Neural identification of Type 1 Diabetes Mellitus for care and forecasting of risk events

Abstract Glucose–insulin models, testing glucose sensors and support systems for health care decisions play an important role in synthesis of glucose control algorithms. In this work we propose an online glucose–insulin identification using the Recurrent High Order Neural Network (RHONN). Then, the model obtained is used to predict n -steps forward of glucose levels, also by RHONN. The used data for identification is from a Type 1 Diabetes Mellitus (T1DM) patient, it was collected from the Continuous Monitoring Glucose System (CMGS) by MiniMed Inc ® and an insulin pump by Paradigm Real-time Insulin Pump ®. RHONN is trained online by Extended Kalman Filter (EKF). The results suggest that it is possible to make a prediction of up to 35 min in the future, which it would help to prevent risky events (hypoglycemia and hyperglycemia). Also shows that, it could be directly connected to a CGMS to help the patient improve the glucose control and even an automatic glucose control algorithm. The proposed Neural Network shows good performance compared to baseline methods in terms of evaluation criteria.

[1]  S. R. Dikondwar.  Design and development of insulin delivery system prototype , 2011, 2011 IEEE 3rd International Conference on Communication Software and Networks.

[2]  Junhwa Chi,et al.  Prediction of Arctic Sea Ice Concentration Using a Fully Data Driven Deep Neural Network , 2017, Remote. Sens..

[3]  Alma Y. Alanis,et al.  Neuro-fuzzy inverse optimal control incorporating a multistep predictor as applied to T1DM patients , 2020 .

[4]  Giovanni Sparacino,et al.  Continuous glucose monitoring and hypo/hyperglycaemia prediction , 2006 .

[5]  Marko V. Jankovic,et al.  Predicting Blood Glucose with an LSTM and Bi-LSTM Based Deep Neural Network , 2018, 2018 14th Symposium on Neural Networks and Applications (NEUREL).

[6]  V. Jain,et al.  Advances in management of type 1 diabetes mellitus. , 2014, World journal of diabetes.

[7]  P. Werbos,et al.  Long-term predictions of chemical processes using recurrent neural networks: a parallel training approach , 1992 .

[8]  Apurv Kamath,et al.  Methods of Evaluating the Utility of Continuous Glucose Monitor Alerts , 2010, Journal of diabetes science and technology.

[9]  P. Cortez,et al.  Multi-step time series prediction intervals using neuroevolution , 2019, Neural Computing and Applications.

[10]  Geoffrey McGarraugh.  Alarm Characterization for a Continuous Glucose Monitor That Replaces Traditional Blood Glucose Monitoring , 2010, Journal of diabetes science and technology.

[11]  Konstantina S. Nikita,et al.  Comparative assessment of glucose prediction models for patients with type 1 diabetes mellitus applying sensors for glucose and physical activity monitoring , 2015, Medical & Biological Engineering & Computing.

[12]  Ricardo Femat,et al.  Weighting Restriction for Intravenous Insulin Delivery on T1DM Patient via $H_{\infty}$ Control , 2009, IEEE Transactions on Automation Science and Engineering.

[13]  Yen-Ming Chiang,et al.  Multi-step-ahead neural networks for flood forecasting , 2007 .

[14]  B Wayne Bequette,et al.  Continuous Glucose Monitoring: Real-Time Algorithms for Calibration, Filtering, and Alarms , 2010, Journal of diabetes science and technology.

[15]  Sue Simpson,et al.  Artificial Pancreas Device Systems for the Closed-Loop Control of Type 1 Diabetes , 2015, Journal of diabetes science and technology.

[16]  D. Gough,et al.  Is blood glucose predictable from previous values? A solicitation for data. , 1999, Diabetes.

[17]  Edgar N. Sánchez,et al.  Inverse Optimal Control Using A Neural Multi-Step Predictor For T1DM Treatment , 2018, 2018 International Joint Conference on Neural Networks (IJCNN).

[18]  Alma Y. Alanis,et al.  Germinal Center Optimization Applied to Neural Inverse Optimal Control for an All-Terrain Tracked Robot , 2017 .

[19]  Renate Sitte,et al.  Neural Networks Approach to the Random Walk Dilemma of Financial Time Series , 2002, Applied Intelligence.

[20]  Alexander G. Loukianov,et al.  Discrete-Time Adaptive Backstepping Nonlinear Control via High-Order Neural Networks , 2007, IEEE Transactions on Neural Networks.

[21]  C. Cobelli,et al.  The UVA/PADOVA Type 1 Diabetes Simulator , 2014, Journal of diabetes science and technology.

[22]  Alexander G. Loukianov,et al.  Discrete-Time High Order Neural Control - Trained with Kaiman Filtering , 2010, Studies in Computational Intelligence.

[23]  C. C. Palerm,et al.  Hypoglycemia prediction and detection using optimal estimation. , 2005, Diabetes technology & therapeutics.

[24]  C. Cobelli,et al.  Artificial neural network algorithm for online glucose prediction from continuous glucose monitoring. , 2010, Diabetes technology & therapeutics.

[25]  Giuseppe De Nicolao,et al.  Neural Network Incorporating Meal Information Improves Accuracy of Short-Time Prediction of Glucose Concentration , 2012, IEEE Transactions on Biomedical Engineering.

[26]  Li-Chiu Chang,et al.  Reinforced recurrent neural networks for multi-step-ahead flood forecasts , 2013 .

[27]  W. A. Sandham,et al.  Neural network and neuro-fuzzy systems for improving diabetes therapy , 1998, Proceedings of the 20th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Vol.20 Biomedical Engineering Towards the Year 2000 and Beyond (Cat. No.98CH36286).

[28]  Ana Gabriela Gallardo-Hernández,et al.  A Discrete-Time Recurrent Neurofuzzy Network for Black-Box Modeling of insulin Dynamics in Diabetic Type-1 Patients , 2010, Int. J. Neural Syst..

[29]  R. Potts,et al.  Physiological differences between interstitial glucose and blood glucose measured in human subjects. , 2003, Diabetes care.

[30]  R. Hovorka,et al.  Partitioning glucose distribution/transport, disposal, and endogenous production during IVGTT. , 2002, American journal of physiology. Endocrinology and metabolism.

[31]  D Rodbard,et al.  Computer Simulation of Plasma Insulin and Glucose Dynamics After Subcutaneous Insulin Injection , 1989, Diabetes Care.

[32]  E D Lehmann,et al.  A physiological model of glucose-insulin interaction in type 1 diabetes mellitus. , 1992, Journal of biomedical engineering.

[33]  Giovanni Sparacino,et al.  Continuous glucose monitoring time series and hypo/hyperglycemia prevention: requirements, methods, open problems. , 2008, Current diabetes reviews.

[34]  Marzena Kryszkiewicz,et al.  Machine Intelligence and Big Data in Industry , 2016 .

[35]  S.G. Mougiakakou,et al.  Neural Network based Glucose - Insulin Metabolism Models for Children with Type 1 Diabetes , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[36]  B De Moor,et al.  An adaptive input–output modeling approach for predicting the glycemia of critically ill patients , 2006, Physiological measurement.

[37]  Howard C. Zisser,et al.  Identification of empirical dynamic models from type 1 diabetes subject data , 2008, 2008 American Control Conference.

[38]  Ali Cinar,et al.  Adaptive control strategy for regulation of blood glucose levels in patients with type 1 diabetes , 2009 .

[39]  Giovanni Sparacino,et al.  “Smart” Continuous Glucose Monitoring Sensors: On-Line Signal Processing Issues , 2010, Sensors.

[40]  Manolis A. Christodoulou,et al.  Adaptive Control with Recurrent High-order Neural Networks , 2000, Advances in Industrial Control.

[41]  A. El-Jabali.  Neural network modeling and control of type 1 diabetes mellitus , 2005, Bioprocess and biosystems engineering.

[42]  Aboul Ella Hassanien,et al.  Identification of Diabetes Disease Using Committees of Neural Network-Based Classifiers , 2016 .

[43]  Scott M. Pappada,et al.  Neural network-based real-time prediction of glucose in patients with insulin-dependent diabetes. , 2011, Diabetes technology & therapeutics.

[44]  Edgar N. Sánchez,et al.  Blood glucose Level Neural Model for Type 1 Diabetes Mellitus Patients , 2011, Int. J. Neural Syst..

[45]  V. Machiulskiene,et al.  Severity of periodontal disease in adult patients with diabetes mellitus in relation to the type of diabetes. , 2014, Biomedical papers of the Medical Faculty of the University Palacky, Olomouc, Czechoslovakia.

[46]  Ali Cinar,et al.  Adaptive system identification for estimating future glucose concentrations and hypoglycemia alarms , 2012, Autom..

[47]  Srinivasan Rajaraman,et al.  Predicting Subcutaneous Glucose Concentration in Humans: Data-Driven Glucose Modeling , 2009, IEEE Transactions on Biomedical Engineering.

[48]  J. Proietto,et al.  Diabetes Research: A Guide for Postgraduates , 2000 .

[49]  Patricio Colmegna,et al.  Analysis of three T1DM simulation models for evaluating robust closed-loop controllers , 2014, Comput. Methods Programs Biomed..