Jump neural network for online short-time prediction of blood glucose from continuous monitoring sensors and meal information

• Published in: Computer methods and programs in biomedicine Vol. 113; no. 1; pp. 144 - 152
• Main Authors: Zecchin, C., Facchinetti, A., Sparacino, G., Cobelli, C.
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier Ireland Ltd 01.01.2014; Elsevier
• Subjects: Algorithms; Biological and medical sciences; Blood Glucose - analysis; Blood Glucose Self-Monitoring - instrumentation; Continuous glucose monitoring; Diabetes; Diabetes Mellitus, Type 1 - blood; Diabetes Mellitus, Type 1 - physiopathology; Diabetes. Impaired glucose tolerance; Endocrine pancreas. Apud cells (diseases); Endocrinopathies; Exact sciences and technology; Food; Forecast; Glucose; Humans; Inference from stochastic processes; time series analysis; Internal Medicine; Management. Various non-drug treatments. Langerhans islet grafts; Mathematical models; Mathematics; Meals; Medical sciences; Monitoring; Neural networks; Neural Networks (Computer); Non-linear modeling; Other; Patients; Probability and statistics; Sciences and techniques of general use; Sensors; Signal processing; Statistics; Time series; Continuous glucose monitoring; Non-linear modeling; Signal processing; Diabetes; Time series; Forecast; Short term; Therapy; Measurement sensor; Meal; Concentration; Glucose; Neural network; Real time; Modeling; Timed system; Blood; Type 1 diabetes; Carbohydrate; Non linear effect; Glycemia; Monitoring
• ISSN: 0169-2607; 1872-7565; 1872-7565
• DOI: 10.1016/j.cmpb.2013.09.016

Several real-time short-term prediction methods, based on time-series modeling of past continuous glucose monitoring (CGM) sensor data have been proposed with the aim of allowing the patient, on the basis of predicted glucose concentration, to anticipate therapeutic decisions and improve therapy of type 1 diabetes. In this field, neural network (NN) approaches could improve prediction performance handling in their inputs additional information. In this contribution we propose a jump NN prediction algorithm (horizon 30min) that exploits not only past CGM data but also ingested carbohydrates information. The NN is tuned on data of 10 type 1 diabetics and then assessed on 10 different subjects. Results show that predictions of glucose concentration are accurate and comparable to those obtained by a recently proposed NN approach (Zecchin et al. (2012) [26]) having higher structural and algorithmical complexity and requiring the patient to announce the meals. This strengthen the potential practical usefulness of the new jump NN approach.