DIAdvisor™-Personal mobile short-term blood glucose predictor and treatment advisor
Researchers: Marzia Cescon, Fredrik Ståhl, Meike Stemmann, Rolf Johansson, Dawn Tilbury
Project leader: Rolf Johansson
|Novo Nordisk A/S, Bagsværd, Denmark||www.novonordisk.com||Coordinator|
|Johannes Kepler University, Linz, Austria||www.jku.at||Control & advisory algorithms|
|Lunds University, Lund, Sweden||www.lu.se||Data based models|
|University of Padova, Padova, Italy||www.unipd.it||Physiological models and clinical trials|
|Centre Hospitalier Universitaire de Montpellier, Montpellier, France||www.chu-montpellier.fr||Clinical trials|
|Toumaz Technology Ltd, Abingdon, UK||www.toumaz.com||Device integration|
|Sensor Technology and Devices Ltd, Belfast, UK||www.stnd.com||Vital sign sensors|
|Ondalys, Montpellier, France||www.ondalys.fr||Non invasive glucose measurement|
|RomSoft, Iasi, Romania||www.rms.ro||Software|
|Institute for Clinical and Experimental Medicine , Prague, Czech Republic||www.ikem.cz||Clinical trials|
|RICAM, Linz, Austria||www.ricam.oeaw.ac.at||Mathematics|
|Ramboll, Virum, Denmark||www.ramboll.com||Risk management|
|Federation Internationale du Diabete Region Europe, Brussels, Belgium||www.idf-europe.org||Validation and Dissemination
Sponsoring Organization: the European Commision through the Information Society Technologies (IST) programme under the Seventh Framework Programme (FP7) n° 216592
Official Website: www.diadvisor.eu
Diabetes Mellitus is a chronic disease of disordered glucose metabolism due to defects in either insulin secretion from the pancreatic beta-cells or insulin action. Type-1 diabetes (T1DM), also called insulin-dependent diabetes mellitus (IDDM) is characterized by no production of insulin what so ever, whereas type-2 diabetes is caused by decreased sensitivity of the tissues to the metabolic effect of insulin. The basic effect of insulin lack or insulin resistance is to prevent the efficient uptake and utilization of glucose by most cells of the body, resulting in abnormally high blood sugar levels (hyperglycemia). Sustained hyperglycemia is associated with acute ketoacidosis, nephropaty, rethinopaty,neuropathy and damages to the cardio-vascular system, therefore intensive insulin therapy aiming at near-normoglycemia(80-100 mg/dL) has been strongly promoted during the last decade, following the results of the major Diabetes Control and ComplicationsTrial (DCCT) and follow-up Epidemiology of Diabetes Interventions and Complications (EDIC) studies. Focusing on tight blood glucose targets, the strategy comprises test of blood glucose levels at least four times a day, taking insulin at least three times a day by injections or using a pump and patient assistance by healthcare team through visits and phone calls. Meanwhile, the lack of improved quality of life and above all, the occurence of induced hypoglycemic events which may result in seizure, coma and eventually death preclude the feasibilityof such a DCCT-like intensive therapy.
The problem of maintaining glucose levels within a predefined range by acting on insulin delivery is a control poblem, whose controlled variable is glucose utilization, measured output is either the subcutaneous glucose provided by the CGMS or the capillary glucose provided by the fingerprick, control input is the insulin intake, and the clinical criterion for success is plasma glucose. The system is subject to disturbances, the most important one being the meals. Control strategies involving the regulation of blood glucose levels in type 1 diabetes subjects range from classic PID feedback controller, run-to-run strategies to MPC algorithms.
The DIAdvisor™ project:
The DIAdvisor™is a large-scale integrating project (IP) aiming at the development of a prediction based tool which uses past and easily available information to optimise the therapy of type 1 and developed type 2 diabetes. The DIAdvisor™ is not dependent on specific sensor technologies and can be adapted to technologies like standard strip sensing, minimally-invasive continuous glucose sensors and emerging non-invasive methods.
For safety reason, the DIAdvisor™ system will be able to self-assess the confidence of its proposed decisions. For safety reasons as well as for the sake of therapy improvements, the system connects and provides information and trends to the Health Care Provider.
Glucose prediction is difficult and requires advanced science within the fields of physiological modelling, identification theory, control theory, medical device technology, risk management theory, sensor science and user understanding. It can be achieved only by a well balanced group of eminent experts, including academics, clinicians, user representatives and leading companies.
The expected impact of DIAdvisor™ will be improved diabetes control and quality of life in large populations of insulin treated patients, leading to fewer diabetic complications and lower Health Care costs. Moreover, the project will constitute a valuable opportunity for European companies to build up a special know-how leading to products that profoundly and positively have an impact on the lives of millions of people with other indications than diabetes.
Jens Ulrik Poulsen
Novo Nordisk A/S
Phone: +45 4443 9692
Press Releases and Articles
- LUM May 2008
- Ny Teknik 2008-06-04 (.pdf)
Marzia Cescon: Modeling and Prediction in Diabetes Physiology. PhD Thesis ISRN LUTFD2/TFRT--1099--SE, Department of Automatic Control, Lund University, Sweden, November 2013.
Fredrik Ståhl, Rolf Johansson, Eric Renard: "Model-Based Estimates of the Post-Prandial Response to Carbohydrate and Insulin and of the Carbohydrate-to-Insulin Ratio". In American Diabetes Association 73rd Scientific Sessions, Chicago, IL, June 2013.
Fredrik Ståhl, Rolf Johansson, Eric Renard: "Investigation of the relationship between elevated levels of insulin antibodies and prolonged insulin action". In 6th International Conference on Advanced Technologies and Treatments for Diabetes, Paris, France, February 2013.
Meike Stemmann: "Predictive Control of Diabetic Glycemia". Licentiate Thesis ISRN LUTFD2/TFRT--3258--SE, Department of Automatic Control, Lund University, Sweden, 2013.
Fredrik Ståhl: "Diabetes Mellitus Glucose Prediction by Linear and Bayesian Ensemble Modeling". Licentiate Thesis ISRN LUTFD2/TFRT--3255--SE, Department of Automatic Control, Lund University, Sweden, December 2012.
Fredrik Ståhl, Rolf Johansson: "Receding Horizon Prediction by Bayesian Combination of Multiple Predictors". In 51st IEEE Conference on Decision and Control, Maui, Hawaii, USA, December 2012.
Dawn M. Tilbury, Eric Renard, Rolf Johansson: "Integrating Multiple Controllers to Balance Competing Performance Objectives: Application to Blood Glucose Management". In 2012 ASME Dynamic Systems and Control Conference, Fort Lauderdale, FL, USA, October 2012.
Meike Stemmann, Rolf Johansson: "Diabetic Blood Glucose Control Via Optimization Over Insulin and Glucose Doses". In 8th IFAC Symposium on Biological and Medical Systems, Budapest, Hungary, August 2012.
Fredrik Ståhl, Rolf Johansson, Eric Renard: "Bayesian Combination of Multiple Plasma Glucose Predictors". In 34th Annual Conference of the IEEE EMBS, San Diego, CA, USA, August 2012.
Marzia Cescon, Rolf Johansson, Eric Renard: "Personalized short-term blood glucose prediction in T1DM". In 5th International Conference on Advanced Technologies and Treatments for Diabetes, Barcelona, Spain, February 2012.
Meike Stemmann, Rolf Johansson: "Control of Type 1 Diabetes via Risk-Minimization for Multi Dose Injection Patients". In 5th International Conference on Advanced Technologies and Treatments for Diabetes, Barcelona, Spain, February 2012.
Marzia Cescon, Rolf Johansson: "Patient-specific Glucose Metabolism Models for Model Predictive Control of T1DM Glycemia". In 5th International Conference on Advanced Technologies and Treatments for Diabetes, Barcelona, Spain, January 2012.
Marzia Cescon, Eric Renard: "Adaptive Subspace-based prediction of T1DM glycemia". In 50th IEEE Conference on Decision and Control and European Control Conference, Orlando, Florida, USA, December 2011.
Harald Kirchsteiger, Stephan Pölzer, Rolf Johansson, Eric Renard, Luigi del Re: "Direct Continuous Time System Identification of MISO Transfer Function Models applied to Type 1 Diabetes". In 50th IEEE Conference on Decision and Control and European Control Conference, Orlando, Florida, USA, December 2011.
Marzia Cescon, Rolf Johansson: "On Data-driven Multistep Subspace-based Linear Predictors". In 18th IFAC World Congress, Milano, Italy, August 2011.
Marzia Cescon: "Linear Modeling and Prediction in Diabetes Physiology". Licentiate Thesis ISRN LUTFD2/TFRT--3250--SE, 3250, Department of Automatic Control, Lund University, Sweden, June 2011.
Marzia Cescon, Rolf Johansson: "Multi-step-ahead Multivariate Predictors: A Comparative Analysis". In 49th IEEE Conference on Decision and Control, Atlanta, Georgia, USA, December 2010.
Fredrik Ståhl, Rolf Johansson: "Observer Based Plasma Glucose Prediction in Type I Diabetes". In 2010 IEEE Multi-Conference on Systems and Control, Yokohama, Japan , September 2010.
Fredrik Ståhl, Rolf Johansson, Eric Renard: "Post-prandial plasma glucose prediction in type I diabetes based on Impulse Response Models.". In 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Buenos Aires, Argentina, August 2010.
Meike Stemmann, Fredrik Ståhl, Jordane Lallemand, Eric Renard, Rolf Johansson: "Sensor Calibration Models for a Non-Invasive Blood Glucose Measurement Sensor". In 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Buenos Aires, Argentina, August 2010.
Rolf Johansson: "Continuous-Time Model Identification and State Estimation Using Non-Uniformly Sampled Data". In Proc. 19th International Symposium on Mathematical Theory of Networks and Systems, July 2010.
Marzia Cescon, Rolf Johansson: "Glycemic Trend Prediction Using Empirical Model Identification". In Proc. Joint 48th IEEE Conference on Decision and Control & Chinese Control Conference (CDC2009 & CCC 2009), Shanghai, China, December 16-18, 2009, pp. 3501-3506 , December 2009.
Rolf Johansson: "Focus on: Subspace-based Identification and Its Applications". International Journal of Adaptive Control and Signal Processing, 23:12, pp. 1051–1052, December 2009.
Fredrik Ståhl, Marzia Cescon, Rolf Johansson, Eric Renard: "Infinite Horizon Prediction of Post Prandial Breakfast Glucose Excursion". In Diabetes Technology Meeting 2009, San Francisco, November 2009.
Rolf Johansson: "Continuous-Time Model Identification and State Estimation Using Non-Uniformly Sampled Data". In Proc. 15th IFAC Symposium on System Identification (SYSID2009), July 6 - 8, 2009, Saint-Malo, France, July 2009.
Marzia Cescon, Fredrik Ståhl, Rolf Johansson: "Subspace-Based Model Identification of Diabetic Blood Glucose Dynamics". In Proc. 15th IFAC Symposium on System Identification, July 6 - 8, 2009, Saint-Malo, France, July 2009.
Marzia Cescon, Fredrik Ståhl, Mona Landin-Olsson, Rolf Johansson: "Short-Term Diabetes Blood Glucose Prediction Based On Blood Glucose Measurements". In 2nd Int. Conference on Advanced Technologies and Treatments for Diabetes 2009, Athens, Greece, February 2009.
Fredrik Ståhl, Rolf Johansson: "Diabetes mellitus modeling and short-term prediction based on blood glucose measurements.". Mathematical Bioscienses, 217, pp. 101–117, January 2009.
Fredrik Ståhl, Rolf Johansson: "Short-Term Diabetes Blood Glucose Prediction Based On Blood Glucose Measurements". In 30th Annual International IEEE EMBS Conference Vancouver, British Columbia, Canada, August 20-24, August 2008.
Rolf Johansson: "Subspace-Based Continuous-Time Identification". In (Eds.): Identification of Continuous-time Models from Sampled Data, Springer-Verlag, London, March 2008.
Rolf Johansson: "Continuous-Time Model Identification Using Spectrum Analysis with Passivity-Preserving Model Reduction". In (Eds.): Identification of Continuous-time Models from Sampled Data, Springer-Verlag, London, March 2008.