Reducing Exercise-Related Hypoglycemia in Automated Insulin Delivery with Reinforcement Learning


Loading...

Date

2024

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric

Data

Abstract

Exercise is an important component for glucose management in type 1 diabetes, but remains challenging for automated insulin delivery systems as altered glucose dynamics are difficult to model explicitly. Glucose monitoring data might enable data-driven approaches for learning these dynamics implicitly. We propose combining model predictive control with a reinforcement learning component to adjust basal insulin infusion rates for exercise. We train our model on a variety of exercise scenarios and demonstrate improved glucose control using a generically trained model. We further improve performance by personalizing the model with a small number of additional individual-specific training episodes.

Publication status

published

Book title

12th IFAC Symposium on Biological and Medical Systems (BMS 2024)

Volume

58 (24)

Pages / Article No.

239 - 244

Publisher

Elsevier

Event

12th IFAC Symposium on Biological and Medical Systems (BMS 2024)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Diabetes management; Decision support; Exercise; Hypoglycemia; Automated insulin delivery; Reinforcement learning

Organisational unit

03699 - Stelling, Jörg / Stelling, Jörg check_circle

Notes

Conference presentation on September 12, 2024

Funding

Related publications and datasets