DIACON, ice-cream in a cone

DIACON: The Artificial Pancreas

According to International Diabetes Federation, 415 million people had diabetes in 2015. These numbers are expected to rise to 642 million by 2040. The objective of the DIACON project is to develop an artificial pancreas for people with Type 1 diabetes.



Type 1 diabetes is treated with multiple daily injections of insulin or continuous infusion of insulin using a pump. The stress of calculating the needed amount of insulin based on food intake, exercise and insulin sensitivity has led to research in creating an artificial pancreas (AP). A basic AP is a closed-loop (CL) system consisting of an insulin pump, a continuous glucose monitor (CGM) and a control algorithm to adjust insulin dosage through the pump based on CGM sensor readings [Wendt et al. 2016][1].


Project Objectives

The DIACON project aims to develop advanced mathematical models for prediction and simulation of the insulin-blood glucose dynamics.

The project explores the use of linear and non-linear model predictive control (MPC) strategies to control the blood glucose level. MPC is a useful control method for the artificial pancreas due to its ability to handle constraints on insulin administration and glucose level in a systematic and proactive way. 

Some features being researched are critical safeguards like an insulin-on-board constraint and the trade-offs associated with a meal announcement strategy. Supervisory levels of control - known as monitoring strategies - are also being investigated. 

The statistical simulation models are very useful for pre-clinical testing of the control algorithms. These models combine physiological knowledge about the human insulin-blood glucose system with statistical information from data obtained in-clinic at the DIACON clinical research unit at Hvidovre University Hospital.





Kirsten Nørgaard
Hvidovre Hospital


John Bagterp Jørgensen
DTU Compute
+45 45 25 30 88
27 JANUARY 2023