.74; RF=0.74; k-NN=0.72), with the random forest model showing better calibration (Brier=0.151). Predictors of the good glycemic response included baseline HbA1c, entering carbohydrates, and following the recommended bolus dose, with models comparable in discriminatory power (LR=0.81, RF=0.80, k-NN=0.78) but the random forest model being better calibrated (Brier=0.099).

Conclusion

These proof-of-concept analyses demonstrate the feasibility of using SMLAs to develop clinically relevant predictive models of adherence with IPSMB criteria and glycemic control within six months. Subject to further study, non-linear prediction models may perform better.

Keywords

insulin pump

Diabetes type 1

behaviors

self-care

machine learning

prediction models

View full text

© 2023 Published by Elsevier Inc. on behalf of the AACE.

留言 (0)

沒有登入
gif