Abstract
Basal insulin infusion rate which should be adjusted to increase or decrease insulin delivery with the varying blood sugar level plays a key role in type 1 diabetes mellitus (T1DM) patients for maintaining the blood glucose level approximately steady within reference range in order to avoid the complications developed from diabetes. This paper proposes an effective hybrid particle swarm optimization (HPSO) algorithm for solving the basal insulin infusion rate problem. In HPSO, bad experience lesson learning scheme and local search based on chaotic dynamics are proposed to make a good balance between global exploration and local exploitation. Simulation results based on a set of well-known optimization benchmark instances and the basal insulin infusion rate adjustment problem for T1DM demonstrate the effectiveness of the proposed HPSO. In silico tests on standard virtual subject via HPSO show that under nominal condition the blood glucose concentrations could be kept within a range of 80–150 mg/dL within less than 5 days; meanwhile, in case of random variations in meal timings within \(\pm \)60 min or meal amounts within \(\pm \)75 % deviation from the nominal values, the blood glucose concentrations could be kept within the safe regions.
Similar content being viewed by others
References
American Diabetes Association (2014) Standards of medical care in diabetes—2014. Diabetes Care 37(Supplement 1):S14–S80
Bequette BW (2012) Challenges and recent progress in the development of a closed-loop artificial pancreas. Annu Rev Control 36(2):255–266
Chen X, Ong Y-S, Lim M-H, Tan KC (2011) A multi-facet survey on memetic computation. IEEE Trans Evol Comput 15(5):591–607
Clerc M, Kennedy J (2002) The particle swarm—explosion, stability, and convergence in a multidimensional complex space. IEEE Trans Evol Comput 6(1):58–73
Cobelli C, Dalla Man C, Sparacino G, Magni L, De Nicolao G, Kovatchev BP (2009) Diabetes: models, signals, and control. IEEE Rev Biomed Eng 2:54–96
Dalla Man C, Camilleri M, Cobelli C (2006) A system model of oral glucose absorption: validation on gold standard data. IEEE Trans Biomed Eng 53(12):2472–2478
Dalla Man C, Raimondo DM, Rizza RA, Cobelli C (2007a) GIM, simulation software of meal glucose-insulin model. J Diabetes Sci Technol 1(3):323–330
Dalla Man C, Rizza RA, Cobelli C (2007b) Meal simulation model of the glucose-insulin system. IEEE Trans Bio-Med Eng 54(10):1740–1749
Deepa, Sugumaran (2011) Model order formulation of a multivariable discrete system using a modified particle swarm optimization approach. Swarm Evolut Comput 1(4):204–212
Deng W, Chen R, He B, Liu YQ, Yin LF, Guo JH (2012) A novel two-stage hybrid swarm intelligence optimization algorithm and application. Soft Comput 16(10):1707–1722
Dorigo M, Birattari M, Stutzle T (2006) Ant colony optimization. IEEE Comput Intell Mag 1(4):28–39
Elbeltagi E, Hegazy T, Grierson D (2005) Comparison among five evolutionary-based optimization algorithms. Adv Eng Inf 19(1):43–53
Hovorka R, Kumareswaran K, Harris J, Allen JM, Elleri D, Xing DY, Kollman C, Nodale M, Murphy HR, Dunger DB, Amiel SA, Heller SR, Wilinska ME, Evans ML (2011) Overnight closed loop insulin delivery (artificial pancreas) in adults with type 1 diabetes: crossover randomised controlled studies. Br Med J 342:d1855
Kennedy J, Eberhart R (1995) Particle swarm optimization. In: Proceedings of IEEE international conference on neutral networks, vol 2, Australia, pp 1942–1948
Kovatchev BP, Cox DJ, Gonder-Frederick LA, Clarke W (1997) Symmetrization of the blood glucose measurement scale and its applications. Diabetes Care 20(11):1655–1658
Kovatchev BP, Breton M, Man CD, Cobelli C (2009) In silico preclinical trials: a proof of concept in closed-loop control of type 1 diabetes. J Diabetes Sci Technol 3(1):44–55
Le MN, Ong YS, Jin YC, Sendhoff B (2012) A unified framework for symbiosis of evolutionary mechanisms with application to water clusters potential model design. IEEE Comput Intell Mag 7(1):20–35
Li YY, Xiang RR, Jiao LC, Liu RC (2012) An improved cooperative quantum-behaved particle swarm optimization. Soft Comput 16(6):1061–1069
Liu B, Wang L, Jin Y-H, Tang F (2005) Improved particle swarm optimization combined with chaos. Chaos Solitons Fractals 25:1261–1271
Liu B, Wang L, Jin YH (2008) An effective hybrid PSO-based algorithm for flow shop scheduling with limited buffers. Comput Oper Res 35(9):2791–2806
Liu B, Wang L, Liu Y, Qian B, Jin YH (2010) An effective hybrid particle swarm optimization for batch scheduling of polypropylene processes. Comput Chem Eng 34(4):518–528
Liu B, Wang L, Liu Y, Wang SY (2011) A unified framework for population-based metaheuristics. Ann Oper Res 186(1):231–262
Liu LZ, Wu FX, Zhang WJ (2012) Inference of biological S-system using the separable estimation method and the genetic algorithm. IEEE/ACM Trans Comput Biol Bioinf 9(4):955–965
Moussouni F, Brisset S, Brochet P (2008) Comparison of two multi-agent algorithms: ACO and PSO for the optimization of a brushless DC wheel motor. In: Intelligent computer techniques in applied electromagnetics. Springer, Berlin, pp 3–10
Nguyen QH, Ong Y-S, Lim MH (2009) A probabilistic memetic framework. IEEE Trans Evol Comput 13(3):604–623
Ong YS, Keane AJ (2004) Meta-Lamarckian learning in memetic algorithms. IEEE Trans Evol Comput 8(2):99–110
Ong Y-S, Lim MH, Chen X (2010) Research frontier-memetic computation–past, present & future. IEEE Comput Intell Mag 5(2):24–36
Owens C, Zisser H, Jovanovic L, Srinivasan B, Bonvin D, Doyle FJ (2006) Run-to-run control of blood glucose concentrations for people with type 1 diabetes mellitus. IEEE Trans Biomed Eng 53(6):996–1005
Palerm CC, Zisser H, Jovanovic L, Doyle FIJ (2008) A run-to-run control strategy to adjust basal insulin infusion rates in type 1 diabetes. J Process Control 18(3–4):258–265
Percival MW, Bevier WC, Wang Y, Dassau E, Zisser HC, Jovanovic L, Doyle FJ 3rd (2010) Modeling the effects of subcutaneous insulin administration and carbohydrate consumption on blood glucose. J Diabetes Sci Technol 4(5):1214–1228
Percival MW, Wang Y, Grosman B, Dassau E, Zisser H, Jovanovic L, Doyle FJ (2011) Development of a multi-parametric model predictive control algorithm for insulin delivery in type 1 diabetes mellitus using clinical parameters. J Process Control 21(3):391–404
Rana S, Jasola S, Kumar R (2011) A review on particle swarm optimization algorithms and their applications to data clustering. Artif Intell Rev 35(3):211–222
Socha K, Dorigo M (2008) Ant colony optimization for continuous domains. Eur J Oper Res 185(3):1155–1173
Wang YQ, Dassau E, Doyle FJ (2010a) Closed-loop control of artificial pancreatic beta-cell in type 1 diabetes mellitus using model predictive iterative learning control. IEEE Trans Biomed Eng 57(2):211–219
Wang YQ, Dassau E, Zisser H, Jovanovic L, Doyle FJ III (2010b) Automatic bolus and adaptive basal algorithm for the artificial pancreatic beta-cell. Diabetes Technol Ther 12(11):879–887
Wang YQ, Zisser H, Dassau E, Jovanovic L, Doyle FJ (2010c) Model predictive control with learning-type set-point: application to artificial pancreatic beta-cell. AIChE J 56(6):1510–1518
Xu WX, Geng ZQ, Zhu QX, Gu XB (2013) A piecewise linear chaotic map and sequential quadratic programming based robust hybrid particle swarm optimization. Inf Sci 218:85–102
Zhan ZH, Zhang J, Li Y, Chung HSH (2009) Adaptive particle swarm optimization. IEEE Trans Syst Man Cybern Part B Cybern 39(6):1362–1381
Acknowledgments
This work is supported by National Natural Science Foundation of China (61374099 and 71101139), Beijing Nova Program (2011025), an EFSD/CDS/Lilly grant, and the Fok Ying-Tong Education Foundation (131060).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by V. Loia.
Appendix
Appendix
Five well-known benchmark problems used for the purpose of comparison are:
-
1.
Sphere model
$$\begin{aligned} f_1 (x)=\sum \limits _{i=1}^{30} x_i^2,\quad x\in [-100,100]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
2.
Schwefel’ Problem 2.22
$$\begin{aligned} f_2 (x)=\sum _{i=1}^{30} |x_i|+\prod \limits _{i=1}^{30} |x_i|,\quad x\in [-10,10]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
3.
Schwefel’ Problem 1.2
$$\begin{aligned} f_3 (x)=\sum _{i=1}^{30} \left( \sum _{i=1}^i x_i\right) ^2,\quad x\in [-100,100]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
4.
Schwefel’ Problem 2.21
$$\begin{aligned} f_4 (x)=\hbox {MAX}\{|x_i|,\ 1\le i\le 30\},\quad x\in [-100,100]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
5.
Generalized Rosenbrock’s function
$$\begin{aligned} f_5 (x)&= \sum \limits _{i=1}^{30} {\left[ {100(x_{i+1} -x_i^2 )^2+(x_i -1)^2} \right] },\\&\quad x\in [-30,30]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
6.
Step function
$$\begin{aligned} f_6 (x)=\sum \limits _{i=1}^{30} (\lfloor x_i +0.5 \rfloor )^2,\quad x\in [-100,100]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
7.
Quadric noise function
$$\begin{aligned} f_7 (x)=\sum \limits _{i=1}^{30} {ix_i^4 +\hbox {random}[0,1]},\quad x\in [-1.28,1.28]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
8.
Generalized Schwefel’s Problem 2.26
$$\begin{aligned} f_8 (x)=-\sum \limits _{i=1}^{30} {\left( {x_i \sin \sqrt{\left| {x_i} \right| }}\right) },\quad x\in [-500,500]^{30} \end{aligned}$$Its global minimum is equal to \(-12{,}569.5.\)
-
9.
Generalized Rastrigin’s function
$$\begin{aligned} f_9 (x)&= \sum \limits _{i=1}^{30} {\left[ {x_i^2 -10\cos (2\pi x_i )+10} \right] },\\&\quad x\in [-5.12,5.12]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
10.
Ackley’s function
$$\begin{aligned} f_{10} (x)&= -20\exp \left( -0.2\sqrt{1/30\sum _{i=1}^{30}{x_i^2}}\right) \nonumber \\&\quad -\,\exp \left( 1/30\sum _{i=1}^{30} {\cos (2\pi x_i)}\right) +20+e \\&\quad \quad x\in \left[ {-32,32} \right] ^{30} \end{aligned}$$Its global minimum is equal to 0.
-
11.
Generalized Griewank’s function
$$\begin{aligned} f_{11} (x)&= \frac{1}{4{,}000}\sum \limits _{i=1}^{30} {x_i^2 } -\prod \limits _{i=1}^{30} \cos \left( \frac{x_i}{\sqrt{i}}\right) +1,\\&\quad x\in [-600,600]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
12.
Generalized Penalized function
$$\begin{aligned}&f_{12} (x)=\frac{\pi }{30}\left\{ 10\sin ^2(\pi y_1)+ \sum \limits _{i=1}^{29} (y_i-1)^2\right. \\&\qquad \qquad \quad \left. \times [1+10\sin ^2(\pi y_{i+1} )]+(y_{30}-1)^2 \right\} \\&\qquad \qquad \quad \,+\sum \limits _{i=1}^{30} {u(x_i,10,100,4)}, \\&\qquad y_i =1+\frac{(x_i +1)}{4}, \\&\qquad u(x_i,a,k,m)=\left\{ {\begin{array}{l@{\quad }l} k(x_i -a)^m&{} x_i >a \\ 0&{} -a\le x_i \le a \\ k(-x_i-a)^m&{} x_i<-a \\ \end{array}} \right. \\&\qquad \qquad \qquad \quad \qquad \qquad x\in [-50,50]^{30} \end{aligned}$$Its global minimum is equal to 0.
-
13.
Generalized penalized function
$$\begin{aligned}&f_{13} (x)=0.1\left\{ \sin ^2(3\pi x_1)+\sum \limits _{i=1}^{29} (x_i-1)^2\right. \\&\qquad \qquad \quad \times \,[1+\sin ^2(3\pi x_{i+1})]+(x_{30}-1)^2\\&\qquad \qquad \quad \times \, \left. [1+\sin ^2(2\pi x_{30})]\right\} \\&\qquad \qquad \quad +\,\sum \limits _{i=1}^{30} {u(x_i,5,100,4)}, \\&\qquad u(x_i,a,k,m)=\left\{ {\begin{array}{l@{\quad }l} k(x_i-a)^m&{}\quad x_i <a \\ 0&{}\quad -a\le x_i \le a \\ k(-x_i-a)^m&{}\quad x_i \ge a \\ \end{array}} \right. \\&\qquad \qquad \qquad \qquad \qquad \, x\in [-50,50]^{30} \end{aligned}$$Its global minimum is equal to 0.
Rights and permissions
About this article
Cite this article
Lou, Z., Liu, B., Xie, H. et al. Adjustment of basal insulin infusion rate in T1DM by hybrid PSO. Soft Comput 19, 1921–1937 (2015). https://doi.org/10.1007/s00500-014-1378-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00500-014-1378-6