Closed-loop control with unannounced exercise for adults with type 1 diabetes using the Ensemble Model Predictive Control

Jose Garcia-Tirado, John P. Corbett, Dimitri Boiroux, John Bagterp Jørgensen, Marc D. Breton*

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

This paper presents an individualized Ensemble Model Predictive Control (EnMPC) algorithm for blood glucose (BG) stabilization and hypoglycemia prevention in people with type 1 diabetes (T1D) who exercise regularly. The EnMPC formulation can be regarded as a simplified multi-stage MPC allowing for the consideration of Nen scenarios gathered from the patient's recent behavior. The patient's physical activity behavior is characterized by an exercise-specific input signal derived from the deconvolution of the patient's continuous glucose monitor (CGM), accounting for known inputs such as meal, and insulin pump records. The EnMPC controller was tested in a cohort of in silico patients with representative inter-subject and intra-subject variability from the FDA-accepted UVA/Padova simulation platform. Results show a significant improvement on hypoglycemia prevention after 30 min of mild to moderate exercise in comparison to a similarly tuned baseline controller (rMPC); with a reduction in hypoglycemia occurrences (<70 mg/dL), from 3.08 % ±3.55 with rMPC to 0.78 % ±2.04 with EnMPC (P < 0.05).

Original languageEnglish
JournalJournal of Process Control
Volume80
Pages (from-to)202-210
ISSN0959-1524
DOIs
Publication statusPublished - 1 Aug 2019

Keywords

  • Artificial pancreas
  • Exercise
  • Hypoglycemia
  • Model Predictive Control
  • Type 1 diabetes

Cite this

@article{f9e074b5eb924720be9772cfb2d82168,
title = "Closed-loop control with unannounced exercise for adults with type 1 diabetes using the Ensemble Model Predictive Control",
abstract = "This paper presents an individualized Ensemble Model Predictive Control (EnMPC) algorithm for blood glucose (BG) stabilization and hypoglycemia prevention in people with type 1 diabetes (T1D) who exercise regularly. The EnMPC formulation can be regarded as a simplified multi-stage MPC allowing for the consideration of Nen scenarios gathered from the patient's recent behavior. The patient's physical activity behavior is characterized by an exercise-specific input signal derived from the deconvolution of the patient's continuous glucose monitor (CGM), accounting for known inputs such as meal, and insulin pump records. The EnMPC controller was tested in a cohort of in silico patients with representative inter-subject and intra-subject variability from the FDA-accepted UVA/Padova simulation platform. Results show a significant improvement on hypoglycemia prevention after 30 min of mild to moderate exercise in comparison to a similarly tuned baseline controller (rMPC); with a reduction in hypoglycemia occurrences (<70 mg/dL), from 3.08 {\%} ±3.55 with rMPC to 0.78 {\%} ±2.04 with EnMPC (P < 0.05).",
keywords = "Artificial pancreas, Exercise, Hypoglycemia, Model Predictive Control, Type 1 diabetes",
author = "Jose Garcia-Tirado and Corbett, {John P.} and Dimitri Boiroux and J{\o}rgensen, {John Bagterp} and Breton, {Marc D.}",
year = "2019",
month = "8",
day = "1",
doi = "10.1016/j.jprocont.2019.05.017",
language = "English",
volume = "80",
pages = "202--210",
journal = "Journal of Process Control",
issn = "0959-1524",
publisher = "Elsevier",

}

Closed-loop control with unannounced exercise for adults with type 1 diabetes using the Ensemble Model Predictive Control. / Garcia-Tirado, Jose; Corbett, John P.; Boiroux, Dimitri; Jørgensen, John Bagterp; Breton, Marc D.

In: Journal of Process Control, Vol. 80, 01.08.2019, p. 202-210.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Closed-loop control with unannounced exercise for adults with type 1 diabetes using the Ensemble Model Predictive Control

AU - Garcia-Tirado, Jose

AU - Corbett, John P.

AU - Boiroux, Dimitri

AU - Jørgensen, John Bagterp

AU - Breton, Marc D.

PY - 2019/8/1

Y1 - 2019/8/1

N2 - This paper presents an individualized Ensemble Model Predictive Control (EnMPC) algorithm for blood glucose (BG) stabilization and hypoglycemia prevention in people with type 1 diabetes (T1D) who exercise regularly. The EnMPC formulation can be regarded as a simplified multi-stage MPC allowing for the consideration of Nen scenarios gathered from the patient's recent behavior. The patient's physical activity behavior is characterized by an exercise-specific input signal derived from the deconvolution of the patient's continuous glucose monitor (CGM), accounting for known inputs such as meal, and insulin pump records. The EnMPC controller was tested in a cohort of in silico patients with representative inter-subject and intra-subject variability from the FDA-accepted UVA/Padova simulation platform. Results show a significant improvement on hypoglycemia prevention after 30 min of mild to moderate exercise in comparison to a similarly tuned baseline controller (rMPC); with a reduction in hypoglycemia occurrences (<70 mg/dL), from 3.08 % ±3.55 with rMPC to 0.78 % ±2.04 with EnMPC (P < 0.05).

AB - This paper presents an individualized Ensemble Model Predictive Control (EnMPC) algorithm for blood glucose (BG) stabilization and hypoglycemia prevention in people with type 1 diabetes (T1D) who exercise regularly. The EnMPC formulation can be regarded as a simplified multi-stage MPC allowing for the consideration of Nen scenarios gathered from the patient's recent behavior. The patient's physical activity behavior is characterized by an exercise-specific input signal derived from the deconvolution of the patient's continuous glucose monitor (CGM), accounting for known inputs such as meal, and insulin pump records. The EnMPC controller was tested in a cohort of in silico patients with representative inter-subject and intra-subject variability from the FDA-accepted UVA/Padova simulation platform. Results show a significant improvement on hypoglycemia prevention after 30 min of mild to moderate exercise in comparison to a similarly tuned baseline controller (rMPC); with a reduction in hypoglycemia occurrences (<70 mg/dL), from 3.08 % ±3.55 with rMPC to 0.78 % ±2.04 with EnMPC (P < 0.05).

KW - Artificial pancreas

KW - Exercise

KW - Hypoglycemia

KW - Model Predictive Control

KW - Type 1 diabetes

U2 - 10.1016/j.jprocont.2019.05.017

DO - 10.1016/j.jprocont.2019.05.017

M3 - Journal article

VL - 80

SP - 202

EP - 210

JO - Journal of Process Control

JF - Journal of Process Control

SN - 0959-1524

ER -