A learning automata-based blood glucose regulation mechanism in type 2 diabetes

Abstract This paper proposes a learning automata-based mechanism for blood glucose regulation in type 2 diabetics. The proposed mechanism takes into account the past history of the blood glucose level to determine the correct dosage of the insulin. This method uses the learning automata theory to predict the required dosage of insulin and records the patient history in parameters of a Gaussian probability distribution function. The parameters of the distribution function are updated based on the difference between the actual glucose level regulated by the learning automata and the normal range in such a way that the gap between the actual glucose level and the normal one is minimized. As the proposed algorithm proceeds, it can be seen that it converges to the optimal insulin dosage that keeps the glucose level in normal range for a long time. Convergence of the proposed algorithm to the optimal insulin dosage is theoretically proven. A clinical study is conducted to show the performance of the proposed insulin therapy system for regulation of the blood glucose level of type 2 diabetics.

[1]  Athanasios V. Vasilakos,et al.  ANASA-a stochastic reinforcement algorithm for real-valued neural computation , 1996, IEEE Trans. Neural Networks.

[2]  Javad Akbari Torkestani.  A new approach to the job scheduling problem in computational grids , 2011, Cluster Computing.

[3]  A. V. Vasilakos,et al.  ASAFES: adaptive stochastic algorithm for fuzzy computing/function estimation , 1994, Proceedings of 1994 IEEE 3rd International Fuzzy Systems Conference.

[4]  Ricardo Femat,et al.  Blood glucose control for type I diabetes mellitus: A robust tracking H∞ problem , 2004 .

[5]  Javad Akbari Torkestani.  Energy-efficient backbone formation in wireless sensor networks , 2013, Comput. Electr. Eng..

[6]  A. Farmer,et al.  Addition of biphasic, prandial, or basal insulin to oral therapy in type 2 diabetes. , 2007, The New England journal of medicine.

[7]  Javad Akbari Torkestani.  Degree constrained minimum spanning tree problem: a learning automata approach , 2012, The Journal of Supercomputing.

[8]  MS Ibbini,et al.  A semiclosed-loop optimal control system for blood glucose level in diabetics , 2004, Journal of medical engineering & technology.

[9]  Javad Akbari Torkestani,et al.  An energy-efficient topology construction algorithm for wireless sensor networks , 2013, Comput. Networks.

[10]  H. L. Le Roy,et al.  Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability; Vol. IV , 1969 .

[11]  Javad Akbari Torkestani.  An adaptive focused Web crawling algorithm based on learning automata , 2012, Applied Intelligence.

[12]  Kaddour Najim,et al.  Learning automata and stochastic optimization , 1997 .

[13]  Javad Akbari Torkestani,et al.  An adaptive learning to rank algorithm: Learning automata approach , 2012, Decis. Support Syst..

[14]  Javad Akbari Torkestani.  A MULTI-ATTRIBUTE RESOURCE DISCOVERY ALGORITHM FOR PEER-TO-PEER GRIDS , 2013, Appl. Artif. Intell..

[15]  Vijaykumar Gullapalli,et al.  A stochastic reinforcement learning algorithm for learning real-valued functions , 1990, Neural Networks.

[16]  Javad Akbari Torkestani,et al.  A NEW APPROACH TO THE VERTEX COLORING PROBLEM , 2013, Cybern. Syst..

[17]  F. Chee,et al.  Expert PID control system for blood glucose control in critically ill patients , 2003, IEEE Transactions on Information Technology in Biomedicine.

[18]  R. Holman,et al.  A Practical Guide to Basal and Prandial Insulin Therapy , 1985, Diabetic medicine : a journal of the British Diabetic Association.

[19]  Javad Akbari Torkestani.  An adaptive learning automata-based ranking function discovery algorithm , 2012, Journal of Intelligent Information Systems.

[20]  B.W. Bequette,et al.  Model predictive control of blood glucose in type I diabetics using subcutaneous glucose measurements , 2002, Proceedings of the 2002 American Control Conference (IEEE Cat. No.CH37301).

[21]  Z. Benyo,et al.  LPV fault detection of glucose-insulin system , 2006, 2006 14th Mediterranean Conference on Control and Automation.

[22]  D. Dazzi,et al.  The control of blood glucose in the critical diabetic patient: a neuro-fuzzy method. , 2001, Journal of diabetes and its complications.

[23]  A. Dvoretzky.  On Stochastic Approximation , 1956 .

[24]  Kumpati S. Narendra,et al.  Learning automata - an introduction , 1989 .

[25]  M Ibbini.  A PI-fuzzy logic controller for the regulation of blood glucose level in diabetic patients , 2006, Journal of medical engineering & technology.

[26]  Javad Akbari Torkestani.  Mobility-Based Backbone Formation in Wireless Mobile Ad-hoc Networks , 2013, Wirel. Pers. Commun..

[27]  R. Hovorka,et al.  Nonlinear model predictive control of glucose concentration in subjects with type 1 diabetes. , 2004, Physiological measurement.

[28]  Javad Akbari Torkestani,et al.  An adaptive energy-efficient area coverage algorithm for wireless sensor networks , 2013, Ad Hoc Networks.

[29]  B. R. Harita,et al.  Learning automata with changing number of actions , 1987, IEEE Transactions on Systems, Man, and Cybernetics.

[30]  Francis J. Doyle,et al.  Glucose control design using nonlinearity assessment techniques , 2005 .

[31]  Balázs Benyó,et al.  Past and present of automatic glucose-insulin control research at BME , 2009 .

[32]  Javad Akbari Torkestani.  A NEW DISTRIBUTED JOB SCHEDULING ALGORITHM FOR GRID SYSTEMS , 2013, Cybern. Syst..

[33]  Z Benyó,et al.  Optimal glucose-insulin control in /spl Hscr//sub 2/ space , 2004, The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[34]  Javad Akbari Torkestani,et al.  A distributed resource discovery algorithm for P2P grids , 2012, J. Netw. Comput. Appl..

[35]  J. Wolfowitz.  On the Stochastic Approximation Method of Robbins and Monro , 1952 .

[36]  Timothy Gordon,et al.  Continuous action reinforcement learning applied to vehicle suspension control , 1997 .

[37]  Francis J. Doyle,et al.  Robust H∞ glucose control in diabetes using a physiological model , 2000 .

[38]  B. Bequette.  A critical assessment of algorithms and challenges in the development of a closed-loop artificial pancreas. , 2005, Diabetes technology & therapeutics.

[39]  P. S. Sastry,et al.  Continuous action set learning automata for stochastic optimization , 1994 .