Zone Model Predictive Control with Glucose- and Velocity-Dependent Control Penalty Adaptation for an Artificial Pancreas
Зонное управление с прогнозирующей моделью и адаптацией штрафа за управление в зависимости от уровня и скорости изменения глюкозы для искусственной поджелудочной железы
2018-06-01
SCID: 54.1/wbva4vbc
Discuss with AI
Universities of Virginia/Padova simulatorartificial pancreasglucose-dependent control penalty adaptationtype 1 diabetes mellituszone model predictive control
Figures from the paper
Abstract (AI)
An adaptive zone model predictive control design problem is considered for enhanced blood glucose regulation in patients with type 1 diabetes mellitus. The key contribution of this work is the development of a zone MPC with a dynamic cost function that updates its control penalty parameters based on the predicted glucose and its rate of change. A parameter adaptation law is proposed by explicitly constructing maps from glucose state and velocity spaces to control penalty parameter spaces. The proposed controller is tested on the to-patient cohort of the US Food and Drug Administration accepted Universities of Virginia/Padova simulator and compared with the zone model predictive control without parameter adaptation. The obtained in-silico results indicate that for unannounced meals, the controller leads to statistically significant improvements in terms of mean glucose level (154.2 mg/dL vs. 160.7 mg/dL; and percentage time in the safe euglycemic range of [70, 180] mg/dL (72.7% vs. 67.5%; ) without increasing the risk of hypoglycemia (percentage time below 70 mg/dL, 0.0% vs. 0.0%; p=0.788). For announced meals, the obtained performance is similar (and slightly superior) to that of the zone model predictive control without adaptation in terms of mean glucose level (135.6 mg/dL vs. 136.5 mg/dL; ), percentage time in [70, 180] mg/dL (91.2% vs. 90.9%; p=0.04), and percentage time below 70 mg/dL (0.0% vs. 0.0%; p=0.346).
Key Findings
1
Developed an adaptive zone MPC whose control penalty dynamically updates using predicted glucose levels and their rate of change.
2
For announced meals, adapted control slightly improved mean glucose and time in range versus standard zone MPC, while hypoglycemia remained 0.0%.
3
For unannounced meals, time in the 70–180 mg/dL range increased from 67.5% to 72.7% without increasing time below 70 mg/dL, which remained 0.0%.
4
In FDA-accepted UVA/Padova simulator tests with unannounced meals, adaptation reduced mean glucose from 160.7 to 154.2 mg/dL.
5
The adaptation law explicitly maps glucose state and velocity spaces to control penalty parameter spaces.
Research Object
Artificial pancreas system for blood glucose regulation in patients with type 1 diabetes mellitus
Research Subject
Blood glucose regulation performance and control-penalty adaptation based on predicted glucose level and rate of change, including glycemic safety and time-in-range
Publication Details
Publication Date
2018-06-01
Journal
Proceedings of the ... American Control Conference/Proceedings of the American Control Conference
Publisher
Institute of Electrical and Electronics Engineers
ISSN
0743-1619
Cited by
8
Access Type
Author Information
Download PDF
Subscribe to digest