An Actor-Critic based controller for glucose regulation in type 1 diabetes

Authors:
Elena Daskalaki;Peter Diem;Stavroula G. Mougiakakou
Affiliations:
ARTORG Center for Biomedical Engineering Research, Diabetes Technology Research Group, University of Bern, Murtenstrasse 50, 3010 Bern, Switzerland;Division of Endocrinology, Diabetes and Clinical Nutrition, University Hospital, Inselspital, University of Bern, 3010 Bern, Switzerland;ARTORG Center for Biomedical Engineering Research, Diabetes Technology Research Group, University of Bern, Murtenstrasse 50, 3010 Bern, Switzerland
Venue:
Computer Methods and Programs in Biomedicine
Year:
2013

Citing 11
Cited 0

Reinforcement Learning

Reinforcement Learning
Neuro-Dynamic Programming

Neuro-Dynamic Programming
Learning to Predict by the Methods of Temporal Differences

Machine Learning
On Actor-Critic Algorithms

SIAM Journal on Control and Optimization
Genetic network programming with actor-critic and its application to stock trading model

Proceedings of the 9th annual conference on Genetic and evolutionary computation
A survey of insulin-dependent diabetes-part II: control methods

International Journal of Telemedicine and Applications - Regular issue
Fuzzy-PID Control for the Regulation of Blood Glucose in Diabetes

GCIS '09 Proceedings of the 2009 WRI Global Congress on Intelligent Systems - Volume 02
ACE (Actor-Critic-Explorer) paradigm for reinforcement learning in basal ganglia: Highlighting the role of subthalamic and pallidal nuclei

Neurocomputing
Anticipating the next meal using meal behavioral profiles: A hybrid model-based stochastic predictive control algorithm for T1DM

Computer Methods and Programs in Biomedicine
Physiologic insulin delivery with insulin feedback: A control systems perspective

Computer Methods and Programs in Biomedicine
Brief paper: Average cost temporal-difference learning

Automatica (Journal of IFAC)

Quantified Score

Hi-index	0.00

Visualization

Abstract

A novel adaptive approach for glucose control in individuals with type 1 diabetes under sensor-augmented pump therapy is proposed. The controller, is based on Actor-Critic (AC) learning and is inspired by the principles of reinforcement learning and optimal control theory. The main characteristics of the proposed controller are (i) simultaneous adjustment of both the insulin basal rate and the bolus dose, (ii) initialization based on clinical procedures, and (iii) real-time personalization. The effectiveness of the proposed algorithm in terms of glycemic control has been investigated in silico in adults, adolescents and children under open-loop and closed-loop approaches, using announced meals with uncertainties in the order of +/-25% in the estimation of carbohydrates. The results show that glucose regulation is efficient in all three groups of patients, even with uncertainties in the level of carbohydrates in the meal. The percentages in the A+B zones of the Control Variability Grid Analysis (CVGA) were 100% for adults, and 93% for both adolescents and children. The AC based controller seems to be a promising approach for the automatic adjustment of insulin infusion in order to improve glycemic control. After optimization of the algorithm, the controller will be tested in a clinical trial.