Diabetes: Models, Signals and Control

On July 14, at 2.30 pm, Claudio Cobelli, Emeritus Professor of Bioengineering at the University of Padova, will present the Seminar "Diabetes: Models, Signals and Control". The seminar is part of Phd in BioRobotics (Seminar Cycle on Prostheses and Artificial Organs) and is hosted by Prof. Leonardo Ricotti.
Join the seminar (Webex platform) at the following link: 
https://sssup.webex.com/sssup-en/j.php?MTID=m0444ecf1e2cf8f64f9541fb3c8435346

ABSTRACT

Diabetes mellitus is a life-threatening disease and its global prevalence has dramatically increased over the past years, with 22 new cases every 100.000 youths per year. Between 1980 and 2017, the number of people with diabetes has more than doubled reaching 425 million people worldwide and is projected to rise to 629 million by 2045.
The last decades have seen growing attention due to the diabetes pandemic and biomedical engineering has allowed important achievements in the areas of technology, modeling, signal processing and control of both type 2 and type 1 diabetes.

We first discuss oral minimal models, i.e. parsimonious system descriptions capable of measuring non-accessible parameters like insulin sensitivity and beta-cell responsivity from a physiological oral test, e.g. a meal or an OGTT. Adding a tracer to the oral dose allows a more detailed system portrait. PET modeling allows to obtain an even deeper understanding of insulin action in the skeletal muscle, i.e. quantitating the individual steps of glucose transport and phosphorylation.
Next, we move to maximal models which are very comprehensive descriptions, i.e. large, nonlinear model of high order with several parameters, which allow to perform simulation and to conduct in silico trials. Here tracers are indispensable. The gold standard is the triple tracer meal technique which allows measuring meal/OGTT glucose fluxes, i.e. glucose rate of appearance, glucose production and disposal. This rich database has allowed building the UVA/Padova Type 1 diabetes and the Padova Type 2 diabetes large-scale simulators. In particular, the UVA/Padova Type 1 simulator has been accepted by the U.S. FDA as a substitute to animal trials for in silico testing of various insulin treatments

In the last ten years there has been incredible progress of subcutaneous glucose sensors. The ability to measure virtually in continuous time glucose concentration has allowed to accelerate the development of an artificial pancreas (AP), i.e. a system combining a glucose sensor, a control algorithm, and an insulin infusion device. We will discuss our AP experience both in terms control methodology and clinical results. In particular, the value of the UVA/Padova Type 1 diabetes simulator in designing robust control algorithms is stressed. In addition, its role to prove the non-adjunctive use of glucose sensors (no need of blood fingerpricks) and to explore new insulin molecules is discussed.

Finally, we briefly mention a radically new approach for the treatment of type 1 diabetes which is being developed in the FET project FORGETDIABETES by employing the intraperitoneal route for both glucose sensing and insulin delivery and an innovative insulin infusion pump.

BIOSKETCH

Claudio Cobelli is Emeritus Professor of Bioengineering at the University of Padova. His research activity is in modeling and control of the glucose system in diabetes. His research is mainly supported by NIH, JDRF and EU. He published 724 papers in refereed journals, co-authored 8 books and hold 10 patents with an h-index of 87 (Scopus). He serves as Ass. Ed.of IEEE TBME and JDST and is on the Ed.Board  of DTT. He is a IEEE Fellow and received the Diabetes Technology Society Artificial Pancreas Research Award.